vvEPA
EPA/635/R13/138a
Public Comment 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)
August 2013
NOTICE
This document is a Public Comment 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ii DRAFT—DO NOT CITE OR QUOTE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 viii
PREFACE xi
PREAMBLE TO IRIS TOXICOLOGICAL REVIEW xiv
EXECUTIVE SUM MARY xxxii
LITERATURE SEARCH STRATEGY| STUDY SELECTION xxxix
1. HAZARD IDENTIFICATION 1-1
1.1. SYNTHESIS OF EVIDENCE 1-1
1.1.1. Developmental Toxicity 1-1
1.1.2. Reproductive Toxicity 1-21
1.1.3. Immunotoxicity 1-36
1.1.4. Other Toxicity 1-43
1.1.5. Carcinogenicity 1-51
1.2. SUMMARY AND EVALUATION 1-80
1.2.1. Weight of Evidence for Effects Other than Cancer 1-80
1.2.2. Weight of Evidence for Carcinogenicity 1-81
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 Extrapolations Methods 2-28
2.3.3. Derivation of the Oral Slope Factor 2-29
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 Extrapolations 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 Extrapolations 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 xxxiv
Table ES-2. Organ/system-specific RfCs and proposed overall RfCfor benzo[a]pyrene xxxv
Table LS-1. Summary of the search strategy employed for benzo[a]pyrene xxxix
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 1-24
Table 1-6. Evidence pertaining to the female reproductive effects of benzo[a]pyrene in humans 1-31
Table 1-7. Evidence pertaining to the female reproductive effects of benzo[a]pyrene in adult
animals 1-31
Table 1-8. Evidence pertaining to the immune effects of benzo[a]pyrene in animals 1-40
Table 1-9. Evidence pertaining to other toxicities of benzo[a]pyrene in animals 1-48
Table 1-10. Cancer sites for PAH-related agents reviewed by IARC 1-53
Table 1-11. Summary of epidemiologic studies of benzo[a]pyrene (direct measures) in relation
to lung cancer risk: Tier 1 studies 1-54
Table 1-12. Summary of epidemiologic studies of benzo[a]pyrene (direct measures) in relation
to lung cancer risk: Tier 2 studies 1-55
Table 1-13. Summary of epidemiologic studies of benzo[a]pyrene (direct measures) in relation
to bladder cancer risk 1-58
Table 1-14. Tumors observed in chronic oral animal bioassays 1-62
Table 1-15. Tumors observed in chronic inhalation animal bioassays 1-65
Table 1-16. Tumor observations in dermal animal bioassays 1-67
Table 1-17. Experimental support for the postulated key events for mutagenic mode of action 1-74
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 2-46
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 xli
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 1-27
Figure 1-4. Exposure-response array for female reproductive effects following oral exposure in
adult animals 1-33
Figure 1-5. Exposure-response array for immune effects following oral exposure 1-41
Figure 1-6. Proposed metabolic activation pathways and key events in the carcinogenic mode of
action for benzo[a]pyrene 1-69
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
ADAF age-dependent adjustment factor
AhR aryl hydrocarbon receptor
BMC benchmark concentration
BMCL benchmark concentration lower
confidence limit
BMD benchmark dose
BMDL benchmark dose lower confidence limit
BMDS Benchmark Dose Software
BMR benchmark response
BPDE benzo[a]pyrene-7,8-diol-9,10-epoxide
BW body weight
CA chromosomal aberration
CASRN Chemical Abstracts Service Registry
Number
CERCLA Comprehensive Environmental
Response, Compensation, and Liability
Act
CI confidence interval
CYP450 cytochrome P450
DAF dosimetric adjustment factor
DNA deoxyribonucleic acid
EPA Environmental Protection Agency
EROD ethoxyresorufin-o-deethylase
ETS environmental tobacco smoke
FSH follicle stimulating hormone
GD gestation day
HEC human equivalent concentration
HED human equivalent dose
HERO Health and Environmental Research
Online
IHD ischemic heart disease
i.p. intraperitoneal
IRIS Integrated Risk Information System
LH luteinizing hormone
LOAEL lowest-observed-adverse-effect level
MMAD mass median aerodynamic diameter
MN micronuclei
MPPD Multi-Path Particle Deposition
NCEA National Center for Environmental
Assessment
no-observed-adverse-effect level
National Priorities List
National Toxicology Program
odds ratio
Office of Research and Development
polycyclic aromatic hydrocarbon
physiologically based pharmacokinetic
postnatal day
point of departure
red blood cell
regional deposited dose ratio for
extrarespiratory effects
inhalation reference concentration
oral reference dose
reactive oxygen species
relative risk
subcutaneous
squamous cell carcinoma
sister chromatid exchange
sperm chromatin structure assay
standard deviation
standardized incidence ratio
standardized mortality ratio
sheep red blood cells
single strand break
terminal deoxynucleotidyl transferase
dUTP nick end labeling
time-weighted average
uncertainty factor
interspecies uncertainly factor
intraspecies uncertainly factor
LOAEL-to-NOAEL uncertainly factor
subchronic-to-chronic uncertainly
factor
database deficiencies uncertainly factor
white blood cell
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
1
2
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.
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
4
5
6
7
Scientific Support Team
Lynn Flowers, Ph.D.(Team Leader)
John Fox, Ph.D.
Martin Gehlhaus, MHS
Paul White, Ph.D.
Reeder Sams, Ph.D.
John Cowden, Ph.D.
Connie Kang-Sickel, 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
Production Team
This document is a draft for review purposes only and does not constitute Agency policy.
viii DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
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.
Megan Riccardi
Kelly Salinas
Joe Santodonato
Julie Stickney, Ph.D.
George Holds worth, Ph.D.
Lutz W.Weber, Ph.D.
Janusz Z. Byczkowski, Ph.D., D.Sc.
SRC, Inc., Syracuse, NY
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)
RitaSchoeny, Ph.D.
U.S. EPA
National Health and Environmental
Effects Research Laboratory
Research Triangle Park, NC
U.S. EPA
Office of Water
Washington, DC
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
Reviewers
1 This assessment was provided for review to scientists in EPA's Program and Region Offices.
2 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
3 This assessment was provided for review to other federal agencies and Executive Offices of the
4 President. Comments were submitted by:
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.
x DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
PREFACE
This Toxicological Review, prepared under the auspices of EPA's Integrated Risk
Information System (IRIS) program, critically reviews the publicly available studies on
benzo[a]pyrene in order to identify potential adverse health effects and to characterize exposure-
response relationships. Benzo[a]pyrene is found in the environment and in food. Benzo[a]pyrene
occurs in conjunction with other structurally related chemical compounds known as polycyclic
aromatic hydrocarbons (PAHs).1 Benzo[a]pyrene is universally present in these mixtures and is
routinely analyzed and detected in environmental media contaminated with PAH mixtures, thus it
is often used as an indicator chemical to measure exposure to PAH mixtures [Bostrometal.. 2002].
Benzo[a]pyrene is listed as a hazardous substance under the Comprehensive Environmental
Response, Compensation, and Liability Act of 1980 (CERCLA), is found at 524 hazardous waste sites
on the National Priorities List (NPL) and is ranked number 9 out of 275 chemicals on the Priority
List of Hazardous Substances for CERCLA [ATSDR. 2011). Benzo[a]pyrene is also listed as a
drinking water contaminant under the Safe Drinking Water Act and a Maximum Contaminant Level
Goal (MCLG) and enforceable Maximum Contaminant Level (MCL) have been established. In air,
benzo[a]pyrene is regulated as a component in a class of chemicals referred to as Polycyclic Organic
Matter, defined as a Hazardous Air Pollutant by the 1990 amendments to the Clean Air Act
This assessment updates IRIS assessment of benzo[a]pyrene that was developed in 1987.
The previous assessment included a cancer descriptor and oral slope factor. New information has
become available, and this assessment reviews information on all health effects by all exposure
routes. Organ/system-specific reference values are calculated based on developmental,
reproductive and immune system toxicity data. These reference values may be useful for
cumulative risk assessments that consider the combined effect of multiple agents acting on the
same biological system. In addition, in consideration of the Agency's need to estimate the potential
for skin cancer from dermal exposure [U.S. EPA, 2004], especially in children exposed to
contaminated soil, this assessment includes the IRIS Program's first dermal slope factor.
This assessment was conducted in accordance with EPA guidance, which is cited and
summarized in the Preamble to Toxicological Reviews. The findings of this assessment and related
documents produced during its development are available on the IRIS website
[http://www.epa.gov/iris]. Appendices for chemical and physical properties, toxicokinetic
xPAHs 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.
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
information, and summaries of toxicity studies are provided as Supplemental Information to this
assessment.
For additional information about this assessment or for general questions regarding IRIS,
please contact EPA's IRIS Hotline at 202-566-1676 (phone), 202-566-1749 (fax), or
hotline.iris@epa.gov.
Chemical Properties and Uses
Benzo[a]pyrene is a five-ring PAH. It is a pale yellow crystalline solid with a faint aromatic
odor. It is relatively insoluble in water and has low volatility. Benzo[a]pyrene is released to the air
from both natural and anthropogenic sources and removed from the atmosphere by photochemical
oxidation; reaction with nitrogen oxides, hydroxy and hydroperoxy radicals, ozone, sulfur oxides,
and peroxyacetyl nitrate; and dry deposition to land or water. In air, benzo[a]pyrene is
predominantly adsorbed to particulates but may also exist as a vapor at high temperatures (ATSDR,
1995).
There is no known commercial use for benzo[a]pyrene; it is only produced as a research
chemical. Benzo[a]pyrene is ubiquitous in the environment primarily as a result of incomplete
combustion emissions. It is released to the environment via both natural sources (such as forest
fires) and anthropogenic sources including stoves/furnaces burning fossil fuels (especially wood
and coal), motor vehicle exhaust, cigarettes, and various industrial combustion processes (ATSDR,
1995). Benzo[a]pyrene is also found in soot and coal tars. Mahler etal. (2005) has reported that
urban run-off from asphalt-paved car parks treated with coats of coal-tar emulsion seal could
account for the majority of PAHs in many watersheds. Benzo[a]pyrene exposure can also occur to
workers involved in the production of aluminum, coke, graphite, and silicon carbide, and in coal tar
distillation. The major sources of non-occupational exposure are cigarettes and food. Additional
information on benzo[a]pyrene exposure and chemical properties can be found in Appendix A.
Implementation of the 2011 National Research Council Recommendations
On December 23, 2011, The Consolidated Appropriations Act, 2012, was signed into law
(U.S. Congress, 2011). The report language included direction to EPA for the IRIS Program related
to recommendations provided by the National Research Council (NRC) in their review of EPA's
draft IRIS assessment of formaldehyde (NRC. 2011). The report language included the following:
The Agency shall incorporate, as appropriate, based on chemical-specific datasets
and biological effects, the recommendations of Chapter 7 of the National Research
Council's Review of the Environmental Protection Agency's Draft IRIS Assessment of
Formaldehyde into the IRIS process...For draft assessments released in fiscal year
2012, the Agency shall include documentation describing how the Chapter 7
recommendations of the National Academy of Sciences (NAS) have been
implemented or addressed, including an explanation for why certain
recommendations were not incorporated.
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
The NRC's recommendations, provided in Chapter 7 of their review report, offered
suggestions to EPA for improving the development of IRIS assessments. Consistent with the
direction provided by Congress, documentation of how the recommendations from Chapter 7 of the
NRC report have been implemented in this assessment is provided in the table below. Where
necessary, the documentation includes an explanation for why certain recommendations were not
incorporated.
The IRIS Program's implementation of the NRC recommendations is following a phased
approach that is consistent with the NRC's "Roadmap for Revision" as described in Chapter 7 of the
formaldehyde review report. The NRC stated that "the committee recognizes that the changes
suggested would involve a multi-year process and extensive effort by the staff at the National
Center for Environmental Assessment and input and review by the EPA Science Advisory Board and
others."
Phase 1 of implementation has focused on a subset of the short-term recommendations,
such as editing and streamlining documents, increasing transparency and clarity, and using more
tables, figures, and appendices to present information and data in assessments. Phase 1 also
focused on assessments near the end of the development process and close to final posting. The
IRIS benzo[a]pyrene assessment is in Phase 2 and represents a significant advancement in
implementing the NRC recommendations shown in Table F-l in Appendix F. The Program is
implementing all of these recommendations but recognizes that achieving full and robust
implementation of certain recommendations will be an evolving process with input and feedback
from the public, stakeholders, and external peer review committees. Phase 3 of implementation
will incorporate the longer-term recommendations made by the NRC as outlined below in Table F-2
in Appendix F, including the development of a standardized approach to describe the strength of
evidence for noncancer effects . On May 16, 2012, EPA announced [U.S. EPA. 2012c] that as a part
of a review of the IRIS Program's assessment development process, the NRC will also review
current methods for weight-of-evidence analyses and recommend approaches for weighing
scientific evidence for chemical hazard identification. This effort is included in Phase 3 of EPA's
implementation plan.
Assessments by Other National and International Health Agencies
Toxicity information on benzo[a]pyrene has been evaluated by California EPA (CalEPA), the
World Health Organization, Health Canada, the International Agency for Research on Cancer, and
the European Union. The results of these assessments are presented in Appendix B. It is important
to recognize that these assessments were prepared at different times, for different purposes, using
different 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.
xiii DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
PREAMBLE TO IRIS TOXICOLOGICAL REVIEWS
1 1. Scope of the IRIS Program
2 Soon after the EPA was established in
3 1970, it was at the forefront of developing
4 risk assessment as a science and applying it
5 in decisions to protect human health and the
6 environment The Clean Air Act, for example,
7 mandates that the EPA provide "an ample
8 margin of safety to protect public health";
9 the Safe Drinking Water Act, that "no
10 adverse effects on the health of persons may
11 reasonably be anticipated to occur, allowing
12 an adequate margin of safety." Accordingly,
13 the EPA uses information on the adverse
14 effects of chemicals and on exposure levels
15 below which these effects are not
16 anticipated to occur.
17 IRIS assessments critically review the
18 publicly available studies to identify adverse
19 health effects from exposure to chemicals
20 and to characterize exposure-response
21 relationships. In terms set forth by the
22 National Research Council fNRC. 19831 IRIS
23 assessments cover the hazard identification
24 and dose-response assessment steps of risk
25 assessment, not the exposure assessment or
26 risk characterization steps that are
27 conducted by the EPA's program and
28 regional offices and by other federal, state,
29 and local health agencies that evaluate risk
30 in specific populations and exposure
31 scenarios. IRIS assessments are distinct from
32 and do not address political, economic, and
33 technical considerations that influence the
34 design and selection of risk management
35 alternatives.
36 An IRIS assessment may cover a single
37 chemical, a group of structurally or
38 toxicologically related chemicals, or a
39 complex mixture. These agents may be found
40 in air, water, soil, or sediment. Exceptions
41 are chemicals currently used exclusively as
42 pesticides, ionizing and non-ionizing
43 radiation, and criteria air pollutants listed
44 under section 108 of the Clean Air Act
45 (carbon monoxide, lead, nitrogen oxides,
46 ozone, particulate matter, and sulfur oxides).
47 Periodically, the IRIS Program asks other
48 EPA programs and regions, other federal
49 agencies, state health agencies, and the
50 general public to nominate chemicals and
51 mixtures for future assessment or
52 reassessment. Agents may be considered for
53 reassessment as significant new studies are
54 published. Selection is based on program
55 and regional office priorities and on
56 availability of adequate information to
57 evaluate the potential for adverse effects.
58 Other agents may also be assessed in
59 response to an urgent public health need.
60 2. Process for developing and peer-
61 reviewing IRIS assessments
62 The process for developing IRIS
63 assessments (revised in May 2009 and
64 enhanced in July 2013) involves critical
65 analysis of the pertinent studies,
66 opportunities for public input, and multiple
67 levels of scientific review. The EPA revises
68 draft assessments after each review, and
69 external drafts and comments become part
70 of the public record (U.S. EPA. 2009).
71 Before beginning an assessment, the IRIS
72 Program discusses the scope with other EPA
73 programs and regions to ensure that the
74 assessment will meet their needs. Then a
75 public meeting on problem formulation
76 invites discussion of the key issues and the
77 studies and analytical approaches that might
78 contribute to their resolution.
79 Step 1. Development of a draft
80 Toxicological Review. The draft
81 assessment considers all pertinent
82 publicly available studies and applies
83 consistent criteria to evaluate study
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 quality, identify health effects, identify
2 mechanistic events and pathways,
3 integrate the evidence of causation for
4 each effect, and derive toxicity values. A
5 public meeting prior to the integration of
6 evidence and derivation of toxicity
7 values promotes public discussion of the
8 literature search, evidence, and key
9 issues.
10 Step 2. Internal review by scientists in
11 EPA programs and regions. The draft
12 assessment is revised to address the
13 comments from within the EPA.
14 Step 3. Interagency science consultation
15 with other federal agencies and the
16 Executive Offices of the President.
17 The draft assessment is revised to
18 address the interagency comments. The
19 science consultation draft, interagency
20 comments, and the EPA's response to
21 major comments become part of the
22 public record.
23 Step 4. Public review and comment,
24 followed by external peer review. The
25 EPA releases the draft assessment for
26 public review and comment. A public
27 meeting provides an opportunity to
28 discuss the assessment prior to peer
29 review. Then the EPA releases a draft for
30 external peer review. The peer reviewers
31 also receive written and oral public
32 comments, and the peer review meeting
33 is open to the public. The peer reviewers
34 assess whether the evidence has been
35 assembled and evaluated according to
36 guidelines and whether the conclusions
37 are justified by the evidence. The peer
38 review draft, written public comments,
39 and peer review report become part of
40 the public record.
41 Step 5. Revision of draft Toxicological
42 Review and development of draft IRIS
43 summary. The draft assessment is
44 revised to reflect the peer review
45 comments, public comments, and newly
46 published studies that are critical to the
47 conclusions of the assessment. The
48 disposition of peer review comments
49 and public comments becomes part of
50 the public record.
51 Step 6. Final EPA review and interagency
52 science discussion with other federal
53 agencies and the Executive Offices of
54 the President The draft assessment and
55 summary are revised to address the EPA
56 and interagency comments. The science
57 discussion draft, written interagency
58 comments, and EPA's response to major
59 comments become part of the public
60 record.
61 Step 7. Completion and posting. The
62 Toxicological Review and IRIS summary
63 are posted on the IRIS website [http://
64 www. ep a. go v/ir is].
65 The remainder of this Preamble
66 addresses step 1, the development of a draft
67 Toxicological Review. IRIS assessments
68 follow standard practices of evidence
69 evaluation and peer review, many of which
70 are discussed in EPA guidelines [U.S. EPA.
71 2005a. b. 2000. 1998. 1996. 1991. 1986a. b]
72 and other methods [U.S. EPA. 2012a. b. 2011.
73 2006a. b, 2002. 1994b). Transparent
74 application of scientific judgment is of
75 paramount importance. To provide a
76 harmonized approach across IRIS
77 assessments, this Preamble summarizes
78 concepts from these guidelines and
79 emphasizes principles of general
80 applicability.
si 3. Identifying and selecting
82 pertinent studies
83 3.1. Identifying studies
84 Before beginning an assessment, the EPA
85 conducts a comprehensive search of the
86 primary scientific literature. The literature
87 search follows standard practices and
88 includes the PubMed and ToxNet databases
89 of the National Library of Medicine, Web of
90 Science, and other databases listed in the
91 EPA's HERO system (Health and
92 Environmental Research Online, http://
93 hero.epa.gov/]. Searches for information on
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
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 web site 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 2000, §2.1,1986b, §2.2):
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
81 consideration for
82 experimental animal
83 clinical studies.
is a key design
selecting pertinent
studies or human
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
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 information. They also may
2 be useful for identifying effects in
3 animals if deposition or absorption is
4 problematic (for example, for particles
5 and fibers).
6 Exposure duration is also a key design
7 consideration for selecting pertinent
8 experimental animal studies.
9 - Studies of effects from chronic exposure
10 are most pertinent to lifetime human
11 exposure.
12 - Studies of effects from less-than-chronic
13 exposure are pertinent but less
14 preferred for identifying effects from
15 lifetime human exposure. Such studies
16 may be indicative of effects from less-
17 than-lifetime human exposure.
18 Short-duration studies involving animals
19 or humans may provide toxicokinetic or
20 mechanistic information.
21 For developmental toxicity and
22 reproductive toxicity, irreversible effects
23 may result from a brief exposure during a
24 critical period of development Accordingly,
25 specialized study designs are used for these
26 effects [U.S. EPA. 2006b. 1998. 1996. 19911.
27 4. Evaluating the quality of
28 individual studies
29 After the subsets of pertinent
30 epidemiologic and experimental studies
31 have been selected from the literature
32 searches, the assessment evaluates the
33 quality of each individual study. This
34 evaluation considers the design, methods,
35 conduct, and documentation of each study,
36 but not whether the results are positive,
37 negative, or null. The objective is to identify
38 the stronger, more informative studies based
39 on a uniform evaluation of quality
40 characteristics across studies of similar
41 design.
42 4.1. Evaluating the quality of
43 epidemiologic studies
44 The assessment evaluates design and
45 methodological aspects that can increase or
46 decrease the weight given to each
47 epidemiologic study in the overall evaluation
48 [U.S. EPA. 2005a. 1998. 1996. 1994b. 19911:
49 - Documentation of study design,
50 methods, population characteristics, and
51 results.
52
53
54
55
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
Definition and selection of the study
group and comparison group.
Ascertainment of exposure
chemical or mixture.
to the
56 - Ascertainment of disease or health effect.
57 - Duration of exposure and follow-up and
58 adequacy for assessing the occurrence of
59 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 fU.S. EPA. 2005a. 1998.1996.19911.
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 4.2. Evaluating the quality of
2 experimental studies
3 The assessment evaluates design and
4 methodological aspects that can increase or
5 decrease the weight given to each
6 experimental animal study, in-vitro study, or
7 human clinical study [U.S. EPA. 2005a. 1998.
8 1996, 1991]. Research involving human
9 subjects is considered only if conducted
10 according to ethical principles.
11 - Documentation of study design, animals
12 or study population, methods, basic data,
13 and results.
14 - Nature of the assay and validity for its
15 intended purpose.
16 - Characterization of the nature and extent
17 of impurities and contaminants of the
18 administered chemical or mixture.
19 - Characterization of dose and dosing
20 regimen (including age at exposure) and
21 their adequacy to elicit adverse effects,
22 including latent effects.
23 - Sample sizes and statistical power to
24 detect dose-related differences or trends.
25 - Ascertainment of survival, vital signs,
26 disease or effects, and cause of death.
27 - Control of other variables that could
28 influence the occurrence of effects.
29 The assessment uses statistical tests to
30 evaluate whether the observations may be
31 due to chance. The standard for determining
32 statistical significance of a response is a
33 trend test or comparison of outcomes in the
34 exposed groups against those of concurrent
35 controls. In some situations, examination of
36 historical control data from the same
37 laboratory within a few years of the study
38 may improve the analysis. For an uncommon
39 effect that is not statistically significant
40 compared with concurrent controls,
41 historical controls may show that the effect
42 is unlikely to be due to chance. For a
43 response that appears significant against a
44 concurrent control response that is unusual,
45 historical controls may offer a different
46 interpretation [U.S. EPA. 2005a. §2.2.2.1.3).
47 For developmental toxicity, reproductive
48 toxicity, neurotoxicity, and cancer there is
49 further guidance on the nuances of
50 evaluating experimental studies of these
51 effects [U.S. EPA. 2005a. 1998. 1996. 1991).
52 In multi-generation studies, agents that
53 produce developmental effects at doses that
54 are not toxic to the maternal animal are of
55 special concern. Effects that occur at doses
56 associated with mild maternal toxicity are
57 not assumed to result only from maternal
58 toxicity. Moreover, maternal effects may be
59 reversible, while effects on the offspring may
60 be permanent [U.S. EPA. 1998. §3.1.1.4,
61 1991 §3.1.2.4.5.4).
62 4.3. Reporting study results
63 The assessment uses evidence tables to
64 present the design and key results of
65 pertinent studies. There may be separate
66 tables for each site of toxicity or type of
67 study.
68 If a large number of studies observe the
69 same effect, the assessment considers the
70 study quality characteristics in this section
71 to identify the strongest studies or types of
72 study. The tables present details from these
73 studies, and the assessment explains the
74 reasons for not reporting details of other
75 studies or groups of studies that do not add
76 new information. Supplemental information
77 provides references to all studies
78 considered, including those not summarized
79 in the tables.
80 The assessment discusses strengths and
81 limitations that affect the interpretation of
82 each study. If the interpretation of a study in
83 the assessment differs from that of the study
84 authors, the assessment discusses the basis
85 for the difference.
86 As a check on the selection and
87 evaluation of pertinent studies, the EPA asks
88 peer reviewers to identify studies that were
89 not adequately considered.
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 5. Evaluating the overall evidence
2 of each effect
3 5.1. Concepts of causal inference
4 For each health effect, the assessment
5 evaluates the evidence as a whole to
6 determine whether it is reasonable to infer a
7 causal association between exposure to the
8 agent and the occurrence of the effect. This
9 inference is based on information from
10 pertinent human studies, animal studies, and
11 mechanistic studies of adequate quality.
12 Positive, negative, and null results are given
13 weight according to study quality.
14
Causal
inference involves scientific
15 judgment, and the considerations are
16 nuanced and complex. Several health
17 agencies have developed frameworks for
18 causal inference, among them the U.S.
19 Surgeon General [CDC. 2004: HEW. 1964).
20 the International Agency for Research on
21 Cancer [2006] , the Institute of Medicine
22 (2008), and the EPA fU.S. EPA. 2010. §1.6,
23 2005a. §2.5). Although developed for
24 different purposes, the frameworks are
25 similar in nature and provide an established
26 structure and language for causal inference.
27 Each considers aspects of an association that
28 suggest causation, discussed by Hill [1965]
29 and elaborated by Rothman and Greenland
30 [1998] [U.S. EPA. 2005a. §2.2.1.7, 1994b.
31 app. C).
32 Strength of association: The finding of a
33 large relative risk with narrow
34 confidence intervals strongly suggests
35 that an association is not due to chance,
36 bias, or other factors. Modest relative
37 risks, however, may reflect a small range
38 of exposures, an agent of low potency, an
39 increase in an effect that is common,
40 exposure misclassification, or other
41 sources of bias.
42 Consistency of association: An inference of
43 causation is strengthened if elevated
44 risks are observed in independent
45 studies of different populations and
46 exposure scenarios. Reproducibility of
47 findings constitutes one of the strongest
48 arguments for causation. Discordant
49 results sometimes reflect differences in
50 study design, exposure, or confounding
51 factors.
52 Specificity of association: As originally
53 intended, this refers to one cause
54 associated with one effect. Current
55 understanding that many agents cause
56 multiple effects and many effects have
57 multiple causes make this a less
58 informative aspect of causation, unless
59 the effect is rare or unlikely to have
60 multiple causes.
61 Temporal relationship: A causal
62 interpretation requires that exposure
63 precede development of the effect.
64 Biologic gradient (exposure-response
65 relationship): Exposure-response
66 relationships strongly suggest causation.
67 A monotonic increase is not the only
68 pattern consistent with causation. The
69 presence of an exposure-response
70 gradient also weighs against bias and
71 confounding as the source of an
72 association.
73 Biologic plausibility: An inference of
74 causation is strengthened by data
75 demonstrating plausible biologic
76 mechanisms, if available. Plausibility
77 may reflect subjective prior beliefs if
78 there is insufficient understanding of the
79 biologic process involved.
80 Coherence: An inference of causation is
81 strengthened by supportive results from
82 animal experiments, toxicokinetic
83 studies, and short-term tests. Coherence
84 may also be found in other lines of
85 evidence, such as changing disease
86 patterns in the population.
87 "Natural experiments": A change in
88 exposure that brings about a change in
89 disease frequency provides strong
90 evidence, as it tests the hypothesis of
91 causation. An example would be an
92 intervention to reduce exposure in the
93 workplace or environment that is
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 followed by a reduction of an adverse
2 effect
3 Analogy: Information on structural
4 analogues or on chemicals that induce
5 similar mechanistic events can provide
6 insight into causation.
7 These considerations are consistent with
8 guidelines for systematic reviews that
9 evaluate the quality and weight of evidence.
10 Confidence is increased if the magnitude of
11 effect is large, if there is evidence of an
12 exposure-response relationship, or if an
13 association was observed and the plausible
14 biases would tend to decrease the magnitude
15 of the reported effect. Confidence is
16 decreased for study limitations,
17 inconsistency of results, indirectness of
18 evidence, imprecision, or reporting bias
19 (Guyattetal..2008a: Guyattetal.. 2008b).
20 5.2. Evaluating evidence in humans
21 For each effect, the assessment evaluates
22 the evidence from the epidemiologic studies
23 as a whole. The objective is to determine
24 whether a credible association has been
25 observed and, if so, whether that association
26 is consistent with causation. In doing this,
27 the assessment explores alternative
28 explanations (such as chance, bias, and
29 confounding) and draws a conclusion about
30 whether these alternatives can satisfactorily
31 explain any observed association.
32 To make clear how much the
33 epidemiologic evidence contributes to the
34 overall weight of the evidence, the
35 assessment may select a standard descriptor
36 to characterize the epidemiologic evidence
37 of association between exposure to the agent
38 and occurrence of a health effect.
39 Sufficient epidemiologic evidence of an
40 association consistent with causation:
41 The evidence establishes a causal
42 association for which alternative
43 explanations such as chance, bias, and
44 confounding can be ruled out with
45 reasonable confidence.
46 Suggestive epidemiologic evidence of an
47 association consistent with causation:
48 The evidence suggests a causal
49 association but chance, bias, or
50 confounding cannot be ruled out as
51 explaining the association.
52 Inadequate epidemiologic evidence to
53 infer a causal association: The available
54 studies do not permit a conclusion
55 regarding the presence or absence of an
56 association.
57 Epidemiologic evidence consistent with no
58 causal association: Several adequate
59 studies covering the full range of human
60 exposures and considering susceptible
61 populations, and for which alternative
62 explanations such as bias and
63 confounding can be ruled out, are
64 mutually consistent in not finding an
65 association.
66 5.3. Evaluating evidence in animals
67 For each effect, the assessment evaluates
68 the evidence from the animal experiments as
69 a whole to determine the extent to which
70 they indicate a potential for effects in
71 humans. Consistent results across various
72 species and strains increase confidence that
73 similar results would occur in humans.
74 Several concepts discussed by Hill [1965]
75 are pertinent to the weight of experimental
76 results: consistency of response, dose-
77 response relationships, strength of response,
78 biologic plausibility, and coherence [U.S.
79 EPA. 2005a. §2.2.1.7.1994. app. C).
80 In weighing evidence from multiple
81 experiments, U.S. EPA f2005a. §2.5)
82 distinguishes
83 Conflicting evidence (that is, mixed positive
84 and negative results in the same sex and
85 strain using a similar study protocol)
86 from
87 Differing results (that is, positive results
88 and negative results are in different
89 sexes or strains or use different study
90 protocols).
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 Negative or null results do not invalidate
2 positive results in a different experimental
3 system. The EPA regards all as valid
4 observations and looks to explain differing
5 results using mechanistic information (for
6 example, physiologic or metabolic
7 differences across test systems) or
8 methodological differences (for example,
9 relative sensitivity of the tests, differences in
10 dose levels, insufficient sample size, or
11 timing of dosing or data collection).
12 It is well established that there are
13 critical periods for some developmental and
14 reproductive effects (U.S. EPA. 2006b.
15 2005a. b, 1998. 1996. 1991). Accordingly,
16 the assessment determines whether critical
17 periods have been adequately investigated.
18 Similarly, the assessment determines
19 whether the database is adequate to
20 evaluate other critical sites and effects.
21 In evaluating evidence of genetic
22 toxicity:
23 - Demonstration of gene mutations,
24 chromosome aberrations, or aneuploidy
25 in humans or experimental mammals
26 [in vivo] provides the strongest evidence.
27 - This is followed by positive results in
28 lower organisms or in cultured cells
29 [in vitro] or for other genetic events.
30 - Negative results carry less weight, partly
31 because they cannot exclude the
32 possibility of effects in other tissues
33 (IARC. 2006).
34 For germ-cell mutagenicity, The EPA has
35 defined categories of evidence, ranging from
36 positive results of human germ-cell
37 mutagenicity to negative results for all
38 effects of concern (U.S. EPA. 1986a. 52.3).
39 5.4. Evaluating mechanistic data
40 Mechanistic data can be useful in
41 answering several questions.
42 - The biologic plausibility of a causal
43 interpretation of human studies.
44 - The generalizability of animal studies to
45 humans.
46 - The susceptibility of particular
47 populations or lifestages.
48 The focus of the analysis is to describe, if
49 possible, mechanistic pathways that lead to a
50 health effect. These pathways encompass:
51 - Toxicokinetic processes of absorption,
52 distribution, metabolism, and
53 elimination that lead to the formation of
54 an active agent and its presence at the
55 site of initial biologic interaction.
56 - Toxicodynamic processes that lead to a
57 health effect at this or another site (also
58 known as a mode of action].
59 For each effect, the assessment discusses
60 the available information on its modes of
61 action and associated key events [key events
62 being empirically observable, necessary
63 precursor steps or biologic markers of such
64 steps; mode of action being a series of key
65 events involving interaction with cells,
66 operational and anatomic changes, and
67 resulting in disease). Pertinent information
68 may also come from studies of metabolites
69 or of compounds that are structurally similar
70 or that act through similar mechanisms.
71 Information on mode of action is not
72 required for a conclusion that the agent is
73 causally related to an effect (U.S. EPA. 2005a.
74 §2.5).
75 The assessment addresses several
76 questions about each hypothesized mode of
77 action fU.S. EPA. 2005a. 52.4.3.41
78 Is the hypothesized mode of action
79 sufficiently supported in test animals?
80 Strong support for a key event being
81 necessary to a mode of action can come
82 from experimental challenge to the
83 hypothesized mode of action, in which
84 studies that suppress a key event
85 observe suppression of the effect
86 Support for a mode of action is
87 meaningfully strengthened by consistent
88 results in different experimental models,
89 much more so than by replicate
90 experiments in the same model. The
91 assessment may consider various
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 aspects of causation in addressing this
2 question.
3 Is the hypothesized mode of action
4 relevant to humans? The assessment
5 reviews the key events to identify critical
6 similarities and differences between the
7 test animals and humans. Site
8 concordance is not assumed between
9 animals and humans, though it may hold
10 for certain effects or modes of action.
11 Information suggesting quantitative
12 differences in doses where effects would
13 occur in animals or humans is
14 considered in the dose-response
15 analysis. Current levels of human
16 exposure are not used to rule out human
17 relevance, as IRIS assessments may be
18 used in evaluating new or unforeseen
19 circumstances that may entail higher
20 exposures.
21 Which populations or lifestages can be
22 particularly susceptible to the
23 hypothesized mode of action? The
24 assessment reviews the key events to
25 identify populations and lifestages that
26 might be susceptible to their occurrence.
27 Quantitative differences may result in
28 separate toxicity values for susceptible
29 populations or lifestages.
30 The assessment discusses the likelihood
31 that an agent operates through multiple
32 modes of action. An uneven level of support
33 for different modes of action can reflect
34 disproportionate resources spent
35 investigating them [U.S. EPA. 2005a.
36 §2.4.3.3). It should be noted that in clinical
37 reviews, the credibility of a series of studies
38 is reduced if evidence is limited to studies
39 funded by one interested sector [Guyatt et
40 al.. 2008b1.
41 For cancer, the assessment evaluates
42 evidence of a mutagenic mode of action to
43 guide extrapolation to lower doses and
44 consideration of susceptible lifestages. Key
45 data include the ability of the agent or a
46 metabolite to react with or bind to DNA,
47 positive results in multiple test systems, or
48 similar properties and structure-activity
49 relationships to mutagenic carcinogens [U.S.
50 EPA. 2005a. 32.3.51.
51 5.5. Characterizing the overall weight
52 of the evidence
53 After evaluating the human, animal, and
54 mechanistic evidence pertinent to an effect,
55 the assessment answers the question: Does
56 the agent cause the adverse effect? [NRG.
57 2009. 19831 In doing this, the assessment
58 develops a narrative that integrates the
59 evidence pertinent to causation. To provide
60 clarity and consistency, the narrative
61 includes a standard hazard descriptor. For
62 example, the following standard descriptors
63 combine epidemiologic, experimental, and
64 mechanistic evidence of carcinogenicity [U.S.
65 EPA. 2005a. 32.51.
66 Carcinogenic to humans: There is
67 convincing epidemiologic evidence of a
68 causal association (that is, there is
69 reasonable confidence that the
70 association cannot be fully explained by
71 chance, bias, or confounding); or there is
72 strong human evidence of cancer or its
73 precursors, extensive animal evidence,
74 identification of key precursor events in
75 animals, and strong evidence that they
76 are anticipated to occur in humans.
77 Likely to be carcinogenic to humans: The
78 evidence demonstrates a potential
79 hazard to humans but does not meet the
80 criteria for carcinogenic. There may be a
81 plausible association in humans,
82 multiple positive results in animals, or a
83 combination of human, animal, or other
84 experimental evidence.
85 Suggestive evidence of carcinogenic
86 potential: The evidence raises concern
87 for effects in humans but is not sufficient
88 for a stronger conclusion. This
89 descriptor covers a range of evidence,
90 from a positive result in the only
91 available study to a single positive result
92 in an extensive database that includes
93 negative results in other species.
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 Inadequate information to assess
2 carcinogenic potential: No other
3 descriptors apply. Conflicting evidence
4 can be classified as inadequate
5 information if all positive results are
6 opposed by negative studies of equal
7 quality in the same sex and strain.
8 Differing results, however, can be
9 classified as suggestive evidence or as
10 likely to be carcinogenic.
11 Not likely to be carcinogenic to humans:
12 There is robust evidence for concluding
13 that there is no basis for concern. There
14 may be no effects in both sexes of at least
15 two appropriate animal species; positive
16 animal results and strong, consistent
17 evidence that each mode of action in
18 animals does not operate in humans; or
19 convincing evidence that effects are not
20 likely by a particular exposure route or
21 below a defined dose.
22 Multiple descriptors may be used if there
23 is evidence that carcinogenic effects differ by
24 dose range or exposure route [U.S. EPA.
25 2005a. 32.51.
26 Another example of standard descriptors
27 comes from the EPA's Integrated Science
28 Assessments, which evaluate causation for
29 the effects of the criteria pollutants in
30 ambient air [U.S. EPA. 2010. 31.61.
31 Causal relationship: Sufficient evidence to
32 conclude that there is a causal
33 relationship. Observational studies
34 cannot be explained by plausible
35 alternatives, or they are supported by
36 other lines of evidence, for example,
37 animal studies or mechanistic
38 information.
39 Likely to be a causal relationship:
40 Sufficient evidence that a causal
41 relationship is likely, but important
42 uncertainties remain. For example,
43 observational studies show an
44 association but co-exposures are difficult
45 to address or other lines of evidence are
46 limited or inconsistent; or multiple
47 animal studies from different
48 laboratories demonstrate effects and
49 there are limited or no human data.
50 Suggestive of a causal relationship: At
51 least one high-quality epidemiologic
52 study shows an association but other
53 studies are inconsistent
54 Inadequate to infer a causal relationship:
55 The studies do not permit a conclusion
56 regarding the presence or absence of an
57 association.
58 Not likely to be a causal relationship:
59 Several adequate studies, covering the
60 full range of human exposure and
61 considering susceptible populations, are
62 mutually consistent in not showing an
63 effect at any level of exposure.
64 The EPA is investigating and may on a
65 trial basis use these or other standard
66 descriptors to characterize the overall
67 weight of the evidence for effects other than
68 cancer.
69 6. Selecting studies for derivation
70 of toxicity values
71 For each effect where there is credible
72 evidence of an association with the agent,
73 the assessment derives toxicity values if
74 there are suitable epidemiologic or
75 experimental data. The decision to derive
76 toxicity values may be linked to the hazard
77 descriptor.
78 Dose-response analysis requires
79 quantitative measures of dose and response.
80 Then, other factors being equal:
81 - Epidemiologic studies are preferred over
82 animal studies, if quantitative measures
83 of exposure are available and effects can
84 be attributed to the agent
85 - Among experimental animal models,
86 those that respond most like humans are
87 preferred, if the comparability of
88 response can be determined.
89 - Studies by a route of human
90 environmental exposure are preferred,
91 although a validated toxicokinetic model
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 can be used to extrapolate across
2 exposure routes.
3 - Studies of longer exposure duration and
4 follow-up are preferred, to minimize
5 uncertainty about whether effects are
6 representative of lifetime exposure.
7 - Studies with multiple exposure levels are
8 preferred for their ability to provide
9 information about the shape of the
10 exposure-response curve.
11 - Studies with adequate power to detect
12 effects at lower exposure levels are
13 preferred, to minimize the extent of
14 extrapolation to levels found in the
15 environment
16 Studies with non-monotonic exposure-
17 response relationships are not necessarily
18 excluded from the analysis. A diminished
19 effect at higher exposure levels may be
20 satisfactorily explained by factors such as
21 competing toxicity, saturation of absorption
22 or metabolism, exposure misclassification,
23 or selection bias.
24 If a large number of studies are suitable
25 for dose-response analysis, the assessment
26 considers the study characteristics in this
27 section to focus on the most informative
28 data. The assessment explains the reasons
29 for not analyzing other groups of studies. As
30 a check on the selection of studies for dose-
31 response analysis, the EPA asks peer
32 reviewers to identify studies that were not
33 adequately considered.
34 7. Deriving toxicity values
35 7.1. General framework for dose-
36 response analysis
37 The EPA uses a two-step approach that
38 distinguishes analysis of the observed dose-
39 response data from inferences about lower
40 doses [U.S. EPA. 2005a. S3).
41 Within the observed range, the preferred
42 approach is to use modeling to incorporate a
43 wide range of data into the analysis. The
44 modeling yields a point of departure (an
45 exposure level near the lower end of the
46 observed range, without significant
47 extrapolation to lower doses) (sections 7.2-
48 7.3).
49 Extrapolation to lower doses considers
50 what is known about the modes of action for
51 each effect (Sections 7.4-7.5). If response
52 estimates at lower doses are not required, an
53 alternative is to derive reference values,
54 which are calculated by applying factors to
55 the point of departure in order to account
56 for sources of uncertainty and variability
57 (section 7.6).
58 For a group of agents that induce an
59 effect through a common mode of action, the
60 dose-response analysis may derive a relative
61 potency factor for each agent A full dose-
62 response analysis is conducted for one well-
63 studied index chemical in the group, then the
64 potencies of other members are expressed in
65 relative terms based on relative toxic effects,
66 relative absorption or metabolic rates,
67 quantitative structure-activity relationships,
68 or receptor binding characteristics (U.S. EPA.
69 2005a. 33.2.6. 2000. 34.41.
70 Increasingly, the EPA is basing toxicity
71 values on combined analyses of multiple
72 data sets or multiple responses. The EPA
73 also considers multiple dose-response
74 approaches if they can be supported by
75 robust data.
76 7.2. Modeling dose to sites of biologic
77 effects
78 The preferred approach for analysis of
79 dose is toxicokinetic modeling because of its
80 ability to incorporate a wide range of data.
81 The preferred dose metric would refer to the
82 active agent at the site of its biologic effect or
83 to a close, reliable surrogate measure. The
84 active agent may be the administered
85 chemical or a metabolite. Confidence in the
86 use of a toxicokinetic model depends on the
87 robustness of its validation process and on
88 the results of sensitivity analyses (U.S. EPA.
89 2006a. 2005a. §3.1.1994b. §4.3).
90 Because toxicokinetic modeling can
91 require many parameters and more data
92 than are typically available, the EPA has
93 developed standard approaches that can be
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 applied to typical data sets. These standard
2 approaches also facilitate comparison across
3 exposure patterns and species.
4 - Intermittent study exposures are
5 standardized to a daily average over the
6 duration of exposure. For chronic effects,
7 daily exposures are averaged over the
8 lifespan. Exposures during a critical
9 period, however, are not averaged over a
10 longer durationfU.S. EPA. 2005a. §3.1.1,
11 1991. §3.2).
12 - Doses are standardized to equivalent
13 human terms to facilitate comparison of
14 results from different species.
15 - Oral doses are scaled allometrically
16 using mg/kg3/4-d as the equivalent
17 dose metric across species.
18 Allometric scaling pertains to
19 equivalence across species, not
20 across lifestages, and is not used to
21 scale doses from adult humans or
22 mature animals to infants or children
23 [U.S. EPA. 2011. 2005a. §3.1.3).
24 - Inhalation exposures are scaled
25 using dosimetry models that apply
26 species-specific physiologic and
27 anatomic factors and consider
28 whether the effect occurs at the site
29 of first contact or after systemic
30 circulation [U.S. EPA. 2012a. 1994b.
31 §3).
32 It can be informative to convert doses
33 across exposure routes. If this is done, the
34 assessment describes the underlying data,
35 algorithms, and assumptions [U.S. EPA.
36 2005a. 33.1.41.
37 In the absence of study-specific data on,
38 for example, intake rates or body weight, the
39 EPA has developed recommended values for
40 use in dose-response analysis [U.S. EPA,
41 19881.
42 7.3. Modeling response in the range
43 of observation
44 Toxicodynamic ("biologically based")
45 modeling can incorporate data on biologic
46 processes leading to an effect Such models
47 require sufficient data to ascertain a mode of
48 action and to quantitatively support model
49 parameters associated with its key events.
50 Because different models may provide
51 equivalent fits to the observed data but
52 diverge substantially at lower doses, critical
53 biologic parameters should be measured
54 from laboratory studies, not by model fitting.
55 Confidence in the use of a toxicodynamic
56 model depends on the robustness of its
57 validation process and on the results of
58 sensitivity analyses. Peer review of the
59 scientific basis and performance of a model
60 is essential [U.S. EPA. 2005a. §3.2.2).
61 Because toxicodynamic modeling can
62 require many parameters and more
63 knowledge and data than are typically
64 available, the EPA has developed a standard
65 set of empirical ("curve-fitting") models
66 (http://www.epa.gov/ncea/bmds/) that can
67 be applied to typical data sets, including
68 those that are nonlinear. The EPA has also
69 developed guidance on modeling dose-
70 response data, assessing model fit, selecting
71 suitable models, and reporting modeling
72 results (U.S. EPA. 2012b). Additional
73 judgment or alternative analyses are used if
74 the procedure fails to yield reliable results,
75 for example, if the fit is poor, modeling may
76 be restricted to the lower doses, especially if
77 there is competing toxicity at higher doses
78 fU.S. EPA. 2005a. 33.2.31.
79
80
81
Modeling is used to derive a point of
departure (U.S. EPA. 2012b. 2005a. §3.2.4).
(See section 7.6 for alternatives if a point of
82 departure cannot be derived by modeling.)
83 - If linear extrapolation is used, selection
84 of a response level corresponding to the
85 point of departure is not highly
86 influential, so standard values near the
87 low end of the observable range are
88 generally used (for example, 10% extra
89 risk for cancer bioassay data, 1% for
90 epidemiologic data, lower for rare
91 cancers).
92 - For nonlinear approaches, both
93 statistical and biologic considerations
94 are taken into account.
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
2
3
4
5
6
7
8
9
10
11
12
13
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
minimally adverse effects, one-half
standard deviation for more severe
effects.
14 The point of departure is the 95% lower
15 bound on the dose associated with the
16 selected response level.
17 7.4. Extrapolating to lower doses and
18 response levels
19 The purpose of extrapolating to lower
20 doses is to estimate responses at exposures
21 below the observed data. Low-dose
22 extrapolation, typically used for cancer data,
23 considers what is known about modes of
24 action fU.S. EPA. 2005a. §3.3.1. §3.3.21
25 1) If a biologically based model has been
26 developed and validated for the agent,
27 extrapolation may use the fitted model
28 below the observed range if significant
29 model uncertainty can be ruled out with
30 reasonable confidence.
31 2) Linear extrapolation is used if the dose-
32 response curve is expected to have a
33 linear component below the point of
34 departure. This includes:
35 - Agents or their metabolites that are
36 DN A-re active and have direct
37 mutagenic activity.
38 - Agents or their metabolites for which
39 human exposures or body burdens
40 are near doses associated with key
41 events leading to an effect
42 Linear extrapolation is also used when
43 data are insufficient to establish mode of
44 action and when scientifically plausible.
45 The result of linear extrapolation is
46 described by an oral slope factor or an
47 inhalation unit risk, which is the slope of
48 the dose-response curve at lower doses
49 or concentrations, respectively.
50 3) Nonlinear models are used for
51 extrapolation if there are sufficient data
52 to ascertain the mode of action and to
53 conclude that it is not linear at lower
54 doses, and the agent does not
55 demonstrate mutagenic or other activity
56 consistent with linearity at lower doses.
57 Nonlinear approaches generally should
58 not be used in cases where mode of
59 action has not ascertained. If nonlinear
60 extrapolation is appropriate but no
61 model is developed, an alternative is to
62 calculate reference values.
63 4) Both linear and nonlinear approaches
64 may be used if there a multiple modes of
65 action. For example, modeling to a low
66 response level can be useful for
67 estimating the response at doses where a
68 high-dose mode of action would be less
69 important.
70 If linear extrapolation is used, the
71 assessment develops a candidate slope
72 factor or unit risk for each suitable data set
73 These results are arrayed, using common
74 dose metrics, to show the distribution of
75 relative potency across various effects and
76 experimental systems. The assessment then
77 derives or selects an overall slope factor and
78 an overall unit risk for the agent, considering
79 the various dose-response analyses, the
80 study preferences discussed in section 6, and
81 the possibility of basing a more robust result
82 on multiple data sets.
83 7.5. Considering susceptible
84 populations and lifestages
85 The assessment analyzes the available
86 information on populations and lifestages
87 that may be particularly susceptible to each
88 effect A tiered approach is used [U.S. EPA.
89 2005a. 33.51.
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
2)
1
2
3
4
5
6
7
8
9
10
11
12
13 3)
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
If an epidemiologic or experimental
study reports quantitative results for a
susceptible population or lifestage, these
data are analyzed to derive separate
toxicity values for susceptible
individuals.
If data on risk-related parameters allow
comparison of the general population
and susceptible individuals, these data
are used to adjust the general-population
toxicity values for application to
susceptible individuals.
In the absence of chemical-specific data,
the EPA has developed age-dependent
adjustment factors for early-life exposure
to potential carcinogens that have a
mutagenic mode of action. There is
evidence of early-life susceptibility to
various carcinogenic agents, but most
epidemiologic studies and cancer
bioassays do not include early-life
exposure. To address the potential for
early-life susceptibility, the EPA
recommends (U.S. EPA. 2005b. §5):
10-fold adjustment for exposures
before age 2 years.
3-fold adjustment for exposures
between ages 2 and 16 years.
29 7.6. Reference values and uncertainty
30 factors
31 An oral reference dose or an inhalation
32 reference concentration is an estimate of an
33 exposure (including in susceptible
34 subgroups) that is likely to be without an
35 appreciable risk of adverse health effects
36 over a lifetime [U.S. EPA. 2002. §4.2).
37 Reference values are typically calculated for
38 effects other than cancer and for suspected
39 carcinogens if a well characterized mode of
40 action indicates that a necessary key event
41 does not occur below a specific dose.
42 Reference values provide no information
43 about risks at higher exposure levels.
44 The assessment characterizes effects
45 that form the basis for reference values as
46 adverse, considered to be adverse, or a
47 precursor to an adverse effect For
48 developmental toxicity, reproductive
49 toxicity, and neurotoxicity there is guidance
50 on adverse effects and their biologic markers
51 fU.S. EPA. 1998.1996. 19911
52 To account for uncertainty and
53 variability in the derivation of a lifetime
54 human exposure where adverse effects are
55 not anticipated to occur, reference values are
56 calculated by applying a series of uncertainty
57 factors to the point of departure. If a point of
58 departure cannot be derived by modeling, a
59 no-observed-adverse-effect level or a
60 lowest-observed-adverse-effect level is used
61 instead. The assessment discusses scientific
62 considerations involving several areas of
63 variability or uncertainty.
64 Human variation. The assessment accounts
65 for variation in susceptibility across the
66 human population and the possibility
67 that the available data may not be
68 representative of individuals who are
69 most susceptible to the effect. A factor of
70 10 is generally used to account for this
71 variation. This factor is reduced only if
72 the point of departure is derived or
73 adjusted specifically for susceptible
74 individuals (not for a general population
75 that includes both susceptible and non-
76 susceptible individuals) (U.S. EPA. 2002.
77 §4.4.5, 1998. §4.2, 1996. §4, 1994b.
78 §4.3.9.1,1991, §3.4).
79 Animal-to-human extrapolation. If animal
80 results are used to make inferences
81 about humans, the assessment adjusts
82 for cross-species differences. These may
83 arise from differences in toxicokinetics
84 or toxicodynamics. Accordingly, if the
85 point of departure is standardized to
86 equivalent human terms or is based on
87 toxicokinetic or dosimetry modeling, a
88 factor of 101/2 (rounded to 3) is applied
89 to account for the remaining uncertainty
90 involving toxicokinetic and
91 toxicodynamic differences. If a
92 biologically based model adjusts fully for
93 toxicokinetic and toxicodynamic
94 differences across species, this factor is
95 not used. In most other cases, a factor of
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 10 is applied [U.S. EPA. 2011. 2002.
2 §4.4.5, 1998. §4.2, 1996. §4, 1994b.
3 §4.3.9.1,1991 §3.4).
4 Adverse-effect level to no-observed-
5 adverse-effect level. If a point of
6 departure is based on a lowest-
7 observed-adverse-effect level, the
8 assessment must infer a dose where
9 such effects are not expected. This can be
10 a matter of great uncertainty, especially
11 if there is no evidence available at lower
12 doses. A factor of 10 is applied to
13 account for the uncertainty in making
14 this inference. A factor other than 10
15 may be used, depending on the
16 magnitude and nature of the response
17 and the shape of the dose-response
18 curve [U.S. EPA. 2002. §4.4.5, 1998. §4.2,
19 1996, §4,1994b, §4.3.9.1,1991 §3.4).
20 Subchronic-to-chronic exposure. If a point
21 of departure is based on subchronic
22 studies, the assessment considers
23 whether lifetime exposure could have
24 effects at lower levels of exposure. A
25 factor of 10 is applied to account for the
26 uncertainty in using subchronic studies
27 to make inferences about lifetime
28 exposure. This factor may also be
29 applied for developmental or
30 reproductive effects if exposure covered
31 less than the full critical period. A factor
32 other than 10 may be used, depending
33 on the duration of the studies and the
34 nature of the response [U.S. EPA. 2002.
35 §4.4.5,1998, §4.2,1994b, §4.3.9.1).
36 Incomplete database. If an incomplete
37 database raises concern that further
38 studies might identify a more sensitive
39 effect, organ system, or lifestage, the
40 assessment may apply a database
41 uncertainty factor [U.S. EPA.
42 2002334.4.5. 1998. §4.2, 1996. §4,
43 1994b. §4.3.9.1, 1991. §3.4). The size of
44 the factor depends on the nature of the
45 database deficiency. For example, the
46 EPA typically follows the suggestion that
47 a factor of 10 be applied if both a
48 prenatal toxicity study and a two-
49 generation reproduction study are
50 missing and a factor of 101/2 if either is
51 missing [U.S. EPA. 2002. §4.4.5).
52 In this way, the assessment derives
53 candidate values for each suitable data set
54 and effect that is credibly associated with the
55 agent These results are arrayed, using
56 common dose metrics, to show where effects
57 occur across a range of exposures [U.S. EPA.
58 1994b. 34.3.91.
59 The assessment derives or selects an
60 organ- or system-specific reference value for
61 each organ or system affected by the agent
62 The assessment explains the rationale for
63 each organ/system-specific reference value
64 (based on, for example, the highest quality
65 studies, the most sensitive outcome, or a
66 clustering of values). By providing these
67 organ/system-specific reference values, IRIS
68 assessments facilitate subsequent
69 cumulative risk assessments that consider
70 the combined effect of multiple agents acting
71 at a common site or through common
72 mechanisms fNRC. 20091.
73 The assessment then selects an overall
74 reference dose and an overall reference
75 concentration for the agent to represent
76 lifetime human exposure levels where
77 effects are not anticipated to occur. This is
78 generally the most sensitive organ/system-
79 specific reference value, though
80 consideration of study quality and
81 confidence in each value may lead to a
82 different selection.
83 7.7. Confidence and uncertainty in the
84 reference values
85 The assessment selects a standard
86 descriptor to characterize the level of
87 confidence in each reference value, based on
88 the likelihood that the value would change
89 with further testing. Confidence in reference
90 values is based on quality of the studies used
91 and completeness of the database, with more
92 weight given to the latter. The level of
93 confidence is increased for reference values
94 based on human data supported by animal
95 data fU.S. EPA. 1994b. §4.3.9.21
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 High confidence: The reference value is not
2 likely to change with further testing,
3 except for mechanistic studies that might
4 affect the interpretation of prior test
5 results.
6 Medium confidence: This is a matter of
7 judgment, between high and low
8 confidence.
9 Low confidence: The reference value is
10 especially vulnerable to change with
11 further testing.
12 These criteria are consistent with
13 guidelines for systematic reviews that
14 evaluate the quality of evidence. These also
15 focus on whether further research would be
16 likely to change confidence in the estimate of
17 effect fGuyatt et al.. 2008a).
18 All assessments discuss the significant
19 uncertainties encountered in the analysis.
20 The EPA provides guidance on
21 characterization of uncertainty [U.S. EPA.
22 2005a, §3.6). For example, the discussion
23 distinguishes model uncertainty (lack of
24 knowledge about the most appropriate
25 experimental or analytic model) and
26 parameter uncertainty (lack of knowledge
27 about the parameters of a model).
28 Assessments also discuss human variation
29 (interpersonal differences in biologic
30 susceptibility or in exposures that modify
31 the effects of the agent).
32
References
33 CDC. (Centers for Disease Control and
34 Prevention). (2004). The health
35 consequences of smoking: A report of the
36 Surgeon General. Washington, DC: U.S.
37 Department of Health and Human
38 Services.
39 http://www.surgeongeneral.gov/library
40 /smokingconsequences/
41 Guyatt. GH: Oxman. AD: Vist. GE: Kunz. R:
42 Falck-Ytter. Y: Alonso-Coello. P:
43 Schiinemann. HI. (2008a). GRADE: An
44 emerging consensus on rating quality of
45 evidence and strength of
46 recommendations. BMJ 336: 924-926.
47 http://dx.doi.0rg/10.1136/bmj.39489.4
48 70347.AD
49 Guyatt. GH: Oxman. AD: Kunz. R: Vist. GE:
50 Falck-Ytter. Y: Schiinemann. HI. (2008b).
51 GRADE: What is "quality of evidence"
52 and why is it important to clinicians?
53 [Review]. BMJ 336: 995-998.
54 http://dx.doi.0rg/10.1136/bmj.39490.5
55 51019.BE
56 HEW. (U.S. Department of Health, Education
57 and Welfare). (1964). Smoking and
58 health: Report of the advisory committee
59 to the surgeon general of the public
60 health service. Washington, DC: U.S.
61 Department of Health, Education, and
62 Welfare.
63 http://profiles.nlm.nih.gov/ps/retrieve/
64 ResourceMetadata/NNBBMQ
65 Hill. AB. (1965). The environment and
66 disease: Association or causation? Proc R
67 SocMed 58: 295-300.
68 I ARC. (International Agency for Research on
69 Cancer). (2006). Preamble to the IARC
70 monographs. Lyon, France.
71 http://monographs.iarc.fr/ENG/Preamb
72 le/
73 IOM. (Institute of Medicine), (2008)
74 Improving the presumptive disability
75 decision-making process for veterans. In
76 JM Samet; CC Bodurow (Eds.).
77 Washington, DC: National Academies
78 Press.
79 NRC. (National Research Council). (1983).
80 Risk assessment in the federal
81 government: Managing the process.
82 Washington, DC: National Academies
83 Press.
84 NRC. (National Research Council). (2009).
85 Science and decisions: Advancing risk
86 assessment Washington, DC: National
87 Academies Press.
88 Rothman. KT: Greenland. S. (1998). Modern
89 epidemiology (2nd ed.). Philadelphia, PA:
90 Lippincott, Williams, & Wilkins.
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 U.S. EPA. (U.S. Environmental Protection
2 Agency). (1986a). Guidelines for
3 mutagenicity risk assessment [EPA
4 Report]. (EPA/630/R-98/003).
5 Washington, DC.
6 http: //www. ep a. go v/ir is/backgr d. html
7 U.S. EPA. (U.S. Environmental Protection
8 Agency). (1986b). Guidelines for the
9 health risk assessment of chemical
10 mixtures. Fed Reg 51: 34014-34025.
11 U.S. EPA. (U.S. Environmental Protection
12 Agency). (1988). Recommendations for
13 and documentation of biological values
14 for use in risk assessment [EPA Report].
15 (EPA/600/6-87/008). Cincinnati, OH.
16 http://cfpub.epa.gov/ncea/cfm/recordis
17 play.cfm?deid=34855
18 U.S. EPA. (U.S. Environmental Protection
19 Agency). (1991). Guidelines for
20 developmental toxicity risk assessment
21 [EPA Report]. (EPA/600/FR-91/001).
22 Washington, DC: U.S. Environmental
23 Protection Agency, Risk Assessment
24 Forum.
25 http://www.epa.gov/iris/backgrd.html
26 U.S. EPA. (U.S. Environmental Protection
27 Agency). (1994b). Methods for
28 derivation of inhalation reference
29 concentrations and application of
30 inhalation dosimetry [EPA Report].
31 (EPA/600/8-90/066F). Research
32 Triangle Park, NC.
33 http://cfpub.epa.gov/ncea/cfm/recordis
34 play.cfm?deid=71993
35 U.S. EPA. (U.S. Environmental Protection
36 Agency). (1996). Guidelines for
37 reproductive toxicity risk assessment
38 [EPA Report]. (EPA/630/R-96/009).
39 Washington, DC.
40 http://www.epa.gOV/raf/publications/p
41 dfs/REPR051.PDF
42 U.S. EPA. (U.S. Environmental Protection
43 Agency). (1998). Guidelines for
44 neurotoxicity risk assessment [EPA
45 Report]. (EPA/630/R-95/001F).
46 Washington, DC.
47 http://www.epa.gOV/raf/publications/p
48 dfs/NEUROTOX.PDF
49 U.S. EPA. (U.S. Environmental Protection
50 Agency). (2000). Supplementary
51 guidance for conducting health risk
52 assessment of chemical mixtures [EPA
53 Report]. (EPA/630/R-00/002).
54 Washington, DC.
55 http://cfpub.epa.gov/ncea/cfm/recordis
56 play.cfm?deid=20533
57 U.S. EPA. (U.S. Environmental Protection
58 Agency). (2002). A review of the
59 reference dose and reference
60 concentration processes [EPA Report].
61 (EPA/630/P-02/002F). Washington, DC.
62 http://cfpub.epa.gov/ncea/cfm/recordis
63 play.cfm?deid=51717
64 U.S. EPA. (U.S. Environmental Protection
65 Agency). (2 005 a). Guidelines for
66 carcinogen risk assessment [EPA
67 Report]. (EPA/630/P-03/001F).
68 Washington, DC.
69 http://www.epa.gov/cancerguidelines/
70 U.S. EPA. (U.S. Environmental Protection
71 Agency). (2005b). Supplemental
72 guidance for assessing susceptibility
73 from early-life exposure to carcinogens
74 [EPA Report] (Vol. 113). (EPA/630/R-
75 03/003F). Washington, DC.
76 http://www.epa.gOV/cancerguidelines/g
77 uidelines-carcinogen-supplementhtm
78 U.S. EPA. (U.S. Environmental Protection
79 Agency). (2006a). Approaches for the
80 application of physiologically based
81 pharmacokinetic (PBPK) models and
82 supporting data in risk assessment (Final
83 Report) [EPA Report]. (EPA/600/R-
84 05/043F). Washington, DC.
85 http://cfpub.epa.gov/ncea/cfm/recordis
86 play.cfm?deid=157668
87 U.S. EPA. (U.S. Environmental Protection
88 Agency). (2006b). A framework for
89 assessing health risk of environmental
90 exposures to children [EPA Report].
91 (EPA/600/R-05/093F). Washington, DC.
92 http://cfpub.epa.gov/ncea/cfm/recordis
93 play.cfm?deid=158363
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 U.S. EPA. (U.S. Environmental Protection
2 Agency). (2009). EPAs Integrated Risk
3 Information System: Assessment
4 development process [EPA Report].
5 Washington, DC.
6 http://epa.gov/iris/process.htm
7 U.S. EPA. (U.S. Environmental Protection
8 Agency). (2010). Integrated science
9 assessment for carbon monoxide [EPA
10 Report]. (EPA/600/R-09/019F).
11 Research Triangle Park, NC.
12 http://cfpub.epa.gov/ncea/cfm/recordis
13 play.cfm?deid=218686
14 U.S. EPA. (U.S. Environmental Protection
15 Agency). (2011). Recommended use of
body weight 3/4 as the default method
in derivation of the oral reference dose
[EPA Report]. (EPA/100/R11/0001).
16
17
18
36
37
38
August 2013
19 Washington, DC.
20 http://www.epa.gOV/raf/publications/i
21 nterspecies-extrapolation.htm
22 U.S. EPA. (U.S. Environmental Protection
23 Agency). (2012a). Advances in inhalation
24 gas dosimetry for derivation of a
25 reference concentration (rfc) and use in
26 risk assessment [EPA Report].
27 (EPA/600/R-12/044). Washington, DC.
28 http://cfpub.epa.gov/ncea/cfm/recordis
29 play.cfm?deid=244650
30 U.S. EPA. (U.S. Environmental Protection
31 Agency). (2012b). Benchmark dose
32 technical guidance. (EPA/100/R-
33 12/001). Washington, DC.
34 http://www.epa.gOV/raf/publications/p
35 dfs/benchmark_dose_guidance.pdf
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 EXECUTIVE SUMMARY
3 Occurren ce an d 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.
xxxii 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-DNA adducts, or external measures of benzo[a]pyrene exposure. Overall, the
4 human studies report developmental and reproductive effects that are generally analogous to those
5 observed in animals, and provide qualitative, supportive evidence for hazards associated with
6 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, was chosen as the basis
13 for the proposed overall oral reference dose (RfD) as the available data indicate that
14 neurobehavioral changes representthe most sensitive hazard of benzo[a]pyrene exposure. The
15 neurodevelopmental study by Chenetal. (2012) and the observed neurobehavioral changes were
16 used to derive the RfD. The endpoint of altered anxiety-like behavior, as measured in the elevated
17 plus maze, was selected as the critical effect due to the sensitivity of this endpoint and the observed
18 dose-response relationship of effects across dose groups. Benchmark dose (BMD) modeling was
19 utilized to derive the BMDLiso of 0.09 mg/kg-day that was used as the point of departure (POD) for
20 RfD derivation.
21 The proposed overall RfD was calculated by dividing the POD for altered anxiety-like
22 behavior as measured in the elevated plus maze by a composite uncertainty factor (UF) of 300 to
23 account for the extrapolation from animals to humans (10), for inter individual differences in
24 human susceptibility (10), and for deficiencies in the toxicity database (3).
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
Table ES-1. Organ/system-specific RfDs and proposed overall RfD for
benzo[a]pyrene
Effect
Developmental
Reproductive
Immunological
Proposed Overall
RfD
Basis
Neurobehavioral changes
Gavage neurodevelopmental study in rats (PND 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 9
De Jong etal. (1999)
Developmental toxicity
RfD
(mg/kg-d)
3 x 10'4
4 x 10"4
2 x 10"3
3 x 10'4
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
Confidence in the Overall Oral RfD
The overall confidence in the RfD is medium. Confidence in the principal study [Chen etal..
2012] is medium-to-high. The design, conduct, and reporting of this neurodevelopmental study
was good and a wide variety of neurotoxicity endpoints were measured. Some informative
experimental details were, however, omitted including the sensitivity of some assays at the
indicated developmental ages and lack of reporting gender-specific data for all outcomes. Several
subchronic and developmental studies covering a wide variety of endpoints are also available;
however, the lack of a multigeneration toxicity study with exposure throughout development is not
available. Therefore, confidence in the database is medium.
Effects Other Than Cancer Observed Following Inhalation Exposure
In animals, inhalation exposure to benzo[a]pyrene has been shown to result in
developmental and reproductive toxicity. Studies in rats following inhalation exposure show
decreased fetal survival and brain effects in offspring, and decreased testes weight and sperm
counts in adult animals. Overall, the available human PAH mixtures studies report developmental
and reproductive effects that are generally analogous to those observed in animals, and provide
qualitative, supportive evidence for the hazards associated with benzo[a]pyrene exposure.
Inhalation Reference Concentration (RfC) for Effects Other Than Cancer
An attempt was made to derive organ or system-specific RfCs for hazards associated with
benzo[a]pyrene exposure where data were amenable (see Table ES-2). 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.
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
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
Developmental toxicity, represented by decreased fetal survival, was chosen as the basis for
the proposed inhalation reference concentration (RfC) as the available data indicate that
developmental effects represent a sensitive hazard of benzo[a]pyrene exposure. The
developmental inhalation study in rats by Archibong et al. [2002] and the observed decreased fetal
survival following exposure to benzo[a]pyrene on gestation days (CDs) 11-20 were used to derive
the overall RfC. The lowest-observed-adverse-effect level (LOAEL) of 25 [ig/m3 based on decreased
fetal survival was selected as the POD. The LOAEL was adjusted to account for the discontinuous
daily exposure to derive the PODADj and the human equivalent concentration (HEC) was calculated
from the PODADj by multiplying by the regional deposited dose ratio (RDDRER) for extrarespiratory
(i.e., systemic) effects, as described in Methods for Derivation of Inhalation Reference Concentrations
and Application of Inhalation Dosimetry [U.S. EPA, 1994]. These adjustments resulted in a PODHEc of
4.6 [ig/m3, which was used as the POD for RfC derivation.
The RfC was calculated by dividing the POD by a composite UF of 3,000 to account for
toxicodynamic differences between animals and humans (3], interindividual differences in human
susceptibility (10], LOAEL-to-no-observed-adverse-effect level (NOAEL] extrapolation (10], and
deficiencies in the toxicity database (10].
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 (GD 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
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
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 developmental effects) and similar effects observed in human populations exposed to PAH
2 mixtures.
3 Evidence for Human Carcinogenicity
4 Under EPA's Guidelines for Carcinogen Risk Assessment [U.S. EPA, 2005a], benzo[a]pyrene is
5 "carcinogenic to humans" based on strong and consistent evidence in animals and humans. The
6 evidence includes an extensive number of studies demonstrating carcinogenicity in multiple animal
7 species exposed via all routes of administration and increased cancer risks, particularly in the lung
8 and skin, in humans exposed to different PAH mixtures containing benzo[a]pyrene. Mechanistic
9 studies provide strong supporting evidence that links the metabolism of benzo[a]pyrene to DNA-
10 reactive agents with key mutational events in genes that can lead to tumor development. These
11 events include formation of specific DNA adducts and characteristic mutations in oncogenes and
12 tumor suppressor genes that have been observed in humans exposed to PAH mixtures. This
13 combination of human, animal, and mechanistic evidence provides the basis for characterizing
14 benzo[a]pyrene as "carcinogenic to humans."
15 Quantitative Estimate of Carcinogenic Risk From Oral Exposure
16 Lifetime oral exposure to benzo[a]pyrene has been associated with forestomach, liver, oral
17 cavity, jejunum or duodenum, and auditory canal tumors in male and female Wistar rats,
18 forestomach tumors in male and female Sprague-Dawley rats, and forestomach, esophagus, tongue,
19 and larynx tumors in female B6C3Fi mice (male mice were not tested). Less-than-lifetime oral
20 exposure to benzo[a]pyrene has also been associated with forestomach tumors in more than
21 10 additional bioassays with several strains of mice. The Kroese etal. [2001) and Beland and Gulp
22 [1998) studies were selected as the best available studies for dose-response analysis and
23 extrapolation to lifetime cancer risk following oral exposure to benzo[a]pyrene. These studies
24 included histological examinations for tumors in many different tissues, contained three exposure
25 levels and controls, contained adequate numbers of animals per dose group (~50/sex/group),
26 treated animals for up to 2 years, and included detailed reporting methods and results (including
27 individual animal data).
28 Time-weighted, average daily doses were converted to human equivalent doses on the basis
29 of (body weight)3/4 scaling [U.S. EPA. 1992). EPA then used the multistage-Weibull model for the
30 derivation of the oral slope factor. This model was used because it incorporates the time at which
31 death-with-tumor occurred and can account for differences in mortality observed between the
32 exposure groups. Using linear extrapolation from the BMDLio, human equivalent oral slope factors
33 were derived for each gender/tumor site combination (slope factor = 0.1/BMDLio) reported by
34 Kroese etal. (2001) and Beland and Gulp (1998). The oral slope factor of 1 per mg/kg-day based
35 on the tumor response in the alimentary tract (forestomach, esophagus, tongue, and larynx) of
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 female B6C3Fi mice [Beland and Gulp, 1998] was selected as the factor with the highest value
2 (most sensitive) among a range of slope factors derived.
3 Quantitative Estimate of Carcinogenic Risk From Inhalation Exposure
4 Inhalation exposure to benzo[a]pyrene has been associated with squamous cell neoplasia in
5 the larynx, pharynx, trachea, esophagus, and forestomach of male Syrian golden hamsters exposed
6 to benzo[a]pyrene condensed onto NaCl particles [Thyssenetal., 1981]. Supportive evidence for
7 the carcinogenicity of inhaled benzo[a]pyrene comes from additional studies with hamsters
8 exposed to benzo[a]pyrene via intratracheal instillation. The Thyssen et al. [1981] bioassay
9 represents the only available data that exhibit a dose-response relationship for cancer from inhaled
10 benzo[a]pyrene.
11 A time-to-tumor dose-response model was fit to the time-weighted average (TWA]
12 continuous exposure concentrations and the individual animal incidence data for tumors in the
13 larynx, pharynx, trachea, esophagus, and forestomach. The inhalation unit risk of 5 x 10-* per
14 Hg/m3 was calculated by linear extrapolation (slope factor = 0.1/BMCLio] from a BMCLio of 0.20
15 mg/m3 for the occurrence of upper respiratory and upper digestive tract tumors in male hamsters
16 chronically exposed by inhalation to benzo[a]pyrene (Thyssen et al., 1981].
17 Quantitative Estimate of Carcinogenic Risk From Dermal Exposure
18 Skin cancer in humans has been documented to result from occupational exposure to
19 complex mixtures of PAHs including benzo[a]pyrene, such as coal tar, coal tar pitches, unrefined
20 mineral oils, shale oils, and soot In animal models, numerous dermal bioassays have demonstrated
21 an increased incidence of skin tumors with increasing dermal exposure of benzo[a]pyrene in all
22 species tested (mice, rabbits, rats, and guinea pigs], although mostbenzo[a]pyrene bioassays have
23 been conducted in mice. Due to the evidence supporting a hazard from exposure to
24 benzo[a]pyrene by the dermal route (see Section 1.1.5] and the availability of quantitative
25 information, a cancer slope factor for the dermal route was developed. The analysis in this
26 assessment focuses on chronic carcinogenicity bioassays in several strains of mice demonstrating
27 increasing incidence of benign and malignant skin tumors following repeated dermal exposure to
28 benzo[a]pyrene for the animals' lifetime.
29 The Poel (1959] and Sivaketal. (1997] studies were selected as the best available studies
30 for dose-response analysis and extrapolation to lifetime cancer risk following dermal exposure to
31 benzo[a]pyrene. Both studies included at least three exposure levels (including several low doses],
32 group sizes of 30-50 mice, and reporting of intercurrent mortality.
33 Both mouse skin tumor incidence data sets were modeled using the multistage-cancer
34 model. Following the modeling, the BMDLio was adjusted for interspecies differences by
35 allometric scaling. The dermal slope factor of 0.005 per u.g/day was calculated by linear
36 extrapolation (slope factor = 0.1/BMDLio-HEo] from the human equivalent POD for the occurrence of
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 skin tumors in male mice chronically exposed dermally to benzo[a]pyrene. As this slope factor has
2 been developed for a local effect, it is not intended to estimate systemic risk of cancer following
3 dermal absorption of benzo[a]pyrene into the systemic circulation.
4 Susceptible Populations and Lifestages
5 Benzo[a]pyrene has been determined to be carcinogenic by a mutagenic mode of action in
6 this assessment. According to the Supplemental Guidance for Assessing Susceptibility from Early Life
7 Exposure to Carcinogens [U.S. EPA, 2005b], individuals exposed during early life to carcinogens with
8 a mutagenic mode of action are assumed to have an increased risk for cancer. The oral slope factor
9 of 1 per mg/kg-day, inhalation unit risk of 0.0005 per ug/m3, and dermal slope factor of 0.005 per
10 ug/day for benzo[a]pyrene, calculated from data applicable to adult exposures, do not reflect
11 presumed early life susceptibility to this chemical. Although some chemical-specific data exist for
12 benzo[a]pyrene that demonstrate increased early life susceptibility to cancer, these data were not
13 considered sufficient to develop separate risk estimates for childhood exposure. In the absence of
14 adequate chemical-specific data to evaluate differences in age-specific susceptibility, the
15 Supplemental Guidance [U.S. EPA. 2005b) recommends that age-dependent adjustment factors
16 (ADAFs) be applied in estimating cancer risk. The ADAFs are 10- and 3-fold adjustments that are
17 combined with age specific exposure estimates when estimating cancer risks from early life
18 (<16 years of age) exposures to benzo[a]pyrene.
19 Regarding effects other than cancer, there are epidemiological studies that report
20 associations between developmental effects (decreased postnatal growth, decreased head
21 circumference, and neurodevelopmental delays), reproductive effects and internal biomarkers of
22 exposure to benzo[a]pyrene. Studies in animals also indicate alterations in neurological
23 development and heightened susceptibility to reproductive effects following gestational or early
24 postnatal exposure to benzo[a]pyrene.
25 Key Issues Addressed in Assessment
26 The dermal slope factor was developed based on data in animals. Because there is no
27 established methodology for extrapolating dermal toxicity from animals to humans, several
28 alternative approaches were evaluated (See Appendix D in Supplemental Information). Allometric
29 scaling using body weight to the % power was selected based on known species differences in
30 dermal metabolism and penetration of benzo[a]pyrene.
31
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
LITERATURE SEARCH STRATEGY | STUDY SELECTION
3
4
5
6
7
The literature search strategy used to identify primary, peer-reviewed literature pertaining
to benzo[a]pyrene was conducted using the databases listed in Table LS-1 (see Appendix C for the
complete list of keywords used). References from previous assessments by EPA and other national
and international health organizations were also examined. A comprehensive literature search was
last conducted in February 2012.
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)a
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)
9
10
11
12
13
14
15
16
17
Primary 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,
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 abstracts, and/or papers was then conducted. Notable exclusions from the Toxicological Review
2 are large numbers of animal in vivo or in vitro studies designed to identify potential therapeutic
3 agents that would prevent the carcinogenicity or genotoxicity of benzo[a]pyrene and toxicity
4 studies of benzo[a]pyrene in nonmammalian species (e.g., aquatic species, plants).
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
References identified based on initial keyword search (see Table LS-1): ~21,000 references
Secondary keyword searching (see Table
LS-1): ~14,600 references excluded
References identified based on
secondary keyword search (see Table
LS-1): ~6,100 references
30 references submitted by American
Petroleum Institute
Manual screen of titles/abstracts:
~4,900 references excluded
• Not relevant toBaP 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
• Not available in English
Approximately 700 references cited in the August 2012 Draft Toxicological Review
• Developmental toxicity: 37 references
• Reproductive toxicity: 70 references
• Immunotoxicity: 58 references
• OtherToxicological Effects: 27 references
• Forestomach toxicity: 5 references
• Hematological toxicity: 3 references
• Livertoxicity: 3 references
• Kidney toxicity: 3 references
• Cardiovasculartoxicity: 11 references
• Neurological toxicity: 12 references
• Carcinogenicity: 171 references
•Toxicokinetic: 115 references
•Genotoxicity: 196 references
2
3
Figure LS-1. Study selection strategy.
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 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 (A 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. In
24 addition, animal toxicity studies involving short-term duration and other routes of exposure were
25 evaluated to inform conclusions about health hazards.
26 The references considered and cited in this document, including bibliographic information
27 and abstracts, can be found on the Health and Environmental Research Online (HERO] website2
28 (http://hero.epa.gov/benzoapyrene].
29
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.
xlii DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1
2 1. HAZARD IDENTIFICATION
3 1.1. SYNTHESIS OF EVIDENCE
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 no 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 as an
10 indicator chemical to measure exposure to PAH mixtures [Bostrometal., 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 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)-DNA adducts in relation to measures of child growth
23 following exposure to PAH mixtures [Tangetal.. 2006: Pereraetal.. 2005b: Perera etal.. 2004]
24 (Table 1-1]. In the Chinese cohort, high benzo[a]pyrene-adduct levels were associated with
25 reduced weight at 18, 24, and 30 months of age, but not at birth [Tangetal., 2006]. In the U.S.
26 cohort, an independent effect on birth weight was not observed with either benzo[a]pyrene-
27 adducts or environmental tobacco smoke (ETS] exposure; however, a doubling of cord blood
28 adducts in combination with ETS exposure in utero was seen, corresponding to an 8% reduction in
29 birth weight [Perera et al., 2005b]. ETS, also called secondhand smoke, is the smoke given off by a
30 burning tobacco product and the smoke exhaled by a smoker that contains over 7,000 chemicals
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 including benzo[a]pyrene. No associations were seen with birth length (or height at later ages) in
2 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 not observe
14 any difference in fetal survival (Tules etal.. 2012: McCallister etal.. 2008: Brown etal.. 2007). By
15 the inhalation route, fetal survival was decreased by 19% following exposure to 25 |J.g/m3
16 benzo[a]pyrene on CDs 11-20 in F344 rats (Archibongetal.. 2002). Another study from the same
17 group of collaborators Wu etal. (2003a) evaluated fetal survival as part of a study analyzing
18 metabolites of benzo[a]pyrene and activation of the aryl hydrocarbon receptor (AhR) and
19 cytochrome P450 (CYP450) 1A1 (Wuetal.. 2003a). This study did not report the 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 |ig/m3 benzo[a]pyrene on CDs 11-20. An apparent
22 decrease in fetal survival was also seen at 2 5 |ig/m3, but it was unclear whether this change was
23 statistically significant.
24 In animals (Table 1-2), reduced bodyweight in offspring has also been noted in some
25 developmental studies. Decreases in body weight (up to 13%) were observed in mice following
26 prenatal gavage exposure (gestation days [CDs] 7-16), and as time from exposure increased
27 (postnatal days [PNDs] 20-42) the dose at which effects were observed decreased (from 40 to
28 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 (Tules etal.. 2012: McCallister etal.. 2008). Maternal
33 toxicity was not observed in mouse or rat dams exposed to up to 160 mg/kg-day benzo[a]pyrene
34 (Tules etal.. 2012: McCallister etal.. 2008: Brown etal.. 2007: Kristensen etal.. 1995: Mackenzie and
35 Angevine. 1981).
36 Fertility in Offspring
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 Several studies suggest that gestational exposure to maternal tobacco smoke decreases the
2 future fertility of female offspring [Ye etal.. 2010: Tensen etal.. 1998: Weinbergetal.. 1989] (Table
3 1-1). In animal models, marked effects on the development of male and female reproductive organs
4 and the fertility of animals exposed gestationally has also been demonstrated [Kris tensen etal.,
5 1995: Mackenzie and Angevine, 1981] (Table 1-2]. In two studies examining reproductive effects in
6 mice, decreased fertility and fecundity in Fl animals was observed following exposure to doses
7 >10 mg/kg-day during gestation [Kristensenetal., 1995: Mackenzie and Angevine, 1981]. When Fl
8 females were mated with untreated males, a dose-related decrease in fertility of >30% was
9 observed, in addition to a 20% decrease in litter size starting at the lowest dose tested of 10 mg/kg-
10 day [Mackenzie and Angevine. 1981]. A dose-related decrease in fertility was also observed in male
11 mice treated gestationally with benzo[a]pyrene. At the lowest dose tested (10 mg/kg-day], a 35%
12 decrease in fertility was observed when gestationally exposed animals were mated with untreated
13 females [Mackenzie and Angevine, 1981]. Similar effects on fertility were observed in another
14 developmental study in mice [Kristensen et al., 1995]. Fl females (bred continuously for 6 months]
15 in this study had 63% fewer litters, and litters were 30% smaller as compared to control animals.
16 The fertility of male offspring was not assessed in this study.
17 Reproductive Organ Effects in Offspring
18 The above mentioned studies also demonstrated dose-related effects on male and female
19 reproductive organs in animals exposed gestationally to benzo[a]pyrene (Table 1-2]. Testicular
20 weight was decreased and atrophic seminiferous tubules and vacuolization were increased at
21 >10 mg/kg-day in male mice exposed to benzo[a]pyrene gestationally from GD 7 to 16; severe
22 atrophic seminiferous tubules were observed at 40 mg/kg-day [Mackenzie and Angevine. 1981].
23 In female mice treated with doses >10 mg/kg-day during gestation, ovarian effects were
24 observed including decreases in ovary weight, numbers of follicles, and corpora lutea [Kristensen et
25 al., 1995: Mackenzie and Angevine, 1981]. Specifically, ovary weight in Fl offspring was reduced
26 31% following exposure to 10 mg/kg-day benzo[a]pyrene [Kristensen et al., 1995] while in another
27 gestational study at the same dose level, ovaries were so drastically reduced in size (or absent] that
28 they were not weighed [Mackenzie and Angevine. 1981]. Hypoplastic ovaries with few or no
29 follicles and corpora lutea (numerical data not reported], and ovaries with few or no small,
30 medium, or large follicles and corpora lutea (numerical data not reported] have also been observed
31 in mouse offspring exposed gestationally to benzo[a]pyrene [Kristensenetal.. 1995: Mackenzie and
32 Angevine. 1981].
3 3 Cardiovascular Effects in Offspring
34 Increased systolic and diastolic blood pressure was observed in adult animals following
35 gestational treatment with benzo[a]pyrene [Tules etal., 2012] (Table 1-2]. Approximate elevations
36 in systolic and diastolic blood pressure of 20-30% and 50-80% were noted in the 0.6 and
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
2
3
4
5
6
7
1.2 mg/kg-day dose groups, respectively. Heart rate was decreased at 0.6 mg/kg-day, but was
increased at 1.2 mg/kg-day.
Immune Effects in Offspring
Several injection studies in laboratory animals suggest that immune effects may occur
following gestational or early postnatal exposure to benzo[a]pyrene. These studies are discussed in
Section 1.1.3.
8
9
Table 1-1. Evidence pertaining to developmental effects of benzo[a]pyrene in
humans
Study Design and Reference
Tang et al. (2006)
Tongliang, China
Birth cohort
150 non-smoking 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
Perera et al. (2005b); Perera et al. (2004)
New York, United States
Results
Relation between cord blood benzo[a]pyrene-DNA adducts and log-
transformed weight and height
Birth
18 mo
24 mo
30 mo
Weight
Beta (p-value)
-0.007 (0.73)
-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)
Relation between cord blood benzo[a]pyrene-DNA adducts and log-
transformed weight and length
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
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)
Results
Interaction
term
Benzo[a]-
pyrene-DNA
adducts
ETS in home
Weight
Beta (p-value)
-0.088 (0.05)
-0.020 (0.49)
-0.003 (0.90)
Length
Beta (p-value)
-0.014 (0.39)
-0.005 (0.64)
-0.007 (0.32)
Adjusted for ethnicity, sex of newborns, maternal body mass index,
dietary PAHs, and gestational age
Benzo[a]pyrene adduct levels (/l(f 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
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
Kristensen et al. (1995)
NMRI mice, 9 FO females/dose
0 or 10 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*
^ Fl body weight at PND 20
% change from control: 0, 4, -7*, and -13*
4, Fl body weight at PND 42
% change from control: 0, -6*, -6*, and -10*
(no difference in pup weight at PND 4)
Exposed FO females showed no gross signs of toxicity and no effects on
fertility (data not reported)
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
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
McCallister et al. (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 PND 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
Archibongetal. (2002)
F344 rats, 10 females/group
0, 25, 75, or 100 u.g/m3 nose-only
inhalation for 4 hrs/d
CDs 11-20
fetal survival ([pups/litter]/[implantation sites/litter] x 100)
% fetal survival: 97, 78*, 38*, and 34*%
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
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*
4, 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)
4, testicular weight in Fl offspring
% change from control: 0, -42, -82, and ND (statistical significance not
reported)
1" atrophic seminiferous tubules and vacuolization at >10 mg/kg-d; severe
atrophic seminiferous tubules at 40 mg/kg-d (numerical data not reported)
Kristensen et al. (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
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)
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-day
8%* decrease at 1.2 mg/kg-day
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 :
1 n
•— • 1U :
ra :
T3 :
00
^ 1 --
60 :
_§ ;
oi
0 0.1 :
Q :
n m
U.Ul ~
• LOAEL ANOAEL * Doses > LOAEL 0 Doses < NOAO.
• • • • •
A * • • •
•
A A
0 0
o A o A .
o A o A
0 0
rH rsl ^ CO rv.
00 .H ^ O Q v/l
2 SH Sa 82
- ° ^4- 2 r-T ~J ^
o "F ° ri ° T1 S "*
- c. S G 03 c -^
< «j _D 13 .^ ^ g Q
c h- 2 o m
So ^
'N 13 ^
o
^
j^
s
^ number
viable litters
rH O 00 rH
CO rr; O ° h-
0 IN ,/, n 0 "1 _,
rt ;n^QOJ;/,Q ^
rgO^Z- ,;:;_, -
.Son ^" c 2 H,- t o ^ S
g-yZ^OOHo^. Q
^^toSES^-p-rlz
^ 2-u ^5« 3 g|^
•S S S " 5 E 5 - |
c o ^
op 7
^ ^
(J ^
5
•i. body weight
in offspring
Birth Oilcomes
f-\ t/i
00 C^
s s.s
O — JI
.± "J
> .y o 3
C£ £ C 'i
c • - o r^-
IS P
•i@ 5
o
v;
r^
5
4/F1 female
fertility
_,
CO
o
o
.± o
> u
O r-
C£ t
< ^
0 &
O
v;
J^
5
4-F1
male
fertility
• t
• •
•
£
^ m
00 C
^
0 C1
rf i d
o —
.£ o
3 -^
> y o '•o
C£ C C
- ft ^
j i —
O Ci ^
o
v:
^
1
4. ovarunsize
^H in
CO O
0 (Ti „
,_, ^H 0
c o Q ~
'> ° o ^b"
Op i-H
r- r. 5 Is-
1? p
o Q ^
o
Ni
^
5
•J/ ovarian
follicles
»H
CO
d
iH
.= O
> V
O r-
j?
o Q
o
^:
^
5
xb
esticulai
weight
and t
atrophy
Reproductive Effects
inOffspr
1R
no
^1
2 2 Q
o p z
^ r, CL
r h-" r-
0 T1 0
-j ^J "O
O rH p
^ Q 3
3 13 0
—i O
~
Cardio-
vascular
Iffects 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
1 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 a period of rapid brain growth during the last 3 months of
12 pregnancy through the first 2 years of life in humans [Dobbingand Sands. 1979.1973] and the first
13 1-2 weeks of life in the rat and mouse neonate [Chen etal.. 2011). This period is characterized by
14 the maturation of axonal and dendritic outgrowth and the establishment of neuronal connections.
15 Also during this critical period, animals acquire many new motor and sensory abilities [Kolb and
16 Whishaw, 1989]. There is a growing literature of animal studies that shows subtle changes in
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 motor and cognitive function following acute or repeated perinatal or lactational exposure to
2 benzo[a]pyrene [Bouayed et al.. 2009a: McCallister et al.. 2008: Wormley etal.. 2004). These effects
3 are described below.
4 Cognitive function
5 Head circumference at birth is associated with measures of intelligence in children, even
6 among term infants [Broekman etal., 2009: Gale etal., 2006]. The two birth cohort studies that
7 examined maternal or cord blood levels of benzo[a]pyrene-DNA adducts in relation to head
8 circumference provide some evidence of an association, most strongly within the context of an
9 interaction with ETS [Tang etal.. 2006: Pereraetal.. 2005b: Perera etal.. 2004] (Table 1-3]. The
10 cohort in Tongliang, China also examined intelligence quotient scores at age 5 years [Pereraetal..
11 2012a]. An interaction with ETS was seen in this analysis, with larger decrements seen on the full
12 scale and verbal scales with increased benzo[a]pyrene-DNA adduct levels in the presence of
13 prenatal exposure to environmental tobacco smoke compared to the effects seen in the absence of
14 prenatal exposure to environmental tobacco smoke.
15 Animal studies have also provided evidence of altered learning and memory behaviors
16 following lactational or direct postnatal exposure to benzo[a]pyrene [Chen etal., 2012: Bouayed et
17 al.. 2009a] (Table 1-4]. In mice, spatial working memory was measured using the Y-maze
18 spontaneous alternation test [Bouayed et al., 2009a]. This test records alternations between arm
19 entries in a Y-shaped maze as a measure of memory, as rodents typically prefer to investigate a new
20 arm of the maze. To a lesser extent, this test can also reflect changes in sensory processing, novelty
21 preference, and anxiety-related responses in rodents. An improvement in working memory was
22 evident in mice, as exhibited by significant increases in spontaneous alternations in the Y-maze test
23 in mice on PND 40 following lactational exposure to 2 mg/kg-day benzo[a]pyrene (but not 20
24 mg/kg-day] from PND 0 to 14 [Bouayed etal.. 2009a]. The total number of arm entries in the
25 Y-maze was unaffected by lactational exposure, suggesting that changes in motor function were not
26 driving this response. In rats, spatial learning and memory was measured using the Morris water
27 maze, which measures the ability of a rat to navigate to a target platform using external spatial cues
28 [Chen etal.. 2012]. Increased escape latency(time to find the hidden platform], as well as
29 decreased time in the target quadrant and decreased number of platform crossings during a probe
30 trial with the platform removed were observed in PND 39-40 rats following postnatal exposure to
31 2 mg/kg-day benzo[a]pyrene [Chen etal.. 2012]. These effects were more pronounced in animals
32 tested at PNDs 74-75. No difference in swim speed was observed between treatment groups,
33 suggesting the observed changes are not attributable to general motor impairment These
34 observations may indicate primary effects of benzo[a]pyrene on learning and/ or memory
35 processes; however, the presented data were insufficient to attribute these findings to learning and
36 memory processes alone. Specifically, visual examinations of the improvements in escape latency
37 (slopes] over the four learning trial days were not noticeably affected by treatment dose, suggesting
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 that all groups learned at a similar rate. As 4 trials/ day were averaged for each animal at each trial
2 day, it is unclear whether the dose-related increases in escape latency already observable at trial
3 day 1 reflect effects on learning or other effects (e.g., altered anxiety or vision responses).
4 Additionally, as it is not clear that the groups learned to a comparable extent in hidden platform
5 tests, the results of the probe trial cannot be conclusively attributed to memory retention and likely
6 involve other contributing factors.
7 Neuromuscular function, coordination, and sensorimotor development
8 Motor behavior and coordination, assessed by locomotion, reaching, balance,
9 comprehension, drawing, and hand control was one of the specific domains assessed in the Chinese
10 birth cohort evaluated by Tangetal. [2008]. In children aged 2 years, decreased scores were seen
11 in relation to increasing benzo[a]pyrene-DNA adducts measured in cord blood, with a Beta per unit
12 increase in adducts of-16 (p = 0.04), and an approximate twofold increased risk of development
13 delay per unit increase in adducts (Table 1-3).
14 In laboratory animals (Table 1-4), impaired performance in neuromuscular and
15 sensorimotor tests have been consistently observed in mice lactationally exposed to >2 mg/kg-day
16 benzo[a]pyrene from PND 0 to 14 (Bouayed et al., 2009a] and in rat pups postnatally exposed to
17 >0.02 mg/kg-day benzo[a]pyrene from PND 5 to 11 (Chenetal.. 2012). In the righting reflex test,
18 significant increases in righting time were observed in PNDs 3-5 mice and in PNDs 12-16 rats.
19 These decrements did not show a monotonic dose response. In another test of sensorimotor
20 function and coordination, dose-dependent increases in latency in the negative geotaxis test were
21 observed in PNDs 5-9 mice and in PNDs 12-14 rats. The forelimb grip strength test of
22 neuromuscular strength was also evaluated in both mice and rats, but alterations were only
23 observed in mice. In mice, a dose-dependent increase in duration of forelimb grip was observed on
24 PNDs 9 and 11 during lactational exposure to benzo[a]pyrene. The Water Escape Pole Climbing
25 test was also used to evaluate neuromuscular function and coordination in mice (Bouayed etal.,
26 2009a). No effect on climbing time was observed, suggesting no change in muscle strength.
27 However, increased latency in pole grasping and pole escape in PND 20 male pups was observed,
28 highlighting potential decrements in visuomotor integration and/or coordination, although anxiety-
29 related responses cannot be ruled out Treatment-dependant increases in pup body weight around
30 the testing period complicate the interpretation of these results.
31 Negative geotaxis and surface righting are discrete endpoints routinely used as part of a
32 neurobehavioral test battery to assess acquisition of developmental milestones. In typical
33 protocols, animals are tested on successive days (e.g. PNDs 3-12) and successful acquisition of
34 these phenotypes is indicated when righting occurs. In rats, both phenotypes are nearly always
35 established before PND 12. Chenetal. (2012) performed these tests as quantitative measures of
36 sensorimotor function at PND 12 and beyond, with control animals already able to right within
37 0.8-1.8 seconds and orient 180° within 5-9 seconds. Although informative in terms of possible
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 delays in sensory motor development, the sensitivity of these measures at these later postnatal ages
2 is difficult to interpret Specifically, statistically significant differences observed by Chenetal.
3 [2012] in the surface righting test were on the order of ~0.2-0.3 seconds and in the negative
4 geotaxis test, ~3-4 seconds, with no automated recording of latency (such as use of video
5 recordings). Additionally, male and female rats (which sometimes exhibit differences in the
6 maturation of these developmental landmarks following challenge) were pooled for these
7 measures.
8 Anxiety and activity
9 Anxiety, attention, and hyperactivity in children ages 6-7 years were examined in relation
10 to benzo[a]pyrene-DNA adducts measured at birth in a follow-up of a birth cohort study conducted
11 in New York City (Pereraetal., 2012b). The associations were stronger using the measures in cord
12 blood compared with maternal samples, with indications of a fourfold increased risk (p = 0.051) of
13 attention problems associated with cord blood adduct levels above the median compared with 2-
14 fold increased risk associated with maternal blood adduct levels (Table 1-3). Exposure was treated
15 as a dichotomy (i.e., detectable compared with non-detectable levels) in these analyses.
16 Decreased anxiety-like behavior was reported in both rat and mice following postnatal oral
17 exposure to benzo[a]pyrene (Chenetal.. 2012: Bouayedetal.. 2009a] (Table 1-4). Anxiety-like
18 behaviors were measured in both species using an elevated plus maze, where an increase in the
19 time spent in the closed arms of the maze is considered evidence of anxious behavior. In mice,
20 significant increases in the entries and time spent in open arms of the maze, as well as significantly
21 decreased entries into closed arms of the maze (in the 2 mg/kg-day group), were observed on PND
22 32 followinglactational exposure to >2 mg/kg-day benzo[a]pyrene (Bouayedetal.. 2009a). The
23 mice also exhibited decreased latency of the first entry into an open arm following lactational
24 exposure to 20 mg/kg-day benzo[a]pyrene. There was no exposure-related effect on the total
25 number of times the mice entered arms of the maze, indicating the lack of an effect on general
26 locomotor activity. Decreased anxiety-like behavior was also reported in rats following oral
27 benzo[a]pyrene exposure from PND 5 to 11, although sex-specific differences were observed (Chen
28 etal.. 2012). In females, postnatal exposure to >0.2 mg/kg-day benzo[a]pyrene was associated with
29 a significant increase in the number of open arm entries and significant decreases in the number of
30 closed arm entries on PND 70. Significantly increased time in open arms of the maze was reported
31 in PND 70 female rats following postnatal exposure to >0.02 mg/kg-day. Male rats also showed
32 decreased anxiety-like behavior on PND 70, although the doses needed to detect these responses
33 were higher than females (i.e., increases at >2 mg/kg-day for open arm entries and >0.2 mg/kg-day
34 for time spent in open arms). A significant decrease in latency to enter an open arm of the maze
35 was observed in both male and female rat pups exposed to 2 mg/kg-day benzo[a]pyrene.
36 In contrast to results from the elevated plus maze, no altered reactions were observed in
37 cliff aversion test paradigms (testing stress responses and integration of sensorimotor processes)
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 following postnatal oral exposure to benzo[a]pyrene [Chenetal.. 2012: Bouayedetal.. 2009a].
2 Increased spontaneous locomotor activity in the open field on PND 69 has been reported in rats
3 postnatally exposed to >0.2 mg/kg-day [Chenetal.. 2012). Elevated activity in an open field is
4 attributable primarily to either increased motor activity or decreased anxiety-like behavior,
5 although the relative contributions of these two components could not be separated.
6 Increased locomotor activity on PND 69, measured using the open field test, has been
7 reported in rats postnatally exposed to >0.2 mg/kg-day benzo[a]pyrene on PNDs 5-11 [Chenetal.,
8 2012], but not in mice exposed lactationally to doses up 20 mg/kg-day and tested on PND 15
9 fBouayedetal..2009a).
10 Electrophysiological changes
11 Electrophysiological effects of gestational exposure to benzo[a]pyrene have been examined
12 in two studies (by the same research group) through implanted electrodes in the rat cortex and
13 hippocampus (Table 1-4). Maternal inhalation exposure to 0.1 mg/m3 resulted in reduced long-
14 term potentiation in the dentate gyrus of male offspring between PND 60 and 70 (Wormley etal..
15 2004). Oral exposure of dams to 0.3 mg/kg-day for four days during late gestation resulted in
16 decreased evoked neuronal activity in male offspring following mechanical whisker stimulation
17 between PND 90 and 120 fMcCallister etal.. 20081 Specifically, the authors noted reduced spike
18 numbers in both short and long latency responses following whisker stimulation. These effects
19 were observed several months post-exposure, suggesting that gestational benzo[a]pyrene exposure
20 has 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 non-smoking 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 years of
age (4 domains: motor, adaptive,
language, and social); standardized mean
score = 100 ± SD 15 (score < 85 =
developmental delay)
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/108 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
Beta (p-value)
Birth -0.011 (0.057)
18 mo -0.012 (0.085)
24 mo -0.006(0.19)
30 mo -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 et al. (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
years of age (4 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
Cl = confidence level; OR = odds ratio
Pereraetal. (2012a); Tang et al. (2008);
Tangetal. (2006)
(See above for population and exposure
details); 132 (83%) followed through age
5; 100 of these had complete data for
Environmental tobacco smoke (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
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
analysis. No differences between the 100
participants in this analysis and the
nonparticipants with respect to adduct
levels, environmental tobacco smoke
exposure, IQ measures, maternal age,
gestational age, or infant gender. Higher
maternal education (60% and 35% with >
high school, respectively, in participants
and non-participants, p < 0.05)
Outcomes: Wechsler Preschool and
Primary Intelligence Quotient scale
(Shanghai version)
Beta (95% Cl)
Main Effect
With ETS Interaction term
Full scale
Verbal
Performance
-2.42 (-7.96, 3.13)
-1.79 (-7.61, 4.03)
-2.57 (-8.92, 3.79)
-10.10 (-18.90,-1.29)
-10.35 (-19.61, -1.10)
-7.78 (-18.03,2.48)
Beta per 1 unit increase in log-transformed cord adducts, adjusted for
ETS exposure, gestational age, maternal education, cord lead, maternal
age, and gender
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 non-smoking women that
delivered babies between April 1998-
October 2002 (253 and 207 for behavior
and head circumference analysis,
respectively); approximately 40% with
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
Beta (p-value)
Interaction term
benzo[a]pyrene-DNA adducts
ETS in home
-0.032 (0.01)
-0.007 (0.39)
-0.005 (0.43)
High versus low, dichotomized at 0.36 adducts/10"8 nucleotides,
adjusted for ethnicity, sex of newborns, maternal body mass
index, dietary PAHs, and gestational age
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, environmental tobacco smoke
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 to detectable levels of benzo[a]pyrene adducts
Maternal Cord blood
Prevalence OR (95% Cl) OR (95% Cl)
Anxious/depressed 6.32 % 1.4 (0.38, 5.4) 2.6 (0.69, 9.4)
Attention problems 6.72% 2.2(0.74,6.8) 4.1(0.99,16.6)
Anxiety (DSM) 9.48% 2.2(0.79,6.1) 2.5(0.84,7.7)
Attention deficit- 1.8(0.66,5.1) 2.6(0.68,10.3)
hyperactivity (DSM)
Exposure dichotomized as detectable (n = 87 maternal, 56 cord
blood samples) versus non-detectable; adjusted for sex,
gestational age, maternal education, maternal IQ, prenatal ETS,
ethnicity, age, heating season, prenatal demoralization, and
HOME inventory
1
2
*Statistically significantly different from the control (p < 0.05).
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
1
2
Table 1-4. Evidence pertaining to the neurodevelopmental effects of
benzo[a]pyrene in animals
Reference and Study Design
Results3
Cognitive function
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
Hidden Platform test in Morris water maze (day 4 of testing):
PND 39: significant increase in escape latency at 2 mg/kg-d only
PND 74: significant increase in escape latency at >0.2 mg/kg-d
Increases in latency were observed in trials 1-3 ( PND 36-38 or PND
71-73): ~30+% greater than controls at 2 mg/kg-day (note: all
experimental groups exhibited similar decreases in escape latency-
slopes were visually equivalent, across the four trial days)
Probe test in the Morris water maze (day 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
2 mg/kg-d (in males)
Bouayed etal. (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 sensorimotor 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
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
Bouaved etal. (2009a)
Female Swiss albino mice, 5/group
0, 2, or 20 mg/kg-d maternal gavage
PNDs 0-14 (lactational exposure)
Results3
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
Bouaved etal. (2009a)
Female Swiss albino mice, 5/group
0, 2, or 20 mg/kg-d maternal gavage
PNDs 0-14 (lactational exposure)
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 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)
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)
Cliff aversion test
No effect on the latency to retract from the edge
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
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
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
Open field test
No significant change in activity on PND 15
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
Wormlev et al. (2004)
F344 rats, 10 females/group
0 or 100 u.g/m3 nose-only inhalation
for 4 hrs/d
CDs 11-21
Statistically significant decreases in stimulus-evoked cortical
activity on PNDs 90-120
Reduction in the number of spikes in both the short and long
on PNDs 90-120 (numerical data not presented)
Electrophysiological changes in the hippocampus:
Consistently lower long term potentiation following
exposure (statistical analysis not reported)
% change relative to control: -26%
neuronal
; latency periods
gestational
% 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,
1-17 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
100
10 =
1 =
o
Q
0.1 -.
0.01
> Doses > LOACL O Doses < NOACL
1" motor
2 Figure 1-2. Exposure-response array for neurodevelopmental effects
3 following oral exposure.
4
5 Mode of Action Analysis- Developmental Toxicity
6 Data regarding the potential mode of action for the various manifestations of developmental
7 toxicity associated with benzo[a]pyrene exposure are limited, and the mode of action for
8 developmental toxicity is not known. General hypothesized modes of action for the various
9 observed developmental effects include, but are not limited to altered cell signaling, genotoxicity,
10 cytotoxicity, and oxidative stress.
11 It is plausible that developmental effects of benzo[a]pyrene may be mediated by altered cell
12 signaling through the Ah receptor (AhR). Benzo[a]pyrene is a ligand for the AhR and activation of
13 this receptor regulates downstream gene expression including the induction of GYP enzymes
14 important in the conversion of benzo[a]pyrene into reactive metabolites. Studies in AhR knock-out
15 mice indicate that AhR signaling during embryogenesis is essential for normal liver, kidney,
16 vascular, hematopoietic, and immune development [Schmidt etal., 1996: Fernandez-Salguero etal.,
17 1995]. In experiments in AhR responsive and less-responsive mice, the mice with the less
18 responsive AhR were protected from renal injury as adults following gavage treatment with 0.1 or
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 0.5 mg/kg-day benzo[a]pyrene from GD 10-13. Renal injury was indicated by an increase in
2 urinary albumin and a decrease in glomerular number [Nanez etal.. 2011).
3 Low birth weight has been associated with prenatal exposure to PAHs in human
4 populations [Pereraetal., 2005b]. Several epidemiology studies have revealed an inverse
5 association between low birth weight and increased blood pressure, hypertension, and measures of
6 decreased renal function as adults [Zandi-Nejadetal., 2006]. It has been hypothesized that this
7 may be attributable to a congenital nephron deficit associated with intrauterine growth restriction
8 fZandi-Neiadetal.. 20061.
9 No clear mode(s) of action for the observed neurodevelopmental and neurobehavioral
10 changes following benzo[a]pyrene exposure have been demonstrated. General hypothesized
11 mechanisms with limited evidentiary support are related to altered CNS neurotransmission. These
12 mechanisms involve altered neurotransmitter gene expression, and neurotransmitter levels, in
13 regions associated with spatial learning, anxiety, and aggression, such as the hippocampus,
14 striatum, and hypothalamus [Li etal.. 2012: Oiu etal.. 2011: Tang etal.. 2011: Xia etal.. 2011:
15 Bouayedetal.. 2009a: Grovaetal.. 2008: Brown etal.. 2007: Grovaetal.. 2007: Stephanouetal..
16 1998). Specifically, benzo[a]pyrene exposure caused changes associated with potential
17 modifications to the dopaminergic andserotonergic systems [Bouayedetal.. 2009a: Stephanou et
18 al., 1998], as well as NMDA receptor signaling [Qiuetal., 2011]. Increased oxidative stress in these
19 same regions has also been proposed as a mechanism [Saunders etal., 2006].
20 Summary of Developmental Effects
21 Developmental effects following in utero exposure to PAH mixtures or benzo[a]pyrene
22 alone have been reported in humans and in animal models. In human populations, decreased head
23 circumference, decreased birth weight, and decreased postnatal weight have been reported.
24 Analogous effects in laboratory animals, including decreased pup weight and decreased fetal
25 survival, have been noted following gestational or early postnatal exposure to benzo[a]pyrene by
26 the oral or inhalation route [Chen etal., 2012: Archibongetal., 2002: Mackenzie and Angevine,
27 1981]. Reproductive function is also altered in mice treated gestationally with benzo[a]pyrene
28 [Kristensenetal.. 1995: Mackenzie and Angevine. 1981]. These effects include impaired
29 reproductive performance in Fl offspring (male and female] and alterations of the weight and
30 histology of reproductive organs (ovaries and testes].
31 The available human and animal data also support the conclusion thatbenzo[a]pyrene is a
32 developmental neurotoxicant Human studies of environmental PAH exposure in two cohorts have
33 observed neurotoxic effects, including suggestions of reduced head circumference (Tang etal.,
34 2006: Pereraetal.. 2005b: Pereraetal.. 2004], impaired cognitive ability (Pereraetal.. 2009: Tang
35 etal.. 2008]. impaired neuromuscular function (Tang etal.. 2008]. and increased attention
36 problems and anxious/depressed behavior following prenatal exposure (Pereraetal.. 2012b].
37 These effects were seen in birth cohort studies in different populations (New York City and China],
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 in studies using specific benzo[a]pyrene measures (i.e., adduct levels measured in cord blood
2 samples) [Pereraetal.. 2012b: Tang etal.. 2008: Tangetal.. 2006: Pereraetal.. 2005b: Pereraetal..
3 2004). This type of measure covers a relevant time window of exposure with respect to gestational
4 development The analytical method was the same in the two studies (with a common set of
5 investigators). The coefficient of variation of the exposure measures was relatively small (12%),
6 but a high proportion of samples were below the detection limit; thus, these studies were limited in
7 terms of ability to examine a broad range of exposure. The available evidence from mice and rats
8 also demonstrates significant and persistent developmental impairments following exposure to
9 benzo[a]pyrene. Impaired learning and memory behaviors and impaired neuromuscular function
10 were consistently observed in multiple neurobehavioral tests in two separate species at
11 comparable doses in the absence of maternal or neonatal toxicity (Chen etal., 2012: Bouayedetal.,
12 2009a).
13 In conclusion, the available human and animal data suggest that developmental toxicity and
14 developmental neurotoxicity are hazards of benzo[a]pyrene exposure.
15 Suscep tible Pop illations an d Lifestages
16 Childhood susceptibility to benzo[a]pyrene toxicity is indicated by epidemiological studies
17 reporting associations between adverse birth outcomes and developmental effects and internal
18 biomarkers of exposure to benzo[a]pyrene, presumably via exposure to complex PAH mixtures
19 (Pereraetal.. 2012b: Pereraetal.. 2009: Tangetal.. 2008: Tangetal.. 2006: Pereraetal.. 2005b:
20 Pereraetal., 2005a: Perera etal., 2004). The occurrence of benzo[a]pyrene-specific DNAadducts in
21 maternal and umbilical cord blood in conjunction with exposure to ETS was associated with
22 reduced birth weight and head circumference in offspring of pregnant women living in New York
23 City (Pereraetal., 2005b). In other studies, elevated levels of BPDE-DNA adducts in umbilical cord
24 blood were associated with: (1) reduced birth weights or reduced head circumference (Perera et
25 al.. 2005a: Pereraetal.. 2004): and (2) decreased body weight at 18, 24, and 30 months (Tangetal..
26 2008: Tangetal.. 2006).
27 Studies in humans and experimental animals indicate that exposure to PAHs in general, and
28 benzo[a]pyrene in particular, may impact neurological development. Observational studies in
29 humans have suggested associations between gestational exposure to PAHs and later measures of
30 neurodevelopment (Perera etal.. 2009: Tangetal.. 2008). In the Perera etal. (2009) study, the
31 exposure measures are based on a composite of 8 PAHs measured in air. In Tangetal.. (2008).
32 increased levels of benzo[a]pyrene-DNA adducts in cord blood were associated with decreased
33 developmental quotients in offspring (Tangetal., 2008).
34 Evidence in animals of the effects of benzo[a]pyrene on neurological development includes:
35 (1) decreased electrophysiological response to electrical stimulation of the dentate gyrus of the
36 hippocampus and increased brain concentrations of benzo[a]pyrene metabolites in offspring of
37 F344 rats exposed by inhalation to benzo[a]pyrene:carbon black aerosols on CDs 11-21 (Wormley
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 etal.. 2004: Wu etal.. 2003a]: (2) decreased evoked response in the field cortex and decreased
2 cerebrocortical levels of messenger RNA for the N-methyl-D-aspartate receptor subunit in offspring
3 of Long-Evans rats exposed to 300 [J.g/kg on CDs 14-17 [McCallister et al.. 2008): and (3) decreased
4 righting reflex and altered anxiety-like behavior in offspring of lactating rats exposed to oral doses
5 of 2 or 20 mg/kg-day on PNDs 1-14 (Bouayed etal.. 2009a).
6 1.1.2. Reproductive Toxicity
7 Human and animal studies provide evidence for benzo[a]pyrene-induced male and female
8 reproductive toxicity. Effects on sperm quality and male fertility have been demonstrated in human
9 populations highly exposed to PAH mixtures [Spares and Melo, 2008: Hsu etal., 2006]. The use of
10 internal biomarkers of exposure in humans (e.g., BPDE-DNA adducts) support associations between
11 benzo[a]pyrene exposure and these effects. In females, numerous epidemiological studies indicate
12 that cigarette smoking reduces fertility; however, few studies have specifically examined levels of
13 benzo[a]pyrene exposure and female reproductive outcomes. Animal studies demonstrate
14 decrements in sperm quality, changes in testicular histology, and hormone alterations following
15 benzo[a]pyrene exposure in adult male animals, and decreased fertility and ovotoxic effects in adult
16 females following exposure to benzo[a]pyrene.
17 Male Reproductive Effects
18 Fertility
19 Effects on male fertility have been demonstrated in populations exposed to mixtures of
20 PAHs. Spermatozoa from smokers have reduced fertilizing capacity, and embryos display lower
21 implantation rates [Spares and Melo. 2008). Occupational PAH exposure has been associated with
22 higher levels of PAH-DNA adducts in sperm and male infertility [Gaspari et al.. 2003). In addition,
23 men with higher urinary levels of PAH metabolites have been shown to be more likely to be infertile
24 [Xiaetal., 2009]. Studies were not identified that directly examined the reproductive capacity of
25 adult animals following benzo[a]pyrene exposure. However, a dose-related decrease in fertility
26 was observed in male mice treated in utero with benzo[a]pyrene, as discussed in Section 1.1.1.
27 Sperm parameters
28 Effects on semen quality have been demonstrated in populations exposed to mixtures of
29 PAHs including coke oven workers and smokers [Spares and Melo, 2008: Hsu etal., 2006]. Coke
30 oven workers had higher frequency of oligospermia (19 versus 0% in controls] and twice the
31 number of morphologically abnormal sperm (Hsu etal., 2006]. Elevated levels of BPDE-DNA
32 adducts have been measured in the sperm of populations exposed to PAHs occupationally (Gaspari
33 etal.. 2003] and through cigarette smoke (Phillips. 2002: Zenzes etal.. 1999]. A higher
34 concentration of BPDE-DNA adducts was observed in sperm not selected for intrauterine
35 insemination or in vitro fertilization based on motility and morphology in patients of fertility clinics
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 [Perrinetal.. 2011b: Perrinetal.. 2011a]. An association between benzo[a]pyrene exposure levels
2 and increased sperm DNA fragmentation using the sperm chromatin structure assay was observed
3 by Rubes etal. [2010]. However, it is currently unclear whether the sperm chromatin structure
4 assay, which measures sperm fragmentation following denaturation, is predictive of fertility
5 [Sakkas and Alvarez. 2010: ASRM. 20081.
6 In several studies in rats and mice, a decrease in sperm count, motility, and production and
7 an increase in morphologically abnormal sperm have been reported (Table 1-5). Alterations in
8 these sperm parameters have been observed in different strains of rats and mice and across
9 different study designs and routes of exposure.
10 Decreases in epididymal sperm counts (25-50% compared to controls) have been reported
11 in Sprague-Dawley rats and C57BL6 mice treated with 1-5 mg/kg-day benzo[a]pyrene by oral
12 exposure for 42 or 90 days (Chen etal., 2011: Mohamedetal., 2010). Additionally, a 15% decrease
13 in epididymal sperm count was observed at a dose 100-fold lower in Sprague-Dawley rats exposed
14 to benzo[a]pyrene for 90 days (Chung etal., 2011). However, confidence in this study is limited
15 because the authors dosed the animals with 0.001, 0.01, and 0.1 mg/kg-day benzo[a]pyrene, but
16 only reported on sperm parameters at the mid-dose, and no other available studies demonstrated
17 findings in the range of the mid- and high-dose. In rats, an oral short-term study and a subchronic
18 inhalation study lend support for the endpoint of decreased sperm count (Arafaetal., 2009:
19 Archibong et al., 2008: Rameshetal., 2008). Significantly decreased sperm count and daily sperm
20 production (~40% decrease from control in each parameter) were observed following 10 days of
21 gavage exposure to 50 mg/kg-day benzo[a]pyrene in rats (Arafaetal., 2009). In addition, a 69%
22 decrease from controls in sperm count was observed in rats following inhalation exposure to 75
23 Mg/m3benzo[a]pyrene for 60 days (Archibongetal.. 2008: Rameshetal.. 2008).
24 Both oral and inhalation exposure of rodents to benzo[a]pyrene have been shown to lead to
25 decreased epididymal sperm motility and altered morphology. Decreased motility of 20-30%
26 compared to controls was observed in C57BL6 mice (>1 mg/kg-day) and Sprague-Dawley rats
27 (0.01 mg/kg-day) (Chung etal., 2011: Mohamed etal., 2010). The effective doses spanned two
28 orders of magnitude; however, as noted above, reporting is limited in the study that observed
29 effects at 0.01 mg/kg-day benzo[a]pyrene (Chung etal., 2011). A short-term oral study in rats also
30 reported a significantly decreased number of motile sperm (~40% decrease) following 10 days of
31 gavage exposure to 50 mg/kg-day benzo[a]pyrene (Arafa et al.. 2009). In addition, decreased
32 sperm motility was observed following inhalation exposure to 75 |ig/m3 benzo[a]pyrene in rats for
33 60 days (Archibong etal.. 2008: Rameshetal.. 2008) and to >75 [ig/m3 for 10 days (Inyangetal..
34 2003). Abnormal sperm morphology was observed in Sprague-Dawley rats treated with 5 mg/kg-
35 day benzo[a]pyrene by gavage for 84 days (Chen etal., 2011] and in rats exposed to 75 |ig/m3
36 benzo[a]pyrene by inhalation for 60 days (Archibong etal., 2008: Rameshetal., 2008).
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 Testicular changes
2 Several studies have demonstrated dose-related effects on male reproductive organs in
3 adult animals exposed subchronically to benzo[a]pyrene (Table 1-5). Decreases in testicular
4 weight of approximately 35% have been observed in a 10-day gavage study in adult Swiss albino
5 rats exposed to 50 mg/kg-day benzo[a]pyrene [Arafa etal., 2009] and following subchronic
6 inhalational exposure of adult F344 rats to 75 |J.g/m3 [Archibongetal., 2008: Rameshetal., 2008].
7 No effects on testes weight were observed in Wistar rats exposed for 35 days to gavage doses up to
8 50 mg/kg-day [Kroese etal., 2001] F344 rats exposed for 90 days to dietary doses up to 100
9 mg/kg-day [Knuckles etal.. 2001]: or Sprague-Dawley rats exposed for 90 days to gavage doses up
10 to 0.1 mg/kg-day [Chung etal.. 2011]. Strain differences may have contributed to differences in
11 response, however, F344 rats exposed to benzo[a]pyrene via inhalation showed effects on
12 testicular weight [Archibong et al., 2008: Ramesh et al., 2008]. In addition, decreased testicular
13 weight has also been observed in offspring following in utero exposure to benzo[a]pyrene as
14 discussed in Section 1.1.1.
15 Histological changes in the testis have often been reported to accompany decreases in
16 testicular weight Apoptosis, as evident by increases in terminal deoxynucleotidyl transferase dUTP
17 nick end labeling (TUNEL] positive germ cells and increases in caspase-3 staining, was evident in
18 seminiferous tubules of Sprague-Dawley rats following 90 days of exposure to >0.001 and
19 0.01 mg/kg-day, respectively, benzo[a]pyrene by gavage [Chung etal., 2011]. However, the study
20 authors did not observe testicular atrophy or azospermia in any dose group. Seminiferous tubules
21 were reported to look qualitatively similar between controls and animals exposed to
22 benzo[a]pyrene by inhalation doses of 75 |ig/m3 for 60 days [Archibong et al., 2008: Ramesh etal.,
23 2008]. However, when histologically examined, statistically significantly reduced tubular lumen
24 size and length were observed in treated animals. Seminiferous tubule diameters also appeared to
25 be reduced in exposed animals, although this difference did not reach statistical significance
26 [Archibongetal., 2008: Rameshetal., 2008]. In addition, histological changes in the seminiferous
27 tubules have also been observed in offspring following in utero exposure to benzo[a]pyrene as
28 discussed in Section 1.1.1.
29 Epididymal changes
30 In addition to testicular effects, histological effects in the epididymis have been observed
31 following 90-day gavage exposure to benzo[a]pyrene [Chung etal., 2011] (Table 1-5]. Specifically,
32 statistically significant decreased epididymal tubule diameter (for caput and cauda] was observed
33 at doses >0.001 mg/kg-day. Atthe highest dose tested (0.1 mg/kg-day], diameters were reduced
34 approximately 25%. No changes in epididymis weights were observed following an 84-day
35 treatment in Sprague-Dawley rats of 5 mg/kg-day benzo[a]pyrene (Chen etal.. 2011].
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 Hormone changes
2 Several animal models have reported decreases in testosterone following both oral and
3 inhalation exposure to benzo[a]pyrene (Table 1-5). In male Sprague-Dawley rats, decreases in
4 testosterone have been observed following 90-day oral exposures [Chung etal., 2011: Zheng etal.,
5 2010]. Statistically significant decreases of 15% in intratesticular testosterone were observed at
6 5 mg/kg-day in one study [Zheng etal., 2010], while a second study in the same strain of rats
7 reported statistically significant decreases of approximately 40% in intratesticular testosterone and
8 70% in serum testosterone at 0.1 mg/kg-day [Chung et al., 2011 ]. Statistically significant decreases
9 in intratesticular testosterone (80%] and serum testosterone (60%] were also observed following
10 inhalation exposure to 75 |J.g/m3 benzo[a]pyrene in F344 rats for 60 days (Archibongetal.. 2008:
11 Ramesh et al., 2008]. Statistically significant increases in serum luteinizing hormone (LH] have also
12 been observed in Sprague-Dawley rats following gavage exposure to benzo[a]pyrene at doses of
13 >0.01 mg/kg-day (Chung etal., 2011] and in F344 rats following inhalation exposure to 75 |ig/m3
14 benzo[a]pyrene for 60 days (Archibongetal.. 2008: Ramesh etal.. 20081
15
16
Table 1-5. Evidence pertaining to the male reproductive toxicity of
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, daily gavage (FO males
only)
42 d
epididymal sperm count in FO mice
Approximate % change from control:
0, -50*, and -70*
(data reported graphically)
i, epididymal sperm motility in FO mice
Approximate % change from control:
0, -20*, and -50*
(data reported graphically)
•^ 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
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
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
Reference and Study Design
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
Results
•i, epididymal sperm motility (% change from control)
0 and -73*
•^ epididymal sperm count (% change from control)
0 and -69*
/T" % abnormal epididymal sperm
33 and 87*
•i, 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, daily gavage (FO males
only)
42 d
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
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
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
•i, 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
1" number of apoptotic germ cells per tubule (TUNEL or caspase 3
positive)
No change in testis weight or histology
/T" testicular lesions characterized as irregular arrangement of germ
cells and absence of spermatocytes (numerical data not reported)
No change in testis weight
•i, decreased testis weight (% change from control)
0 and 34*
•^ 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
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
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
60 d
Results
•^ 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*
•i, Serum testosterone (approximate % change from control ;
numerical data not reported)
0, 0, -35, and -70*
1" serum LH (approximate % change from control; numerical data not
reported)
0, 33, 67*, and 87*
•i, hCG or dbcAMP-stimulated testosterone production in Leydig cells
•^ Intratesticular testosterone (approximate % change from control;
numerical data not reported)
0, -15, and -15*
•^ intratesticular testosterone (approximate % change from control;
numerical data not reported)
0 and -80*
•i, serum testosterone (approximate % change from control)
0 and -60*
1" serum LH (approximate % change from control)
0 and 50*
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,
1-26 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1000 3
100 ~;
10 .
1
o
Q
0.1 :
0.01 ;
0.001
o y
Osl 'r-
^ c
\b cpididymal
sperm count
and motility
o >.
o "9
l—1 00
4, cpididymal
sperm count
T*in abnormal
sperm
Sperm Quality
5 "9
Jz o
u en
-i- opididymal
sperm motility
Irregular
germ cell
organization
Testicular
Effects
= T3
.E: o
u en
4/ TT (serum
and
intratesticular)
5 "9
Jz o
u en
t LH
o
r-t
O
xb TT
(intratesticular)
Hormone Changes
2 Figure 1-3. Exposure-response array for male reproductive effects following
3 oral exposure in adult animals.
4 Mode of action analysis—male reproductive effects
5 Exposure to benzo[a]pyrene in laboratory animals induces male reproductive effects
6 including decreased levels of testosterone and increased levels of LH, decreased sperm quality, and
7 histological changes in the testis. Decrements in sperm quality and decreased fertility have also
8 been demonstrated in populations highly exposed to PAH mixtures [Spares and Melo, 2008: Hsuet
9 al.. 20061
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 Numerous studies have indicated that benzo[a]pyrene reduces testosterone levels following
2 oral or inhalation exposure [Chung etal.. 2011: Zheng etal.. 2010: Archibongetal.. 2008: Ramesh et
3 al.. 2008). It is plausible that the effects on sperm quality and histological changes of the
4 reproductive organs are secondary to an insufficiency of testosterone [Inyangetal., 2003]. One
5 study has hypothesized that benzo[a]pyrene perturbs the production of testosterone by Leydig
6 cells [Chung etal., 2011]. This study found a statistically significant reduction in testicular
7 testosterone in rats treated with 0.1 mg/kg-day benzo[a]pyrene for 90 days and found that
8 testosterone production in isolated Leydig cells was also inhibited approximately 50%, even in
9 cultures stimulated with human chorionic gonadotropin and dibutryl cyclic adenosine
10 monophosphate.
11 Leydig cell function is thought to be regulated by testicular macrophages [Hales, 2002].
12 When testicular macrophages are activated and produce inflammatory mediators, Leydig cell
13 testosterone production is inhibited [Hales, 2002]. Zheng etal. [2010] treated rats with 5 mg/kg-
14 day benzo[a]pyrene for 90 days and reported a statistically significant increase in ED-1 type
15 testicular macrophages and a statistically significant decrease in intratesticular testosterone.
16 Arafaetal. [2009] reported that male reproductive effects observed following
17 benzo[a]pyrene exposure could be ameliorated by antioxidant pre-treatment This study reported
18 decreased sperm count, motility, and production, in addition to decreased testis weight following a
19 10 day oral administration in rats of 50 mg/kg-day benzo[a]pyrene. Pretreatment with the citris
20 flavonoid hesperidin protected rats from all of these effects except the decrease in sperm motility.
21 A study in tobacco smokers suggests that direct DNA damage from the reactive metabolite
22 BPDE may decrease sperm motility [Perrinetal., 201 la]. In this study, motile sperm were
23 separated from non-motile sperm using a "swim-up" self-migration technique. The investigators
24 found that the motile sperm selected by this method had significantly fewer BPDE-adducts than
25 non-selected sperm.
26 Other hypothesized modes of action of the observed male reproductive effects include
27 benzo[a]pyrene-mediated DNA damage to male germ cells leading to genotoxicity, cytotoxicity, and
28 apoptosis [Chungetal.. 2011: Perrinetal.. 2011b: Perrinetal.. 2011a: Olsenetal.. 2010: Revel et
29 al., 2001], compromised function of Sertoli cells [Raychoudhury and Kubinski, 2003], and
30 decreased embryo viability post-fertilization associated with sperm DNA damage [Borini etal..
31 2006: Seli etal.. 20041
32 Female Reproductive Effects
33 Fertility
34 In women, exposure to cigarette smoke has been shown to affect fertility, including effects
35 related to pregnancy, ovulatory disorders, and spontaneous abortion (as reviewed in Waylenetal..
36 2009: Cooper and Moley, 2008: Spares and Melo, 2008]. In addition, several studies suggest that in
37 utero exposure to maternal tobacco smoke also decreases the future fertility of female offspring [Ye
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 etal.. 2010: Jensen etal.. 1998: Weinbergetal.. 1989]. Benzo[a]pyrene levels in follicular fluid and
2 benzo[a]pyrene-DNA adducts in granulosa-lutein cells and oocytes and in human cervical cells have
3 been associated with smoking status and with amount smoked [Neal etal.. 2008: Mancini etal..
4 1999: Melikian etal.. 1999: Zenzes etal.. 1998: Shamsuddin and Can. 1988].
5 Few epidemiological studies have examined the specific influence of components of PAH
6 mixtures on fertility or other reproductive outcomes; EPA identified only two studies with specific
7 data on benzo[a]pyrene (Table 1-6]. One of these studies addressed the probability of conception
8 among women undergoing in vitro fertilization [Neal etal., 2008]. Follicular fluid benzo[a]pyrene
9 levels were significantly higher among the women who did not conceive compared with women
10 who did get pregnant No association was seen between conception and serum levels of
11 benzo[a]pyrene. The other study examined risk of delayed miscarriage (fetal death before 14 weeks
12 of gestation], using a case-control design with controls selected from women undergoing elective
13 abortion (Wu etal., 2010]. A strong association was seen between maternal blood benzo[a]pyrene-
14 DNA adduct levels and risk of miscarriage, with a fourfold increased risk for levels above compared
15 with below the median. Benzo[a]pyrene-DNA adduct levels were similar in the aborted tissue of
16 cases compared with controls.
17 Experimental studies in mice also provide evidence that benzo[a]pyrene exposure affects
18 fertility (Table 1-7). Decreased fertility and fecundity (decreased number of FO females producing
19 viable litters at parturition) was statistically significantly reduced by about 35% in adult females
20 exposed to 160 mg/kg-day of benzo[a]pyrene (Mackenzie and Angevine, 1981]. In another study,
21 FO females showed no signs of general toxicity or effects on fertility following gavage exposure to
22 10 mg/kg-day on CDs 7-16 (Kristensenetal., 1995]. Decrements in fertility were more striking in
23 the offspring from these studies, as described in Section 1.1.1.
24 Ovarian effects
25 Human epidemiological studies that directly relate ovotoxicity and benzo[a]pyrene
26 exposure are not available; however, smoking, especially during the time of the peri-menopausal
27 transition, has been shown to accelerate ovarian senescence (Midgette and Baron, 1990].
28 Benzo[a]pyrene-induced ovarian toxicity has been demonstrated in animal studies. In adult female
29 rats treated by gavage, statistically significant, dose-related decreases in ovary weight has been
30 observed in female rats treated for 60 days at doses >5 mg/kg (2.5 mg/kg-day adjusted) (Xu etal.,
31 2010). At 10 mg/kg in adult rats (5 mg/kg-day adjusted), ovary weight was decreased 15% (Xuet
32 al., 2010). Changes in ovary weight were not observed in two subchronic studies in rats.
33 Specifically, no changes in ovary weight were seen in Wistar rats exposed for 35 days to gavage
34 doses up to 50 mg/kg-day (Kroese etal., 2001] or in F344 rats exposed for 90 days to dietary doses
35 up to 100 mg/kg-day (Knuckles etal.. 2001).
36 In adult female rats treated by gavage, dose-related decreases in the number of primordial
37 follicles have been observed in female rats treated for 60 days at doses >2.5 mg/kg-day, with a
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 statistically significant decrease of approximately 20% at the high dose [Xu etal.. 2010] (Table 1-7).
2 No notable differences in other follicle populations and corpora lutea were observed. However, in
3 utero studies exposing dams to the same doses produced offspring with few or no follicles or
4 corpora lutea [Kristensenetal., 1995: Mackenzie and Angevine, 1981]. Additional support for the
5 alteration of female reproductive endpoints comes from intraperitoneal (i.p.] experiments in
6 animals and in vitro experiments. Several studies have observed ovarian effects (decreased
7 numbers of ovarian follicles and corpora lutea, absence of folliculogenesis, oocyte degeneration,
8 and decreased fertility] in rats and mice exposed via i.p. injection (Bormanetal., 2000: Miller etal.,
9 1992: Swartz and Mattison. 1985: Mattison et al.. 1980]. Further evidence is available from in vitro
10 studies showing inhibition of antral follicle development and survival, as well as decreased
11 production of estradiol, in mouse ovarian follicles cultured with benzo[a]pyrene for 13 days (Sadeu
12 and Foster, 2011]. Likewise, follicle stimulating hormone (FSH]-stimulated growth of cultured rat
13 ovarian follicles was inhibited by exposure to benzo[a]pyrene (Neal etal., 2007].
14 Hormone levels
15 Alteration of hormone levels has been observed in female rats following oral or inhalation
16 exposure to benzo[a]pyrene (Table 1-7]. Inhalation exposure to benzo[a]pyrene:carbon black
17 particles during gestation resulted in decreases in plasma progesterone, estradiol, and prolactin in
18 pregnant rats (Archibongetal., 2002]. In addition, statistically significant, dose-related decreases
19 in estradiol along with altered estrus cyclicity was observed in female rats treated for 60 days at
20 doses >2.5 mg/kg-day by gavage (Xu etal., 2010]. Mechanistic experiments have also noted
21 decreased estradiol output in murine ovarian follicles cultured with benzo[a]pyrene in vitro for
22 13 days, but did not find any decrease in progesterone (Sadeu and Foster. 2011].
23 Cervical effects
24 One subchronic animal study is available that investigated effects in the cervix following
25 oral exposure to benzo[a]pyrene (Table 1-7]. Statistically-significant dose-related increases in the
26 incidence of cervical inflammatory cells were observed in mice exposed twice a week for 98 days to
27 benzo[a]pyrene via gavage at doses >2.5 mg/kg (Gao etal.. 2011a: Gao etal.. 2010]. Cervical effects
28 of increasing severity, including epithelial hyperplasia, atypical hyperplasia, apoptosis, and
29 necrosis, were observed at higher doses. There are no data on cervical effects in other species or in
30 other mouse strains. Gao etal. (2011a] considered the hyperplasia responses to be preneoplastic
31 lesions. Cervical neoplasia was not reported in the available chronic bioassays, but this tissue was
32 not subjected to histopathology examination in either bioassay (Kroese etal., 2001: Beland and
33 Culp, 1998]. Thus, the relationship of the cervical lesions to potential development of neoplasia is
34 uncertain.
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
2
Table 1-6. Evidence pertaining to the female reproductive effects of
benzo[a]pyrene in humans
Reference and Study Design
Probability of conception
Neal et al. (2008)
36 women undergoing in vitro fertilization
(19 smokers, 7 passive smokers, and 10 non-
smokers)
Exposure: benzo[a]pyrene in serum and follicular
fluid
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)
Results
Benzo[a]pyrene levels (ng/mL)
Did not
Conceived Conceive (p-value)
Follicular fluid 0.1 1.7 (<0.001)
Serum 0.01 0.05 (not reported)
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
3
4
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
•i, number of FO females with viable litters
46/60, 21/30, 44/60, and 13/30*
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
Reference and Study Design
CDs 7-16
Kristensen et al. (1995)
NMRI mice, 9 females/dose
0 or 10 mg/kg-d by gavage
CDs 7-16
Results3
No changes in fertility of FO females
Ovarian effects (weight, histology, follicle numbers)
Xu et al. (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
Knuckles etal. (2001)
F344 rats, 20/sex/dose
0, 5, 50, or 100 mg/kg-d in diet
90 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
•i, ovary weight (% change from control)
0, -11*, and -15*
•^ number of primordial follicles (20%* decrease at high dose)
/T" increased apoptosis of ovarian granulosa cells (approximate %
apoptosis)
2, 24*, and 14*
No changes in ovary weight
No changes in ovary weight
Hormone levels
Xu et al. (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
Archibongetal. (2002)
F344 rats, 10 females/group
0, 25, 75, or 100 u.g/m3by inhalation
4 hrs/d
GDs 11-20 (serum hormones tested
at GD 15 and 17 in 0, 25, and
75 ug/m3dose groups)
•i, serum estradiol (approximate % change from control)
0, -16, and -25*
Altered estrous cyclicity
4> FO estradiol, approximately 50% decrease at 75 u.g/m3at GD 17
4> 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
Cervical effects
Gaoetal. (2011a)
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*
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
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 -,
-
re 100 -
-
aa I
M
J.
01
i/>
o
Q
10 :
-
-
1 -
• LOAEL
ANOAEL
• Doses > LOAEL
T t
o Doses < NOAEL
T ft
f
I T ^ T
• o o A A •
i
A
i
^H in
OO CTi
O C~^
.~ O ° . .
> <-> 71 LO
0 -F J vH
CC E C .
c • - o r^
•
C3 >yi
-j >-
O tj
in
•i- ovarian
follicles
Ovarian Effects
O ?3
t — 1 5~~
o y,
™ -D
G ra
3 TJ
X o
LO
sb serum
Estradiol
Hormone
Changes
o
O f-_
_T ~5
o >.
o _g
U CO
en
1s cervical
epithelial
hyperplasia
Cervical
Effects
5 Figure 1-4. Exposure-response array for female reproductive effects following
6 oral exposure in adult animals.
7 Mode-of-action analysis—female reproductive effects
8 Although the mechanisms underlying female reproductive effects following benzo[a]pyrene
9 exposure are not fully established, associations with stimulation of apoptosis, impairment of
10 steroidogenesis, and cytotoxicity have been made. Ovarian lesions in benzo[a]pyrene-exposed rats
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
1 have been associated with increased apoptosis in ovarian granulosa cells and alteration in
2 hormone-mediated regulation of folliculogenesis [Xu etal.. 2010). and results from in vitro
3 experiments provide support for an association between benzo[a]pyrene exposure and impaired
4 folliculogenesis, steroidogenesis, and oocyte maturation [Sadeu and Foster, 2011: Neal etal., 2007].
5 A growing body of research suggests that benzo[a]pyrene triggers the induction of apoptosis in
6 oocytes through AhR-driven expression of pro-apoptotic genes, including Bax [Kee etal., 2010: Neal
7 etal.. 2010: Pru etal.. 2009: Matikainen etal.. 2002: Matikainen etal.. 2001: Robles etal.. 20001.
8 Other proposed mechanisms include the impairment of folliculogenesis from reactive metabolites
9 [Takizawaetal.. 1984: Mattison and Thorgeirsson. 1979.1977] or by a decreased sensitivity to
10 FSH-stimulated follicle growth [Neal etal.. 2007]. Based on findings that an ERa antagonist
11 counteracted effects of subcutaneously administered benzo[a]pyrene on uterine weight (decreased
12 in neonatal rats and increased in immature rats], interactions with ERa have been proposed,
13 possibly via occupation of ERa binding sites or via AhR-ER-crosstalk [Kummer et al., 2 00 8: Kummer
14 etal., 2007]. However, several in vitro studies have demonstrated low affinity binding of
15 benzo[a]pyrene to the estrogen receptor and alteration of estrogen-dependent gene expression
16 [Liu etal.. 2006: van Lipzigetal.. 2005: Vondraceketal.. 2002: Fertucketal.. 2001: Charles etal..
17 2000]. so the role of the ER in benzo[a]pyrene-induced reproductive toxicity is unclear.
18 Summary of Reproductive Effects
19 Male reproductive effects
20 Exposure to benzo[a]pyrene in laboratory animals induces male reproductive effects
21 including decreased levels of testosterone and increased levels of LH, decreased sperm count and
22 motility, histological changes in the testis, and decreased reproductive success. These findings in
23 animals are supported by decrements in sperm quality and decreased fertility in human
24 populations exposed to PAH mixtures [Spares and Melo, 2008: Hsu etal., 2006]. In laboratory
25 animals, male reproductive toxicity has been observed after oral and inhalation exposure to rats or
26 mice. Effects seen after oral exposures include impaired fertility, effects on sperm parameters,
27 decreased reproductive organ weight, testicular lesions, and hormone alterations [Chen etal.. 2011:
28 Chung etal.. 2011: MohamedetaL 2010: Zheng etal.. 2010: Mackenzie andAngevine. 19811 In
29 addition to oral exposure, male reproductive effects of benzo[a]pyrene have also been observed
30 following inhalation exposure in rats [Archibongetal.. 2008: Rameshetal.. 2008: Inyangetal..
31 2003]. The male reproductive effects associated with benzo[a]pyrene exposure are considered to
32 be biologically plausible and adverse. The evidence for male reproductive toxicity seen across
33 multiple human and animal studies identifies the male reproductive system effects as a potential
34 hazard associated with exposure to benzo[a]pyrene.
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 Female reproductive effects
2 A large body of mechanistic data, both in vivo and in vitro, suggests that benzo [a]pyrene
3 impacts fertility through the disruption of folliculogenensis. This finding is supported, albeit
4 indirectly, by observations of premature ovarian senescence in women exposed to cigarette smoke
5 [Midgette and Baron. 1990]. Evidence for female reproductive toxicity of benzo[a]pyrene comes
6 from studies of human populations exposed to PAH mixtures as well as laboratory animal and in
7 vitro studies. In addition, two human studies observed associations specifically between
8 benzo[a]pyrene measures and two fertility-related endpoints: decreased ability to conceive [Neal et
9 al.. 2008: Neal etal.. 2007] and increased risk of early fetal death (i.e., before 14 weeks gestation]
10 [Wu etal.. 2010]. Studies in multiple strains of rats and mice indicate fertility-related effects
11 including decreases in ovarian follicle populations and decreased fecundity. Decreased serum
12 estradiol has also been noted in two different strains of rats exposed by oral or inhalation exposure.
13 The reproductive effects associated with benzo[a]pyrene exposure are biologically supported and
14 relevant to humans. In consideration of the evidence from human, animal, and mechanistic studies,
15 female reproductive effects are identified as a potential hazard associated with exposure to
16 benzo[a]pyrene.
17 Suscep tible Pop illations an d Lifestages
18 Epidemiological studies indicate that exposure to complex mixtures of PAHs, such as
19 through cigarette smoke, is associated with measures of decreased fertility in humans [Neal etal.,
20 2008: El-Nemr etal., 1998] and that prenatal exposure to cigarette smoking is associated with
21 reduced fertility of women later in life [Weinbergetal.. 1989]. A case-control study in a Chinese
22 population has also indicated that women with elevated levels of benzo [a]pyrene-DNA adducts in
23 maternal blood were 4 times more likely to have experienced a miscarriage [Wuetal., 2010].
24 Inhalation exposure of pregnant female rats to benzo[a]pyrene:carbon black aerosols on
25 CDs 11-20 caused decreased fetal survival and number of pups per litter associated with decreased
26 levels of plasma progesterone, estradiol, and prolactin [Archibongetal., 2002]. Decreased numbers
27 of live pups were also seen in pregnant mice following i.p. exposure to benzo[a]pyrene [Mattison et
28 al.. 1980]. These results indicate that benzo [ajpyrene exposure can decrease the ability of females
29 to maintain pregnancy.
30 Oral multigenerational studies of benzo [ajpyrene exposure in mice demonstrated effects on
31 fertility and the development of reproductive organs (decreased ovary and testes weight] in both
32 male and female offspring of pregnant mice exposed to 10-160 mg/kg-dayon CDs 7-16
33 (Kristensenetal.. 1995: Mackenzie and Angevine, 1981].
34 Reductions in female fertility associated with decreased ovary weight and follicle number
35 following gestational exposure (as discussed in Section 1.1.1] are supported by observations of:
36 (1] destruction of primordial follicles (Bormanetal.. 2000: Mattison etal.. 1980] and decreased
37 corpora lutea (Miller etal., 1992: Swartz and Mattison, 1985] in adult female mice following i.p.
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 exposure; (2) decreased ovary weight in adult female rats following oral exposure [Xuetal.. 2010]:
2 and (3) stimulation of oocyte apoptosis [Matikainenetal.. 2002: Matikainen etal.. 2001] or by a
3 decreased sensitivity to FSH-stimulated follicle growth [Neal etal.. 2007].
4 Reductions in male fertility associated with decreased testes weight following gestational
5 exposure (as discussed in Section 1.1.1] are supported by observations of: (1] decreased sperm
6 count, altered serum testosterone levels, testicular lesions, and/or increased numbers of apoptotic
7 germ cells in adult rats following repeated oral exposure to benzo[a]pyrene [Chung etal., 2011:
8 Chen etal.. 2010: Zheng etal.. 2010: Arafa etal.. 2009]: (2] decreased epididymal sperm counts in
9 adult FO and Fl generations of male mice following 6 weeks of oral exposure of the FO animals to
10 benzo[a]pyrene [Mohamedetal.. 2010]: and (3] decreased testis weight, decreased testicular or
11 plasma testosterone levels, and/or decreased sperm production, motility, and density in adult male
12 rats following repeated inhalation exposure to aerosols of benzo[a]pyrene:carbon black [Archibong
13 etal.. 2008: Ramesh etal.. 2008: Inyangetal.. 2003].
14 1.1.3. Immunotoxicity
15 Human studies evaluating immune effects following exposure to benzo[a]pyrene alone are
16 not available for any route of exposure. However, a limited number of occupational human studies,
17 particularly in coke oven workers [Zhang etal.. 2012: Wuetal.. 2003b: Winker etal.. 1997:
18 Szczekliketal., 1994], show effects on immune parameters associated with exposure to PAH
19 mixtures. These studies are of limited utility because effects associated specifically with
20 benzo[a]pyrene cannot be distinguished from other constituents of the PAH mixture. Subchronic
21 and short-term animal studies have reported immunotoxic effects of benzo[a]pyrene by multiple
22 routes of exposure (Table 1-8]. Effects include changes in thymus weight and histology, decreased
23 B cell percentages and other alterations in the spleen, and immune suppression. Data obtained
24 from subchronic oral gavage studies are supported by short-term, i.p., intratracheal, and
25 subcutaneous (s.c.] studies. Additionally, there is evidence in animals for effects of benzo[a]pyrene
26 on the developing immune system. No studies were located that examined immune system
27 endpoints following inhalation exposure of animals to benzo[a]pyrene.
28 Thymus Effects
29 Decreased thymus weights (up to 62% compared to controls] were observed in male and
30 female Wistar rats exposed by gavage to 10-90 mg/kg-day benzo[a]pyrene for 35 or 90 days
31 (Kroese etal.. 2001: De long etal., 1999]. This effect may be due to thymic atrophy. The incidence
32 of slight thymic atrophy was increased in males (6/10] and females (3/10] at a dose of 30 mg/kg-
33 day in a 90-day study, although there was no evidence of atrophy at any lower dose (Kroese etal.,
34 2001]. Additionally, at the highest dose tested (90 mg/kg-day] in one of the 35-day studies, the
35 relative cortex surface area of the thymus and thymic medullar weight were significantly reduced
36 (De long etal.. 1999]. Other histopathological changes in the thymus (increased incidence of brown
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 pigmentation of red pulp; hemosiderin) were observed in Wistar rats of both sexes at 5 0 mg/kg-
2 day in a 35-day study; however, this tissue was not examined in intermediate-dose groups [Kroese
3 etal.. 2001). Consistent with the effects observed in these studies, decreased thymus weights and
4 reduced thymic cellularity were observed in i.p. injection studies that exposed mice to doses
5 ranging from 50 to 150 mg/kg in utero [Holladay and Smith, 1995,1994: Urso and Tohnson, 1988].
6 Spleen Effects
7 Reduced splenic cellularity indicated by decreased relative and absolute number of B cells
8 in the spleen (decreased up to 41 and 61% compared to controls, respectively) and decreased
9 absolute number of splenic cells (31% decrease at the highest dose) was observed in a subchronic
10 study in male Wistar rats administered 3-90 mg/kg-day benzo[a]pyrene by gavage for 35 days (De
11 long etal., 1999). While the effect on the relative number of B cells was dose-related, the lower
12 doses did not affect the number of B cells or the absolute splenic cell number. The reduced splenic
13 cell count at the highest dose was attributed by the study authors to the decreased B cells, and
14 suggests a possible selective toxicity of benzo[a]pyrene to B cell precursors in the bone marrow.
15 The spleen effects observed in De long etal. (1999).
16 are supported by observations of reduced spleen cellularity and decreased spleen weights
17 following i.p. injection or in utero benzo[a]pyrene exposure to doses ranging from 50 to 150 mg/kg
18 (Holladay and Smith. 1995: Urso etal.. 1988).
19 In addition to physical effects on the spleen, several studies have demonstrated functional
20 suppression of the spleen following benzo[a]pyrene exposure. Dose-related decreases in sheep red
21 blood cell (SRBC) specific serum IgM levels after SRBC challenge were reported in rats (10 or
22 40 mg/kg-day) and mice (5, 20,or 40 mg/kg-day) following s.c. injection of benzo[a]pyrene for
23 14 days (Temple etal., 1993). Similarly, reduced spleen cell responses, including decreased
24 numbers of plaque forming cells and reduced splenic phagocytosis to SRBC and lipopolysaccharide
25 challenge, were observed in B6C3Fi mice exposed to doses >40 mg/kg-day benzo[a]pyrene by i.p.
26 or s.c. injection for 4-14 days (Lyte and Bick. 1985: Dean etal.. 1983: Munson and White. 1983] or
27 by intratracheal instillation for 7 days (Schnizleinetal., 1987).
28 Immun oglob ulin Alterations
29 Alterations in immunoglobulin levels have been associated with exposure to PAH mixtures
30 in a limited number of human studies. Some occupational studies have reported evidence of
31 immunosuppression following PAH exposure. For example, reductions in serum IgM and/or IgA
32 titers were reported in coke oven workers (Wu et al., 2003b: Szczeklik et al., 1994]. Conversely,
33 immunostimulation of immunoglobulin levels has also been observed in humans, specifically
34 elevated IgG fKarakaya etal.. 19991 and elevated IgE fWuetal.. 2003bl following occupational PAH
35 exposure.
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 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 et al.. 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 B6C3Fimice 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 40.9 ±
27 28.4% that of controls) exposed to 90 mg/kg-day by gavage for 35 days (De long etal., 1999):
28 however, splenic natural killer cell activity was not affected in B6C3Fi mice after s.c. injection of
29 40mg/kg-day benzo[a]pyrene for 14 days (Munson etal.. 1985). The magnitude of the dose and
30 duration of the exposure may account for the discrepancy between these two studies. Single i.p.
31 injections of 50 mg/kg benzo[a]pyrene decreased pro- and/or pre-B-lymphocytes and neutrophils
32 in the bone marrow of C57BL/6J mice without affecting the numbers of immature and mature B-
33 lymphocytes or GR-1+ myeloid cells (Galvanetal., 2006).
34 In contrast to studies that have shown immunosuppression, benzo[a]pyrene may also
35 induce sensitization responses. Epicutaneous abdominal application of 100 [igbenzo[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.
1-38 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, CD8) 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 andTohnson. 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 etal.. 1990: Silverstone etal.. 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 ConA and a weak response compared to controls in an allegeneic mixed lymphocyte
31 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[a]pyrene by subcutaneous injection for 14 days (Matiasovic etal.. 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.
1-39 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):
Males (% change from control):
1" slight thymic atrophy
Females (incidence):
Males (incidence):
4> thymus weight
% change from control:
4> thymus weight
Females (% change from control):
Males (% change from control):
0, -3, -6, and -28*
0, 0, -13, and -29*
0/10, 0/10, 0/10, and 3/10
0/10, 2/10, 1/10, and 6/10*
0, -9, -15*, -25*, and -62*
0, 13, 8, -3, and -17*
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
4, relative number (%) of B cells in spleen
% change from control: 0, -8, -13*, -18*, and -41*
•^ 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:
4, serum IgA
% change from control:
0, -13, -14, -33*, and -19
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,
1-40 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 nn
1UU
-
'
-
10 -
TO
TJ
ao
OO
_£
t/>
O
o
1
±
• LOAEL
ANOAEL • • 1
• Doses > LOAEL
o Doses < NOAEL
• • • A • • L
A •
A A • : ° A • <
o A • 1 •
o o A o 6 A <
i •
\ A 1
> o L
) O C
• •
1 A
\ 0
) O
o o
,n vo ^ vo vo vo
o 75 ° 75 §75 o 75 ° 75
OF S E -IF OF S F
- ~- _T >• 75 -^ ^ ^ — r ~~
I^-O T' 2 o3 I^^J CJ-2
s-^>- O^3 O^5 ^l~^ O^
Or-1 Ow — ^^J CJm O
2-3 Jo cij, o-S 2j,
s^o O Qro ^LTt ^
CTi ro
•vHhymus Weight
^ ^
o "^ ° "^
o E S E
_r T: -^ .g-
1^ TJ 0-2
O t/i ^2"
^n >- O
2 T3 ^ O
cn
'tSlightThvmic
Atrophy
Thymus Effects
o ^
CTi H3
^H £
75 -g"
0 5
Qfi '-^
I I"
O u^
Q ro
si/ in
relative
number
(%)of B
cells in
spleen
^ ^
CPi — CTi -^
SI S E
75 T3~ "^ -o"
O £ ^J ^
oc ^ oc '^
o 13 o 13
— , ^D — , T3
o i_n ° Ln
Q ro Q ro
sUin total Change in
number of total cell
B cells in number in
spleen the spleen
Spleen Effects
ci "> cn ">
m ^ CT ^
rt F -i_ F
75 ^ 75 ^
o 3 o 5
1-3 1-3
O uS O uS
Q ro Q ro
-sU Serum sb Serum
IgM IgA
Immunoglobultn
Alterations
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 Postlindetal., 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 [Hardinetal., 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,
1-41 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 mice [Hardinetal.. 1992]. Addition of the AhR antagonist and CYP450 inhibitor, a-napthaflavone,
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 Summary of Immune 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 oral gavage studies are supported by a larger database of in vivo studies
20 of 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: Blantonetal.. 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. Overall, the weight of evidence in animals indicates that
29 immunotoxicity may be a hazard associated with benzo[a]pyrene exposure.
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 Suscep tible Pop ulations an d Lifestages
2 The severity and persistence of immune effects observed during in utero studies suggests
3 that immunotoxicity may be greater during gestation than adulthood [Dietert and Piepenbrink.
4 2006: Holladay and Smialowicz, 2000: Urso and Gengozian, 1982]. Urso and Gengozian [1982]
5 provide experimental support demonstrating that immunosuppression from benzo[a]pyrene
6 exposure during gestation was greater than for mice exposed after birth to a 2 5-fold higher dose.
7 There is also substantial literature indicating that disruption of the immune system during certain
8 critical periods of development (e.g., initiation of hematopoiesis, migration of stem cells, expansion
9 of progenitor cells] may have significant and lasting impacts on lifetime immune function (e.g.,
10 Burns-Naasetal.,2008: Dietert. 2008: Landreth. 2002: Dietert etal.. 2000]. In addition, chemical-
11 specific studies show increased dose sensitivity and disease persistence from developmental versus
12 adult chemical exposure (reviewed in Luebke etal., 2006].
13 Thymus toxicity is a sensitive and specific effect of benzo[a]pyrene and has been observed
14 in both prenatal and adult exposure studies. The thymus serves as a major site of thymocyte
15 proliferation and selection for maturation, and impairment can lead to cell-mediated immune
16 suppression fKuper etal.. 2002: De Waal etal.. 1997: Kuper etal.. 19921 The thymus is believed to
17 be critical for T lymphocyte production during early life and not in adulthood (Hakim etal.. 2005:
18 Schonlandetal., 2003: Petrie, 2002: Mackalletal., 1995]. Therefore, the decreases in thymus
19 weight observed in studies of adult animals exposed to benzo[a]pyrene suggest that
20 immunosuppression may be a heightened concern for individuals developmentally exposed to
21 benzo[a]pyrene.
22 1.1.4. Other Toxicity
23 There is some evidence thatbenzo[a]pyrene can produce effects in the forestomach, liver,
24 kidney, and cardiovascular system, as well as alter hematological parameters. However, there is
25 less evidence for these effects compared to organ systems described earlier in Section 1.1.1 to 1.1.3.
26 Forestomach Toxicity
27 Lesions have been observed in the forestomach following subchronic and chronic oral
28 exposure to benzo[a]pyrene (Table 1-10). Increases in the incidence of forestomach hyperplasia
29 have been observed in Wistar rats following shorter-term, subchronic, and chronic gavage exposure
30 (Kroese etal.. 2001: De long et al., 1999] and in B6C3Fi mice following chronic dietary exposure
31 (Beland and Gulp. 1998: Gulp etal.. 1998].
32 Following chronic gavage exposure, increased incidences of forestomach hyperplasia were
33 observed in male and female rats at 3 and 10 mg/kg-day; at the highest dose, a lower incidence of
34 hyperplasia was reported (Kroese etal.. 2001). However, only the highest-level lesion (hyperplasia,
35 papilloma, or carcinoma) observed in each organ was scored, such that hyperplasia observed in the
36 forestomach, in which tumors were also observed, was not scored. The majority of animals in the
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 high-dose group exhibited forestomach tumors; therefore, the hyperplasia was not scored and the
2 incidence of forestomach hyperplasia in the study is more uncertain at the highest dose. Shorter-
3 term studies [Kroese etal.. 2001: De long et al.. 1999] showed dose-related increases in
4 forestomach hyperplasia at doses >10 mg/kg-day in Wistar rats. In addition, following chronic
5 dietary exposure, a dose-dependent increase in the incidence of forestomach hyperplasia and
6 hyperkeratosis was observed in female mice at >0.7 mg/kg-day [Beland and Gulp, 1998: Gulp etal.,
7 1998]. Forestomach tumors were also observed at >0.7 mg/kg-day by Beland and Gulp [1998] and
8 Gulp etal. [1998].
9 Although humans do not have a forestomach, forestomach effects observed in rodents are
10 believed to be supportive of a human hazard as humans have similar squamous epithelial tissue in
11 their oral cavity [IARC, 2003: Wester and Kroes, 1988]. Mechanistic investigations suggest that
12 bioactivation of benzo[a]pyrene leads to reactive intermediates that can lead to mutagenic events,
13 as well as to cytotoxic and apoptotic events. The available human, animal, and in vitro evidence
14 best supports a mutagenic mode of action as the primary mode by which benzo[a]pyrene induces
15 carcinogenesis. Available data indicate that forestomach hyperplasia may be a histological
16 precursor to neoplasia observed at this site after chronic exposure to benzo[a]pyrene [Kroese etal..
17 2001: De long et al.. 1999]. Dose-response data show that forestomach hyperplasia occurs at
18 shorter durations and at lower doses than tumors in rats and mice exposed to benzo[a]pyrene for
19 up to 2 years [Kroese etal.. 2001: Beland and Gulp. 1998]. Kroese etal. [2001] reported that the
20 forestomach lesions demonstrated a progression over the course of intercurrent sacrifices; the
21 authors described early lesions as focal or confluent basal hyperplasia, followed by more advanced
22 hyperplasia with squamous cell papilloma, culminating in squamous cell carcinoma. The
23 description of the progression of forestomach lesions provided by Kroese et al. [2001]. coupled
24 with the observation that hyperplasia occurs before tumors and at lower doses than tumors,
25 suggests that forestomach hyperplasia induced by benzo[a]pyrene is likely a preneoplastic lesion.
26 Hematological Toxicity
27 Altered hematological parameters, including decreases in red blood cell [RBC] count,
28 hemoglobin, and hematocrit have been observed in laboratory animals following benzo[a]pyrene
29 exposure (Table 1-9. Statistically significant decreases in RBC count, hemoglobin, and hematocrit
30 were observed in male Wistar rats at doses >10 mg/kg-day for 35 days [De long etal., 1999]. A
31 minimal, but statistically significant increase in mean cell volume and a decrease in mean cell
32 hemoglobin were observed at the highest dose (90 mg/kg-day], which may indicate dose-related
33 toxicity for the RBCs and/or RBC precursors in the bone marrow [De long etal., 1999]. Similarly,
34 male and female F344 rats also showed maximal decreases in RBC counts, hematocrit, and
35 hemoglobin levels between 10-12% in a 90-day dietary study [Knuckles etal.. 2001]. Findings
36 were significant for RBC counts and hematocrit in males at > 50 mg/kg-day, while decreased RBC
37 counts and hematocrit in females and hemoglobin levels in both sexes were only significant in the
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
1 100 mg/kg-day group [Knuckles etal.. 2001]. Small, but not statistically significant, decreases in
2 RBC counts and hemoglobin were observed in both 35- and 90-day studies in Wistar rats [Kroese et
3 al.. 2001). It should be noted that when observed, the magnitudes of the decreases in RBCs,
4 hemoglobin, and hematocrit were generally small; about 18% at 90 mg/kg-day and <10% at lower
5 doses [Belong etal., 1999] and about 10% in F344 rats [Knuckles etal., 2001]. A decrease in white
6 blood cells (WBCs], attributed to reduced numbers of lymphocytes and eosinophils, was also
7 observed at 90 mg/kg-day following gavage exposure for 35 days [De long etal., 1999]. The mode
8 of action by which benzo[a]pyrene exposure may lead to altered hematological parameters is
9 undetermined.
10 Liver Toxicity
11 Liver effects other than cancer associated with benzo[a]pyrene exposure primarily include
12 changes in liver weight and abnormal histopathology (Table 1-10]. Increased liver weight was
13 reported in a 90-day study in both male and female Wistar rats given benzo[a]pyrene by gavage
14 [Kroese etal.. 2001]. Both females (17% increase] and males (29% increase] demonstrated
15 statistically significant increased liver weights at the highest dose tested (30 mg/kg-day]; a
16 statistically significant increase (15%] was also reported in males at 10 mg/kg-day. Similar to the
17 findings in the 90-day study by Kroese etal. (2001]. increased liver:body weight ratios were
18 observed at the highest dose in a 90-day dietary study in male F344 rats, although there was no
19 change observed in female liver weights (Knuckles etal., 2001]. Increased liver:body weight ratios
20 were also observed in both sexes at high doses (600 and 1,000 mg/kg] in an accompanying acute
21 study (Knuckles etal.. 2001]. A statistically significant increase in liver weight was also observed in
22 male Wistar rats given 90 mg/kg-day benzo[a]pyrene by gavage for 35 days (De long et al.. 1999].
23 Consistent with the findings by De long et al. (1999], a statistically significant increased liver weight
24 (about 18%] was also observed in both male and female Wistar rats at the highest dose (50 mg/kg-
25 day] given by gavage in a 35-day study (Kroese etal., 2001].
26 Limited exposure-related differences in clinical chemistry parameters associated with liver
27 toxicity were observed; no differences in alanine aminotransferase or serum aspartate
28 transaminase levels were observed, and a small dose-related decrease in y-glutamyl transferase
29 was observed in males only exposed to benzo[a]pyrene for 90 days (Kroese etal.. 2001].
30 Treatment-related lesions in the liver (oval cell hyperplasia] were identified as statistically
31 significantly increased following exposure to 90 mg/kg-day benzo[a]pyrene for 35 days; however,
32 incidence data were not reported (De Jong etal., 1999]. A 2-year carcinogenicity study (Kroese et
33 al., 2001] observed some histopathological changes in the liver; however, organs with tumors were
34 not evaluated. Since many of the animals in the highest two doses developed liver tumors, the dose
35 responsiveness of the histological changes is unclear.
36 A dose-dependent increase in liver microsomal ethoxyresorufin-o-deethylase (EROD]
37 activity, indicative of CYP1A1 induction, was observed in both sexes at doses >1.5 mg/kg-day in a
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 3 5-day study [Kroese etal.. 2001]. However, at the highest dose tested, with the greatest fold
2 induction in EROD activity, there was no evidence of associated adverse histopathologic findings.
3 The finding of increased liver weight across multiple studies of varying exposure durations, as well
4 as histopathological changes in the liver provide evidence of the liver as a target of benzo[a]pyrene-
5 induced toxicity. The mode of action by which benzo[a]pyrene induces these effects is unknown.
6 Kidney Toxicity
7 There is minimal evidence of kidney toxicity following exposure to benzo[a]pyrene
8 (Table 1-9). Statistically significant decreases in kidney weight were observed at doses of 3, 30,
9 and 90 mg/kg-day, but not at 10 mg/kg-day, in a 35-day gavage study in male Wistar rats [De long
10 etal.. 1999). In a 35-day gavage study with a similar dose range in male and female Wistar rats, no
11 statistically significant changes in kidney weights were observed at any dose. [Kroese etal., 2001].
12 Histopathological analysis of kidney lesions revealed an apparent dose-responsive increase in the
13 incidence of abnormal tubular casts in the kidney in male F344 rats exposed by diet for 90 days
14 [Knuckles etal.. 2001]. The casts were described as molds of distal nephrons lumen and were
15 considered by the study authors to be indicative of renal dysfunction. However, the statistical
16 significance of the kidney lesions is unclear. Several gaps and inconsistencies in the reporting make
17 interpretation of the kidney effects difficult, including: (1] no reporting of numerical data; (2] no
18 indication of statistical significance in the accompanying figure for kidney lesions; (3] discrepancies
19 between the apparent incidences and sample sizes per dose group; and (4] uncertainty in how
20 statistical analysis of histopathological data was applied. As such, the significance of the abnormal
21 tubular casts is unclear. While there are some findings to suggest that the kidneys may be affected
22 by benzo[a]pyrene exposure, the results are inconsistent, and there are insufficient data to suggest
23 that the kidneys may be a primary target of benzo[a]pyrene-induced toxicity.
24 Cardiovascular Toxicity
25 Atherosclerotic vascular disease and increased risk of cardiovascular mortality have been
26 associated with cigarette smoking [Ramos and Moorthy, 2005: Miller and Ramos, 2001: Thirman et
27 al.. 1994] and, to a more limited degree, occupational exposure to PAH mixtures [Friesenetal..
28 2010: Friesenetal.. 2009: Burstynetal.. 2005: Chauetal.. 1993]. Elevated mortality due to
29 cardiovascular disease was observed in a PAH-exposed occupational population (coke oven plant
30 workers], but elevated cardiovascular mortality was also observed in the non-exposed or slightly
31 exposed populations (Chau etal., 1993]. Elevated risks of ischemic heart disease (IHD] were
32 associated with past cumulative benzo[a]pyrene exposure among aluminum smelter workers (with
33 a 5-year lag], although the trend was not statistically significant; there was no observed association
34 with more recent benzo[a]pyrene exposure (Friesenetal.. 2010]. Elevated risk of mortality from
35 IHD was also associated with cumulative benzo[a]pyrene exposure in a cohort of male asphalt
36 workers (although not statistically significant]; the trend in average benzo[a]pyrene exposure and
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 association with IHD was statistically significant, with an approximately 60% increase in risk
2 between the lowest and highest exposure groups [Burstyn et al.. 2005). The two studies that
3 associate benzo[a]pyrene exposure with cardiovascular effects [Friesenetal.. 2010: Burstyn etal..
4 2005] rely on statistical models to create exposure groups rather than direct measurement of the
5 cohort under examination. Additionally, while these studies used benzo[a]pyrene exposure
6 groupings for analysis, they cannot address co-exposures that may have occurred in the
7 occupational setting (asphalt or aluminum smelters) or exposures that occurred outside the
8 workplace.
9 Increased systolic and diastolic blood pressure has been observed in the offspring of dams
10 exposed to increasing concentrations of benzo[a]pyrene (Tules etal.. 2012] (Table 1-1]. At the
11 highest dose tested (1.2 mg/kg body weight by gavage to the dams], systolic pressures were
12 elevated approximately 50% and diastolic pressures were elevated approximately 80% above
13 controls. An intranasal exposure of 0.01 mg/kg-day benzo[a]pyrene in adult male rats also
14 produced an increase in blood pressure following a 7 day exposure (Centner and Weber, 2011].
15 Reduced endothelial integrity and increased smooth muscle cell mass, both related to
16 atherosclerosis, have been observed in Sprague-Dawley rats exposed to 10 mg/kg benzo[a]pyrene
17 by i.p. injection (once/week for 8 weeks] (Zhang and Ramos. 1997]. The molecular mechanisms
18 underlying PAH-induced vascular injury and the development of atherosclerosis are not well
19 established, but current hypotheses include cell proliferative responses to injury of endothelial cells
20 from reactive metabolites (including reactive oxygen species [ROS]] and genomic alterations in
21 smooth muscle cells from reactive metabolites leading to transformed vasculature cells and
22 eventual plaque formation (Ramos and Moorthy, 2005]. However, while the link between PAHs
23 and atherosclerotic disease has been studied, experiments specifically looking at the relationship
24 between levels of exposure to benzo[a]pyrene (via environmentally relevant routes] and the
25 development of aortic wall lesions related to atherosclerosis have not generally been performed.
26 One exception to this observation comes from a series of experiments on Apolipoprotein E
27 knock-out (ApoE -/-] mice exposed orally to benzo[a]pyrene. ApoE -/- mice develop spontaneous
28 atherosclerosis, which is thought to be due to enhanced oxidative stress from the lack of ApoE
29 (Godschalk et al., 2003]. Overall, these studies suggest that benzo[a]pyrene exposure in ApoE-/-
30 mice enhances the progression of atherosclerosis through a general local inflammatory process.
31 Neurological Toxicity
32 Impaired learning and memory, as well as neurochemical alterations, have been observed in
33 humans following occupational exposure to PAH mixtures (Niuetal., 2010]. Male coke oven
34 workers were analyzed for alterations in neurobehavioral function using the World Health
35 Organization Neurobehavioral Core Test Battery (WHO-NCTB], as well as changes in
36 neurotransmitter concentrations in blood. Urinary levels of the PAH metabolite, 1-hydroxypyrene,
37 were used as markers of PAH exposure. In the WHO-NCTB, coke workers had lower scores in the
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 digit span and forward digit span tests than matched control subjects, suggesting that short-term
2 memory was impaired. The authors also reported that the digit span and forward digit span scores
3 significantly decreased with increasing 1-hydroxypyrene levels in urine. PAH exposure also altered
4 the blood levels of several neurotransmitters. As in the functional assays, the authors reported that
5 alterations in neurochemical measures were associated with urinary levels of 1-hydroxypyrene.
6 Alterations in neuromuscular, autonomic, sensorimotor, aggression, and
7 electrophysiological endpoints have been reported in rats and mice following acute or short-term
8 exposure to benzo[a]pyrene [Bouayed et al., 2009b: Grovaetal., 2008: Grovaetal., 2007: Saunders
9 etal.. 2006: LiuetaL 2002: Saunders etal.. 2002: Saunders etal.. 20011 Impaired Morris water
10 maze performance was observed following subchronic oral gavage in adult rats [Chen etal.. 2011:
11 Chengzhietal., 2011] and following short-term i.p. exposure in adult mice [Grovaetal., 2007]:
12 however, the former study was conducted with only a single dose group, while the latter did not
13 evaluate possible changes in locomotion and reported unexplained decreases in escape latency on
14 trial day 1 following benzo[a]pyrene exposure. Decreased anxiety-like behavior in hole board and
15 elevated plus maze tests has been observed following short-term i.p. exposure [Grova etal., 2008].
16 In addition, a 28-day gavage study in male mice observed an increase in aggressive behavior (as
17 measured by the resident intruder test] and an increase in consummatory sexual behavior in mice
18 treated with 0.02 mg/kg-day [Bouayed etal., 2009b]. These data suggest that benzo[a]pyrene
19 exposure could be neurotoxic in adults; however, only limited data are available to inform the
20 neurotoxic potential of repeated subchronic or chronic exposure to benzo[a]pyrene via the oral
21 route (Table 1-9].
22
23
Table 1-9. Evidence pertaining to other toxicities of benzo[a]pyrene in
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 bw 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*
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
Reference and Study Design
Results3
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
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 for35 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)
4, 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)
4, 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)
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
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
Results3
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*
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
/T" 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
/T" abnormal tubular casts
Females: not statistically significant (numerical data not
reported)
Males: apparent dose-dependent increase (numerical data
not reported)
•^ kidney weight
% change from control: 0, -11*, -4, -10*, and -18*
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
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
Results3
kidney weight: no change (data not reported)
Neurological toxicity
Chengzhietal. (2011)
Sprague-Dawley rats, male, 32/dose
0 or 2 mg/kg-d by gavage
90 d
Bouayedetal. (2009b)
Swiss albino mice, male, 9/group
0, 0.02, or 0.2 mg/kg by gavage
28 d
1" time required for treated rats to locate platform in
water maze (data reported graphically)
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 Evidence in Humans
12 There are many epidemiologic studies involving exposure to PAH mixtures that contain
13 benzo[a]pyrene (e.g., studies of coke oven workers, asphalt workers). This discussion primarily
14 focuses on epidemiologic studies that included a direct measure of benzo[a]pyrene exposure. All
15 identified studies have co-exposures to other PAHs. The identified studies were separated into
16 tiers according to the extent 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
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
1 studies addressed in the analysis the potential for confounding by smoking [Armstrong and Gibbs,
2 2009: Spinellietal..20Q6: Xuetal.. 1996] (Table 1-11). These three studies represent different
3 geographic locations and two different industries. The pattern of results in the Tier 2 studies was
4 mixed, as would be expected for studies with less precise exposure assessments or smaller sample
5 sizes: one of the standardized mortality ratio (SMR) estimates was <1.0, with the other eight
6 estimates ranging from 1.2 to 2.9 (Table 1-12). In considering all of the available studies,
7 particularly those with the strongest methodology, there is considerable support for an association
8 between benzo[a]pyrene exposure and lung cancer, although the relative contributions of
9 benzo[a]pyrene and of other PAHs cannot be established.
10 For bladder cancer, the cohort and nested case-control studies observed a much smaller
11 number of cases compared with lung cancer; this limits their ability to examine exposure-response
12 relationships. Three cohort studies with detailed exposure data, however, identified 48 - 90 cases
13 (Burstynetal.. 2007: Gibbs and Sevigny. 2007a: Gibbs etal.. 2007: Gibbs and Sevigny. 2007b]
14 (Spinelli etal., 2006] (Tier 1 studies, Table 1-13). Although cumulative exposure (up to
15 approximately 2 [ig/m3-years) was not related to increasing risk in the study of asphalt workers by
16 Burstynetal. (2007). an exposure-response was seen with the wider exposure range (i.e., >80
17 [ig/m3-years) examined in two studies of aluminum smelter workers by Gibbs and Sevigny (2007a):
18 Gibbs etal. (2007): Gibbs and Sevigny (2007b) and (Spinelli etal.. 2006). This difference in
19 response is not surprising, given that the highest exposure group in the asphalt worker studies
20 corresponded to the exposures seen in the lowest exposure categories in the studies of aluminum
21 smelter workers. The five studies with more limited exposure information or analyses each
22 included between 2 and 16 bladder cancer cases, with relative risk estimates ranging from 0.6 to
23 2.9. None of these individual effect estimates was statistically significant (Table 1-13).
24 Two of the identified studies contained information on risk of mortality from melanoma.
25 Neither of these studies observed increased risks of this type of cancer, with an SMR of 0.91 (95%
26 confidence interval [CI] 0.26, 2.48) [22 cases) in (Spinelli etal.. 2006) and 0.58 (95% CI 0.12,1.7) in
27 Gibbs et al. (2007][3 cases]. Of additional interest is non-melanoma skin cancer, particularly with
28 respect to dermal exposures. The literature pertaining to this kind of cancer and PAH exposure
29 goes back to the 18th century work of Sir Percival Pott describing scrotal cancer, a squamous cell
30 skin cancer, in chimney sweeps (Brown and Thornton. 1957). One of the identified studies
31 reported an increased risk of mortality from non-melanoma skin cancer among asphalt workers
32 (roofers), with an SMR of 4.0 (95% CI: 1.0,10.9) among workers with >20 years (Hammond etal..
33 1976}. In addition to this study, three studies in Scandinavian countries examined non-melanoma
34 skin cancer risk in relation to occupations with likely dermal exposure to creosote (i.e., timber
35 workers, brickmakers, and power linesmen) using incidence data from population registries
36 (Pukkala. 1995: Karlehagenetal.. 1992: Tornqvist etal.. 1986).
37 The standardized incidence ratio (SIR) estimates were 1.5 (95% CI: 0.7, 2.6) based on five
38 exposed cases, 2.37 (95% CI: 1.08, 4.50) based on nine cases, and 4.64 (95% CI: 1.51,10.8) based
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 on five cases, respectively, in Tornqvist et al. [1986]. Karlehagen et al. [1992]. and Pukkala [1995].
2 These studies provide support for the association between dermal PAH exposure, including
3 benzo[a]pyrene exposure, and skin cancer.
4 In addition to cohorts of workers occupationally exposed to PAH mixtures, populations
5 exposed to benzo[a]pyrene through topical coal tar formulations for the treatment of psoriasis,
6 eczema, and dermatitis have also been studied. In the majority of studies with greater than 20
7 years of follow-up, coal tar treatment was not significantly associated with skin cancer [Roelofzen
8 etal., 2010: Pittelkowetal., 1981: Maughanetal., 1980]. However, in populations of patients with
9 co-exposure to psoralen and ultraviolet-A light therapy (later determined to be carcinogenic], high
10 exposure to coal tar treatments was associated with an increased risk of non-melanoma skin cancer
11 [Stern etal.. 1998: Stern etal.. 1980].
12 Lung, bladder, and skin cancers are the cancers that have been observed in occupational
13 studies of PAH mixtures [Benbrahim-Tallaaetal.. 2012: Baan etal.. 2009: Secretan etal.. 2009]..
14 The reproducibility of lung, bladder, and skin cancers in different populations and exposure
15 settings after occupational exposure to PAH mixtures (see Table 1-10] adds plausibility to the
16 hypothesis that common etiologic factors may be operating. The potential role that benzo[a]pyrene
17 may play as a causal agent is further supported by the observation that these same sites are also
18 increased in the studies that included a direct measure of benzo[a]pyrene.
19 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 biomass fuel (primarily
wood)
Indoor emissions from household
combustion of coal
Sites with Sufficient
Evidence in Humans
Lung
Urinary bladder
Lung
Skin
Lung
Lung
Lung
Lung
Sites with Limited
Evidence in Humans
Lung
Urinary bladder
Skin
Urinary bladder
Lung
Reference
Baan etal. (2009)
IARC (2010)
Baan etal. (2009)
Baan etal. (2009)
Baan etal. (2009)
Baan etal. (2009)
IARC (2010)
Benbrahim-Tallaa et al.
(2012)
Secretan et al. (2009)
Secretan et al. (2009)
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
PAH-Related Mixture or
Occupation
Mineral oils, untreated or mildly
treated
Shale oils
Soot (chimney sweeping)
Sites with Sufficient
Evidence in Humans
Skin
Skin
Lung
Skin
Sites with Limited
Evidence in Humans
Urinary bladder
Reference
Baanetal. (2009)
Baanetal. (2009)
Baanetal. (2009)
1
2
3
4
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.
(1994)
SMR 1.32 (1.22, 1.42) [677 cases]
Lung cancer risk by cumulative benzo[a]pyrene exposure
Median
benzo[a]-
pyrene n
Hg/m3-years cases SMR (95% Cl)
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)
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)
RR = relative risk
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-years (0.0035 per
u.g/m3-years increase); other shapes of exposure-
response curve examined.
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 ~24yrs)
Smoking information from self-administered
questionnaire
Exposure: Job exposure matrix using 1,275 personal
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
ug/m3-years
n cases
RR (95% Cl)a
0-0.5
0.5-20
25
42
1.0 (referent)
1.23 (0.74,2.03)
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
benzo[a]pyrene measures from 1977 to 2000 (69%
for compliance monitoring)
Related references: Friesen et al. (2006) (exposure
data); Spinellietal. (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
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)
a Adjusting for smoking category; trend p < 0.001
Lung cancer risk by cumulative benzo[a]pyrene exposure
Benzo[a]-pyrene
(u.g/m3-years) 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 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)
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
0
>0-0.41
0.41-10.9
>10.9
aPoisson regression
n cases RR
20 1.0
6 0.7
6 1.4
7 1.7
, adjusting for smoking; trend
(95% Cl)a
(referent)
(0.3, 1.8)
(0.6, 3.5)
(0.7, 4.2)
p = 0.22.
Proxy measure
Olsson etal. (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;
Lung cancer risk by
Coal tar
unit-yrsa
0.39-4.29
4.30-9.42
9.43-16.88
cumulative coal tar exposure3
n
cases RR
43 1.31
32 0.98
30 0.97
(95% Cl)
(0.87, 2.0)
(0.62, 1.6)
(0.61, 1.6)
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
Reference and Study Design
Results
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 et
al. (2000)
16.89-196.48
(trend p-value)
54
1.60
(1.09, 2.4)
(0.07)
Adjusting for set, age, country, tobacco pack-years.
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)
SMR 1.95 (1.59, 2.33) [255 cases]
Lung cancer risk by cumulative exposure
Coal tar pitch
volatiles
(mg/m3-mo)
n
cases
RR (95%CI)a
0
1-49
50-199
200-349
350-499
500-649
>650
203
34
43
59
39
27
56
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)
Adjusting for age, race, coke plant, period of follow-up;
trend p< 0.001.
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
Mean
benzo[a]-
pyrene
Mg/m3
n cases SMR (95% Cl)a
None - 13
Low - 6
Moderate 0.30 5
High 1.19 26
'Calculated by EPA from data in paper
1.49 (0.83,2.5)
1.19 (0.48,2.5)
1.52 (0.55,3.4)
4.30 (2.9,6.2)
Bergerand Manz (1992) (Germany)
SMR 2.88 (2.28, 3.59) [78 cases]
Cohort, coke oven workers
789 (all men)
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
Results
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 (range 0.9-89)
u.g/m3
Hansen (1991); (Hansen, 1989) (Denmark)
SMR 2.90 (1.88, 4.3) [25 cases] (ages 40 to 89)
SMR 2.46 (1.59, 3.6) [25 cases] (with smoking adjustment)
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 (range 0.5-
260) mg/m3
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 ~15 yrs)
Smoking information from interviews with older
workers
Exposure: area sampling - top of ovens.
Benzo[a]pyrene, 1,964 mean 4.3 (range 0.007-33);
1,965 mean 0.52, (0.021-1.29) |jg/m3
SMR 0.82 (0.22, 2.1) [4 cases] (referent group = employed
men)
SIR 1.35(0.36,3.5) [4 cases]
Moulin et al. (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
Exposure: benzo[a]pyrene, 19 area samples and 16
personal samples in Plant A (personal sample mean
2.7; range 0.59-6.2 u.g/m3); 10 area samples and
7 personal samples in Plant B; personal sample mean
0.17, range 0.02-0.57 u.g/m3
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]
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
Hammond et al. (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
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 eta I., 2003); Burstyn
et al. (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 n cases
RR (95% Cl)
(no lag)b
RR (95% Cl)
(15 yr lag)0
0-0.253 12 1.0 (referent) 1.0 (referent)
0.253-0.895 12 0.69 (0.29, 1.6) 1.1 (0.44, 2.9)
0.895-1.665 12 1.21(0.45,3.3) 1.7(0.62,4.5)
>1.665 12 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); Gibbsetal. (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, 7 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
Benzo[a]-
pyrene
u.g/m3-yrsa n cases SMR (95% Cl)
Smoking-
adjusted
RRb
0
10
30
60
120
3 0.73(0.15,2.1) 1.0 (referent)
14 0.93 (0.45, 1.4) 1.11
3 1.37 (0.28,4.0) 1.97
1 0.35 (0.9, 1.9) 0.49
15 4.2 (2.4,6.9) 8.49
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
data); Armstrong et al. (1994); Gibbs (1985); Gibbs
and Horowitz (1979)
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)
Results
240 30 6.4 (4.3, 9.2)
480 12 23.9 (12.2, 41.7)
aCategory midpoint
bCls not reported; highest category is >80 u.g/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.
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
|jg/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
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)
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)
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
Reference and Study Design
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
Plant A: 0 observed cases;
Plant B:SMR 1.94 (0.40, 5
expected <1.0
0) (3 cases)
SMR 2.85 (0.30, 10.3) (2 cases) (referent group
men)
= employed
2 Evidence in Animals
3 Oral exposure
4 Evidence of tumorigenicity following oral exposure to benzo[a]pyrene has been
5 demonstrated in rats and mice. As summarized in Table 1-14, oral exposure to benzo[a]pyrene has
6 resulted in an increased incidence of tumors in the alimentary tract in male and female rats [Kroese
7 etal.. 2001: Brune etal.. 19811 and female mice fBeland and Gulp. 1998: Gulp etal.. 19981 liver
8 carcinomas in male and female rats, kidney adenomas in male rats [Kroese etal., 2001], and
9 auditory canal tumors in both sexes [Kroese etal.. 2001).
10 Forestomach tumors have been observed in several lifetime cancer bioassays in rats and
11 mice following both gavage and dietary exposure to benzo[a]pyrene at doses ranging from
12 0.016 mg/kg-day in Sprague-Dawley rats to 3.3 and 10 mg/kg-day in B6C3Fi mice and Wistar rats,
13 respectively [Kroese etal.. 2001: Belandand Gulp. 1998: Gulp etal.. 1998: Brune etal.. 1981]. In
14 addition, multiple less-than-lifetime oral exposure cancer bioassays in mice provide supporting
15 evidence that oral exposure to benzo[a]pyrene is associated with an increased incidence of
16 forestomach tumors [Weyandetal.. 1995: Benjamin etal.. 1988: Robinson etal.. 1987: El-Bayoumy.
17 1985: Triolo etal.. 1977: Wattenberg. 1974: Roe etal.. 1970: Biancifiori etal.. 1967: Chouroulinkov
18 etal.. 1967: Fedorenko and Yansheva. 1967: Neal and Rigdon. 1967: Berenblum and Haran. 1955].
19 Although humans do not have a forestomach, similar squamous epithelial tissue is present in the
20 oral cavity [IARC. 2003: Wester and Kroes. 1988]: therefore, EPA concluded that forestomach
21 tumors observed in rodents following benzo[a]pyrene exposure are relevant in the assessment of
22 carcinogenity. For further discussion, see Sections 1.2 and 2.3.4.
23 Elsewhere in the alimentary tract, dose-related increases of benign and malignant tumors
24 were observed. In rats, oral cavity tumors were induced in both sexes and adenocarcinomas of the
25 jejunum were induced in males [Kroese etal., 2001]. In mice, tumors were induced in the tongue,
26 esophagus, and larynx of females (males were not tested] [Beland and Gulp, 1998: Gulp etal., 1998].
27 Chronic oral exposure to benzo[a]pyrene resulted in a dose-dependent increased incidence
28 of liver carcinomas in both sexes of Wistar rats, with the first liver tumors detected in week 35 in
29 high-dose male rats; liver tumors were described as complex, with a considerable proportion
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
1 (59/150 tumors) metastasizing to the lungs [Kroese etal.. 2001]. Treatment-related hepatocellular
2 tumors were not observed in mice [Beland and Gulp. 1998: Gulp etal.. 1998).
3 A statistically significantly increased incidence of kidney tumors (cortical adenomas) was
4 observed in male Wistar rats following chronic gavage exposure (Kroese etal., 2001] (Table 1-14).
5 The kidney tumors were observed at the mid- and high-dose groups. Treatment-related kidney
6 tumors were not observed in two other studies (Brune etal., 1981).
7 Lung tumors were also observed following almost nine months of dietary exposure to
8 approximately 10 mg/kg-day in female AJ mice (Weyandetal., 1995). Other lifetime exposure
9 studies did not report treatment-related increases in lung tumors (Kroese etal.. 2001: Beland and
10 Gulp. 1998: Gulp etal.. 19981
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
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,
1-62 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 blncidences are for 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 et al., 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 one to two years following cessation
19 of 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,
1-63 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 Chronic inhalation exposure to benzo[a]pyrene resulted in the development of tumors in
2 the respiratory tract and pharynx in Syrian golden hamsters (Table 1-15). Concentration-
3 dependent increased incidences of tumors in the upper respiratory tract, including the larynx and
4 trachea, were reported by Thyssen et al. [1981] at measured exposure concentrations of >9.5
5 mg/m3. In addition, a decrease in tumor latency was observed in the larynx and trachea, and nasal
6 cavity tumors were observed at the mid- and high-concentration, but the incidences were not dose-
7 dependently increased. A concentration-related increase in tumors in the upper digestive tract
8 (pharynx and esophagus) was also reported. In addition, a single forestomach tumor was observed
9 in each of the mid- and high-concentration groups, in animals with a tumor in either the larynx or
10 pharynx; forestomach tumors were not observed in control animals. The study authors presumed
11 that the pharyngeal and esophageal tumors were a consequence of mucociliary particle clearance.
12 The authors stated that the rates of tumors of other organs generally corresponded to the rates in
13 controls.
14 Under contract to the U.S. EPA, Clement and Associates U.S. EPA (1990b) obtained the
15 individual animal data (including individual animal pathology reports, time-to-death data, and
16 exposure chamber monitoring data) collected by Thyssen etal. (1981). A re-analysis of the
17 individual animal pathology reports from the original study supports the concentration-dependent
18 increased incidence of tumors in the larynx and pharynx (U.S. EPA, 1990a, b). The exposure
19 measurements and individual animal data from Thyssen etal. (1981) were used to calculate
20 average continuous lifetime exposures for each individual hamster. Group averages of individual
21 average continuous lifetime exposure concentrations were 0, 0.25,1.01, and 4.29 mg/m3 for the
22 control through high-exposure groups, as described in U.S. EPA (1990b).
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
Table 1-15. Tumors observed in chronic inhalation animal bioassays
Reference and Study Design
Results'3
Thyssenetal. (1981)
Syrian golden hamsters: male
(20-30 animals/group)
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: 0/27; 0/27; 8/26; and 13/25
earliest observation of tumor0:107 and 68 wks
Pharynx
incidences: 0/27; 0/27; 6/26; and 14/25
earliest observation of tumor: 97 and 68 wks
Trachea
incidences: 0/27; 0/27; 1/26; and 3/25
earliest observation of tumor: 115 and 63 wks
Nasal cavity
incidences: 0/27; 0/27; 3/26; and 1/25
earliest observation of tumor: 116 and 79 wks
Esophagus
incidences: 0/27; 0/27; 0/27; and 2/25
earliest observation of tumor: 71 wks
Forestomach
incidences: 0/27; 0/27; 1/26; and 1/25
earliest observation of tumor: 119 and 72 wks
Revised tumor incidence data
Larynx
incidences: 0/27; 0/27; 11/26; and 12/34
Pharynx
incidences: 0/27; 0/27; 9/26; and 18/34
Larynx and pharynx (combined)6
incidences: 0/27; 0/27; 16/26; and 18/34
2
3 aDuration adjusted inhalation concentrations calculated from exposure chamber monitoring data and exposure
4 treatment times obtained by Clement Associates and reported in U.S. EPA (1990b). Daily exposure times:
5 4.5 hours/day, 5 days/week on weeks 1-12; 3 hours/day, 5 days/week on weeks 13-29; 3.7 hours/day,
6 5 days/week on week 30; 3 hours/day, 5 days/week on weeks 31^41; and 3 hours/day, 7 days/week for reminder
7 of the experiment.
8 Statistical significance not reported by study authors.
9 °Earliest observation of tumor provided for 9.5 and 46.5 mg/m3 concentration groups.
10 dRevised tumor incidence data based on original study pathology reports obtained by Clement Associates and
11 reported in U.S. EPA(1990b).
12 eNasal, forestomach, esophageal, and tracheal tumors occurred in hamsters that also had tumors in the larynx or
13 pharynx, except for two animals in the mid-concentration group that displayed nasal tumors (one malignant and
14 one benign) without displaying tumors in the pharynx or larynx.
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
1 Dermal exposure
2 Repeated application of benzo[a]pyrene to skin (in the absence of exogenous promoters)
3 has been demonstrated to induce skin tumors in mice, rats, rabbits, and guinea pigs. These studies
4 have been reviewed by multiple national and international health agencies [IARC, 2010: IPCS, 1998:
5 ATSDR, 1995: IARC, 1983,1973]. Mice have been the most extensively studied species in dermal
6 carcinogenesis studies ofbenzo[a]pyrene because of evidence that they may be more sensitive than
7 other animal species; however, comprehensive comparisons of species differences in sensitivity to
8 lifetime dermal exposure are not available. Systemic tumors in benzo[a]pyrene-treated mice were
9 not increased compared to controls in lifetime dermal bioassays in which macroscopic examination
10 of internal organs was included [Higginbotham etal.. 1993: Habs etal.. 1980: Schmahletal.. 1977:
11 Schmidt etal.. 1973: Roe etal.. 1970: Poel. 19591.
12 The analysis in this document focuses on chronic carcinogenicity bioassays in several
13 strains of mice following repeated dermal exposure to benzo[a]pyrene for the animals' lifetime
14 (Table 1-16). These studies involved 2- or 3-times/week exposure protocols, at least two exposure
15 levels plus controls, and histopathological examinations of the skin and other tissues (Sivaketal.,
16 1997: Grimmer etal.. 1984: Habs etal.. 1984: Grimmer etal.. 1983: Habs etal.. 1980: Schmahletal..
17 1977: Schmidt etal.. 1973: Roe etal.. 1970: Poel. 1960.19591
18 Numerous studies in mice observed skin tumors following benzo[a]pyrene exposure, but
19 were not considered further in this assessment because of the availability of the chronic studies
20 identified above. These studies included several "skin painting" studies in mouse skin that did not
21 report the doses applied (e.g., Wynder and Hoffmann. 1959: Wynderetal.. 1957): several shorter-
22 term studies (Albertetal.. 1991: Nesnowetal.. 1983: Emmettetal.. 1981: Levin etal.. 1977):
23 initiation-promotion studies utilizing acute dosing of benzo[a]pyrene followed by repeated
24 exposure to a potent tumor promoter; and studies involving vehicles expected to interact with or
25 enhance benzo[a]pyrene carcinogenicity (e.g., Bingham and Falk, 1969).
26 One study applied benzo[a]pyrene (topically once a week for 6 months) to
27 immunocompromised mice with human skin xenografts (n=10) and did not observe tumors,
28 whereas all three control mice (mice with no xenografts) developed skin tumors (Urano etal..
29 1995). The authors concluded this result indicates that human skin is much less susceptible to
30 benzo[a]pyrene than mouse skin. However, it is unclear that this human skin xenograft model
31 preserves the physiological and morphological properties of human skin in vivo (Kappes etal..
32 2004).
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
Table 1-16. Tumor observations in dermal animal bioassays
Reference and Study Design
Results3
Poel (1959)
C57L mice: 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
Poel (1960)
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
Roe etal. (1970)
Swiss mice: female (50/dose)
0, vehicle, 0.1, 0.3,1, 3, or 9 u.g
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
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 u.g (papillomas);
2/100 at 0.8 u.g and 30/100 at 2 u.g (carcinomas)
Swiss:
3/80 at 2 u.g (papillomas);
5/80 at 0.8 u.g and 45/80 at 2 u.g (carcinomas)
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)
Habs etal. (1980)
NMRI mice: female (40/group)
0,1.7, 2.8, or 4.6 ug
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%
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
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)
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)
Sivaketal. (1997)
C3H/HeJ mice: male (30/group)
0, 0.05, 0.5, or 5 u.g
Dermal; 2 times/wk for 104 wks
Skin tumors (papillomas and carcinomas)
incidences:
0/30; 0/30; 5/30 (2 papillomas, 3 carcinomas); and
27/30 (1 papilloma, 28 carcinomas)
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: Bostrometal.. 2002: Penning etal.. 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
21 event 3), and these mutations can undergo clonal expansion (key event 4) enabled by multiple
22 mechanisms also induced by benzo[a]pyrene, including AhR binding leading to an upregulation of
23 genes related to biotransformation, growth, and differentiation, and regenerative cell proliferation
24 resulting from cytotoxicity and a sustained inflammatory response. However, there is not sufficient
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
1 evidence that these mechanisms, which contribute to the promotion and progression phases of
2 cancer development, act independently of DNA damage and mutation to produce benzo[a]pyrene-
3 induced tumors (please see Other possible modes of action, below). The available human, animal,
4 and in vitro evidence supports a mutagenic mode of action as the primary mode by which
5 benzo[a]pyrene induces carcinogenesis.
7
8
10
11
12
13
14
15
16
Key events in the mode of action for benzofalpyrene carcinogenicity
Promotion Progression
Upreguiatton of genes
related to
biotransformation,
growth, and
differentiation
Proliferation
of initiated
o-qumone
and ROS
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
(+)-anti- BPDE, one of the most potent DNA-binding metabolites of benzo[a]pyrene.
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 Benzo[a]pyrene diol epoxide metabolites interact preferentially with the exocyclic amino groups of
2 deoxyguanine and deoxyadenine [Geacintovetal.. 1997: Terinaetal.. 1991). Adducts may give rise
3 to mutations unless these adducts are removed by DNA repair processes prior to replication. The
4 stereochemical nature of the diol epoxide metabolite (i.e., anti- versus syn-diol epoxides) affects the
5 number and type of adducts and mutation that occurs [Geacintovetal., 1997]. Transversion
6 mutations (e.g., GC—>TA or AT—>TA] are the most common type of mutation found in mammalian
7 cells following diol epoxide exposure (Bostrom etal., 2002].
8 Radical cation pathway. Radical cation formation involves a one-electron oxidation by GYP
9 or peroxidase enzymes (i.e., horseradish peroxidase, prostaglandin H synthetase] that produce
10 electrophilic radical cation intermediates (Cavalieri and Rogan. 1995.1992]. Radical cations can be
11 further metabolized to phenols and quinones (Cavalieri etal., 1988e: Cavalieri et al., 1988d], or they
12 can form unstable adducts with DNA that ultimately result in depurination. The predominant
13 depurinating adducts occur at the N-3 and N-7 positions of adenine and the C-8 and N-7 positions of
14 guanine (Cavalieri and Rogan, 1995].
15 o-Quinone/ROS pathway. The o-quinone metabolites of PAHs are formed by enzymatic
16 dehydrogenation of dihydrodiols (Bolton etal.. 2000: Penning etal.. 1999: Harvey. 1996: ATSDR.
17 1995] (see Appendix D of the Supplemental Information]. Dihydrodiol dehydrogenase enzymes are
18 members of the a-keto reductase gene superfamily. o-Quinone metabolites are potent cytotoxins,
19 are weakly mutagenic, and are capable of producing a broad spectrum of DNA damage. These
20 metabolites can interact directly with DNA as well as result in the production of ROS (i.e., hydroxyl
21 and superoxide radicals] that may produce further cytotoxicity and DNA damage. The
22 o-quinone/ROS pathway also can produce depurinated DNA adducts from benzo[a]pyrene
23 metabolites. In this pathway, and in the presence of NAD(P]+, aldo-keto reductase oxidizes
24 benzo[a]pyrene-7,8-diol to a ketol, which subsequently forms benzo[a]pyrene-7,8-dione. This and
25 other PAH o-quinones react with DNA to form unstable, depurinating DNA adducts. In the presence
26 of cellular reducing equivalents, o-quinones can also activate redox cycles, which produce ROS
27 (Penning etal., 1996]. DNA damage in in vitro systems following exposure to benzo[a]pyrene-
28 7,8-dione or other o-quinone PAH derivatives occurs through the AKR pathway and can involve the
29 formation of stable DNA adducts (Baluetal., 2004], N-7 depurinated DNA adducts (Mccoulletal.,
30 1999]. DNA damage from ROS (8-oxo-dG] (Park etal.. 2006]. and strand scission (Flowers etal..
31 1997: Flowers etal.. 19961
32 Summary of genotoxicity and mutagenicity
33 The ability of metabolites of benzo[a]pyrene to cause mutations and other forms of DNA
34 damage in both in vivo and in vitro studies is well documented (see genotoxicity tables in
35 Appendix D in Supplemental Information]. With metabolic activation (e.g., the inclusion of S9],
36 benzo[a]pyrene is consistently mutagenic in the prokaryotic Salmonella/Ames and Escherichia coli
37 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.
1-70 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 In vitro mammalian cell assays have been conducted in various test systems, including human cell
6 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 Cancer Guidelines [Section 2.4; (2005a]J describe a procedure for evaluating mode-
16 of-action data for cancer. A framework for analysis of mode of action information is provided and
17 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 (Keohavong et al.,
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: Hainautand Pfeifer. 2001: Marshall etal.. 19841: 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.. 20001 or by i.p. injection fNesnowetal.. 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 thatbenzo[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
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 elevated BPDE-DNA adduct levels in WBCs of groups of coke oven workers and chimney sweeps,
2 occupations with known elevated risks of cancer [Vineis etal.. 2007: Pavanello etal.. 1999). and in
3 lung tissue from tobacco smokers with lung cancer [Rojas etal.. 2004: Godschalketal.. 2002: Rojas
4 etal., 1998: Andreassenetal., 1996: Alexandrovetal., 1992]. Several epidemiological studies have
5 indicated that PAH-exposed individuals who are homozygous for a CYP1A1 polymorphism, which
6 increases the inducibility of this enzyme (thus increasing the capacity to produce benzo[a]pyrene
7 diol epoxide), have increased levels of PAH or BPDE-DNA adducts [Aklillu etal., 2005: Alexandrov
8 etal.. 2002: Bartsch etal.. 2000: Perera and Weinstein. 20001.
9 Additional supporting evidence of a mutagenic mode of action for benzo[a]pyrene
10 carcinogenicity is the extensive database of in vitro and in vivo studies demonstrating the
11 genotoxicity and mutagenicity of benzo[a]pyrene following metabolic activation (Table 1-17). In
12 vitro studies overwhelmingly support the formation of DNA adducts, mutagenesis in bacteria, yeast,
13 and mammalian cells, several measures of cytogenetic damage (CA, SCE, MN), and DNA damage. In
14 vivo systems in animal models are predominantly positive for somatic mutations following
15 benzo[a]pyrene exposure.
16 Support for the radical cation activation pathway contributing to tumor initiation through
17 mutagenic events includes observations that depurinated DNA adducts (expected products from
18 reactions of benzo[a]pyrene radical cations with DNA) accounted for 74% of identified DNA
19 adducts in mouse skin exposed to benzo[a]pyrene (Roganetal., 1993] and that 9 of 13 tumors
20 examined from mice exposed to dermal applications of benzo[a]pyrene had H-ras oncogene
21 mutations attributed to depurinated DNA adducts from benzo[a]pyrene radical cations
22 (Chakravartietal.. 1995).
23 Support for the o-quinone/ROS pathway contributing to tumor initiation via mutagenic
24 events includes in vitro demonstration that several types of DNA damage can occur from
25 o-quinones andROS (Parketal.. 2006: Balu etal.. 2004: Mccoull etal.. 1999: Flowers etal.. 1997:
26 Flowers etal., 1996]. In addition, benzo[a]pyrene-7,8-dione can induce mutations in the p53 tumor
27 suppressor gene using an in vitro yeast reporter gene assay (Park etal., 2008: Shenetal., 2006: Yu
28 etal., 2002], and dominant p53 mutations induced by benzo[a]pyrene-7,8-dione in this system
29 corresponded to p53 mutational hotspots observed in human lung cancer tissue (Park etal., 2008].
30 Dose-response concordance and temporal relationship. Studies in humans demonstrating
31 that benzo[a]pyrene-induced mutational events in p53 or ras oncogenes precede tumor formation
32 are not available, but there is evidence linking benzo[a]pyrene exposure to signature mutational
33 events in humans. In vitro exposure of human p53 knock-in murine fibroblasts to 1 |iM
34 benzo[a]pyrene for 4-6 days induced p53 mutations with similar features to those identified in p53
35 mutations in human lung cancer; i.e., predominance of G—>T transversions with strand bias and
36 mutational hotspots atcodons 157-158 (Liu etal.. 2005).
37 Bennett etal. (1999) demonstrated a dose-response relationship between smoking
38 history/intensity and the types of p53 mutations associated with benzo[a]pyrene (G—>T
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 transversions) in human lung cancer patients (Table 1-17). In lung tumors of non-smokers, 10% of
2 p53 mutations were G^T transversions, versus 40% in lung tumors from smokers with >60 pack-
3 years of exposure.
4 In mice, dose-response and temporal relationships have been described between the
5 formation of BPDE-DNA adducts and skin and forestomach tumors (Table 1-17). In a study using
6 mice treated dermally with benzo[a]pyrene once or twice per week for up to 15 weeks (10, 25, or
7 50 nmol benzo[a]pyrene per application), levels of benzo[a]pyrene-DNA adducts in the skin, lung,
8 and liver increased with increasing time of exposure and increasing dose levels (Talaskaetal.,
9 2006). Levels at the end of the exposure period were highest in the skin; levels in the lung and liver
10 at the same time were 10- and 20-fold lower, respectively. Levels of benzo[a]pyrene-DNA adducts
11 in skin and lung increased in an apparent biphasic manner showing a lower linear slope between
12 the two lowest dose levels, compared with the slope from the middle to the highest dose.
13 Another study examined the dose-response relationship and the time course of
14 benzo[a]pyrene-induced skin damage (Table 1-17), DNA adduct formation, and tumor formation in
15 female mice. Mice were treated dermally with 0,16, 32, or 64 |ig of benzo[a]pyrene once per week
16 for 29 weeks (Albert etal.. 1991). Indices of skin damage and levels of BPDE-DNA adducts in skin
17 reached plateau levels in exposed groups by 2-4 weeks of exposure. With increasing dose level,
18 levels of BPDE-DNA adducts (fmol/ng DNA) initially increased in a linear manner and began to
19 plateau at doses >32 |ig/week. Tumors began appearing after 12-14 weeks of exposure for the
20 mid- and high-dose groups and at 18 weeks for the low-dose group. At study termination
21 (35 weeks after start of exposure), the mean number of tumors per mouse was approximately one
22 per mouse in the low- and mid-dose groups and eight per mouse in the high-dose group. The time-
23 course data indicate thatbenzo[a]pyrene-induced increases in BPDE-DNA adducts preceded the
24 appearance of skin tumors, consistent with the formation of DNA adducts as a precursor event in
25 benzo[a]pyrene-induced skin tumors.
26 (Gulp etal., 1996) compared dose-response relationships for BPDE-DNA adducts and
27 tumors in female B6C3Fi mice exposed to benzo[a]pyrene in the diet at 0,18.5, 90, or 350 |ig/day
28 for 28 days (to examine adducts) or 2 years (to examine tumors) (Table 1-17). The benzo[a]pyrene
29 dose-tumor response data showed a sharp increase in forestomach tumor incidence between the
30 18.5 |ig/day group (6% incidence) and the 90 |ig/day group (78% incidence). The BPDE-DNA
31 adduct levels in forestomach showed a relatively linear dose-response throughout the
32 benzo[a]pyrene dose range tested. The appearance of increased levels of BPDE-DNA adducts in the
33 targettissue at 28 days is temporally consistent with the contribution of these adducts to the
34 initiation of forestomach tumors. Furthermore, about 60% of the examined tumors had mutations
35 in the K-ras oncogene at codons 12 and 13, which were G—>T or G—>C transversions indicative of
36 BPDE reactions with DNA (Gulp etal.. 1996).
37 Biological plausibility and coherence. The evidence for a mutagenic mode of action for
38 benzo[a]pyrene is consistent with the current understanding that mutations in p53 and ras
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 oncogenes are associated with increased risk of tumor initiation (Table 1-17). The benzo[a]pyrene
2 database is internally consistent in providing evidence for BPDE-induced mutations associated with
3 tumor initiation in cancer tissue from humans exposed to complex mixtures containing
4 benzo[a]pyrene [Keohavongetal., 2003: Pfeifer and Hainaut, 2003: Pfeifer etal., 2002: DeMarini et
5 al., 2001: Hainautand Pfeifer, 2001: Bennett etal., 1999], in animals exposed to benzo[a]pyrene
6 [Gulp etal.. 2000: Nesnowetal.. 1998a: Nesnowetal.. 1998b: Nesnowetal.. 1996.1995: Mass etal..
7 1993], and in in vitro systems [Denissenko etal., 1996: Puisieuxetal., 1991]. Consistent
8 supporting evidence includes: (1] elevated BPDE-DNA adduct levels in tobacco smokers with lung
9 cancer fRoias etal.. 2004: Godschalk etal.. 2002: Roias etal.. 1998: Andreassen etal.. 1996:
10 Alexandrovetal.. 1992]: (2] demonstration of dose-response relationships between G—>T
11 transversions in p53 mutations in lung tumors and smoking intensity [Bennett etal., 1999]: (3] the
12 extensive database of in vitro and in vivo studies demonstrating the genotoxicity and mutagenicity
13 of benzo[a]pyrene following metabolic activation; and (4] general concordance between temporal
14 and dose-response relationships for BPDE-DNA adduct levels and tumor incidence in studies of
15 animals exposed to benzo[a]pyrene [Gulp etal., 1996: Albert etal., 1991]. There is also supporting
16 evidence that contributions to tumor initiation through mutagenic events can be made by the
17 radical cation [Chakravarti etal.. 1995: Roganetal.. 1993] and o-quinone/ROS metabolic activation
18 pathways [Park etal.. 2008: Park etal.. 2006: Shen etal.. 2006: Balu etal.. 2004: Yu etal.. 2002:
19 Mccoull etal.. 1999: Flowers etal.. 1997: Flowers etal.. 1996].
20 Table 1-17. Experimental support for the postulated key events for mutagenic
21 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; Pavanelloetal., 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.
1-74 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 (Parketal., 2006; Balu et al., 2004; Mccoull etal., 1999; Flowers etal., 1997; Flowers
etal., 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 et al., 1999; Rojasetal., 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->Ttransversion 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 et al., 2005; Gulp et al., 1996);(Wei et al., 1999)
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->Ttransversion 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 etal.. 2005: Keohavong et al.. 2003: Pfeiferand Hainaut. 2003: Pfeifer et a I.. 2002: DeMarini etal.. 2001:
Hainaut and Pfeifer, 2001; Bennett et al., 1999; Denissenko et al., 1996; Puisieux et al., 1991; Marshall et al.,
1984; Koreeda et al., 1978; Jeffrey et al., 1976)
This document is a draft for review purposes only and does not constitute Agency policy.
1-75 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: Sivaketal.. 1997: ATSDR. 1995: Grimmer et al.. 1984: Habsetal.. 1984:
Grimmer etal., 1983; IARC, 1983,1973; Habsetal., 1980; Schmahl et al., 1977; Schmidt et al., 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 et al., 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)
2 Other possible modes of action
3 The carcinogenic process for benzo[a]pyrene is likely to be related to some combination of
4 molecular events resulting from the formation of several reactive metabolites that interact with
5 DNA to form adducts and produce DNA damage resulting in mutations in cancer-related genes, such
6 as tumor suppressor genes or oncogenes. These events may reflect the initiation potency of
7 benzo[a]pyrene. However, benzo[a]pyrene possesses promotional capabilities that may be related
8 to AhR affinity, immune suppression, cytotoxicity, and the formation of ROS, as well as the
9 inhibition of gap junctional intercellular communication.
10 The ability of certain PAHs to act as initiators and promoters may increase their
11 carcinogenic potency. The promotional effects of PAHs appear to be related to AhR affinity and the
12 upregulation of genes related to growth and differentiation [Bostrometal., 2002]. The genes
13 regulated by this receptor belong to two major functional groups (i.e., induction of metabolism or
14 regulation cell differentiation and proliferation). PAHs bind to the cytosolic AhR in complex with
15 heat shock protein 90. The ligand-bound receptor is then transported to the nucleus in complex
16 with the AhR nuclear translocator protein. The AhR complex interacts with AhR elements of DNA
17 to increase the transcription of proteins associated with induction of metabolism and regulation of
18 cell differentiation and proliferation. Following benzo[a]pyrene exposure, disparities have been
19 observed in the tumor pattern and toxicity of Ah-responsive and Ah-nonresponsive mice, as Ah-
20 responsive mice were more susceptible to tumorigenicity in target tissues such as liver, lung, and
21 skin [Ma and Lu. 2007: Talaska etal.. 2006: Shimizu etal.. 2000).
22 Benzo[a]pyrene has both inflammatory and immunosuppressive effects that may function
23 to promote tumorigenesis. Inflammatory responses to cytotoxicity may contribute to the tumor
24 promotion process; for example, benzo[a]pyrene quinones (1,6-, 3,6-, and 6,12-benzo[a]pyrene-
25 quinone) generated ROS and increased cell proliferation by enhancing the epidermal growth factor
26 receptor pathway in cultured breast epithelial cells [Burdick et al., 2003]. In addition, several
27 studies have demonstrated that exposure to benzo[a]pyrene increases the production of
28 inflammatory cytokines, which may contribute to cancer progression [N'Diaye etal.. 2006: Tamaki
29 etal.. 2004: Garconetal.. 2001b: Garconetal.. 2001a]. Conversely, the immunosuppressive effects
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
1 of benzo[a]pyrene exposure (see Section 1.1.3) may provide an environment where tumor cells can
2 evade detection by immune surveillance mechanisms normally responsible for recognizing and
3 eliminating nascent cancer cells [Hanahan and Weinberg. 2011). In addition, the developing fetus
4 may be even more sensitive to these effects; Urso and Gengozian [1980] found that mice exposed to
5 benzo[a]pyrene in utero not only had a significantly increased tumor incidence as adults but also a
6 persistently suppressed immune system.
7 Gap junctions are channels between cells that are crucial for differentiation, proliferation,
8 apoptosis, and cell death. Interruption of gap junctional intercellular communication is associated
9 with a loss of cellular control of growth and differentiation, and consequently with the two
10 epigenetic steps of tumor formation, promotion and progression. Thus, the inhibition of gap
11 junctional intercellular communication by benzo[a]pyrene, observed in vitro [Sharovskaya et al.,
12 2006: Blahaetal., 2002], provides another mechanism of tumor promotion.
13 In summary, there are tumor promoting effects of PAH exposures that are not mutagenic.
14 Although these effects are observed following benzo[a]pyrene-specific exposures, the occurrence of
15 BPDE-DNA adducts and associated mutations that precede both cytotoxicity and tumor formation
16 and increase with dose provides evidence that mutagenicity is the primary event that initiates
17 tumorigenesis following benzo[a]pyrene exposures. A biologically plausible mode of action may
18 involve a combination of effects induced by benzo[a]pyrene, with mutagenicity as the initiating
19 tumorigenic event. Subsequent AhR activation and cytotoxicity could then lead to increased ROS
20 formation, regenerative cell proliferation, and inflammatory responses, which, along with evasion
21 of immune surveillance and gap junctional intercellular communication, would provide an
22 environment where the selection for mutated cells increases the rate of mutation, allowing clonal
23 expansion and progression of these tumor cells to occur. However, it was determined that, in
24 comparison to the large database on the mutagenicity of benzo[a]pyrene, there were insufficient
25 data to develop a separate mode of action analysis for these promotional effects.
26 Conclusions about the hypothesized mode of action
27 There is sufficient evidence to conclude that the major mode of action for benzo[a]pyrene
28 carcinogenicity involves mutagenicity mediated by DNA reactive metabolites. The evidence for a
29 mutagenic mode of action for benzo[a]pyrene is consistent with the current understanding that
30 mutations in p53 and ras oncogenes are associated with increased risk of tumor initiation. The
31 benzo[a]pyrene database provides strong and consistent evidence for BPDE-induced mutations
32 associated with tumor initiation in cancer tissue from humans exposed to complex mixtures
33 containing benzo[a]pyrene, in animals exposed to benzo[a]pyrene, and in in vitro systems.
34 Supporting evidence suggests that contributions to tumor initiation through potential mutagenic
35 events can be made by the radical cation and o-quinone/ROS metabolic activation pathways. Other
36 processes may contribute to the carcinogenicity of benzo[a]pyrene via the promotion and
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 progression phases of cancer development (e.g., inflammation, cytotoxicity, sustained regenerative
2 cell proliferation).
3 Support for the Hypothesized Mode of Action in Test Animals
4 Benzo[a]pyrene induces gene mutations in a variety of in vivo and in vitro systems and
5 produces tumors in all animal species tested and by all routes of exposure (see Appendix D in
6 Supplemental Information). Strong, consistent evidence in animal models supports the postulated
7 key events: the metabolism of benzo[a]pyrene to DNA-reactive intermediates, the formation of
8 DNA adducts, the subsequent occurrence of mutations in oncogenes and tumor suppressor genes,
9 and the clonal expansion of mutated cells.
10 Relevance of the Hypothesized Mode of Action to Humans
11 A substantial database indicates that the postulated key events for a mutagenic mode of
12 action all occur in human tissues. Evidence is available from studies of humans exposed to PAH
13 mixtures (including coal smoke and tobacco smoke) indicating a contributing role for
14 benzo[a]pyrene diol epoxide in inducing key mutational events in genes that are associated with
15 tumor initiation (mutations in the p53 tumor suppressor gene and H-ras or K-ras oncogenes). The
16 evidence includes observations of a spectrum of mutations in ras oncogenes and the p53 gene in
17 lung tumors of human patients exposed to coal smoke or tobacco smoke) that are similar to the
18 spectrum of mutations caused by benzo[a]pyrene diol epoxide in several biological systems,
19 including tumors from mice exposed to benzo[a]pyrene. Additional supporting evidence includes
20 correspondence between hotspots of p53 mutations in human lung cancers and sites of DNA
21 adduction by benzo[a]pyrene diol epoxide in experimental systems, and elevated BPDE-DNA
22 adduct levels in respiratory tissue of lung cancer patients or tobacco smokers with lung cancer.
23 Populations or Lifestages Particularly Susceptible to the Hypothesized Mode of Action
24 A mutagenic mode of action for benzo[a]pyrene-induced carcinogenicity is considered
25 relevant to all populations and lifestages. The current understanding of biology of cancer indicates
26 that mutagenic chemicals, such as benzo[a]pyrene, are expected to exhibit a greater effect in early
27 life exposure versus later life exposure fU.S. EPA. 2005b: Vesselinovitch etal.. 19791 The EPA's
28 Supplemental Guidance for Assessing Susceptibility from Early-Life Exposure to Carcinogens (U.S. EPA.
29 2005b) recommends the application of age-dependent adjustment factors (ADAFs) for carcinogens
30 that act through a mutagenic mode of action. Given that a determination benzo [ajpyrene acts
31 through a mutagenic mode of carcinogenic action has been made, ADAFs should be applied along
32 with exposure information to estimate cancer risks for early-life exposure.
33 Toxicokinetic information suggest early lifestages may have lower levels of some GYP
34 enzymes than adults (Ginsberg etal.. 2004: Cresteil. 1998). suggesting that lower levels of
35 mutagenic metabolites may be formed in early lifestages. Though expression of bioactivating
36 enzymes is believed to be lower in the developing fetus and children, metabolism of
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 benzo[a]pyrene still occurs, as indicated by the detection of benzo[a]pyrene-DNA or protein
2 adducts or urinary metabolites [Naufal et al.. 2010: Ruchirawat et al.. 2010: Suter etal.. 2010:
3 Mielzynskaetal.. 2006: Pereraetal.. 2005a: Tang etal.. 1999: Whyattetal.. 1998). While
4 expression of GYP enzymes is lower in fetuses and infants, the greater liver to body mass ratio and
5 increased blood flow to liver in fetuses and infants may compensate for the decreased expression of
6 GYP enzymes [Ginsberg etal., 2004]. Activity of Phase II detoxifying enzymes in neonates and
7 children is adequate for sulfation but decreased for glucuronidation and glutathione conjugation
8 [Ginsberg et al., 2004]. The conjugation of benzo[a]pyrene-4,5-oxide with glutathione was
9 approximately one-third less in human fetal than adult liver cytosol [Pacifici etal..1988].
10 In addition, newborn or infant mice develop liver and lung tumors more readily than young
11 adult mice following acute i.p. exposures to benzo[a]pyrene [Vesselinovitch etal., 1975]. These
12 results indicate that exposure to benzo[a]pyrene during early life stages presents additional risk for
13 cancer, compared with exposure during adulthood, despite lower metabolic activity in early
14 lifestages. Population variability in metabolism and detoxification of benzo[a]pyrene, in addition to
15 DNA repair capability, may affect cancer risk. Polymorphic variations in the human population in
16 CYP1A1, CYP1B1, and other GYP enzymes have been implicated as determinants of increased
17 individual cancer risk in some studies [Ickstadtetal.. 2008: Aklillu etal.. 2005: Alexandrov et al..
18 2002: Perera and Weinstein, 2000]. Some evidence suggests that humans lacking a functional
19 GSTM1 gene have higher BPDE-DNA adduct levels and are thus at greater risk for cancer [Binkova
20 etal.. 2007: Vineis etal.. 2007: Pavanello etal.. 2005: Pavanello etal.. 2004: Alexandrov etal.. 2002:
21 Perera and Weinstein, 2000]. In addition, acquired deficiencies or inherited gene polymorphisms
22 that affect the efficiency or fidelity of DNA repair may also influence individual susceptibility to
23 cancer from environmental mutagens [Wang etal.. 2010: Ickstadtetal.. 2008: Binkova etal.. 2007:
24 Matullo etal.. 2003: Shen etal.. 2003: Cheng etal.. 2000: Perera and Weinstein. 2000: Wei etal..
25 2000: Amos etal., 1999]. In general, however, available support for the role of single
26 polymorphisms in significantly modulating human PAH cancer risk from benzo[a]pyrene or other
27 PAHs is relatively weak or inconsistent. Combinations of polymorphisms, on the other hand, may
28 be critical determinants of a cumulative DNA-damaging dose, and thus indicate greater
29 susceptibility to cancer from benzo[a]pyrene exposure [Vineis etal., 2007].
30
31 Analysis of Toxicogenomics Data
32 An analysis of pathway-based transcriptomic data was conducted to help inform the cancer
33 mode of action for benzo[a]pyrene (see the Supplemental Information for details of this analysis].
34 These data support a mutagenic and cellular proliferation mode of action that follows three
35 candidate pathways: aryl hydrocarbon signaling; DNA damage regulation of the Gl/S phase
36 transition; and/or Nrf2 regulation of oxidative stress. Specifically, the analysis showed that
37 benzo[a]pyrene may activate the AhR, leading to the formation of oxidative metabolites and
38 radicals which may lead to oxidative damage and DNA damage. Subsequently, DNA damage can
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 occur and activate p53 andp53 target genes, including p21 andMDM2. In addition, the data
2 indicate that p53 signaling may be decreased under these conditions, as ubiquitin and MDM2 are
3 both upregulated, and work together to degrade p53. Furthermore, the transcriptional
4 upregulation of cyclin D may result in enough cyclin D protein to overcome the p21 inhibitory
5 competition for CDK4, allowing for Gl/S phase transition to occur. The data also supports the
6 hypothesis that an upregulation of PCNA in combination with the upregulation of ubiquitin
7 indicates that cells are moving towards the Gl/S phase transition. Although the alterations to the
8 Nrf2 pathway suggest cells are preparing for a pro-apoptotic environment, there is no
9 transcriptional evidence that the apoptotic pathways are being activated.
10 There are uncertainties associated with the available transcriptomics data. For instance,
11 the available studies only evaluate gene expression following benzo[a]pyrene exposure and do not
12 monitor changes in protein or metabolite expression, which would be more indicative of an actual
13 cellular state change. Further research is required at the molecular level to demonstrate that the
14 cellular signaling events being inferred from such data are actually operative and result in
15 phenotypic changes. In addition, this analysis relied upon two short term studies that evaluated
16 mRNA expression levels in a single tissue (liver) and species (mouse) and were conducted at
17 relatively high doses.
18 1.2. SUMMARY AND EVALUATION
19 1.2.1. Weight of Evidence for Effects Other than Cancer
20 The weight of the evidence from human and animal studies indicates that the strongest
21 evidence for potential hazard following benzo[a]pyrene exposure is for developmental and
22 reproductive toxicity and immunotoxicity. In humans, exposure to PAH mixtures has been shown
23 to result in developmental and reproductive toxicity and immunotoxicity. Most of the available
24 human data on benzo[a]pyrene report associations between particular health endpoints and
25 concentrations of benzo[a]pyrene-DNA adducts, with fewer noncancer studies correlating health
26 effects with external measures of exposure. The available human studies report effects that are
27 generally analogous to the effects observed in animal toxicological studies, and provide qualitative,
28 supportive evidence for the effect-specific hazards identified in Section 1.1.1 to 1.1..4.
29 In animals, evidence of developmental and reproductive toxicity and immunotoxicity has
30 been observed across species and dosing regimens. The available evidence from mice and rats
31 treated by gavage during gestation or in the early postnatal period demonstrate developmental
32 effects including decreased body weight, decreased fetal survival, decreased fertility, atrophy of
33 reproductive organs, and altered neurobehavioral outcomes (Chenetal., 2012: Tules etal., 2012:
34 Bouayedetal.. 2009a: Kristensenetal.. 1995: Mackenzie and Angevine. 1981). Male and female
35 reproductive toxicity, as evidenced by effects on sperm parameters, decreased reproductive organ
36 weights, histological changes, and hormone alterations, have been observed after oral exposure in
37 rats and mice (Chenetal.. 2011: Chung etal.. 2011: Mohamed etal.. 2010: Zheng etal.. 2010:
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 Mackenzie and Angevine. 1981]. Benzo[a]pyrene exposure has also been shown to lead to altered
2 immune cell populations and histopathological changes in immune system organs [Kroese etal..
3 2001: De long et al.. 1999). as well as thymic and splenic effects following subchronic oral exposure.
4 Varying immunosuppressive responses are also observed in short term oral and injection studies.
5 The weight of the evidence indicates that developmental toxicity, reproductive toxicity and
6 immunotoxicity are hazards following oral exposure to benzo[a]pyrene.
7 Following inhalation exposure to benzo[a]pyrene in animals, evidence of developmental
8 and reproductive toxicity has been observed. Decreased fetal survival has been observed in rats
9 exposed to benzo[a]pyrene via inhalation during gestation [Wormley et al.. 2004: Archibongetal..
10 2002). Male reproductive toxicity, as evidenced by effects on sperm parameters, decreased testes
11 weight, and hormone alterations, has also been observed in rats following subchronic inhalation
12 exposure to benzo[a]pyrene [Archibong et al., 2008: Rameshetal., 2008]. Female reproductive
13 toxicity, as evidenced by modified hormone levels in dams, has been observed following inhalation
14 exposure to benzo[a]pyrene during gestation [Archibongetal., 2002]. The weight of the evidence
15 indicates that developmental toxicity and reproductive toxicity are hazards following inhalation
16 exposure to benzo[a]pyrene.
17 Forestomach hyperplasia was observed following oral and inhalation exposure; however,
18 this endpoint most likely reflects early events in the neoplastic progression of forestomach tumors
19 following benzo[a]pyrene exposure (see Section 1.1.4], and was not considered further for dose-
20 response analysis and the derivation of reference values.
21 1.2.2. Weight of Evidence for Carcinogenicity
22 Under EPA's Guidelines for Carcinogen Risk Assessment [U.S. EPA. 2005b]. benzo[a]pyrene is
23 "carcinogenic to humans." EPA's Cancer Guidelines [U.S. EPA, 2005b] emphasize the importance of
24 weighing all of the evidence in reaching conclusions about human carcinogenic potential. The
25 descriptor of "carcinogenic to humans" can be used when the following conditions are met:
26 (a] there is strong evidence of an association between human exposure and either cancer or the key
27 precursor events of the agent's mode of action but not enough for a causal association, (b] there is
28 extensive evidence of carcinogenicity in animals, (c] the mode or modes of carcinogenic action and
29 associated key precursor events have been identified in animals, and (d] there is strong evidence
30 that the key precursor events that precede the cancer response in animals are anticipated to occur
31 in humans and progress to tumors, based on available biological information. The data supporting
32 these four conditions for benzo[a]pyrene are presented below and in Table 1-18.
33 a) Strong human evidence of cancer or its precursors
34 There is a large body of evidence for human carcinogenicity for complex PAH mixtures
35 containing benzo[a]pyrene, including soot, coal tars, coal-tar pitch, mineral oils, shale oils, and
36 smoke from domestic coal burning [IARC. 2010: Baanetal.. 2009]. There is also evidence of
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 carcinogenicity, primarily of the lung and skin, in occupations involving exposure to PAH mixtures
2 containing benzo[a]pyrene, such as chimney sweeping, coal gasification, coal-tar distillation, coke
3 production, iron and steel founding, aluminum production, and paving and roofing with coal tar
4 pitch [IARC, 2010: Baanetal., 2009: Straif etal., 2005]. Increased cancer risks have been reported
5 among other occupations involving exposure to PAH mixtures such as carbon black and diesel
6 exhaust [Benbrahim-Tallaa etal., 2012: Bosettietal., 2007]. There is extensive evidence of the
7 carcinogenicity of tobacco smoke, of which benzo[a]pyrene is a notable constituent. The
8 methodologically strongest epidemiology studies (in terms of exposure assessment, sample size,
9 and follow-up period] provide consistent evidence of a strong association between benzo[a]pyrene
10 exposure and lung cancer. Three large epidemiology studies in different geographic areas,
11 representing two different industries, observed increasing risks of lung cancer with increasing
12 cumulative exposure to benzo[a]pyrene (measured in |ig/m3-years], with approximately a 2-fold
13 increased risk at the higher exposures; each of these studies addressed potential confounding by
14 smoking (Armstrong and Gibbs. 2009: Spinelli etal.. 2006: Xuetal.. 1996] (Table 1-11]. Although
15 the relative contributions of benzo[a]pyrene and of other PAHs cannot be established, the
16 exposure-response patterns seen with the benzo[a]pyrene measures make it unlikely that these
17 results represent confounding by other exposures. Similarly, for bladder cancer, two of the three
18 cohort studies with detailed exposure data observed an increasing risk with exposures above >80
19 [ig/m3-years(Gibbs and Sevigny. 2007a: Gibbs etal.. 2007: Gibbs and Sevigny. 2007b: Spinelli etal..
20 2006] (Table 1-13]. The exposure range was much lower in the third study (Burstyn etal.. 2007:
21 Gibbs and Sevigny. 2007a: Gibbs etal.. 2007: Gibbs and Sevigny. 2007b]. such that the highest
22 exposure group only reached the level of exposure seen in the lowest exposure categories in the
23 other studies. Data pertaining to non-melanoma skin cancer is limited to studies with more indirect
24 exposure measures, e.g., based on occupations with likely dermal exposure to creosote (i.e., timber
25 workers, brick makers, and power linesmen]; the relative risk estimates seen in the four available
26 studies that provide risk estimates for this type of cancer ranged from 1.5 to 4.6, with three of these
27 four estimates greater than 2.5 and statistically significant (Pukkala, 1995: Karlehagenetal., 1992:
28 Tornqvistetal., 1986: Hammond etal., 1976]. These four studies provide support for the
29 association between dermal PAH exposure, including benzo[a]pyrene exposure, and skin cancer.
30 Although it is likely that multiple carcinogens present in PAH mixtures contribute to the
31 carcinogenic responses, strong evidence is available from several studies of humans exposed to
32 PAH mixtures supporting a contributing role for benzo[a]pyrene diol epoxide in inducing key
33 mutagenic precursor cancer events in target tissues. Elevated BPDE-DNA adducts have been
34 reported in smokers compared to non-smokers, and the increased adduct levels in smokers are
35 typically increased twofold compared with non-smokers (Phillips, 2002]. Elevated BPDE-DNA
36 adduct levels have been observed in WBCs of groups of coke oven workers and chimney sweeps,
37 occupations with known elevated risks of cancer (Rojas etal.. 2000: Bartschetal.. 1999: Pavanello
38 etal.. 1999: Bartschetal.. 1998: Rojas etal.. 1998]. and in lung tissue from tobacco smokers with
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 lung cancer (Rojas etal.. 2004: Godschalketal.. 2002: Bartsch etal.. 1999: Rojas etal.. 1998:
2 Andreassen et al.. 1996: Alexandrovetal.. 1992).
3 Mutation spectra distinctive to diol epoxides have been observed in the tumor suppressor
4 gene p53 and the K-ras oncogene in tumor tissues taken from lung cancer patients who were
5 chronically exposed to two significant sources of PAH mixtures: coal smoke and tobacco smoke.
6 Hackmanetal. [2000] reported an increase of GC—>TA transversions and a decrease of GC—>AT
7 transitions at the hprt locus in T-lymphocytes of humans with lung cancer who were smokers
8 compared to non-smokers. Lung tumors from cancer patients exposed to emissions from burning
9 smoky coal showed mutations in p53 and K-ras that were primarily G—>T transversions (76 and
10 86%, respectively) [DeMarini etal.. 2001). [Keohavongetal.. 2003] investigated the K-ras
11 mutational spectra from non-smoking women and smoking men chronically exposed to emissions
12 from burning smoky coal, and smoking men who resided in homes using natural gas; among those
13 with K-ras mutations, 67, 86, and 67%, respectively, were G^T transversions. Lung tumors from
14 tobacco smokers showed a higher frequency of p53 mutations that were G—>T transversions
15 compared with lung tumors in non-smokers [Pfeifer and Hainaut, 2003: Pfeifer etal., 2002: Hainaut
16 and Pfeifer. 2001]. and the frequency of these types of p53 mutations in lung tumors from smokers
17 increased with increasing smoking intensity [Bennett etal.. 1999].
18 Similarly, investigations of mutagenesis following specific exposures to benzo[a]pyrene (as
19 opposed to PAH mixtures] have consistently observed that the benzo[a]pyrene diol epoxide is very
20 reactive with guanine bases in DNA, and that G^T transversions are the predominant type of
21 mutations caused by benzo[a]pyrene diol epoxide in several biological test (Pfeifer and Hainaut,
22 2003: Hainaut and Pfeifer, 2001]. Following treatment of human HeLa cells with benzo[a]pyrene
23 diol epoxide, Denissenko etal. (1996] reported that the distribution of BPDE-DNA adducts within
24 p53 corresponded to mutational hotspots observed in p53 in human lung cancers. Benzo[a]pyrene
25 exposure induced mutations in embryonic fibroblasts from human p53 "knock-in" mice that were
26 similar to those found in smoking related human cancers, with a predominance of G^T
27 transversions that displayed strand bias and were also located in the same mutational hotspots
28 found in p53 in human lung tumors (Liu etal., 2005]. These results, combined with a mechanistic
29 understanding that mutations in p53 (which encodes a key transcription factor in DNA repair and
30 regulation of cell cycle and apoptosis] may be involved in the initiation phase of many types of
31 cancer, are consistent with a common mechanism for mutagenesis following exposures to PAH
32 mixtures and provide evidence of a contributing role of benzo[a]pyrene diol epoxide in the
33 carcinogenic response of humans to coal smoke and tobacco smoke.
34 Therefore, while the epidemiological evidence alone does not establish a causal association
35 between human exposure and cancer, there is strong evidence that the key precursor events of
36 benzo[a]pyrene's mode of action are likely to be associated with tumor formation in humans.
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 b) Extensive animal evidence
2 In laboratory animals (rats, mice, and hamsters), exposures to benzo[a]pyrene via the oral,
3 inhalation, and dermal routes have been associated with carcinogenic responses both systemically
4 and at the site of administration. Three 2-year oral bioassays are available that associate lifetime
5 benzo[a]pyrene exposure with carcinogenicity at multiple sites. These bioassays observed
6 forestomach, liver, oral cavity, jejunum, kidney, auditory canal (Zymbal gland), and skin or
7 mammary gland tumors in male and female Wistar rats [Kroese etal., 2001): forestomach tumors
8 in male and female Sprague-Dawley rats [Brune etal., 1981): and forestomach, esophagus, tongue,
9 and larynx tumors in female B6C3Fi mice [Beland and Gulp. 1998: Gulp etal.. 1998). Repeated or
10 short-term oral exposure to benzo[a]pyrene was associated with forestomach tumors in additional
11 bioassays with several strains of mice [Weyandetal., 1995: Benjamin et al., 1988: Robinson etal.,
12 1987: El-Bayoumy. 1985: Triolo etal.. 1977: Wattenberg. 1974: Roe etal.. 1970: Biancifiori etal..
13 1967: Chouroulinkovetal.. 1967: Fedorenko andYansheva. 1967: Neal andRigdon. 1967:
14 Berenblum and Haran. 1955). EPA has considered the uncertainty associated with the relevance of
15 forestomach tumors for estimating human risk from benzo[a]pyrene exposure. While humans do
16 not have a forestomach, squamous epithelial tissue similar to that seen in the rodent forestomach
17 exists in the oral cavity and upper two-thirds of the esophagus in humans [IARC. 2003: Wester and
18 Kroes, 1988). Human studies, specifically associating exposure to benzo[a]pyrene with the
19 alimentary tract tumors are not currently available. However, benzo[a]pyrene-DNA adducts have
20 been detected in oral and esophageal tissue obtained from smokers [reviewed by Phillips, 2002)
21 and several epidemiological studies have identified increased exposure to PAHs as an independent
22 risk factor for esophageal cancer [Abedi-Ardekani et al., 2010: Szymanska et al., 2010: Gustavsson
23 etal.. 1998: Liu etal.. 1997). Thus, EPA concluded that forestomach tumors in rodents are relevant
24 for assessing the carcinogenic risk to humans.
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 Chronic inhalation exposure to benzo[a]pyrene was associated primarily with tumors in the
2 larynx and pharynx of male Syrian golden hamsters exposed to benzo[a]pyrene:NaCl aerosols
3 [Thyssen et al.. 1981). Additionally, less-than-lifetime oral exposure cancer bioassays in mice
4 provide supporting evidence that exposure to benzo[a]pyrene is associated with an increased
5 incidence of lung tumors in mice [Weyandetal., 1995: Robinson etal., 1987: Wattenberg, 1974]. In
6 additional studies with hamsters, intratracheal instillation of benzo[a]pyrene was associated with
7 upper and lower respiratory tract tumors [Feron and Kruysse, 1978: Ketkar etal., 1978: Feron et
8 al.. 1973: Henry etal.. 1973: Saffiotti etal.. 1972]. Chronic dermal application of benzo[a]pyrene
9 (2-3 times/week] has been associated with mouse skin tumors in numerous bioassays [Sivaketal..
10 1997: Grimmer etal.. 1984: HabsetaL 1984: Grimmer etal.. 1983: Habsetal.. 1980: SchmahletaL
11 1977: Schmidt etal.. 1973: Roe etal.. 1970: Poel. 1960.1959]. Skin tumors in rats, rabbits, and
12 guinea pigs have also been associated with repeated application of benzo[a]pyrene to skin in the
13 absence of exogenous promoters [IPCS. 1998: ATSDR. 1995: IARC. 1983.1973]. When followed by
14 repeated exposure to a potent tumor promoter, acute dermal exposure to benzo[a]pyrene induced
15 skin tumors in numerous studies of mice, indicating that benzo[a]pyrene is a strong tumor-
16 initiating agent in the mouse skin model [Weyandetal.. 1992: Cavalierietal.. 1991: Rice etal..
17 1985: El-Bayoumyetal.. 1982: Lavoieetal.. 1982: Ravehetal.. 1982: Cavalierietal.. 1981: Slagaet
18 al.. 1980: Wood etal.. 1980: Slaga etal.. 1978: Hoffmann etal.. 1972].
19 Carcinogenic responses in animals exposed to benzo[a]pyrene by other routes of
20 administration include: (1] liver or lung tumors in newborn mice given acute postnatal i.p.
21 inj ections [Lavoieetal.. 1994: Busby etal.. 1989: Weyand and Lavoie. 1988: Lavoie etal.. 1987:
22 Wislocki etal.. 1986: Busby etal.. 1984: Buening etal.. 1978: Kapitulnik etal.. 1978]: (2] increased
23 lung tumor multiplicity in A/J adult mice given single i.p. injections [Mass etal.. 1993]: (3] injection
24 site tumors in mice following s.c. injection [Nikonova. 1977: Pfeiffer. 1977: Homburger etal.. 1972:
25 Roe and Waters. 1967: GrantandRoe. 1963: Steiner. 1955: Rask-Nielsen. 1950: Pfeiffer and Allen.
26 1948: Bryan and Shimkin, 1943: Barry etal., 1935]: (4] injection site sarcomas in mice following
27 intramuscular injection[Sugiyama, 1973]: (5] mammary tumors in rats with intramammilary
28 administration [Cavalierietal., 1991: Cavalieri et al., 1988c: Cavalierietal., 1988b: Cavalieri etal.,
29 1988a]: (6] cervical tumors in mice with intravaginal application [Naslundetal., 1987]: and
30 (7] tracheal tumors in rats with intratracheal implantation [Toppingetal.. 1981: Nettesheim etal..
31 19771
32 Therefore, the animal database provides extensive evidence of carcinogenicity in animals.
33 c) Identification of key precursor events have been identified in animals
34 There is sufficient evidence to conclude that benzo[a]pyrene carcinogenicity involves a
35 mutagenic mode of action mediated by DNA reactive metabolites. The benzo[a]pyrene database
36 provides strong and consistent evidence for BPDE-induced mutations associated with tumor
37 initiation in cancer tissue from humans exposed to complex mixtures containing benzo[a]pyrene, 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 animals exposed to benzo[a]pyrene, and in in vitro systems. Other processes may contribute to the
2 carcinogenicity of benzo[a]pyrene via the promotion and progression phases of cancer
3 development (e.g., inflammation, cytotoxicity, sustained regenerative cell proliferation, anti-
4 apoptotic signaling), but the available evidence best supports a mutagenic mode of action as the
5 primary mode by which benzo [a] pyrene acts.
6 d) Strong evidence that the key precursor events are anticipated to occur in humans
7 Mutations in p53 and ras oncogenes have been observed in tumors from mice exposed to
8 benzo[a]pyrene in the diet [Gulp etal.. 2000] or by i.p. injection [Nesnow et al.. 1998a: Nesnow et
9 al., 1998b: Nesnow etal., 1996,1995: Mass etal., 1993]. Mutations in these same genes have also
10 been reported in lung tumors of human cancer patients, bearing distinctive mutation spectra (G^T
11 transversions] that correlate with exposures to coal smoke [Keohavongetal., 2003: DeMarini etal.,
12 2001] or tobacco smoke [Pfeifer and Hainaut. 2003: Pfeifer etal.. 2002: Hainautand Pfeifer. 2001:
13 Bennett etal.. 1999].
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
2
Table 1-18. Supporting evidence for the carcinogenic to humans cancer
descriptor for benzo[a]pyrene
Evidence
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
and chimney sweeps
- BPDE-DNAadducts in smokers
• Benzo[a]pyrene-specific DNA adducts have been
detected in target tissues in humans exposed to PAH
mixtures
cigarette smokers with lung cancer and in skin
eczema patients treated with coal tar
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
- 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 non-smokers
b) Extensive animal evidence
Reference
IARC (2010, 2004); Secretan et al. (2009); Baan et al.
(2009); Benbrahim-Tallaa etal. (2012)
Rnias pt al l~)C\r\r\\- RarKrh pt al MQQQ11 Pauanplln pt al
(1999); Bartsch etal. (1998); Rojas et al. (1998)
Phillips (2002)
Rniac pt al (1C\C\A.\m f^nrlcrhalk pt al OHO?^1 Rartcrh pt al
(1999); Godschalk et al. (1998b); Roias et al. (1998);
Andreassen et al. (1996); Alexandrov et al. (1992)
npniQQpnkn pt al MQQfi^'PiiiQipiiYPtal M QQ'M
Hackman et al. (2000)
Liu et al. (2005)
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,
1-87 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
Evidence
Reference
Oral exposures
• Forestomach tumors in male and female rats and in
female mice following chronic exposure
• Forestomach tumors in mice following less-than-
lifetime exposures
• Alimentary tract and liver tumors in male and female
rats following chronic exposure
• Kidney tumors in male rats following chronic
exposure
• Auditory canal tumors in male and female rats
following chronic exposure
• Esophageal, tongue, and laryngeal tumors in female
mice following chronic exposure
• Lung tumors in mice following less-than-lifetime
exposure
Kroese et al. (2001); Beland and Gulp (1998); Gulp et al.
(1998); Bruneetal. (1981)
Wevand 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)
Wevand et al. (1995); Robinson et al. (1987); Wattenberg
(1974)
Inhalation exposures
• Upper respiratory tract tumors in male hamsters
following chronic exposure
Thvssenetal. (1981)
Dermal exposures
• Skin tumors in mice following chronic exposures
without a promoter or acute exposures with 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
(1960, 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; Wevand 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
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
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
- 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 non-smokers
Gulp et al. (2000); Nesnow et al. (1998a); Nesnow et al.
(1998b); Nesnow et al. (1996, 1995); Massetal.
Keohavong et al. (2003); DeMarini et al. (2001)
Pfeifer and Hainaut (2003); Pfeifer et al. (2002);
and Pfeifer (2001); Bennett et al. (1999)
(1993)
Hainaut
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
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 percent 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 hazards of benzo[a]pyrene exposure by the oral route. Studies within each effect
14 category were evaluated using general study quality characteristics (as discussed in Section 6 of the
15 Preamble) to help inform the selection of studies from which to derive toxicity values. Rationales
16 for selecting the studies and effects to represent each of these hazards are summarized below.
17 Human studies are preferred over animal studies when quantitative measures of exposure
18 are reported and the reported effects are determined to be associated with exposure. For
19 benzo[a]pyrene, human studies of environmental PAH mixtures across multiple cohorts have
20 observed effects following exposure to complex mixtures of PAHs. The available data suggest that
21 benzo[a]pyrene exposure may pose health hazards other than cancer including reproductive and
22 developmental effects such as infertility, miscarriage, and reduced birth weight (Wuetal., 2010:
23 Nealetal..20Q8: Tangetal.. 2008: Pereraetal.. 2005b: Pereraetal.. 2005a) and cardiovascular
24 effects (Friesenetal., 2010: Burstynetal., 2005). However, the available human studies that
25 utilized benzo[a]pyrene-DNA adducts as the exposure metric do not provide external exposure
26 levels of benzo[a]pyrene from which to derive a value, and exposure is likely to have occurred by
27 multiple routes. In addition, uncertainty exists due to concurrent exposure to other PAHs and other
28 components of the mixture (such as metals).
29 Animal studies were evaluated to determine which provided the most relevant routes and
30 durations of exposure; multiple exposure levels to provide information about the shape of the dose-
31 response curve; and power to detect effects at low exposure levels (U.S. EPA, 2002). The oral
32 database for benzo[a]pyrene includes a variety of studies and datasets that are suitable for use in
33 deriving reference values. Specifically, chronic effects associated with benzo[a]pyrene exposure in
34 animals include observations of organ weight and histological changes and hematological
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 parameters observed in several oral cancer bioassays [Kroese etal.. 2001: Beland and Gulp. 1998].
2 Multiple subchronic studies are available which characterize a variety of effects other than cancer.
3 In addition, several developmental studies are available which help inform hazards of exposure
4 during sensitive developmental windows.
5 Developmental Toxicity
6 Numerous animal studies observed endpoints of developmental toxicity following oral
7 exposure during gestational or early postnatal development [Chen etal., 2012: Tules etal., 2012:
8 Bouayedetal.. 2009a: Kristensenetal.. 1995: Mackenzie and Angevine. 1981] and were considered
9 for dose response analysis based on the above criteria. Kristensenetal. [1995], with only one dose
10 group, was not considered further given its concordance with Mackenzie and Angevine [1981],
11 which had multiple groups. From the remaining studies demonstrating developmental toxicity, the
12 studies conducted by Chen etal. [2012] and Tules etal. [2012] were identified as the most
13 informative studies for dose-response analysis. The neurodevelopmental study by Chen et al.
14 [2012] was a well-designed and well-conducted study that evaluated multiple neurobehavioral
15 endpoints and measures of neurotoxicity in adolescent and adult rats. The study randomly
16 assigned a total of 10 male and 10 female pups per treatment group, with no more than one male
17 and one female from each litter for behavioral testing. In addition, the pups were cross-fostered
18 with dams being rotated among litters every 2-3 days to distribute any maternal caretaking
19 differences randomly across litters and treatment groups.
20 Chen etal. [2012] observed increased latency in negative geotaxis, increased motor activity
21 in the open field test, decreased anxiety-like behaviors in the elevated plus maze test, and impaired
22 performance in the Morris water maze test as measured by an increase in latency time to find a
23 hidden platform. Chen etal. [2012] also observed increased latencies for treated pups to right
24 themselves in the surface righting test and the negative geotaxis test The data from the open field
25 test, negative geotaxis test, and surface righting test were considered less informative as male and
26 female data were pooled (male and female rats sometimes show differences in the maturation of
27 these developmental landmarks following challenge]. In addition, 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]. EPA considered the elevated plus maze and Morris water
32 maze tests (which did report responses for each sex] to be the most informative and appropriate
33 measures of neurobehavioral function performed by Chen etal. [2012].
34 Significant, dose-related effects were reported in an established test of spatial learning and
35 memory (Morris water maze]. Specifically, increased escape latency in each of four hidden
36 platform trial days and decreased time spent in the target quadrant during a probe trial were
37 observed following benzo[a]pyrene exposure. Due to the altered baseline performance of treated
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 animals on day 1 of the hidden platform trials these findings cannot be specifically attributed to
2 impaired learning. In fact, the slopes of the lines across trial days are nearly identical for the
3 treatment groups suggesting the lack of an effect on learning. The impaired Morris water maze
4 performance of treated animals could be due to effects on several other components of neurological
5 function besides learning, including anxiety, vision, and locomotion. Similarly, as escape latencies
6 were not comparable across groups after learning acquisition (i.e., the end of the hidden platform
7 trials), differences in probe trial performance are likely to be influenced by factors other than
8 impaired memory retention. As a result, although they identify significant effects of
9 benzo[a]pyrene exposure, the Morris water maze data were considered less informative than the
10 results from the elevated plus maze test. Chenetal. [2012] reported an increase in the number of
11 open arm entries in the elevated plus maze test, an indicator of decreased anxiety-like behavior.
12 These results indicate effects on a single, discreet neurological function which are unlikely to be
13 complicated by changes in other processes such as motor activity (total arm entries calculated by
14 summing open and closed arm entries were unchanged with treatment). This finding is considered
15 adverse, is supported by similar observations in developing (Bouayed et al., 2009a] and adult
16 (Grova etal.. 2008] mice, and may be indirectly related to observations of increased aggression in
17 mice (Bouayed etal.. 2009b] as well as attention and anxiety problems in PAH-exposed children
18 (Pereraetal.. 2012b).
19 Tules etal. (2012] was also identified for dose-response analysis. This study was of
20 sufficient duration, utilized multiple doses, did not observe maternal toxicity, and evaluated
21 multiple cardiovascular endpoints. The study authors reported increases in both systolic
22 (approximately 20-50%) and diastolic (approximately 33-83%) pressure and heart rate in adult
23 rats that were exposed gestationally to benzo[a]pyrene. A limitation of this study is that the
24 authors only reported effects at the two highest doses. However, given the magnitude of the
25 response and the appearance of these effects in adulthood following gestational exposure, these
26 endpoints were selected for dose-response analysis because of their sensitivity and biological
27 plausibility.
28 Bouayed etal. (2009a) and Mackenzie and Ange vine (1981] were not selected for dose-
29 response analysis. Bouayed etal. (2009a) used the same tests as Chen etal. (2012], but at higher
30 doses (2 and 20 mg/kg-day compared to 0.02, 0.2, and 2 mg/kg-day, respectively). Similarly,
31 Mackenzie and Ange vine (1981) demonstrated developmental effects in a multi-dose study with
32 relevant routes and durations of exposure; however, the doses studied (10-160 mg/kg-day) were
33 much higher than those evaluated in other developmental toxicity studies (Chenetal., 2012: Tules
34 etal.. 2012).
35 Reproductive Toxicity
36 Male reproductive toxicity was demonstrated in numerous subchronic studies (Chenetal..
37 2011: Chung etal.. 2011: Mohamedetal.. 2010: Zheng etal.. 2010]. Chung etal. (2011] was not
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 included in the dose-response analysis because numerical data were not reported or only reported
2 for the mid-dose of three doses. Chenetal. [2011] is a subchronic study that applied only a single
3 dose level. This study corroborated other available multi-dose studies and is considered
4 supportive, but it was not considered for dose-response analysis due to the limited reporting of
5 numerical data. The studies conducted by Mohamedetal. [2010] and Zheng etal. [2010] were
6 identified as the most informative male reproductive toxicity studies for dose-response analysis.
7 Decreased sperm count and motility observed by Mohamedetal. [2010] and decreased
8 intratesticular testosterone levels observed by Zheng etal. [2010] were selected for dose-response
9 analysis as both represent sensitive endpoints of male reproductive toxicity and are indicators of
10 potentially decreased fertility. These effects are also consistent with human studies in PAH
11 exposed populations as effects on male fertility and semen quality have been demonstrated in
12 epidemiological studies of smokers [reviewed by Spares and Melo, 2008].
13 Female reproductive toxicity was demonstrated in two subchronic studies [Gao etal.,
14 2011a: Xu etal., 2010]. Specifically, Xu et al., 2010 demonstrated altered ovary weights and follicle
15 numbers, and Gao etal. [201 la] demonstrated cervical epithelial cell hyperplasia following oral
16 exposure to benzo[a]pyrene. These studies were identified as the most informative studies on
17 female reproductive toxicity for dose-response analysis. Gao etal. [201 la] identified statistically-
18 significant, dose-related increases in the incidence of cervical inflammatory cells in mice exposed to
19 low doses ofbenzo[a]pyrene for 98 days [Gao etal., 2 01 la: Gao etal., 2010]. Cervical effects of
20 increasing severity (including epithelial hyperplasia, atypical hyperplasia, apoptosis, and necrosis]
21 were also observed at higher doses [Gao etal., 2011a: Gao etal., 2010]. There are no data on
22 cervical effects in other species or in other mouse strains. However, Gao etal. [2 01 la] also
23 evaluated cervical effects in separate groups of mice exposed via i.p. injection, and observed similar
24 responses in these groups of mice, providing support for the association between effects in this
25 target organ and benzo[a]pyrene exposure. Epidemiological studies have demonstrated an
26 association between cigarette smoking and increased risk of cervical cancer [Pate Capps etal.,
27 2009]. In addition, benzo[a]pyrene metabolites and benzo[a]pyrene-DNA adducts have been
28 detected in human cervical mucus and cervical tissues obtained from smokers [Phillips, 2002:
29 Melikianetal., 1999]. However, data to support that cervical hyperplasia following oral
30 benzo[a]pyrene exposure progresses to cervical tumors were not available (no cervical tumors
31 were noted in the two available chronic oral cancer bioassays]. Thus, in the absence of these data,
32 cervical hyperplasia is presented as a noncancer effect.
33 Xu etal. [2010] identified biologically and statistically significant decreases in ovary weight,
34 estrogen, and primordial follicles, and altered estrus cycling in treated animals. These reductions in
35 female reproductive parameters are supported by a large database of animal studies indicating that
36 benzo[a]pyrene is ovotoxic with effects including decreased ovary weight, decreased primordial
37 follicles, and reduced fertility fBorman etal.. 2000: Kristensen etal.. 1995: Miller etal.. 1992:
38 Swartz and Mattison. 1985: Mackenzie and Angevine. 1981: Mattison etal.. 1980]. Additionally,
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 epidemiology studies indicate that exposure to complex mixtures of PAHs, such as through cigarette
2 smoke, is associated with measures of decreased fertility in humans [Neal etal.. 2008: El-Nemr et
3 al.. 1998). Specific associations have also been made between infertility and increased levels of
4 benzo[a]pyrene in follicular fluid in women undergoing in vitro fertilization [Neal etal., 2008].
5 Immunotoxicity
6 As described in Section 1.1.3, the immune system was identified as a target of
7 benzo[a]pyrene-induced toxicity based on findings of organ weight and immunoglobulin
8 alterations, as well as effects on cellularity and functional changes in the immune system in animals.
9 The only available studies to support development of an RfD were conducted by Kroese et al.
10 [2001] and De long et al. [1999]. These are subchronic studies with multiple exposure levels and
11 adequate power to detect effects. In comparing these studies, the Kroese etal. [2001] study is
12 preferred for dose-response analysis due to its longer duration (90 days].
13 Decreased thymus weight, observed in Kroese etal. [2001], decreased IgM and IgA levels,
14 and decreased relative numbers of B-cells, observed in De long etal. [1999]. were selected for dose-
15 response analysis. It is recognized that thymus weight changes on their own have been noted to be
16 less reliable indicators of immunotoxicity [Luster etal., 1992]. However, there are converging lines
17 of evidence that support the derivation of an organ/system-specific RfD for benzo[a]pyrene
18 immunotoxicity. Alterations in immunoglobulin levels have been noted in humans after exposure to
19 PAHs, as well as in animal studies after exposure to benzo[a]pyrene. Changes in B cell populations
20 in the spleen provide additional evidence of immunotoxicity. Finally, functional effects on the
21 immune system, including dose-related decreases in SRBC-specific IgM levels and dose-dependent
22 decreases in resistance to pneumonia or Herpes simplex type 2 following short-term s.c. injection
23 have been reported [Temple etal., 1993: Munsonetal., 1985]. The observed decreases in thymus
24 weight, IgM and IgA levels, and number of B cells associated with exposure to benzo[a]pyrene were
25 concluded to be representative of immunotoxicity following benzo[a]pyrene exposure and were
26 selected for dose-response analysis.
27 2.1.2. Methods of Analysis
28 No biologically based dose-response models are available for benzo[a]pyrene. In this
29 situation, EPA evaluates a range of dose-response models thought to be consistent with underlying
30 biological processes to determine how best to empirically model the dose-response relationship in
31 the range of the observed data. Consistent with this approach, all models available in EPA's
32 Benchmark Dose Software [BMDS] were evaluated. Consistent with EPA's Benchmark Dose
33 Technical Guidance Document [U.S. EPA. 2012b], the benchmark dose [BMD] and the 95% lower
34 confidence limit on the BMD [BMDL] were estimated using a benchmark response [BMR] of
35 1 standard deviation [SD] from the control mean for continuous data or a BMR of 10% extra risk for
36 dichotomous data in the absence of information regarding what level of change is considered
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 biologically significant, and also to facilitate a consistent basis of comparison across endpoints,
2 studies, and assessments. The estimated BMDLs were used as points of departure (PODs). Further
3 details including the modeling output and graphical results for the best fit model for each endpoint
4 can be found in Appendix E of the Supplemental Information.
5 Among the endpoints identified as representative of the hazards of benzo[a]pyrene
6 exposure, the data for neurobehavioral changes in the elevated plus maze and Morris water maze
7 tests [Chenetal., 2012], decreased ovary weight [Xu etal., 2010], increased cervical hyperplasia
8 [Gao etal. [201 la], and decreased thymus weight [Kroese etal., 2001] were amenable to dose-
9 response modeling. For the water maze escape latency data, the data for male and female rats were
10 combined for dose-response analysis because of the strong similarity in responses and the lack of
11 information available suggesting there would be sex-specific differences in the results of this test
12 (see Appendix E of the Supplemental Information for details of statistical analyses].
13 The data for the remaining endpoints identified in Section 2.1.1 were not modeled.
14 Specifically, the data for cardiovascular effects observed in Jules etal. [2012] were limited due to
15 the reporting of results at only the two highest dose groups. The data for epididymal sperm counts
16 presented in the Mohamedetal. [2010] study were reported graphically only and requests for the
17 raw data were unsuccessful. The observed decrease in IgM and IgA [De long et al.. 1999] was
18 inconsistent and not amenable to dose-response modeling. NOAELs or LOAELs were used as the
19 POD for these endpoints.
20 Human equivalent doses [HEDs] for oral exposures were derived from the PODs estimated
21 from the laboratory animal data as described in EPA's Recommended Use of Body Weight3/4 as the
22 Default Method in Derivation of the Oral Reference Dose [U.S. EPA. 2011]. In this guidance, EPA
23 advocates a hierarchy of approaches for deriving HEDs from data in laboratory animals, with the
24 preferred approach being physiologically-based toxicokinetic modeling. Other approaches can
25 include using chemical-specific information in the absence of a complete physiologically-based
26 toxicokinetic model. As discussed in Appendix D of the Supplemental Information, several animal
27 physiologically based pharmacokinetic [PBPK] models for benzo[a]pyrene have been developed
28 and published, but a validated human PBPK model for benzo[a]pyrene for extrapolating doses from
29 animals to humans is not available. In lieu of either chemical-specific models or data to inform the
30 derivation of human equivalent oral exposures, a body weight scaling to the % power (i.e., BW3/4]
31 approach is applied to extrapolate toxicologically equivalent doses of orally administered agents
32 from adult laboratory animals to adult humans for the purpose of deriving an oral RfD. BW3/4
33 scaling was not employed for deriving HEDs from studies in which doses were administered
34 directly to early postnatal animals because of the absence of information on whether allometric
35 (i.e., body weight] scaling holds when extrapolating doses from neonatal animals to adult humans
36 due to presumed toxicokinetic and/or toxicodynamic differences between lifestages [U.S. EPA,
37 2011: Hattis etal.. 20041
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
Consistent with EPA guidance [U.S. EPA. 20111. the PODs estimated based on effects in adult
animals are converted to HEDs employing a standard dosimetric adjustment factor (DAF) derived
as follows:
where
BWa = animal body weight
BWh = human body weight
Using a BWa of 0.25 kg for rats and 0.035 kg for mice and a 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):
= Laboratory animal dose (mg/kg-day) x DAF
Table 2-1 summarizes the sequence of calculations leading to the derivation of ahuman-
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 BP; 33% -f in diastolic BP)
0.09
0.6
0.09
0.15
Reproductive
Decreased ovary
weight
Xu et al. (2010)
Decreased
intratesticular
testosterone
Zheng etal. (2010)
Decreased sperm
count and motility
Mohamed etal.
(2010)
Female
Sprague-
Dawley rats
Male
Sprague-
Dawley rats
Male C57BL/6
mice
Linear3
ISO
2.3
1.5
NOAEL (1 mg/kg-d)
(15% -^ in testosterone)
LOAELfl mg/kg-d)
(50% 4, in sperm count; 20% 4, in sperm motility)
1.5
1
1
0.37
0.24
0.15
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
Endpoint and
Reference
Cervical epithelial
hyperplasia
Gaoetal. (2011a)
Species/
Sex
Female ICR
mice
Model3
Log-
logistic3
BMR
10%
BMD
(mg/kg-d)
0.58
BMDL
(mg/kg-d)
0.37
PODADJb
(mg/kg-d)
0.37
PODHEDC
(mg/kg-d)
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% ^ in IgM)
NOAEL(30mg/kg-d)
(28% ^ in IgA)
NOAEL(30mg/kg-d)
(7% 'Nn 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
24
25
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., 2011a; Mohamed etal., 2010; Xu etal., 2010; De Jong etal., 1999) or for developmental effects
resulting from in utero exposures. BW scaling was not employed for deriving HEDs from studies in which doses
were administered directly to early postnatal animals (i.e., Chen etal., 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 which incorporates the
best available information on variability in toxicokinetic disposition in the human population
(including sensitive subgroups). In the case of benzo[a]pyrene, insufficient information is available
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 to quantitatively estimate variability in human susceptibility; therefore, the full value for the
2 intraspecies UF was retained.
3 An interspecies uncertainty factor, UFA, of 3 (101/2 = 3.16, rounded to 3) was applied, to all
4 PODs in Table 2-2 except Chen et al. [2012], because BW3/4 scaling is being used to extrapolate oral
5 doses from laboratory animals to humans. Although BW3/4 scaling addresses some aspects of cross-
6 species extrapolation of toxicokinetic and toxicodynamic processes, some residual uncertainty
7 remains. In the absence of chemical-specific data to quantify this uncertainty, EPA's BW3/4
8 guidance [U.S. EPA, 2011] recommends use of an uncertainty factor of 3. BW3/4 scaling was not
9 employed for deriving HEDs from studies in which doses were administered directly to early
10 postnatal animals [i.e. Chenetal.. 2012] because of the absence of information on whether
11 allometric (i.e., body weight] scaling holds when extrapolating doses from neonatal animals to adult
12 humans due to presumed toxicokinetic and/or toxicodynamic differences between lifestages [U.S.
13 EPA, 2011: Hattis etal., 2004]. In this case, a value of 10 was applied because of the absence of
14 quantitative information to characterize either the toxicokinetic or toxicodynamic differences
15 between animals and humans at this lifestage. A subchronic to chronic uncertainty factor, UFs, of 1
16 was applied when dosing occurred during gestation [Tules etal.. 2012] or the early postnatal period
17 [Chen etal.. 2012] that is relevant to developmental effects. The developmental period is
18 recognized as a susceptible lifestage and repeated exposure is not necessary for the manifestation
19 of developmental toxicity [U.S. EPA, 1991c]. A value of 10 was applied when the POD was based on
20 a subchronic study (studies in Table 2-2, other than the two developmental toxicity studies, were
21 42-90 days in duration] to account for the possibility that longer exposure may induce effects at a
22 lower dose.
23 An uncertainty factor for extrapolation from a LOAEL to NOAEL, UFL, of 1 was applied when
24 the POD was based on a NOAEL f Zheng etal.. 2010: De long etal.. 19991 A value of 1 was applied
25 for LOAEL-to-NOAEL extrapolation when a BMR of a 1 SD [Chenetal.. 2012: Kroese etal.. 2001] or
26 10% change [Gao etal., 2011b] from the control was selected under an assumption that it
27 represents a minimal biologically significant response level. A NOAEL was not determined for the
28 most sensitive effects observed in Tules etal. [2012] and Mohamedetal. [2010]. At the LOAEL,
29 Tules etal. [2012] observed statistically significant increases in systolic [15%] and diastolic [33%]
30 blood pressure when measured in adulthood following gestational exposure. Regarding the study
31 by Mohamedetal. [2010]. the authors observed a statistically significant decrease sperm count
32 [50%] and motility [20%] in treated FO males at the LOAEL and the observed decrements in sperm
33 count persisted in untreated Fl male offspring. The data reported in these studies were not
34 amenable to dose-response modeling which would have allowed for extrapolation to a minimally
35 biologically significant response level. Therefore, a full UF of 10 was applied to approximate a
36 NOAEL for these studies which observed a high magnitude of response at the LOAEL. A database
37 uncertainty factor, UFD, of 3 was applied to account for database deficiencies including the lack of a
38 standard multigenerational study or extended 1-generation study that includes exposure from
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 premating through lactation, considering that benzo[a]pyrene has been shown to affect fertility in
2 adult male and female animals by multiple routes of exposure (see Section 1.1.2). Considering that
3 decreased fertility in adult male and female mice is observed following gestational exposure, it is
4 assumed that exposure occurring over this more comprehensive period of development could
5 result in a lower POD. Also, the lack of a study examining functional neurological endpoints
6 following a more comprehensive period of developmental exposure (i.e., gestation through
7 lactation) is a data gap, considering human and animal evidence indicating altered neurological
8 development (see Section 1.1.1).
9 Table 2-2 is a continuation of Table 2-1 and summarizes the application of UFs to each POD
10 to derive a candidate value for each data set. The candidate values presented in the table below are
11 preliminary to the derivation of the organ/system-specific reference values. These candidate
12 values are considered individually in the selection of a representative oral reference value for a
13 specific hazard and subsequent overall RfD for benzo[a]pyrene.
14 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 10'4
2 x 10'4
Reproductive
Decreased ovary weight in
rats
Xu et al. (2010)
Decreased intratesticular
testosterone in rats
Zheng etal. (2010)
Decreased sperm count and
motility in mice
Mohamedetal. (2010)
Cervical epithelial
hyperplasia in mice
Gaoetal. (2011a)
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 10'4
Not calculated
due to UF>
3000a
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
8
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
>
I—
U
U
(D
3
o
Ncurodevelopmontal
alterations in rats
(Chenetjl ,2012)
Cardiovascular effects
in rats (Jules et al., 2012)
•i- Ovary weight in rats
(Xuctal., 2010)
4- Intratosticular
testosterone in rats
(Zhengetal., 2010)
4r Sperm countand
motility in mice
(Mohamedetal., 2010)
Cervical epithelial
hyperplasia in mice
(Gaoetal., 2011)
4- Thymus weight in rats
(Kroeseetal.,2001)
••I/ Serum IgM in rats
(Do Jonget al., 1999)
4- Serum 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
2
3
4
5
6
7
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.
This document is 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 irj4
4 x 10"4
2 x 10"3
3 x If/4
Exposure
description
Critical
window of
development
(postnatal)
Subchronic
Subchronic
Critical
window of
development
(postnatal)
Confidence
MEDIUM
MEDIUM
LOW
MEDIUM
4 Developmental Toxicity
5 The candidate value based on neurobehavioral changes in rats [Chenetal., 2012] was
6 selected as the organ/system-specific RfD representing developmental toxicity. This candidate
7 value was selected because it is associated with the application of the smaller composite UF and
8 because similar effects were replicated across other studies [Bouayedetal., 2009a: Bouayedetal.,
9 2009b:Grovaetal.. 20081
10 Reproductive Toxicity
11 Among the adverse reproductive effects associated with oral benzo[a]pyrene exposure,
12 decrements in sperm parameters, decreases in testosterone, and effects in the ovary were
13 supported by a large body of evidence. The data supporting cervical effects are limited to a single
14 study, and therefore were given less weight compared to the other reproductive effects. The
15 derivation of a candidate value based on decreased sperm count and motility [Mohamedetal..
16 2010] involved too much uncertainty (see Table 2-2] and the study used to derive a candidate value
17 based on decreased testosterone [Zheng et al., 2010] did not observe a dose-response relationship
18 (a 15% decrease in testosterone was seen at the low and high doses, with a statistically significance
19 at the high dose]. The study by Xu etal. [2010] observed a dose-response relationship for
20 decreased ovary weight (both doses were statistically significant]. Additionally, statistically
21 significant decreases in primordial follicles were observed at the high dose; supporting the ovaries
22 as a target of toxicity. Therefore, the candidate value based on decreased ovary weight in rats from
23 the Xuetal. (2010] study was selected as the organ/system-specific RfD representing reproductive
24 toxicity. The ovarian effects are supported by a large database of animal studies and human studies
25 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 et al.. 1999] were equal and provided the most sensitive POD; thus, these candidate
11 values were selected as the organ/system-specific RfDs representing immunotoxicity.
12 2.1.5. Selection of the Proposed Overall Reference Dose
13 For benzo[a]pyrene, multiple organ/system-specific reference doses were derived for
14 effects identified as potential hazards from benzo[a]pyrene including developmental toxicity,
15 reproductive toxicity, and immunotoxicity. To estimate an exposure level below which effects from
16 benzo[a]pyrene exposure are not expected to occur, the lowest organ/system-specific RfD
17 (3 x 10-4 mg/kg-day] is proposed as the overall reference dose for benzo[a]pyrene. This value,
18 based on induction of neurobehavioral changes in rats exposed to benzo[a]pyrene during a
19 susceptible lifestage is supported by several animal and human studies (see Section 1.1.1].
20 The overall reference dose is derived to be protective of all types of effects for a given
21 duration of exposure and is intended to protect the population as a whole including potentially
22 susceptible subgroups [U.S. EPA. 2002]. Decisions concerning averaging exposures over time for
23 comparison with the RfD should consider the types of toxicological effects and specific lifestages of
24 concern. Fluctuations in exposure levels that result in elevated exposures during these lifestages
25 could potentially lead to an appreciable risk, even if average levels over the full exposure duration
26 were less than or equal to the RfD.
27 Furthermore, certain exposure scenarios may require particular attention to the risk-
28 assessment population of interest in order to determine whether a reference value based on
29 toxicity following developmental exposure is warranted. For example, the use of an RfD based on
30 developmental effects may not be appropriate for a risk assessment in which the population of
31 interest is post-reproductive age adults.
32 2.1.6. Confidence Statement
33 A confidence level of high, medium, or low is assigned to the study used to derive the RfD,
34 the overall database, and the RfD itself, as described in Section 4.3.9.2 of EPA's Methods for
35 Derivation of Inhalation Reference Concentrations and Application of Inhalation Dosimetry [U.S. EPA.
36 1994].
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 Confidence in the principal study [Chenetal.. 2012] is medium-to-high. The study design
2 included randomized experimental testing, blinded observations, culling of pups to account for
3 nutritional availability, treatment-randomization, and controls for litter and nursing bias. Some
4 informative experimental details were, however, omitted including the sensitivity of some assays at
5 the indicated developmental ages and lack of reporting gender-specific data for all outcomes.
6 Notably, these study limitations do not apply to the endpoint chosen to derive the RfD, and the
7 overall methods and reporting are considered sufficient Confidence in the database is medium,
8 primarily due to the lack of a multigenerational reproductive toxicity study given the sensitivity to
9 benzo[a]pyrene during development Reflecting medium-to-high confidence in the principal study
10 and medium confidence in the database, confidence in the RfD is medium.
11 2.1.7. Previous IRIS Assessment: Reference Dose
12 An RfD was not derived in the previous IRIS assessment.
13 2.2. INHALATION REFERENCE CONCENTRATION FOR EFFECTS OTHER
14 THAN CANCER
15 The inhalation reference concentration (RfC) (expressed in units of mg/m3) is defined as an
16 estimate (with uncertainty spanning perhaps an order of magnitude) of a continuous inhalation
17 exposure to the human population (including sensitive subgroups) that is likely to be without an
18 appreciable risk of deleterious effects during a lifetime. It can be derived from a NOAEL, LOAEL, or
19 the 95 percent lower bound on the benchmark concentration (BMCL), with UFs generally applied to
20 reflect limitations of the data used.
21 2.2.1. Identification of Studies and Effects for Dose-Response Analysis
22 In Section 1.2.1, developmental and reproductive toxicities were highlighted as hazards of
23 benzo[a]pyrene exposure by the inhalation route. Studies within each effect category were
24 evaluated using general study quality characteristics (as discussed in Section 6 of the Preamble) to
25 help inform the selection of studies from which to derive toxicity values. Rationales for selecting
26 the studies and effects to represent each of these hazards are summarized below.
27 Human studies of environmental PAH mixtures across multiple cohorts have observed
28 developmental and reproductive effects following prenatal exposure. However, these studies are
29 limited by exposure to complex mixtures of PAHs; and, within individual studies, there may have
30 been more than one route of exposure. In addition, the available human studies that utilized
31 benzo[a]pyrene-DNA adducts as the exposure metric do not provide external exposure levels of
32 benzo[a]pyrene from which to derive an RfC. Although preferred for derivation of reference values,
33 human studies were not considered because of the contribution to the observed hazard of multiple
34 PAHs across multiple routes of exposure and uncertainty due to concurrent exposure to other PAHs
35 and other components of the mixtures (such as metals).
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 Animal studies were evaluated to determine which provided the most relevant routes and
2 durations of exposure, multiple exposure levels to provide information about the shape of the dose
3 response curve, and relative ability to detect effects at low exposure levels. The only chronic animal
4 inhalation study available for benzo[a]pyrene, Thyssenetal. [1981], was designed as a cancer
5 bioassay and did not report other effects; however, the inhalation database for benzo[a]pyrene
6 includes several shorter duration studies that are sufficient for use in deriving reference values
7 [U.S. EPA, 2002]. Specifically, several reproductive toxicity studies are available for the inhalation
8 route, including one subchronic study Archibong et al. [2008]. Furthermore, several developmental
9 studies are available which help identify hazards of exposure during sensitive developmental
10 windows [Wormley etal.. 2004: Archibong et al.. 2002]. In addition, a four week inhalation study in
11 rats is available which investigated but did not detect lung injury [Wolff etal., 1989]. The
12 inhalation database for benzo[a]pyrene is less extensive than the database of studies by the oral
13 route, however, the types of non-cancer effects observed are consistent between routes and are
14 supported by studies in human populations (see Sections 1.1.1,1.1.2, and 1.1.3].
15 Developmental Toxicity
16 Developmental toxicity, as represented by decreased fetal survival and developmental
17 neurotoxicity. was observed by Archibong et al. f20021 and fWu etal.. 2003al fWuetal.. 2003al
18 was not considered for dose-response analysis due to lack of study details related to number of
19 dams and litters per group and lack of reporting of numerical data. From the remaining studies
20 demonstrating developmental toxicity, the studies conducted by Archibong et al. [2002] and
21 Wormley etal. [2004] were identified as the most informative studies for dose-response analysis.
22 Archibong et al. [2002] observed decreased fetal survival at the lowest dose tested by the
23 inhalation route on CDs 11-20 (i.e., LOAEL of 25 [ig/m3]. This study indicates that the developing
24 fetus is a sensitive target following inhalation exposure to benzo[a]pyrene. The observed decrease
25 in fetal survival is supported by the oral database for benzo[a]pyrene (e.g., decreased survival of
26 litters in mice following in utero exposure to benzo[a]pyrene on CDs 7-16] [Mackenzie and
27 Angevine, 1981]. In addition, a single exposure inhalation study by Wormley etal. [2004]
28 demonstrated developmental neurotoxicity, represented by electrophysiological changes in the
29 hippocampus, as a result of gestational exposure. This dose group appears to be the high exposure
30 group from Archibong et al. [2002] study (as indicated by identical outcomes for fetal survival], and
31 at this exposure level, a 66% reduction in fetal survival was observed. Due to the apparent overt
32 toxicity at this exposure concentration, Wormley etal. [2004] was not considered for dose-
33 response analysis.
34 Reproductive Toxicity
35 Reproductive toxicity, as represented by reductions in sperm quality, both count and
36 motility, and testis weights in adults, was observed by Archibong et al. [2008] and Ramesh et al.
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 [2008] and Archibong et al. [2002]. Archibongetal. [2008] and Rameshetal. [2008] reported the
2 results of a single exposure, subchronic inhalation exposure study in male rats. This subchronic
3 study was of sufficient duration and possessed adequate power to detect effects, but utilized a
4 single exposure concentration, which is less informative for dose-response analysis than a design
5 using multiple exposure concentrations. However, this single-dose subchronic study is consistent
6 with male reproductive effects observed across multiple studies by the oral route and with human
7 studies in PAH exposed populations (see Section 1.1.2]. The endpoints of decreased testes weight
8 and sperm count and motility reported in Archibongetal. [2008] were selected for dose-response
9 analysis as both represent sensitive endpoints of male reproductive toxicity and are indicators of
10 potentially decreased fertility. These effects are also consistent with human studies in PAH
11 exposed populations as effects on male fertility and semen quality have been demonstrated in
12 epidemiological studies of smokers [reviewed by Spares and Melo, 2008].
13 2.2.2. Methods of Analysis
14 Data for decreased fetal survival from Archibongetal. [2002] were reported as litter means
15 and standard deviations. These data were not amenable to BMD modeling due to the pattern of
16 variability in the data set, and attempts to obtain the raw data from the study authors were
17 unsuccessful. Therefore, the LOAEL from this study was used as the POD for dose-response
18 analysis. The study by Archibongetal. [2008], using only one exposure level, was judged not to
19 support dose-response modeling due to the lack of understanding of the underlying dose-response
20 relationship. LOAELs were also used as the PODs for dose-response analysis.
21 By definition, the RfC is intended to apply to continuous lifetime exposures for humans [U.S.
22 EPA. 1994]. EPA recommends that adjusted continuous exposures be used for inhalation
23 developmental toxicity studies as well as for studies of longer durations [U.S. EPA, 2002]. The
24 LOAELs identified from Archibongetal. [2002] and Archibongetal. [2008] were adjusted to
25 account for the discontinuous daily exposure as follows:
26
27 PODADj = POD x hours exposed per day/24 hours
28 = LOAEL x (duration of exposure/24 hours]
29 = PODAD,
30
31 Next, the human equivalent concentration [HEC] was calculated from the PODADj by
32 multiplying by a DAF, which, in this case, was the regional deposited dose ratio (RDDRER] for
33 extrarespiratory (i.e., systemic] effects as described in Methods for Derivation of Inhalation
34 Reference Concentrations and Application of Inhalation Dosimetry (U.S. EPA, 1994]. The observed
35 developmental effects are considered systemic in nature (i.e., extrarespiratory] and the normalizing
36 factor for extrarespiratory effects of particles is body weight. The RDDRER was calculated as
37 follows:
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
= BWHX(VE)AX(FTOT)A
x - BWA (VE)H (FTOT)H
2 where:
3 BW = body weight (kg)
4 VE = ventilation rate (L/minute)
5 FTOT = total fractional deposition
6
7 The total fractional deposition includes particle deposition in the nasal-pharyngeal,
8 tracheobronchial, and pulmonary regions. FTOT for both animals and humans was calculated using
9 the Multi-Path Particle Dosimetry model, a computational model used for estimating human and rat
10 airway particle deposition and clearance (Multi-Path Particle Dosimetry [MPPD]; Version 2.0 ©
11 2006, publicly available through the Hamner Institute). FTOT was based on the average particle size
12 of 1.7 ± 0.085 |im (mass median aerodynamic diameter [MMAD] ± geometric SD) as reported in Wu
13 etal. (2003a) for the exposure range 25-100 |im3. For the model runs, the Yeh-Schum 5-lobe model
14 was used for the human and the asymmetric multiple path model was used for the rat (see
15 Appendix E for MPPD model output). Both models were run under nasal breathing scenarios with
16 the inhalability adjustment selected. A geometric SD of 1 was used as the default by the model
17 because the reported geometric SD of 0.085 was <1.05.
18 The human parameters used in the model for calculating FTOT and in the subsequent
19 calculation of the PODnEc were as follows: human body weight, 70 kg; VE, 13.8 L/minute; breathing
20 frequency, 16 per minute; tidal volume, 860 mL; functional residual capacity, 3,300 mL; and upper
21 respiratory tract volume, 50 mL. Although the most sensitive population in (Archibongetal., 2002)
22 is the developing fetus, the adult rat dams were directly exposed. Thus, adult rat parameters were
23 used in the calculation of the HEC. The parameters used for the rat were body weight, 0.25 kg (a
24 generic weight for male and female rats); VE, 0.18 L/minute; breathing frequency, 102 per minute;
25 tidal volume, 1.8 mL; functional residual capacity, 4 mL; and upper respiratory tract volume, 4.42
26 mL. All other parameters were set to default values (see Appendix E).
27 Under these conditions, the MPPD model calculated FTOT values of 0.621 for the human and
28 0.181 for the rat. Using the above equation, the RDDRER was calculated to be 1.1.
29 From this, the PODHEc was calculated as follows:
30 PODHEc = PODADj x RDDRER
31
32 Table 2-4 summarizes the sequence of calculations leading to the derivation of ahuman-
33 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
Archibongetal. (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
Archibongetal. (2008)
Decreased sperm count
and motility
Archibong et al. (2008)
Male F344 rats
Male F344 rats
LOAEL(75u.g/m3)34%4,
LOAEL(75u.g/m3)
69% 4/sperm count
73% 4, sperm motility
54% 1" abnormal 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
5 bPODHEc calculated by adjusting the PODADJ by the RDDR calculated using particle size reported in Hood et al. (2000)
6 using MPPD software as detailed in Section 2.2.2 and Appendix E in the Supplemental Information.
7 2.2.3. Derivation of Candidate Values
8 Under EPA's A Review of the Reference Dose and Reference Concentration Processes [U.S. EPA,
9 2002: Section 4.4.5). also described in the Preamble, five possible areas of uncertainty and
10 variability were considered. An explanation of the five possible areas of uncertainty and variability
11 follows:
12 An intraspecies uncertainty factor, UFn, of 10 was applied to account for variability and
13 uncertainty in toxicokinetic and toxicodynamic susceptibility within the subgroup of the human
14 population most sensitive to the health hazards of benzo[a]pyrene [U.S. EPA, 2002]. In the case of
15 benzo[a]pyrene, the PODs were derived from studies in inbred animal strains and are not
16 considered sufficiently representative of the exposure and dose-response of the most susceptible
17 human subpopulations (in this case, the developing fetus). In certain cases, the toxicokinetic
18 component of this factor may be replaced when a PBPK model is available which incorporates the
19 best available information on variability in toxicokinetic disposition in the human population
20 (including sensitive subgroups). In the case of benzo[a]pyrene, insufficient information is available
21 to quantitatively estimate variability in human susceptibility; therefore, the full value for the
22 intraspecies UF was retained.
23 An interspecies uncertainty factor, UFA, of 3 (101/2 = 3.16, rounded to 3) was applied to
24 account for residual uncertainty in the extrapolation from laboratory animals to humans in the
25 absence of information to characterize toxicodynamic differences between rats and humans after
26 inhalation exposure to benzo[a]pyrene. This value is adopted by convention where an adjustment
27 from animal to a human equivalent concentration has been performed as described in EPA's
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 Methods for Derivation of Inhalation Reference Concentrations and Application of Inhalation
2 Dosimetry fU.S. EPA. 19941
3 A subchronic to chronic uncertainty factor, UFs, of 1 was applied when dosing occurred
4 during gestation [Archibongetal., 2002] or the early postnatal period that is relevant to
5 developmental effects [U.S. EPA, 1991a]. A value of 10 was applied when the POD is based on a
6 subchronic study to account for the possibility that longer exposure may induce effects at a lower
7 dose [Archibongetal., 2008] was 60 days in duration]. An uncertainty factor for extrapolation
8 from a LOAEL to a NOAEL, UFL, of 10 was applied when a LOAEL was used as the POD (Archibong et
9 al.. 2008: Archibong et al.. 2002]. The data reported in these studies were not amenable to dose-
10 response modeling which would have allowed for extrapolation to a minimally biologically
11 significant response level. At the LOAEL, these studies observed a high magnitude of response (see
12 Table 2-4]. Therefore, a full UF of 10 was applied to approximate a NOAEL for studies which
13 observed a high magnitude of response at the LOAEL. For example, the LOAEL used as the POD for
14 the developmental effect observed in Archibongetal. [2002] was based on a 19% decrease in fetal
15 survival.
16 A database uncertainty factor, UFo, of 10 was applied to account for database deficiencies
17 including the lack of a standard multigenerational study or extended 1-generation study that
18 includes exposure from premating through lactation, considering that benzo[a]pyrene has been
19 shown to affect fertility in adult male and female animals by multiple routes of exposure and that
20 decrements in fertility are greater following developmental exposure (see Section 1.1.2].
21 In addition, the lack of a study examining functional neurological endpoints following
22 inhalation exposure during development is also a data gap, considering human and animal evidence
23 indicating altered neurological development following exposure to benzo[a]pyrene alone or
24 through PAH mixtures (see Section 1.1.1].
25 The most sensitive point of departure for the RfC candidate values in Table 2-5 is based on
26 the endpoint of decreased fetal survival observed in Archibongetal. (2002]. However, oral
27 exposure studies have demonstrated neurotoxicity at doses lower than those where decreased fetal
28 survival was observed. A statistically significant decrease in fetal survival was observed following
29 treatment with 160 mg/kg-day benzo[a]pyrene, but not at lower doses (Mackenzie and Angevine,
30 1981]: however, other oral studies observed statistically significant neurobehavioral effects at
31 doses ofbenzo[a]pyrene around 0.2 to 2 mg/kg-day (Chenetal.. 2012: Bouayedetal.. 2009a].
32 Considering the relative sensitivity of the systemic health effects observed in the oral database, it is
33 likely that neurodevelopmental toxicity would be expected to occur at exposure concentrations
34 below the POD for the RfC based on decreased fetal survival.
35 According to EPA's A Review of the Reference Dose and Reference Concentration Processes
36 (U.S. EPA, 2002: Section 4.4.5], the UFD is intended to account for the potential for deriving an under
37 protective RfD/RfC as a result of an incomplete characterization of the chemical's toxicity, but also
38 including a review of existing data that may also suggest that a lower reference value might result if
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
8
additional data were available. Therefore, a database UF of 10 for the benzo[a]pyrene 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
Archibongetal. (2002)
4.6
LOAEL
3
10
10
1
10
3,000
2 x 10"6
Reproductive
Decreased testis weight in
rats
Archibongetal. (2008)
Decreased sperm count
and motility in rats
Archibongetal. (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
10
11
12
13
14
15
16
17
aCandidate values were converted from u.g/m3 to mg/m3.
bAs recommended in EPA's/4 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
1
2
3
4
5
6
7
8
9
10
11
12
D-
o
4^ fetal survival in rats
(Archibong etal., 2002)
Composite UF
A Candidate value
• POD(HEC)
4- testis weight in rats
(Archibongetal., 2008)
Q
o
cr
N[/ sperm count and motility
in rats (Archibonget al., 2008)
0.001 0.01 0.1 1
Exposure concentration (|ig/m3)
10
100
Figure 2-2. Candidate values with corresponding PODs and composite UFs.
2.2.4. Derivation of Organ/System-specific Reference Concentrations
Table 2-6 distills the candidate values from Table 2-5 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. The
candidate values for reproductive toxicity derived from Archibongetal. [2008] were not selected to
represent reproductive toxicity because as 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.
Table 2-6. Organ/system-specific RfCs and proposed overall RfC for
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
Exposure
description
Critical window of
development
Subchronic
Critical window of
development
Confidence
LOW-
MEDIUM
NA
LOW-
MEDIUM
13
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 For benzo[a]pyrene, the derivation of multiple organ/system-specific reference doses were
3 considered for effects observed following inhalation exposure and identified as potential hazards of
4 benzo[a]pyrene including developmental and reproductive toxicity. However, an organ/system-
5 specific RfC to represent reproductive toxicity could not be derived due to a composite UF of >3,000
6 from the application of uncertainty factors to cover several areas of extrapolation.
7 An overall RfC of 2 x 10'6 mg/m3 was selected based on the hazard of developmental
8 toxicity. The study by Archibong et al. [2002] was selected as the study used for the derivation of
9 the proposed overall RfC, as it observed biologically significant effects at the lowest dose tested by
10 the inhalation route. This study indicates that the developing fetus is a sensitive target following
11 inhalation exposure to benzo[a]pyrene and the observed decreased fetal survival/litter is the most
12 sensitive noncancer effect observed following inhalation exposure to benzo[a]pyrene. Additional
13 support for this endpoint of decreased fetal survival is provided by a developmental/reproductive
14 study conducted via the oral route [Mackenzie and Angevine. 1981).
15 This overall RfC is derived to be protective of all types of effects for a given duration of
16 exposure and is intended to protect the population as a whole including potentially susceptible
17 subgroups [U.S. EPA. 2002). Decisions concerning averaging exposures over time for comparison
18 with the RfC should consider the types of toxicological effects and specific lifestages of concern.
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.
22 Furthermore, certain exposure scenarios may require particular attention to the risk-
23 assessment population of interest in order to determine whether a reference value based on
24 toxicity following developmental exposure is warranted. For example, the use of an RfC based on
25 developmental effects may not be appropriate for a risk assessment in which the population of
26 interest is post-reproductive age adults.
27 2.2.6. Confidence Statement
28 A confidence level of high, medium, or low is assigned to the study used to derive the RfC,
29 the overall database, and the RfC itself, as described in Section 4.3.9.2 of EPA's Methods for
30 Derivation of Inhalation Reference Concentrations and Application of Inhalation Dosimetry [U.S. EPA,
31 19941.
32 The overall confidence in the RfC is low-to-medium. Confidence in the principal study
33 [Archibong etal., 2002] is medium. The conduct and reporting of this developmental study were
34 adequate; however, a NOAEL was not identified. Confidence in the database is low due to the lack
35 of a multigeneration toxicity study, the lack of studies on developmental neurotoxicity and immune
36 endpoints, and the lack of information regarding subchronic and chronic inhalation exposure.
37 However, confidence in the RfC is bolstered by consistent systemic effects observed by the oral
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 route (including reproductive and developmental effects) and similar effects observed in human
2 populations exposed to PAH mixtures. Reflecting medium confidence in the principal study and low
3 confidence in the database, confidence in the RfC is low-to-medium.
4 2.2.7. Previous IRIS Assessment: Reference Concentration
5 An RfC was not derived in the previous IRIS assessment
6 2.2.8. Uncertainties in the Derivation of the RfD and RfC
7 The following discussion identifies uncertainties associated with the RfD and RfC for
8 benzo[a]pyrene. To derive the RfD, the UF approach (U.S. EPA. 2000.19941 was applied to a POD
9 based on neurobehavioral changes in rats treated developmentally. To derive the RfC, this same
10 approach was applied to a POD from a developmental study for the effect of decreased fetal
11 survival. UFs were applied to the POD to account for extrapolating from an animal bioassay to
12 human exposure, the likely existence of a diverse population of varying susceptibilities, and
13 database deficiencies. These extrapolations are carried out with default approaches given the lack
14 of data to inform individual steps.
15 The database for benzo[a]pyrene contains limited human data. The observation of effects
16 associated with benzo[a]pyrene exposure in humans is complicated by several factors including the
17 existence ofbenzo[a]pyrene in the environment as one component of complex mixtures of PAHs,
18 exposure to benzo[a]pyrene by multiple routes of exposure within individual studies, and the
19 difficulty in obtaining accurate exposure information. Data on the effects of benzo[a]pyrene alone
20 are derived from a large database of studies in animal models. The database for oral
21 benzo[a]pyrene exposure includes two chronic bioassays in rats and mice, two developmental
22 studies in mice, and several subchronic studies in rats.
23 Although the database is adequate for RfD derivation, there is uncertainty associated with
24 the database including that the principal study for the RfD exposed animals during a relatively short
25 period of brain development potentially underestimating the magnitude of resulting neurological
26 effects. Also, the database lacks a comprehensive multi-generation reproductive/developmental
27 toxicity studies and immune system endpoints were not evaluated in the available chronic-duration
28 or developmental studies. Additionally, the only available chronic studies of oral or inhalational
29 exposure to benzo[a]pyrene focused primarily on neoplastic effects leaving non-neoplastic effects
30 mostly uncharacterized.
31 The only chronic inhalation study of benzo[a]pyrene was designed as a lifetime
32 carcinogenicity study and did not examine noncancer endpoints [Thyssenetal., 1981]. In addition,
33 subchronic and short-term inhalation studies are available, which examine developmental and
34 reproductive endpoints in rats. Developmental studies by the inhalation route identified
35 biologically significant reductions in the number of pups/litter and percent fetal survival and
36 possible neurodevelopmental effects (e.g., diminished electrophysiological responses to stimuli in
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 hippocampus) following gestational exposures. Additionally, a 60-day oral study in male rats
2 reported male reproductive effects (e.g., decreased testes weight and sperm production and
3 motility), but provides limited information to characterize dose-response relationships with
4 chronic exposure scenarios.
5 The study selected as the basis of the RfC provided limited information regarding the
6 inhalation exposures of the animals. Specifically, it is not clear whether the reported
7 concentrations were target values or analytical concentrations and the method used to quantify
8 benzo[a]pyrene in the generated aerosols was not provided. Requests to obtain additional study
9 details from the authors were unsuccessful, therefore the assumption was made that the reported
10 concentrations were analytical concentrations.
11 One area of uncertainty in the database pertains to the lack of information regarding
12 fertility in animals exposed gestationally to benzo[a]pyrene, especially in light of developmental
13 studies by the oral route indicating reduced fertility in the Fl generation and decreased
14 reproductive organ weights. The database also lacks a multigenerational reproductive study via the
15 inhalation route. Areas of uncertainty include the lack of chronic inhalation studies focusing on
16 noncancer effects, limited data on dose-response relationships for impaired male or female fertility
17 with gestational exposure or across several generations, and limited data on immune system
18 endpoints with chronic exposure to benzo[a]pyrene.
19 The toxicokinetic and toxicodynamic differences for benzo[a]pyrene between the animal
20 species in which the POD was derived and humans are unknown. PBPK models can be useful for
21 the evaluation of interspecies toxicokinetics; however, the benzo[a]pyrene database lacks an
22 adequate model that would inform potential differences. There is some evidence from the oral
23 toxicity data that mice may be more susceptible than rats to some benzo[a]pyrene effects (such as
24 ovotoxicity) (Borman et al.. 2000). although the underlying mechanistic basis of this apparent
25 difference is not understood. Most importantly, it is unknown which animal species may be more
26 comparable to humans.
27 2.3. ORAL SLOPE FACTOR FOR CANCER
28 The carcinogenicity assessment provides information on the carcinogenic hazard potential
29 of the substance in question and quantitative estimates of risk from oral and inhalation exposure
30 may be derived. Quantitative risk estimates may be derived from the application of a low-dose
31 extrapolation procedure. If derived, the oral slope factor is a plausible upper bound on the estimate
32 of risk per mg/kg-day of oral exposure.
33 2.3.1. Analysis of Carcinogenicity Data
34 The database for benzo[a]pyrene contains numerous cancer bioassays that identify tumors,
35 primarily of the alimentary tract including the forestomach, following oral exposure in rodents.
36 Three 2-year oral bioassays are available that associate lifetime benzo[a]pyrene exposure with
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 carcinogenicity at multiple sites: forestomach, liver, oral cavity, jejunum, kidney, auditory canal
2 (Zymbal gland) tumors, and skin or mammary gland tumors in male and female Wistar rats [Kroese
3 etal.. 2001): forestomach tumors in male and female Sprague-Dawley rats [Brune etal.. 1981): and
4 forestomach, esophageal, tongue, and larynx tumors in female B6C3Fi mice [Beland and Gulp, 1998:
5 Gulp etal.. 19981.
6 In addition to these 2-year cancer bioassays, there are studies available that provide
7 supporting evidence of carcinogenicity but are less suitable for dose-response analysis due to one
8 or more limitations in study design: [1] no vehicle control group, (2) only one benzo[a]pyrene dose
9 group, or (3) a one-time exposure to benzo[a]pyrene [Benjamin etal.. 1988: Robinson etal.. 1987:
10 El-Bayoumy. 1985: Wattenberg. 1974: Roe etal.. 1970: Biancifiori etal.. 1967: ChouroulinkovetaL
11 1967: Berenblum and Haran, 1955]. Of the controlled, multiple dose-group, repeat-dosing studies
12 that remain, most treated animals for <1 year, which is less optimal for extrapolating to a lifetime
13 exposure [Weyandetal., 1995: Triolo etal., 1977: Fedorenko and Yansheva, 1967: Neal and Rigdon,
14 19671.
15 Brune etal. [1981] dosed rats (32/sex/group] with benzo[a]pyrene in the diet or by gavage
16 in a 1.5% caffeine solution, sometimes as infrequently as once every 9th day, for approximately 2
17 years and observed increased forestomach tumors. This study was not selected for quantitation
18 due to the nonstandard treatment protocol in comparison to the GLP studies conducted by Kroese
19 etal. [2001] and Beland and Gulp [1998] and the limited reporting of study methods.
20 The Kroese etal. [2001] and Beland and Gulp [1998]studies were selected as the best
21 available studies for dose-response analysis and extrapolation to lifetime cancer risk following oral
22 exposure to benzo[a]pyrene. The ratbioassay by Kroese etal. [2001] and the mouse bioassay by
23 Beland and Gulp [1998] were conducted in accordance with Good Laboratory Practice as
24 established by the Organisation for Economic Co-operation and Development [OECD]. These
25 studies included histological examinations for tumors in many different tissues, contained three
26 exposure levels and controls, contained adequate numbers of animals per dose group
27 [~50/sex/group], treated animals for up to 2 years, and included detailed reporting of methods
28 and results (including individual animal data].
29 Details of the rat [Kroese et al., 2001] and female mouse [Beland and Gulp, 1998] study
30 designs are provided in Appendix D of the Supplemental Information. Dose-related increasing
31 trends in tumors were noted at the following sites:
32 • Squamous cell carcinomas [SCCs] or papillomas of the forestomach or oral cavity in male
33 and female rats;
34 • SCCs or papillomas of the forestomach, tongue, larynx, or esophagus in female mice;
35 • Auditory canal carcinomas in male and female rats;
36 • Kidney urothelial carcinomas in male rats;
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 • Jejunum/duodenum adenocarcinomas in female and male rats;
2 • Hepatocellular adenomas or carcinomas in male and female rats; and
3 • SCCsor basal cell tumors of the skin or mammary gland in male rats.
4 These tumors were generally observed earlier during the study with increasing exposure
5 levels, and showed statistically significantly increasing trends in incidence with increasing
6 exposure level (Cochran-Armitage trend test, p < 0.001). These data are summarized in Appendix D
7 of the Supplemental Information. As recommended by the National Toxicology Program (NTP)
8 [McConnell etal., 1986] and as outlined in EPA's Cancer Guidelines [U.S. EPA, 2005a], etiologically
9 similar tumor types (i.e., benign and malignant tumors of the same cell type) were combined for
10 these tabulations when it was judged that the benign tumors could progress to the malignant form.
11 In addition, when one tumor type occurred across several functionally related tissues, as with
12 squamous cell tumors in the tongue, esophagus, larynx, and forestomach, or adenocarcinomas of
13 the jejunum or duodenum, these incidences were also aggregated as counts of tumor-bearing
14 animals.
15 In the rat study [Kroese etal., 2001], the oral cavity and auditory canal were examined
16 histologically only if a lesion or tumor was observed grossly at necropsy. Consequently, dose-
17 response analysis for these sites was not straightforward. Use of the number of tissues examined
18 histologically as the number at risk would tend to overestimate the incidence, because the
19 unexamined animals were much less likely to have a tumor. On the other hand, use of all animals in
20 a group as the number at risk would tend to underestimate if any of the unexamined animals had
21 tumors that could only be detected microscopically. The oral cavity squamous cell tumors were
22 combined with those in the forestomach because both are part of the alimentary tract, recognizing
23 that there was some potential for underestimating this cancer risk.
24 The auditory canal tumors from the rat study were not considered for dose-response
25 analysis, for several reasons. First, the control and lower dose groups were not thoroughly
26 examined, similar to the situation described above for oral cavity tumors. Unlike the oral cavity
27 tumors, the auditory canal tumors were not clearly related to any other site or tumor type, as they
28 were described as a mixture of squamous and sebaceous cells derived from pilosebaceous units.
29 The tumors were observed mainly in the high dose groups and were highly coincident with the oral
30 cavity and forestomach tumors. Because the only mid-dose male with an auditory canal tumor did
31 not also have a forestomach or oral cavity squamous cell tumor, and no auditory canal tumors were
32 observed in low-dose male or female rats, the data are insufficient to conclude that the auditory
33 canal tumors occur independently of other tumors. The investigators did not suggest that these
34 tumors were metastases from other sites (in which case, the auditory canal tumors would be
35 repetitions of other tumors, or statistically dependent). Therefore dose-response analysis was not
36 pursued for this site, either separately or in combination with another tumor type.
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 The incidence data that were modeled are provided in Tables E-9, E-10, andE-11 [Kroese et
2 al.. 2001: Beland and Gulp. 19981
3 2.3.2. Dose Response Analysis - Adjustments and Extrapolations Methods
4 EPA's Cancer Guidelines [U.S. EPA, 2005a] recommend that the method used to characterize
5 and quantify cancer risk from a chemical is determined by what is known about the mode of action
6 of the carcinogen and the shape of the cancer dose-response curve. The dose response is assumed
7 to be linear in the low-dose range, when evidence supports a mutagenic mode of action because of
8 DNA reactivity, or if another mode of action that is anticipated to be linear is applicable. In this
9 assessment, EPA concluded that benzo[a]pyrene carcinogenicity involves a mutagenic mode of
10 action (as discussed in Section 1.1.5). Thus, a linear approach to low-dose extrapolation was used.
11 The high-dose groups of both the rat and mouse studies were dead or moribund by week 79
12 for female mice, week 72 for female rats, and week 76 for male rats. Due to the occurrence of
13 multiple tumor types, earlier occurrence with increasing exposure and early termination of the
14 high-dose group in each study, methods that can reflect the influence of competing risks and
15 intercurrent mortality on site-specific tumor incidence rates are preferred. In this case, EPA has
16 used the multistage-Weibull model, which incorporates the time at which death-with-tumor
17 occurred as well as the dose.
18 Adjustments for approximating human equivalent slope factors applicable for continuous
19 exposure were applied prior to dose-response modeling. First, continuous daily exposure for the
20 gavage study in rats [Kroese etal., 2001] was estimated by multiplying each administered dose by
21 (5 days)/(7 days) = 0.71, under the assumption of equal cumulative exposure yielding equivalent
22 outcomes. Dosing was continuous in the mouse diet study [Beland and Gulp. 1998). so no
23 continuous adjustment was necessary. Next, consistent with the EPA's Cancer Guidelines [U.S. EPA,
24 2005a). an adjustment for cross-species scaling was applied to address toxicological equivalence
25 across species. Following EPA's cross-species scaling methodology, the time-weighted daily
26 average doses were converted to HEDs on the basis of (body weight)3/4 (U.S. EPA, 1992}. This was
27 accomplished by multiplying administered doses by (animal body weight (kg)/70 kg)1/4 (U.S. EPA,
28 19921 where the animal body weights were TWAs from each group, and the U.S. EPA T19881
29 reference body weight for humans is 70 kg. It was not necessary to adjust the administered doses
30 for lifetime equivalent exposure prior to modeling for the groups terminated early, because the
31 multistage-Weibull model characterizes the tumor incidence as a function of time, from which it
32 provides an extrapolation to lifetime exposure.
33 Details of the modeling and the model selection process can be found in Appendix E of the
34 Supplemental Information. PODs for estimating low-dose risk were identified at doses at the lower
35 end of the observed data, generally corresponding to 10% extra risk.
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
2
3
4
5
6
7
2.3.3. Derivation of the Oral Slope Factor
The PODs estimated for each tumor site are summarized in Table 2-7. The lifetime oral
cancer slope factor for humans is defined as the slope of the line from the lower 95% bound on the
exposure at the POD to the control response (slope factor = 0.1/BMDLio). This slope, a 95% upper
confidence limit represents a plausible upper bound on the true risk. Using linear extrapolation
from the BMDLio, human equivalent oral slope factors were derived for each gender/tumor site
combination and are listed in Table 2-7.
Table 2-7. Summary of the oral slope factor derivations
Tumor
Forestomach, oral cavity:
squamouscell 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):
squamouscell 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
9
10
11
Human equivalent slope factor = 0.1/BMDL10HED; see Appendix E of the Supplemental Information for details of
modeling results.
bEstimates of 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
2 Oral slope factors derived from rat bioassay data varied by gender and tumor site
3 (Table 2-7). Values ranged from 0.04 per mg/kg-day, based on kidney tumors in males, to 0.4 per
4 mg/kg-day, based on alimentary tract tumors in males. Slope factors based on liver tumors in male
5 and female rats (0.2 per mg/kg-day) were only slightly lower than slope factors based on
6 alimentary tract tumors (0.2-0.3 per mg/kg-day). The oral slope factor for alimentary tract tumors
7 in female mice was highest at 1 per mg/kg-day (Table 2-7), which was approximately twofold
8 higher than the oral slope factor derived from the alimentary tract tumors in male rats.
9 Although the time-to-tumor modeling helps to account for competing risks associated with
10 decreased survival times and other causes of death including other tumors, considering the tumor
11 sites individually still does not convey the total amount of risk potentially arising from the
12 sensitivity of multiple sites—that is, the risk of developing any combination of the increased tumor
13 types. A method, for estimating overall risk, involving the assumption that the variability in the
14 slope factors could be characterized by a normal distribution, is detailed in Appendix E of the
15 Supplemental Information. The resulting composite slope factor for all tumor types for male rats
16 was 0.5 per mg/kg-day, about 25% higher than the slope factor based on the most sensitive tumor
17 site, oral cavity and forestomach, while for female rats, the composite slope factor was equivalent to
18 that for the most sensitive site (Table 2-7; see Appendix E of Supplemental Information for
19 composite slope factor estimates).
20 The overall risk estimates from rats and mice spanned about a threefold range. As there are
21 no data to support any one result as most relevant for extrapolating to humans, the most sensitive
22 result was used to derive the oral slope factor. The recommended slope factor for assessing human
23 cancer risk associated with chronic oral exposure to benzo[a]pyrene is 1 per mg/kg-day, based on
24 the alimentary tract tumor response in female B6C3Fi mice.
25 2.3.4. Uncertainties in the Derivation of the Oral Slope Factor
26 The oral slope factor for benzo[a]pyrene was based on the increased incidence of
27 alimentary tract tumors, including forestomach tumors, observed in a lifetime dietary study in mice
28 (Beland and Gulp, 1998). EPA has considered the uncertainty associated with the relevance of
29 forestomach tumors for estimating human risk from benzo[a]pyrene exposure. The rodent
30 forestomach serves to store foods and liquids for several hours before contents continue to the
31 stomach for further digestion (Clayson etal.. 1990: Grice etal.. 1986). Thus, tissue of the
32 forestomach in rodents may be exposed to benzo[a]pyrene for longer durations than analogous
33 human tissues in the oral cavity and esophagus. This suggests that the rodent forestomach may be
34 quantitatively more sensitive to the development of squamous epithelial tumors in the forestomach
35 compared to oral or esophageal tumors in humans. .
36 Uncertainty in the magnitude of the recommended oral slope factor is reflected to some
37 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
•i, 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
•i, 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
factor
Administered dose
slope
Experimental evidence supports a role for
metabolism in toxicity, but actual responsible
metabolites have not been identified.
Interspecies extrapolation
Alternatives could 4, or /T" slope
factor (e.g., 3.5-fold 4, [scaling by
body weight] or /T" 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 1
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, and whether a tumor
caused the death of the animal), this model was
superior to other available models.
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).
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.
2 2.3.5. Previous IRIS Assessment: Oral Slope Factor
3 The previous cancer assessment for benzo[a]pyrene was posted on the IRIS database in
4 1992. At that time, benzo[a]pyrene was classified as a probable human carcinogen (Group B2)
5 based on inadequate data in humans and sufficient data in animals via several routes of exposure.
6 An oral slope factor was derived from the geometric mean of four slope factor estimates based on
7 studies of dietary benzo[a]pyrene administered in the diet for approximately 2 years in ten week
8 old Sprague-Dawley rats [Brune etal.. 1981] and administered for up 7 months in two week to 5
9 month old CFW-Swiss mice [Neal and Rigdon, 1967]. A single slope factor estimate of 11.7 per
10 mg/kg-day, using a linearized multistage procedure applied to the combined incidence of
11 forestomach, esophageal, and laryngeal tumors, was derived from the Brune etal. [1981] study (see
12 Section 1.1.5 for study details]. Three modeling procedures were used to derive risk estimates
13 from the Neal and Rigdon (1967] bioassay (see Section 1.1.5]. In an analysis by Clement Associates,
14 commissioned by EPA, U.S. EPA (1990b] fit a two-stage response model, based on exposure-
15 dependent changes in both transition rates and growth rates of preneoplastic cells, to derive a value
16 of 5.9 per mg/kg-day. U.S. EPA(1991b] derived a value of 9.0 per mg/kg-day by linear
17 extrapolation from the 10% response point to the background response in a re-analysis of the
18 Clement model. Finally, using a Weibull-type model to reflect less-than-lifetime exposure to
19 benzo[a]pyrene, the same assessment (U.S. EPA, 1991b] derived an upper-bound slope factor
20 estimate of 4.5 per mg/kg-day. The four slope factor estimates, which reflected extrapolation to
21 humans assuming surface area equivalence (BW2/3 scaling] were within threefold of each other and
22 were judged to be of equal merit. Consequently, the geometric mean of these four estimates, 7.3
23 per mg/kg-day, was recommended as the oral slope factor.
24 2.4. INHALATION UNIT RISK FOR CANCER
25 The carcinogenicity assessment provides information on the carcinogenic hazard potential
26 of the substance in question and quantitative estimates of risk from oral and inhalation exposure
27 may be derived. Quantitative risk estimates may be derived from the application of a low-dose
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 extrapolation procedure. If derived, the inhalation unit risk is a plausible upper bound on the
2 estimate of risk per [J.g/m3 air breathed.
3 2.4.1. Analysis of Carcinogenicity Data
4 The inhalation database demonstrating carcinogenicity of benzo[a]pyrene consists of a
5 lifetime inhalation bioassay in male hamsters [Thyssenetal., 1981] and intratracheal instillation
6 studies, also in hamsters [Feron and Kruysse, 1978: Ketkar etal., 1978: Feronetal., 1973: Henry et
7 al., 1973: Saffiotti etal., 1972]. The intratracheal instillation studies provide supporting evidence of
8 carcinogenicity of inhaled benzo[a]pyrene; however, the use of this exposure method alters the
9 deposition, clearance, and retention of substances, and therefore, studies utilizing this exposure
10 technique are not as useful for the quantitative extrapolation of cancer risk from the inhalation of
11 benzo[a]pyrene in the environment [Driscoll etal., 2000].
12 The bioassay by Thyssenetal. [1981] represents the only lifetime inhalation cancer
13 bioassay available for describing exposure-response relationships for cancer from inhaled
14 benzo[a]pyrene. As summarized in Section 1.1.5, increased incidences of benign and malignant
15 tumors of the pharynx, larynx, trachea, esophagus, nasal cavity, or forestomach were seen with
16 increasing exposure concentration. In addition, survival was decreased relative to control in the
17 high-exposure group; mean survival times in the control, low-, and mid-concentration groups were
18 96.4, 95.2, and 96.4 weeks, respectively, and 59.5 weeks in the high-exposure group animals.
19 Overall, tumors occurred earlier in the highest benzo[a]pyrene exposure group than in the mid-
20 exposure group.
21 Strengths of the study included: chronic exposures until natural death, up to 2.5 years;
22 multiple exposure groups; histological examination of multiple organ systems; and availability of
23 individual animal pathology reports with time of death and tumor incidence data by site in the
24 upper respiratory tract In addition, the availability of average weekly continuous chamber air
25 monitoring data and individual times on study allowed the calculation of TWA lifetime continuous
26 exposures for each hamster [U.S. EPA, 1990a]. Group averages of these TWA concentrations were
27 0,0.25,1.01, and 4.29 mg/m3.
28 Several limitations concerning exposure conditions in the Thyssenetal. [1981] study were
29 evaluated for their impact on the derivation of an inhalation unit risk for benzo[a]pyrene. These
30 issues include minimal detail about the particle size distribution of the administered aerosols,
31 variability of chamber concentrations, and the use of a sodium chloride aerosol as a carrier.
32 First, particle distribution analysis of aerosols, in particular the MMAD and geometric SD,
33 was not reported, although the investigators did report that particles were within the respirable
34 range for hamsters, with >99% of the particles having diameters 0.2-0.5 |im and >80% having
35 diameters 0.2-0.3 |im.
36 Second, weekly averages of chamber concentration measurements varied two- to fivefold
37 from the overall average for each group, which exceeds the limit for exposure variability of <20%
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 for aerosols recommended by OECD [2009]. For risk assessment purposes, EPA generally assumes
2 that cancer risk is proportional to cumulative exposure, and therefore to lifetime average exposure
3 as estimated here, when there is no information to the contrary. Under this assumption, the
4 variability of the chamber concentrations has little impact on the estimated exposure-response
5 relationship. The impact of alternative assumptions are considered in Section 2.4.4.
6 Lastly, exposure occurred through the inhalation of benzo[a]pyrene adsorbed onto sodium
7 chloride aerosols, which might have irritant carrier effects, and may have a different deposition
8 than benzo[a]pyrene adsorbed onto carbonaceous particles (as is more typical in the environment).
9 The above study design and reporting issues concerning the particle size composition, exposure
10 variability, and deposition do not negate the robust tumor response following benzo[a]pyrene
11 inhalation exposure. Consequently, EPA concluded thatthe strengths of the study supported the
12 use of the data to derive an inhalation unit risk for benzo[a]pyrene. See Section 2.4.4 for a
13 discussion of uncertainties in the unit risk.
14 2.4.2. Dose Response Analysis—Adjustments and Extrapolation Methods
15 Biologically based dose-response models for benzo[a]pyrene are not available. A simplified
16 version of the two-stage carcinogenesis model proposed by Moolgavkar and Venzon [1979] and
17 Moolgavkar and Knudson [1981] has been applied to the Thyssen et al. [1981] individual animal
18 data [U.S. EPA, 1990a]. However, the simplifications necessary to fit the tumor incidence data
19 reduced that model to an empirical model (i.e., there were no biological data to inform estimates of
20 cell proliferation rates for background or initiated cells]. Sufficient data were available to apply the
21 multistage-Weibull model, as used for the oral slope factor (described in detail in Appendix E of the
22 Supplemental Information], specifically the individual times of death for each animal. Unlike in the
23 oral bioassays, Thyssen etal. [1981] did not determine cause of death for any of the animals. Since
24 the investigators for the oral bioassays considered some of the same tumor types to be fatal at least
25 some of the time, bounding estimates for the Thyssen et al. [1981] data were developed by treating
26 the tumors alternately as either all incidental to the death of an affected animal or as causing the
27 death of an affected animal.
28 The tumor incidence data used for dose-response modeling comprised the benign and
29 malignant tumors in the pharynx, larynx, trachea, esophagus, nasal cavity, or forestomach (tumors
30 of the upper respiratory and digestive tracts; see Table D-16]. The tumors in these sites were
31 judged to be sufficiently similar to combine as joint incidences by the following reasoning. While
32 the pharynx and larynx are associated with the upper digestive tract and the upper respiratory
33 tract, respectively, these sites are close anatomically and in some cases where both tissues were
34 affected, the site of origin could not be distinguished [U.S. EPA, 1990a]. In addition, the benign
35 tumors (e.g., papillomas, polyps, and papillary polyps] were considered early stages of the
36 squamous cell carcinomas in these tissues [U.S. EPA. 1990a]. Consequently, incidence data for
37 animals with malignant or benign tumors originating from the same cell type were selected for
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 dose-response modeling based on the assumption that the benign tumors could develop into
2 malignancies, as outlined in EPA's Cancer Guidelines [Section 2.2.2.1.2: U.S. EPA. 2005a].
3 A toxicokinetic model to assist in cross-species scaling of benzo[a]pyrene inhalation
4 exposure was not available. EPA's RfC default dosimetry adjustments [U.S. EPA, 1994] were
5 utilized in the benzo[a]pyrene RfC calculation (see Section 2.2.2) but could not be applied to the
6 aerosols generated for the inhalation bioassay by Thyssenetal. [1981] as the approaches
7 presented in the RfC methodology guidelines [U.S. EPA, 1994] were developed for insoluble and
8 nonhygroscopic particles, not the sodium chloride particle used in Thyssenetal. [1981].
9 Consequently, without data to inform a basis for extrapolation to humans, it was assumed that
10 equal risk for all species would be associated with equal concentrations in air, at least at anticipated
11 environmental concentrations. This is equivalent to assuming that any metabolism of
12 benzo[a]pyrene is directly proportional to breathing rate and that the deposition rate is equal
13 between species.
14 The multistage-Weibull model was fit to the TWA exposure concentrations and the
15 individual animal tumor and survival data for tumors in the larynx, pharynx, trachea, esophagus, or
16 forestomach (tumors of the upper respiratory and digestive tracts], using the software program
17 MSW [U.S. EPA. 2010]. Modeling results are provided in Appendix E of the Supplemental
18 Information.
19 Because benzo[a]pyrene carcinogenicity involves a mutagenic mode of action, linear low-
20 exposure extrapolation from the BMCLio was used to derive the inhalation unit risk [U.S. EPA,
21 2005a].
22 2.4.3. Inhalation Unit Risk Derivation
23 The results from modeling the inhalation carcinogenicity data from Thyssenetal. [1981]
24 are summarized in Table 2-9. Taking the tumors to have been the cause of death of the
25 experimental animals with tumors, the BMCio and BMCLio were 0.648 and 0.461 mg/m3,
26 respectively. Then, taking all of the tumors to have been incidental to the cause of death for each
27 animal with a tumor, the BMCio and BMCLio values were 0.285 and 0.198 mg/m3, respectively,
28 about twofold lower than the first case. Because the tumors were unlikely to have all been fatal, the
29 lower BMDLio from the incidental deaths analysis, 0.198 mg/m3, is recommended for the
30 calculation of the inhalation unit risk. Using linear extrapolation from the BMCLio of 0.198 mg/m3,
31 an inhalation unit risk of 0.5 per mg/m3, or 5 x 1Q-* per ng/m3 (rounding to one significant digit],
32 was calculated.
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 and digestive tracts;
all treated as cause of death
Thvssenetal. (1981)
Upper respiratory and digestive tracts;
all treated as incidental to death
Thvssenetal. (1981)
Species/
Sex
Male
hamsters
Male
hamsters
Selected
Model
Multistage
Weibull
Multistage
Weibull
BMR
10%
10%
BMC
(mg/ms)
0.648
0.285
POD=
BMCL
(mg/ms)
0.461
0.198
Unit Risk3
(mg/ms)-1
0.22
0.51
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 chronic inhalation exposure to benzo[a]pyrene
9 [Thyssen et al., 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 Thvssenetal. [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 Section
17 1.1.5] suggests that this study may not be ideal for extrapolating to humans. Hamsters have an
18 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=4400] 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.
25 EPA. 2005a] stress that site concordance between animals and humans need not always be
26 assumed. Therefore, the robust tumor response in the upper respiratory tract of Syrian golden
27 hamsters was considered to be supportive of the use of the Thvssenetal. (1981] study for the
28 derivation of an 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 mid
10 and high concentration benzo[a]pyrene groups. Exposure to benzo[a]pyrene in the environment
11 predominantly occurs via non-soluble, non-hygroscopic, carbonaceous particles (such as soot and
12 diesel exhaust particles]. The potential impact of differences in carrier particle on the magnitude of
13 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 1" 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 |Jg/m3 is assumed.
Dose-response modeling
Alternatives could 4, or 1
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% Cl on
administered exposure at 10% extra risk of
respiratory tract tumors.
Sensitive subpopulations
1" 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
l 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 |ig/day of dermal exposure. This derivation
6 provides the first dermal slope factor for the IRIS database.
7 2.5.1. Analysis of Carcinogenicity Data
8 Skin cancer in humans has been documented to result from occupational exposure to
9 complex mixtures of PAHs including benzo[a]pyrene, such as coal tar pitches, non-refined mineral
10 oils, shale oils, and soot [IARC. 2010: Baanetal.. 2009: IPCS. 1998: Boffettaetal.. 1997: ATSDR.
11 1995]. Although studies of human exposure to benzo[a]pyrene alone are not available, repeated
12 application of benzo[a]pyrene to skin (in the absence of exogenous promoters) has been
13 demonstrated to induce skin tumors in guinea pigs, rabbits, rats, and mice. Given the availability of
14 chronic bioassays of dermal benzo[a]pyrene exposure in mice, this analysis focuses on chronic
15 carcinogenicity bioassays in several strains of mice demonstrating predominantly malignant skin
16 tumors, as well as earlier occurrence of tumors with increasing exposure, following repeated
17 dermal exposure to benzo[a]pyrene for the majority of typical two-year chronic study durations.
18 These studies involved 2- or 3-times/week exposure protocols, at least two exposure levels plus
19 controls, and histopathological examinations of the skin and other tissues [Sivaketal., 1997:
20 Grimmer etal.. 1984: Habsetal.. 1984: Grimmer etal.. 1983: Habsetal.. 1980: Schmahletal.. 1977:
21 Schmidt etal.. 1973: Roe etal.. 1970: Poel. 1960.1959] (see Tables D-15 to D-23 in the
22 Supplemental Information for study details].
23 Other carcinogenicity studies in mice were considered as supportive of the studies listed
24 above, but were not considered in the dose-response analysis. These studies included: (1] early
25 "skin painting" studies of benzo[a]pyrene carcinogenicity in mouse skin that did not report
26 sufficient information to estimate the doses applied (e.g., Wynder and Hoffmann, 1959: Wynder et
27 al., 1957]: (2] bioassays with one benzo[a]pyrene dose level or with only dose levels inducing 90-
28 100% incidence of mice with tumors, which provide relatively little information about the shape of
29 the dose-response relationship (e.g.. Wilson and Holland. 1988]: and (3] studies with shorter
30 exposure and observation periods (i.e., <1 year] (Higginbothametal., 1993: Albert etal., 1991:
31 Nesnowetal., 1983: Emmettetal., 1981: Levin etal., 1977], which are less relevant for
32 characterizing lifetime risk.
33 Regarding study design, these data sets varied in terms of number of exposure levels used
34 (two to nine, compared with three in typical NTP bioassays] and in number of mice per group (from
35 ~17 to 100 mice /dose group, compared with 50 used in most NTP bioassays]. While the largest
36 studies would be expected to have greater ability to detect low responses at low doses (Schmidt et
37 al., 1973], studies with smaller group sizes also showed significant dose-response trends involving
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 lower doses [Sivak et al.. 1997: Poel. 1959). One of the data sets. Poel [I960], did not generate data
2 between background and maximal responses (see Table C-16), and thus was not considered further
3 for dose-response modeling, given the availability of other data sets with graded responses which
4 could better support low-dose extrapolation.
5 Additionally, an important consideration was that none of the available studies reported
6 time on study or tumor status for individual animals. When exposure is associated with early
7 mortality, this level of detail helps to understand the number at risk of development of tumors and
8 the extent of exposure associated with tumor development, and helps to minimize under-
9 estimation of cancer risk. Investigators did report that mortality was increased at higher exposures
10 and that tumors occurred earlier with increasing exposure, but reporting was mostly at the level of
11 dose groups rather than individual animals. These details facilitated some refinement of dose-
12 response results through data adjustments (described in Section 2.5.2), allowing for evaluation of
13 results on a more comparable basis across studies.
14 Finally, the available studies included different mouse strains, sexes, and vehicles.
15 However, for any given mouse strain only one sex and only one vehicle was tested. Thus, the
16 studies did not support evaluation of whether any particular vehicle solvent enhanced or
17 diminished carcinogenicity, or whether one sex was more sensitive.
18 Given these considerations, the studies by Roe etal. (1970], Sivak etal. (1997] and Poel
19 (1959] showed the strongest study designs for supporting dose-response analysis, in that they
20 included at least three exposure levels and the lowest doses tested, and reported the actual
21 duration of exposure for each dose group. All but one (Habs etal., 1980] of the remaining studies
22 provided incomplete exposure duration information for approximating lifetime (104-week]
23 equivalent exposures. The study with the most uncertainty for extrapolating to lower exposures
24 may be the Habs etal. (1984] study, which used only two exposure levels, the lower of which was
25 more than tenfold higher than the lowest doses used by Roe etal. (1970], Sivak etal. (1997] and
26 Poel (1959]. The reported duration of exposure in the higher dose group was approximately 80
27 weeks. The high tumor response at the lower dose (~50%] and the uncertainty in characterizing a
28 104-week equivalent exposure suggests that low-dose extrapolation would be relatively uncertain.
29 Nevertheless, all of these studies were included in the dose-response analysis in order to help
30 characterize similarities among the studies on a quantitative basis.
31 2.5.2. Dose Response Analysis - Adjustments and Extrapolation Methods
32 As with the oral and inhalation benzo[a]pyrene carcinogenicity data, benzo[a]pyrene's
33 dermal exposure carcinogenicity data were generally characterized by earlier occurrence of tumors
34 and increased mortality with increasing exposure level. Because individual animal data were not
35 available for any of the identified studies, time-to-tumor modeling was not possible. Each of the
36 dermal data sets was modeled using the multistage model, incorporating data adjustments where
37 appropriate, as summarized below.
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 all studies, administered doses were converted to average daily doses using the
2 equation:
3 Average daily dose/day = (ug/application] x (number of applications/week 4- 7 days/week)
4
5 Next, lifetime equivalent doses were estimated for study groups that were reported to end
6 before 104 weeks by multiplying the relevant average daily doses by (Le/104)3, where Le is the
7 length of exposure, based on observations that tumor incidence tends to increase with age [Doll,
8 1971]. Note that exposure periods <52 weeks would lead to a relatively large adjustment [i.e.,
9 (52/104)3 = 0.125, or an eightfold lower dose than administered], reflecting considerable
10 uncertainty in lifetime equivalent dose estimates generated from relatively short studies. This
11 adjustment was relevant for all dose groups in Poel [1959] and Roe etal. [1970], and the highest
12 dose group in Habsetal. [1980], and in Sivaketal. [1997].
13 Another adjustment made to minimize confounding by mortality was to omit the dose
14 groups with nearly 100% mortality occurring early in the study from dose-response modeling. Poel
15 [1960,1959] exposed multiple strains of male mice to 7-9 levels of benzo[a]pyrene, with all mice
16 in the dose groups with >3.8 ug/application dying by week 44. For three of the strains—C57L
17 [Poel. 1959], SWR and CSHeB [Poel. 1960]—the remaining four dose groups in addition to control
18 survived most of the two-year exposure period and showed a graded dose response, adequate to
19 support derivation of a slope factor for chronic exposure.
20 Concerning the incidence data, some of these studies reported incidences of skin tumor-
21 bearing animals for tumors thought to be malignant only [Roe etal., 1970: Poel, 1959] or without
22 clear designation of the relative percentages of animals with carcinomas and papillomas [Habs et
23 al.. 1980]. In the other studies, incidences of animals with skin papillomas and skin carcinomas
24 were clearly reported, showing that skin tumors from lifetime exposure to benzo[a]pyrene were
25 predominantly malignant [Sivaketal., 1997: Grimmer et al., 1984: Habs etal., 1984: Grimmer etal.,
26 1983: Schmahl et al., 1977: Schmidt etal., 1973]. Following EPA's Guidelines for Carcinogen Risk
27 Assessment [Section 2.2.2.1.2: U.S. EPA, 2005a], incidence data for animals with malignant or benign
28 skin tumors were selected for dose-response modeling based on the evidence that skin papillomas
29 can develop into malignant skin tumors. The data sets as modeled, including adjustments, are
30 presented in Tables E-19 through E-22 in the Supplemental Information.
31 The multistage-cancer model was then fit to each data set. If there was no adequate fit
32 using the multistage-cancer model, then other dichotomous models were used. Because
33 benzo[a]pyrene is expected to cause cancer via a mutagenic mode of action, a linear approach to
34 low dose extrapolation from the PODs (i.e., BMDLio] was used [U.S. EPA, 2005a] for candidate
35 dermal slope factors.
36 Several data sets provided incomplete information about the length of exposure [Habs etal.,
37 1984: Grimmer etal.. 1983: Schmahl etal.. 1977: Schmidt etal.. 1973: Poel. 19601 Accordingly,
38 dermal slope factors derived from those studies may underestimate cancer risk, due to
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 overestimation of exposure associated with the development of tumors for the higher exposure
2 groups. In addition, several study designs generated only relatively high dose-response points
3 [Grimmer etal.. 1984: Habsetal.. 1984: Grimmer etal.. 1983: Habsetal.. 1980). increasing the
4 extent of low-dose extrapolation relative to the other studies. Under the assumption that duration
5 of exposure was not impacted at lower exposures, linear extrapolation from the lowest dose-
6 response point in a study, if the response was higher than 10%, was also used.
7 2.5.3. Derivation of the Dermal Slope Factor
8 Adequate model fits were found using the multistage model for all but one of the mouse
9 skin tumor incidence data sets (see Appendix E of the Supplemental Information). The data from
10 Grimmer etal. [1984] could not be adequately fit by the multistage model initially, and the other
11 dichotomous models available in BMDS were used. Due to the supralinear shape of the dose-
12 response data, only the log-logistic and dichotomous Hill models provided adequate fits. Also due
13 to the supralinear dose-response shape, the POD for slope factor derivation was identified near the
14 lowest response of ~70%, because of the lack of data to inform the dose-response relationship at
15 lower doses. Overall, model fits demonstrated low statistical variability at the PODs, with BMDLs
16 generally less than twofold lower than corresponding BMDs.
17 For data sets designed with higher overall exposure ranges [Grimmer etal.. 1984: Habs et
18 al.. 1984: Grimmer etal.. 1983: Habsetal.. 1980], consideration of PODs based on the lowest
19 response in each study, rather than based on an extrapolated 10% extra risk, led to slope factors
20 less than twofold lower than when based on the BMDLio (see non-shaded portion of Table 2-11).
21 The alternate slope factors (based on BMRs greater than 10%) are less impacted by interpretation
22 of the responses and estimated exposures of the higher exposure groups, but necessarily reflect
23 only an assumption of linearity between the lowest exposure and background responses, which
24 may not be supported.
25 Dermal slope factors, calculated using linear extrapolation from the BMDLioS, ranged from
26 0.25 to 1.8 per |ig/day, a roughly sevenfold range (see Table 2-11). Among the stronger studies
27 (shaded entries in Table 2-11), values for male mice ranged from 0.9 to 1.7 per |ig/day, and for
28 female mice from 0.25 to 0.67 per |ig/day. Results from the remaining studies (shaded entries in
29 Table 2-11) suggest that some female mouse strains maybe as sensitive as some male mouse
30 strains, but the associated uncertainties—e.g., increased extent of low-dose extrapolation and
31 incomplete exposure information—provide less support for relying on many of these values. These
32 four female mice data sets were considered to be the most uncertain because of dose ranges
33 covered and incomplete information regarding length of exposure (Grimmer etal., 1984: Habs etal.,
34 1984: Grimmer etal., 1983: Habs etal., 1980], and are summarized in the shaded portion of Table
35 2-11. In particular, the data setreported by Habs et al. (1984) yielded the most uncertain result
36 with only two dose-response points; the lowest response was 40%, and the slope estimate is
37 determined by the characterization of both exposure estimates.
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 There was insufficient information to conclude that males were more sensitive because
2 both sexes were not tested for any strain. However, among the three better designed and reported
3 studies [Sivaketal.. 1997: Roe etal.. 1970: Poel. 1959). male mice were more sensitive than female
4 mice. Consequently, without any information indicating which data set is more relevant for
5 extrapolation to humans, the male mouse results from the higher quality studies [Sivaketal., 1997:
6 Poel, 1959] were selected for the proposed dermal slope factor. The female mouse results [Roe et
7 al., 1970] were not considered further. The average of the BMDLioS for the two male data sets was
8 0.068 ng/day.
9 Table 2-11. Summary of dermal slope factor derivations -unadjusted for
10 interspecies differences
Reference
Mouse
Strain
Selected
Model3
BM
R
BMD
(Hg/d)
POD=
BMDL
(Hg/d)
Candidate
Dermal
Slope
Factors'5
(Hg/d)-1
Comments
Male mice
Sivak et al.
(1997)
Poel (1959)a'c
Poel (1960)a'c
Poel (1960)a'c
C3H/HeJ
C57L
SWR
CSHeB
Multistage 2°
Multistage 3°
Multistage 3°
Multistage 1°
10%
10%
10%
10%
0.11
0.13
0.13
0.16
0.058
0.078
0.11
0.11
1.7
1.3
0.91
0.91
Grouped survival data reported
Grouped survival data reported
No characterization of
survival/exposure duration
No characterization of
survival/exposure duration
Female mice
Roe etal.
(1970)
Schmidt et al.
(1973)
Schmidt et al.
(1973)
Schmahl et al.
(1977)
Habsetal.
(1980)
(Habsetal.,
1984)
Grimmer et
al. (1983)
Swiss
Swiss
NMRI
NMRI
NMRI
NMRI
CFLP
Multistage 2°
Multistage 3°
Multistage 2°
Multistage 2°
Multistage 4°
Multistage 1°
Multistage 1°
10%
10%
10%
10%
10%
30%
10%
50%
10%
40%
0.69
0.28
0.33
0.23
0.36
0.49
0.078
0.51
0.24
1.2
0.39
0.22
0.29
0.15
0.24
0.44
0.056
0.37
0.21
1.0
0.25
0.45
0.34
0.67
0.42
0.69
1.8
1.4
0.48
0.40
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
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
Grimmer et
al. (1984)a
CFLP
Log-logistic
70%
1.07
0.48
1.5
No characterization of exposure
duration; high response at lowest
exposure limits usefulness of low-
dose extrapolation
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
aSee Appendix E for modeling details.
bUnadjusted for interspecies differences. Slope factor=R/BMDLR, where R is the BMR expressed as a fraction.
cHigh exposure groups omitted prior to dose-response modeling.
2.5.4. Dermal Slope Factor Cross-Species Scaling
Different methodologies have been established for interspecies scaling of PODs used to
derive oral slope factors and inhalation unit risks. Cross-species adjustment of oral doses is based
on allometric scaling using the % power of body weight This adjustment accounts for more rapid
distribution, metabolism, and clearance in small animals [U.S. EPA. 2005a). Cross-species
extrapolation of inhalation exposures is based on standard dosimetry models that consider factors
such as solubility, reactivity, and persistence [U.S. EPA, 1994]. No established methodology exists
to adjust for interspecies differences in dermal toxicity at the point of contact; however, allometric
scaling using body weight to the % power was selected based on known species differences in
dermal metabolism and penetration of benzo[a]pyrene. In vitro skin permeation was highest in the
mouse, compared to rat, rabbit, and human, and was enhanced by induction of CYP enzymes [Kao et
al.. 1985). Using this approach, rodents and humans exposed to the same daily dose of a
carcinogen, adjusted for BW3/4, would be expected to have equal lifetime risks of cancer.
Alternative approaches were also evaluated. A comparison of these alternatives is provided
in Appendix E of the Supplemental Information.
The PODM derived from the mouse studies ofPoelfl9591 and Sivak et al. fl9971 is adjusted
to a HED as follows:
POD HED (u-g/day) = PODM (u_g/day) x (BWH / BWM)3/4
= 0.068 u.g/day x (70 kg/ 0.035 kg)3/4
= 20.3 [ig/day
The resulting PODnED is used to calculate the dermal slope factor for benzo[a]pyrene:
Dermal slope factor = BMR/PODHED = 0.1/(20.3 [ig/day) = 0.005 per Hg/day
Note that the dermal slope factor should only be used with lifetime human exposures
<20 |ig/day, the human equivalent of the PODM, because above this level, the dose-response
relationship may not be proportional to the mass of benzo[a]pyrene applied.
Several assumptions are made in the use of this scaling method. First, it is assumed that the
toxicokinetic processes in the skin will scale similarly to interspecies differences in whole-body
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 toxicokinetics. Secondly, it is assumed that the risk at low doses ofbenzo[a]pyrene is linear.
2 Although one study indicates that at high doses of benzo[a]pyrene carcinogenic potency is related
3 to mass applied per unit skin and not to total mass [Davies. 1969]. this may be due to promotional
4 effects, such as inflammation, that are observed at high doses of benzo[a]pyrene.
5 The dermal slope factor has been developed for a local effect and it is not intended to
6 estimate systemic risk of cancer following dermal absorption of benzo[a]pyrene into the systemic
7 circulation. Although some information suggests thatbenzo[a]pyrene metabolites can enter
8 systemic circulation following dermal exposure in humans [Godschalketal., 1998a], lifetime skin
9 cancer bioassays that have included pathological examination of other organs have not found
10 elevated incidences of tumors at distal sites [Higginbothametal.. 1993: Habs etal.. 1980: Schmahl
11 etal.. 1977: Schmidt etal.. 1973: Roe etal.. 1970: Poel. 19591. This may be because benzo[a]pyrene
12 tends to bind to targets within the skin rather than enter the plasma receptor fluid (a surrogate
13 measure of systemic absorption) in in vitro human skin experiments. These data are consistent
14 with benzo[a]pyrene's metabolism to reactive metabolites within the viable layers of the skin
15 [Wester et al., 1990]. Some studies indicate that the fraction of benzo[a]pyrene left within the
16 viable layers of the skin is a large portion of the applied dose [Moody etal.. 2007: Moody and Chu.
17 1995). Taken together, these data support the conclusion that the risk of skin cancer following
18 dermal exposure likely outweighs cancer risks at distal organs.
19 2.5.5. Uncertainties in the Derivation of the Dermal Slope Factor
20 Uncertainty in the recommended dermal slope factor is partly reflected in the range of POD
21 values derived from the modeled mouse skin tumor data sets: the lowest and highest BMDLio
22 values listed in Table 2-11 show a sevenfold difference (0.058-0.39 |ig/day) in magnitude. There is
23 some indication that the recommended dermal slope factor may underestimate cancer risk, due to
24 inadequate data to take the observed decreasing tumor latency with increasing exposure level into
25 account with a more complex model, such as a time-to-tumor model. Reliance on studies with the
26 lowest exposure levels having low early mortality due to benzo[a]pyrene exposure and exposures
27 continuing for approximately 104 weeks tends to minimize this source of uncertainty.
28 Human dermal exposure to benzo[a]pyrene in the environment likely occurs predominantly
29 through soil contact. The available mouse dermal bioassays of benzo[a]pyrene relied on delivery of
30 benzo[a]pyrene to the skin in a solvent solution (typically acetone or toluene). The use of volatile
31 solvent likely results in a larger dose of benzo[a]pyrene available for uptake into the skin
32 (compared to soil). Consequently, reliance on these studies may overestimate the risk of skin
33 tumors from benzo[a]pyrene contact through soil; however, cancer bioassays delivering
34 benzo[a]pyrene through a soil matrix are not available.
35 There is uncertainty in extrapolating from the intermittent exposures in the mouse assays
36 to daily exposure scenarios. All of the dermal bioassays considered treated animals 2-3 times a
37 week. This assessment makes the assumption that risk is proportional to total cumulative
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 exposure. However, this may overestimate risk if duration adjusted doses are below doses that
2 saturate or diminish detoxifying metabolic steps.
3 The available data were not useful to determine which animal species may be the best
4 surrogate for human dermal response to benzo[a]pyrene. In extrapolation of the animal dermal
5 information to humans, the assumption is made that equal lifetime risks of cancer would follow
6 from exposure to the same daily dose adjusted for BW3/4. Qualitatively, the toxicokinetics and
7 toxicodynamics in mouse and human skin appear to be similar [Knafla etal.. 2011: Bickers etal.,
8 1984]. Specifically, all of the activation pathways implicated in benzo[a]pyrene carcinogenicity
9 have been observed in mouse and human skin, and associations have been made between the
10 spectra of mutations in tumor tissues from benzo[a]pyrene-exposed animals and humans exposed
11 to complex PAH mixtures containing benzo[a]pyrene (see Section 1.1.5).
12 The dermal slope factor for benzo[a]pyrene is based on skin cancer and does not represent
13 systemic cancer risk from dermal exposure. It is unclear whether dermal exposure to
14 benzo[a]pyrene would result in elevated risk of systemic tumors. Some studies in humans suggest
15 that although the skin may be responsible for a "first pass" metabolic effect, benzo[a]pyrene-
16 specific adducts have been detected in WBCs following dermal exposure to benzo[a]pyrene,
17 indicating that dermally applied benzo[a]pyrene enters systemic circulation [Godschalketal..
18 1998a]. Although none of the lifetime dermal bioassays in mice, which included macroscopic
19 examination of internal organs, reported an elevation of systemic tumors in benzo[a]pyrene-
20 treated mice compared to controls [Higginbothametal., 1993: Habs etal., 1980: Schmahletal.,
21 1977: Schmidt etal.. 1973: Roe etal.. 1970: Poel. 19591. most of these studies attempted to remove
22 animals with grossly observed skin tumors from the study before the death of the animal, possibly
23 minimizing the development of more distant tumors with longer latency. The risk of
24 benzo[a]pyrene-induced point-of-contact tumors in the skin possibly competes with systemic risk
25 of tumors. Currently, the potential contribution of dermally absorbed benzo[a]pyrene to systemic
26 cancer risk is unclear.
27
28
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
•^ dermal slope factor if alternative
data set were selected
Sivaketal. (1997); Poel
(1959)
Both studies included lowest doses among
available studies (where intercurrent mortality
was less likely to impact the number at risk) and
used typical group sizes (up to 50/group).
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.
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
Consideration and
Impact on Cancer Risk Value
Decision
Justification and Discussion
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, metabolic
processes were assumed to scale allometrically
without evidence to the contrary.
Dose-response modeling
Alternatives could 4, or
factor
Multistage model
slope
No biologically based models for benzo[a]pyrene
were available. The multistage model is
consistent with biological processes and is the
preferred model for IRIS cancer assessments
(Gehlhausetal.,2011).
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).
Sensitive subpopulations
"T* 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.
2 2.5.6. Previous IRIS Assessment: Dermal Slope Factor
3 A dermal slope factor for benzo[a]pyrene was not previously available on IRIS.
4 2.6. APPLICATION OF AGE-DEPENDENT ADJUSTMENT FACTORS (ADAFS)
5 Based on sufficient support in laboratory animals and relevance to humans, benzo[a]pyrene
6 is determined to be carcinogenic by a mutagenic mode of action. According to the Supplemental
7 Guidance for Assessing Susceptibility from Early Life Exposure to Carcinogens ("Supplemental
8 Guidance"] [U.S. EPA, 2005b], individuals exposed during early life to carcinogens with a mutagenic
9 mode of action are assumed to have increased risk for cancer. The oral slope factor of 1 per mg/kg-
10 day, inhalation unit risk of 0.5 per mg/m3, and dermal slope factor of 0.005 per [ig/day for
11 benzo[a]pyrene, calculated from data applicable to adult exposures, do not reflect presumed early
12 life susceptibility to this chemical. Although chemical-specific data exist for benzo[a]pyrene that
13 quantitatively demonstrate increased early life susceptibility to cancer [Vesselinovitch etal., 1975],
14 these data were not considered sufficient to develop separate risk estimates for childhood
15 exposure, as they used acute i.p. exposures [U.S. EPA, 2005b]. In the absence of adequate chemical-
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
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 AD AFs for three specific age
groups. These AD AFs 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 AD AFs 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
Specific Exposure
Duration
Scenarios
0.0003
0.0006
0.0008
0.002
13
14
15
16
17
18
19
20
21
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 10~4.
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
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 IQ-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,
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
Toxicological Review ofBenzo[a]pyrene
0/70, and 30/70. The age-specific risks for the three groups are 0, 0, and 4 x 1Q-4, which would
result in a total risk estimate of 4 x 1Q-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)
5 x irj4
5 x irj4
5 x irj4
Sample Exposure
Concentration (ug/m3)
1
1
1
Duration
Adjustment
2 yrs/70 yrs
14 yrs/70 yrs
54 yrs/70 yrs
Total risk
Cancer Risk for
Specific Exposure
Duration
Scenarios
0.0001
0.0003
0.0004
0.0008
8
9
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)
0.005
0.005
0.005
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
Specific Exposure
Duration
Scenarios
1 x 10"6
3 x 10"6
4 x 10"6
8 x 10"6
10
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 Alexandrov. K; Cascorbi. I; Rojas. M; Bouvier. G; Kriek. E; Bartsch. H. (2002). CYP1A1 and GSTM1
15 genotypes affect benzo[a]pyrene DNA adducts in smokers' lung: comparison with
16 aromatic/hydrophobic adduct formation [Review]. Carcinogenesis 23: 1969-1977.
17 Alexandrov, K; Rojas, M; Geneste, O; Castegnaro, M; Camus, AM; Petruzzelli, S; Giuntini, C; Bartsch, H.
18 (1992). An improved fluorometric assay for dosimetry of benzo(a)pyrene diol-epoxide-DNA
19 adducts in smokers' lung: Comparisons with total bulky adducts and aryl hydrocarbon
20 hydroxylase activity. Cancer Res 52: 6248-6253.
21 Allan. LL; Schlezinger. JJ; Shansab. M; Sherr. PH. (2006). CYP1A1 in polycyclic aromatic hydrocarbon-
22 induced B lymphocyte growth suppression. Biochem Biophys Res Commun 342: 227-235.
23 http://dx.doi.0rg/10.1016/i.bbrc.2006.01.131
24 Amos. Cl; Xu. W; Spitz. MR. (1999). Is there a genetic basis for lung cancer susceptibility? [Review].
25 Recent Results Cancer Res 151: 3-12.
26 Andreassen. A; Kure. EH; Nielsen. PS; Autrup. H; Haugen. A. (1996). Comparative synchronous
27 fluorescence spectrophotometry and 32P-postlabeling analysis of PAH-DNA adducts in human
28 lung and the relationship to TP53 mutations. Mutat Res 368: 275-282.
29 Arafa. HM; Aly. HA; Abd-Ellah. MF; EI-Refaey. HM. (2009). Hesperidin attenuates benzo[alpha] pyrene-
30 induced testicular toxicity in rats via regulation of oxidant/antioxidant balance. Toxicol Ind
31 Health 25:417-427. http://dx.doi.org/10.1177/0748233709106624
32 Archibong, AE; Inyang, F; Ramesh, A; Greenwood, M; Nayyar, T; Kopsombut, P; Hood, DB; Nyanda, AM.
33 (2002). Alteration of pregnancy related hormones and fetal survival in F-344 rats exposed by
34 inhalation to benzo(a)pyrene. Reprod Toxicol 16: 801-808.
35 Archibong, AE; Ramesh, A; Niaz, MS; Brooks, CM; Roberson, SI; Lunstra, DP. (2008). Effects of
36 benzo(a)pyrene on intra-testicular function in F-344 rats. Int J Environ Res Public Health 5: 32-
37 40. http://dx.doi.org/10.3390/iierph5010032
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. B; Tremblay. C; Baris. D; Theriault. G. (1994). Lung cancer mortality and polynuclear aromatic
2 hydrocarbons: A case-cohort study of aluminum production workers in Arvida, Quebec, Canada.
3 Am J Epidemiol 139: 250-262.
4 Armstrong, BG; Gibbs, G. (2009). Exposure-response relationship between lung cancer and polycyclic
5 aromatic hydrocarbons (PAHs). Occup Environ Med 66: 740-746.
6 http://dx.doi.org/10.1136/oem.2008.043711
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.
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 Benbrahim-Tallaa. L; Baan. RA; Grosse. Y; Lauby-Secretan. B; El ghissassi. F; Bouvard. V; Guha. N; Loomis.
2 D; Straif, K. (2012). Carcinogenicity of diesel-engine and gasoline-engine exhausts and some
3 nitroarenes. Lancet Oncol 13: 663-664. http://dx.doi.org/10.1016/S1470-2045(12)70280-2
4 Benjamin, H; Storkson, J; Tallas, PG; Pariza, MW. (1988). Reduction of benzo[a]pyrene-induced
5 forestomach neoplasms in mice given nitrite and dietary soy sauce. Food Chem Toxicol 26: 671-
6 678. http://dx.doi.org/10.1016/0278-6915(88)90066-X
7 Bennett, WP; Alavanja, MC; Blomeke, B; Vahakangas, KH; Castren, K; Welsh, JA; Bowman, ED; Khan, MA;
8 Flieder, DB; Harris, CC. (1999). Environmental tobacco smoke, genetic susceptibility, and risk of
9 lung cancer in never-smoking women. J Natl Cancer Inst 91: 2009-2014.
10 Berenblum, I; Haran, N. (1955). The influence of croton oil and of polyethylene glycol-400 on
11 carcinogenesis in the forestomach of the mouse. Cancer Res 15: 510-516.
12 Berger. J; Manz. A. (1992). Cancer of the stomach and the colon-rectum among workers in a coke gas
13 plant. Am J Ind Med 22: 825-834. http://dx.doi.org/10.1002/aiim.4700220605
14 Biancifiori. C; Caschera. F; Giornelli-Santulli. F; Bucciarelli. E. (1967). The action of oestrone and four
15 chemical carcinogens in intact and ovariectomized BALB/c/Cb/Se mice. Br J Cancer 21: 452-459.
16 Bickers. PR; Mukhtar. H; Dutta-Choudhury. T; Marcelo. CL; Voorhees. JJ. (1984). Aryl hydrocarbon
17 hydroxylase, epoxide hydrolase, and benzo[a]-pyrene metabolism in human epidermis:
18 Comparative studies in normal subjects and patients with psoriasis. J Invest Dermatol 83: 51-56.
19 http://dx.doi.org/10.llll/1523-1747.epl2261680
20 Bingham. E; Falk. HL. (1969). Environmental carcinogens. The modifying effect of cocarcinogens on the
21 threshold response. Arch Environ Health 19: 779-783.
22 Binkova, B; Chvatalova, I; Lnenickova, Z; Milcova, A; Tulupova, E; Farmer, PB; Sram, RJ. (2007). PAH-DNA
23 adducts in environmentally exposed population in relation to metabolic and DNA repair gene
24 polymorphisms. Mutat Res-Fundam Mol Mech Mutagen 620: 49-61.
25 http://dx.doi.0rg/10.1016/i.mrfmmm.2007.02.022
26 Blaha, L; Kapplova, P; Vondracek, J; Upham, B; Machala, M. (2002). Inhibition of gap-junctional
27 intercellular communication by environmentally occurring polycyclic aromatic hydrocarbons.
28 Toxicol Sci 65: 43-51. http://dx.doi.Org/10.1093/toxsci/65.l.43
29 Blanton, RH; Lyte, M; Myers, MJ; Bick, PH. (1986). Immunomodulation by polyaromatic hydrocarbons in
30 mice and murine cells. Cancer Res 46: 2735-2739.
31 Boffetta. P; Burstyn. I; Partanen. T; Kromhout. H; Svane. O; Langard. S; Jarvholm. B; Frentzel-Bevme. R;
32 Kauppinen. T; Stucker. I; Shaham. J; Heederik. D; Ahrens. W; Bergdahl. IA; Cenee. S; Ferro. G;
33 Heikkila. P; Hooiveld. M; Johansen. C; Randem. BG; Schill. W. (2003). Cancer mortality among
34 European asphalt workers: an international epidemiological study II Exposure to bitumen fume
35 and other agents. Am J Ind Med 43: 28-39. http://dx.doi.org/10.1002/aiim.10182
36 Boffetta. P; Jourenkova. N; Gustavsson. P. (1997). Cancer risk from occupational and environmental
37 exposure to polycyclic aromatic hydrocarbons [Review]. Cancer Causes Control 8: 444-472.
38 http://dx.doi.Org/10.1023/A:1018465507029
39 Bolton. JL; Trush. MA; Penning. TM; Dryhurst. G; Monks. TJ. (2000). Role of quinones in toxicology. Chem
40 Res Toxicol 13: 135-160. http://dx.doi.org/10.1021/tx9902082
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 Borini. A; Tarozzi. N; Bizzaro. D; Bonu. MA; Fava. L; Flamigni. C; Coticchio. G. (2006). Sperm DNA
2 fragmentation: paternal effect on early post-implantation embryo development in ART. Hum
3 Reprod 21: 2876-2881. http://dx.doi.org/10.1093/humrep/del251
4 Borman, SM; Christian, PJ; Sipes, IG; Hoyer, PB. (2000). Ovotoxicity in female Fischer rats and B6 mice
5 induced by low-dose exposure to three polycyclic aromatic hydrocarbons: comparison through
6 calculation of an ovotoxic index. Toxicol Appl Pharmacol 167: 191-198.
7 http://dx.doi.org/10.1006/taap.2000.9006
8 Bosetti, C; Boffetta, P; La Vecchia, C. (2007). Occupational exposures to polycyclic aromatic
9 hydrocarbons, and respiratory and urinary tract cancers: a quantitative review to 2005 [Review].
10 Ann Oncol 18: 431-446. http://dx.doi.org/10.1093/annonc/mdll72
11 Bostrom, CE; Gerde, P; Hanberg, A; Jernstrom, B; Johansson, C; Kyrklund, T; Rannug, A; Tornqvist, M;
12 Victorin. K; Westerholm. R. (2002). Cancer risk assessment, indicators, and guidelines for
13 polycyclic aromatic hydrocarbons in the ambient air [Review]. Environ Health Perspect 110
14 Suppl 3: 451-488.
15 Bouaved. J; Desor. F; Rammal. H; Kiemer. AK; Tybl. E; Schroeder. H; Rychen. G; Soulimani. R. (2009a).
16 Effects of lactational exposure to benzo[alpha]pyrene (B[alpha]P) on postnatal
17 neurodevelopment, neuronal receptor gene expression and behaviour in mice. Toxicology 259:
18 97-106. http://dx.doi.0rg/10.1016/i.tox.2009.02.010
19 Bouaved. J; Desor. F; Soulimani. R. (2009b). Subacute oral exposure to benzo[alpha]pyrene (B[alpha]P)
20 increases aggressiveness and affects consummatory aspects of sexual behaviour in male mice. J
21 Hazard Mater 169: 581-585. http://dx.doi.Org/10.1016/i.ihazmat.2009.03.131
22 Broekman, BF; Chan, YH; Chong, YS; Quek, SC; Fung, D; Low, YL; Ooi, YP; Gluckman, PD; Meaney, MJ;
23 Wong, TY; Saw, SM. (2009). The influence of birth size on intelligence in healthy children.
24 Pediatrics 123: e!011-el016. http://dx.doi.org/10.1542/peds.2008-3344
25 Brown, J. R.; Thornton, JL (1957). Percivall Pott (1714-1788) and chimney sweepers' cancer of the
26 scrotum. Br J Ind Med 14: 68-70.
27 Brown, LA; Khousbouei, H; Goodwin, JS; Irvin-Wilson, CV; Ramesh, A; Sheng, L; Mccallister, MM; Jiang,
28 GC; Aschner, M; Hood, DB. (2007). Down-regulation of early ionotrophic glutamate receptor
29 subunit developmental expression as a mechanism for observed plasticity deficits following
30 gestational exposure to benzo(a)pyrene. Neurotoxicology 28: 965-978.
31 http://dx.doi.0rg/10.1016/i.neuro.2007.05.005
32 Brune. H; Deutsch-Wenzel. RP; Habs. M; Ivankovic. S; Schmahl. D. (1981). Investigation of the
33 tumorigenic response to benzo(a)pyrene in aqueous caffeine solution applied orally to Sprague-
34 Dawley rats. J Cancer Res Clin Oncol 102: 153-157. http://dx.doi.org/10.1007/BF00410666
35 Bryan. WR; Shimkin. MB. (1943). Quantitative analysis of dose-response data obtained with three
36 carcinogenic hydrocarbons in strain C3H male mice. J Natl Cancer Inst 3: 503-531.
37 Buening. MK; Wislocki. PG; Levin. W; Yagi. H; Thakker. PR; Akagi. H; Koreeda. M; Jerina. DM; Conney.
38 AH. (1978). Tumorigenicity of the optical enantiomers of the diastereomeric benzo[a]pyrene
39 7,8-diol-9,10-epoxides in newborn mice: exceptional activity of (+)-7"beta", 8"alpha"-dihydroxy-
40 9"alpha",10"alpha"-epoxy-7,8,9,10-tetrahydrobenzo[a]pyrene. PNAS 75: 5358-5361.
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 Burdick. AD; Davis. JW; Liu. KJ; Hudson. LG; Shi. H; Monske. ML; Burchiel. SW. (2003). Benzo(a)pyrene
2 quinones increase cell proliferation, generate reactive oxygen species, and transactivate the
3 epidermal growth factor receptor in breast epithelial cells. Cancer Res 63: 7825-7833.
4 Burns-Naas, LA; Hastings, KL; Ladies, GS; Makris, SL; Parker, GA; Holsapple, MP. (2008). What's so special
5 about the developing immune system? [Review]. Int J Toxicol 27: 223-254.
6 http://dx.doi.org/10.1080/10915810801978110
7 Burstyn, I; Kromhout, H; Johansen, C; Langard, S; Kauppinen, T; Shaham, J; Ferro, G; Boffetta, P. (2007).
8 Bladder cancer incidence and exposure to polycyclic aromatic hydrocarbons among asphalt
9 pavers. Occup Environ Med 64: 520-526. http://dx.doi.org/10.1136/oem.2006.029801
10 Burstyn, I; Kromhout, H; Kauppinen, T; Heikkila, P; Boffetta, P. (2000). Statistical modelling of the
11 determinants of historical exposure to bitumen and polycyclic aromatic hydrocarbons among
12 paving workers. Ann Occup Hyg 44: 43-56.
13 Burstyn. I; Kromhout. H; Partanen. T; Svane. O; Langard. S; Ahrens. W; Kauppinen. T; Stucker. I; Shaham.
14 J; Heederik. D; Ferro. G; Heikkila. P; Hooiveld. M; Johansen. C; Randem. BG; Boffetta. P. (2005).
15 Polycyclic aromatic hydrocarbons and fatal ischemic heart disease. Epidemiology 16: 744-750.
16 http://dx.doi.org/10.1097/01.ede.0000181310.65043.2f
17 Busby. WF. Jr; Goldman. ME; Newberne. PM; Wogan. GN. (1984). Tumorigenicity of fluoranthene in a
18 newborn mouse lung adenoma bioassay. Carcinogenesis 5: 1311-1316.
19 http://dx.doi.0rg/10.1093/carcin/5.10.1311
20 Busby. WF. Jr; Stevens. EK; Martin. CM; Chow. FL; Garner. RC. (1989). Comparative lung tumorigenicity of
21 parent and mononitro-polynuclear aromatic hydrocarbons in the BLU:Ha newborn mouse assay.
22 Toxicol Appl Pharmacol 99: 555-563. http://dx.doi.org/10.1016/0041-008X(89)90162-2
23 Cavalieri, E; Rogan, E; Cremonesi, P; Higginbotham, S; Salmasi, S. (1988a). Tumorigenicity of 6-
24 halogenated derivatives of benzo[a]pyrene in mouse skin and rat mammary gland. J Cancer Res
25 Clin Oncol 114: 10-15. http://dx.doi.org/10.1007/BF00390479
26 Cavalieri, E; Rogan, E; Higginbotham, S; Cremonesi, P; Salmasi, S. (1988b). Tumor-initiating activity in
27 mouse skin and carcinogenicity in rat mammary gland of fluorinated derivatives of
28 benzo[a]pyrene and 3-methylcholanthrene. J Cancer Res Clin Oncol 114: 16-22.
29 Cavalieri, E; Rogan, E; Sinha, D. (1988c). Carcinogenicity of aromatic hydrocarbons directly applied to rat
30 mammary gland. J Cancer Res Clin Oncol 114: 3-9. http://dx.doi.org/10.1007/BF00390478
31 Cavalieri. E; Rogan. E; Toth. B; Munhall. A. (1981). Carcinogenicity of the environmental pollutants
32 cyclopenteno-[cd]pyrene and cyclopentano[cd]pyrene in mouse skin. Carcinogenesis 2: 277-281.
33 Cavalieri. EL; Devanesan. PD; Rogan. EG. (1988d). Radical cations in the horseradish peroxidase and
34 prostaglandin H synthase mediated metabolism and binding of benzo[a]pyrene to
35 deoxyribonucleic acid. Biochem Pharmacol 37: 2183-2187. http://dx.doi.org/10.1016/0006-
36 2952(88)90579-5
37 Cavalieri. EL; Higginbotham. S; Ramakrishna. NV; Devanesan. PD; Todorovic. R; Rogan. EG; Salmasi. S.
38 (1991). Comparative dose-response tumorigenicity studies of dibenzo[alpha,l]pyrene versus
39 7,12-dimethylbenz[alpha]anthracene, benzo[alpha]pyrene and two dibenzo[alpha,l]pyrene
40 dihydrodiols in mouse skin and rat mammary gland. Carcinogenesis 12: 1939-1944.
41 http://dx.doi.org/10.1093/carcin/12.10.1939
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 Cavalieri. EL; Rogan. EG. (1992). The approach to understanding aromatic hydrocarbon carcinogenesis.
2 The central role of radical cations in metabolic activation [Review]. Pharmacol Ther 55: 183-199.
3 Cavalieri, EL; Rogan, EG. (1995). Central role of radical cations in metabolic activation of polycyclic
4 aromatic hydrocarbons [Review]. Xenobiotica 25: 677-688.
5 http://dx.doi.org/10.3109/00498259509061885
6 Cavalieri, EL; Rogan, EG; Cremonesi, P; Devanesan, PP. (1988e). Radical cations as precursors in the
7 metabolic formation of quinones from benzo[a]pyrene and 6-fluorobenzo[a]pyrene. Fluoro
8 substitution as a probe for one-electron oxidation in aromatic substrates. Biochem Pharmacol
9 37: 2173-2182.
10 Chakravarti, D; Pelling, JC; Cavalieri, EL; Rogan, EG. (1995). Relating aromatic hydrocarbon-induced DNA
11 adducts and c-H-ras mutations in mouse skin papillomas: The role of apurinic sites. PNAS 92:
12 10422-10426.
13 Charles. GD; Bartels. MJ; Zacharewski. TR; Gollapudi. BB; Freshour. ML; Carney. EW. (2000). Activity of
14 benzo[a]pyrene and its hydroxylated metabolites in an estrogen receptor-alpha reporter gene
15 assay. Toxicol Sci 55: 320-326. http://dx.doi.Org/10.1093/toxsci/55.2.320
16 Chau. N; Bertrand. JP; Mur. JM; Figueredo. A; Patris. A; Moulin. JJ; Pham. QT. (1993). Mortality in retired
17 coke oven plant workers. Br J Ind Med 50: 127-135.
18 Chen. C; Tang. Y; Jiang. X; Qi. Y; Cheng. S; Qiu. C; Peng. B; Tu. B. (2012). Early postnatal benzo(a)pyrene
19 exposure in Sprague-Dawley rats causes persistent neurobehavioral impairments that emerge
20 postnatally and continue into adolescence and adulthood. Toxicol Sci 125: 248-261.
21 http://dx.doi.org/10.1093/toxsci/kfr265
22 Chen, G; Gingerich, J; Soper, L; Douglas, GR; White, PA. (2010). Induction of lacZ mutations in
23 MutaMouse primary hepatocytes. Environ Mol Mutagen 51: 330-337.
24 http://dx.doi.org/10.1002/em.20540
25 Chen, X; An, H; Ao, L; Sun, L; Liu, W; Zhou, Z; Wang, Y; Cao, J. (2011). The combined toxicity of dibutyl
26 phthalate and benzo(a)pyrene on the reproductive system of male Sprague Dawley rats in vivo. J
27 Hazard Mater 186: 835-841. http://dx.doi.Org/10.1016/i.ihazmat.2010.ll.078
28 Cheng, L; Spitz, MR; Hong, WK; Wei, Q. (2000). Reduced expression levels of nucleotide excision repair
29 genes in lung cancer: A case-control analysis. Carcinogenesis 21: 1527-1530.
30 Chengzhi, C; Yan, T; Shuqun, C; Xuejun, J; Youbin, Q; Yinyin, X; Qian, T; Baijie, T. (2011). New candidate
31 proteins for benzo(a)pyrene-induced spatial learning and memory deficits. J Toxicol Sci 36: 163-
32 171.
33 Chouroulinkov. I; Gentil. A; Guerin. M. (1967). [Study of the carcinogenic activity of 9,10-dimethyl-
34 benzanthracene and of 3,4-benzopyrene given orally]. Bull Cancer 54: 67-78.
35 Chung. JY: Kim. YJ: Kim. JY: Lee. SG: Park. JE: Kim. WR: Yoon. YD: Yoo. KS: Yoo. YH: Kim. JM. (2011).
36 Benzo[a]pyrene reduces testosterone production in rat Leydig cells via a direct disturbance of
37 testicular steroidogenic machinery. Environ Health Perspect 119: 1569-1574.
38 http://dx.doi.org/10.1289/ehp.1003391
39 Clavson. DB; Iverson. F; Nera. EA; Lok. E. (1990). The significance of induced forestomach tumors. Annu
40 Rev Pharmacol Toxicol 30: 441-463.
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 Cooper. AR; Moley. KH. (2008). Maternal tobacco use and its preimplantation effects on fertility: More
2 reasons to stop smoking [Review]. Seminars in Reproductive Medicine 26: 204-212.
3 http://dx.doi.org/10.1055/s-2008-1042959
4 Costantino, JP; Redmond, CK; Bearden, A. (1995). Occupationally related cancer risk among coke oven
5 workers: 30 years of follow-up. J Occup Environ Med 37: 597-604.
6 Cresteil, T. (1998). Onset of xenobiotic metabolism in children: Toxicological implications. Food Addit
7 Contam 15: 45-51. http://dx.doi.org/10.1080/02652039809374614
8 Gulp, SJ; Gaylor, DW; Sheldon, WG; Goldstein, LS; Beland, FA. (1996). DNA Adduct Measurements in
9 Relation to Small Intestine and Forestomach Tumor Incidence during the Chronic Feeding of
10 Coal TAR or Benzo[ A ]Pyrene to Mice. Polycycl Aromat Compd 11: 161-168.
11 http://dx.doi.org/10.1080/10406639608544662
12 Gulp. SJ; Gaylor. DW; Sheldon. WG; Goldstein. LS; Beland. FA. (1998). A comparison of the tumors
13 induced by coal tar and benzo[a]pyrene in a 2-year bioassay. Carcinogenesis 19: 117-124.
14 Gulp. SJ; Warbritton. AR; Smith. BA; Li. EE; Beland. FA. (2000). DNA adduct measurements, cell
15 proliferation and tumor mutation induction in relation to tumor formation in B6C3F1 mice fed
16 coal tar or benzo[a]pyrene. Carcinogenesis 21: 1433-1440.
17 http://dx.doi.0rg/10.1093/carcin/21.7.1433
18 Davies. RF. (1969). Tumorigenicity of very fresh tobacco smoke condensate to mouse skin. Br J Cancer
19 23:858-860.
20 De Jong. WH; Kroese. ED; Vos. JG; Van Loveren. H. (1999). Detection of immunotoxicity of
21 benzo[a]pyrene in a subacute toxicity study after oral exposure in rats. Toxicol Sci 50: 214-220.
22 http://dx.doi.0rg/10.1093/toxsci/50.2.21
23 De Waal. EJ; Schuurman. HJ; Van Loveren. H; Vos. JG. (1997). Differential effects of 2,3,7,8-
24 tetrachlorodibenzo-p-dioxin, bis(tri-n-butyltin) oxide and cyclosporine on thymus
25 histophysiology [Review]. Crit Rev Toxicol 27: 381-430.
26 http://dx.doi.org/10.3109/10408449709089900
27 Dean, JH; Luster, Ml; Boorman, GA; Lauer, LD; Leubke, RW; Lawson, L (1983). Selective
28 immunosuppression resulting from exposure to the carcinogenic congener of benzopyrene in
29 B6C3F1 mice. Clin Exp Immunol 52: 199-206.
30 DeMarini, DM; Landi, S; Tian, D; Hanley, NM; Li, X; Hu, F; Roop, BC; Mass, MJ; Keohavong, P; Gao, W;
31 Olivier. M; Hainaut. P; Mumford. JL. (2001). Lung tumor KRAS and TP53 mutations in
32 nonsmokers reflect exposure to PAH-rich coal combustion emissions. Cancer Res 61: 6679-6681.
33 Denissenko. MF; Pao. A; Tang. M; Pfeifer. GP. (1996). Preferential formation of benzo[a]pyrene adducts
34 at lung cancer mutational hotspots in P53. Science 274: 430-432.
35 http://dx.doi.org/10.1126/science.274.5286.430
36 Dietert. RR. (2008). Developmental immunotoxicology (DIT): windows of vulnerability, immune
37 dysfunction and safety assessment. J Immunotoxicol 5: 401-412.
38 http://dx.doi.org/10.1080/15476910802483324
39 Dietert. RR; Etzel. RA; Chen. D; Halonen. M; Holladay. SD; Jarabek. AM; Landreth. K; Peden. DB;
40 Pinkerton. K; Smialowicz. RJ; Zoetis. T. (2000). Workshop to identify critical windows of exposure
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 for children's health: immune and respiratory systems work group summary [Review]. Environ
2 Health Perspect 108 Suppl 3: 483-490.
3 Dietert, RR; Piepenbrink, MS. (2006). Perinatal immunotoxicity: Why adult exposure assessment fails to
4 predict risk [Review]. Environ Health Perspect 114: 477-483.
5 http://dx.doi.org/10.1289/ehp.8566
6 Dobbing, J; Sands, J. (1973). Quantitative growth and development of human brain. Arch Dis Child 48:
7 757-767.
8 Dobbing, J; Sands, J. (1979). Comparative aspects of the brain growth spurt. Early Hum Dev 3: 79-83.
9 Doll, R. (1971). The age distribution of cancer: Implication for models of carcinogenesis. Journal of the
10 Royal Statistical Society Series A 134: 133-166.
11 Dong, MH; Redmond, CK; Mazumdar, S; Costantino, JP. (1988). A multistage approach to the cohort
12 analysis of lifetime lung cancer risk among steelworkers exposed to coke oven emissions. Am J
13 Epidemiol 128: 860-873.
14 Dontenwill. W; Chevalier. HJ; Harke. HP; Lafrenz. U; Reckzeh. G; Schneider. B. (1973). Investigations on
15 the effects of chronic cigarette-smoke inhalation in Syrian golden hamsters. J Natl Cancer Inst
16 51: 1781-1832. http://dx.doi.Org/10.1093/inci/51.6.1781
17 Driscoll. KE; Costa. PL; Hatch. G; Henderson. R; Oberdorster. G; Salem. H; Schlesinger. RB. (2000).
18 Intratracheal instillation as an exposure technique for the evaluation of respiratory tract toxicity:
19 Uses and limitations. Toxicol Sci 55: 24-35.
20 EI-Bayoumy. K. (1985). Effects of organoselenium compounds on induction of mouse forestomach
21 tumors by benzo(a)pyrene. Cancer Res 45: 3631-3635.
22 EI-Bayoumy, K; Hecht, SS; Hoffmann, D. (1982). Comparative tumor initiating activity on mouse skin of 6-
23 nitrobenzo[a]pyrene, 6-nitrochrysene, 3-nitroperylene, 1-nitropyrene and their parent
24 hydrocarbons. Cancer Lett 16: 333-337. http://dx.doi.org/10.1016/0304-3835(82)90015-5
25 EI-Nemr, A; Al-Shawaf, T; Sabatini, L; Wilson, C; Lower, AM; Grudzinskas, JG. (1998). Effect of smoking on
26 ovarian reserve and ovarian stimulation in in-vitro fertilization and embryo transfer. Hum
27 Reprod 13: 2192-2198.
28 Emmett, EA; Bingham, EM; Barkley, W. (1981). A carcinogenic bioassay of certain roofing materials. Am J
29 Ind Med 2: 59-64. http://dx.doi.org/10.1002/aiim.4700020110
30 Esser, C. (2009). The immune phenotype of AhR null mouse mutants: not a simple mirror of xenobiotic
31 receptor over-activation [Review]. Biochem Pharmacol 77: 597-607.
32 http://dx.doi.0rg/10.1016/i.bcp.2008.10.002
33 Fedorenko. Z; Yansheva. N. (1967). [Experimental reproduction of tumors of the antral part of the
34 stomach in mice by administration of various dose of 3,4-benzpyrene]. Gig Sanit 32: 168-173.
35 Fernandez-Salguero. P; Pineau. T; Hilbert. DM; McPhail. T; Lee. SS; Kimura. S; Nebert. DW; Rudikoff. S;
36 Ward. JM; Gonzalez. FJ. (1995). Immune system impairment and hepatic fibrosis in mice lacking
37 the dioxin-binding Ah receptor. Science 268: 722-726.
38 Feron. VJ; de Jong. D; Emmelot. P. (1973). Letter: Dose-response correlation for the induction of
39 respiratory-tract tumours in Syrian golden hamsters by intratracheal instillations of
40 benzo(a)pyrene. Eur J Cancer 9: 387-390.
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 Feron. VJ; Kruysse. A. (1978). Effects of exposure to furfural vapour in hamsters simultaneously treated
2 with benzo[alpha] pyrene or diethylnitrosamine. Toxicology 11: 127-144.
3 Perron, GA. (1994). Aerosol properties and lung deposition [Editorial]. Eur Respir J 7: 1392-1394.
4 Fertuck, KC; Matthews, JB; Zacharewski, TR. (2001). Hydroxylated benzo[a]pyrene metabolites are
5 responsible for in vitro estrogen receptor-mediated gene expression induced by
6 benzo[a]pyrene, but do not elicit uterotrophic effects in vivo. Toxicol Sci 59: 231-240.
7 http://dx.doi.0rg/10.1093/toxsci/59.2.231
8 Fine, JS; Silverstone, AE; Gasiewicz, TA. (1990). Impairment of prothymocyte activity by 2,3,7,8-
9 tetrachlorodibenzo-p-dioxin. J Immunol 144: 1169-1176.
10 Flowers, L; Bleczinski, WF; Burczynski, ME; Harvey, RG; Penning, TM. (1996). Disposition and biological
11 activity of benzo[a]pyrene-7,8-dione. A genotoxic metabolite generated by dihydrodiol
12 dehydrogenase. Biochemistry 35: 13664-13672. http://dx.doi.org/10.1021/bi961077w
13 Flowers. L; Ohnishi. ST; Penning. TM. (1997). DNA strand scission by polycyclic aromatic hydrocarbon o-
14 quinones: role of reactive oxygen species, Cu(ll)/Cu(l) redox cycling, and o-semiquinone anion
15 radicals. Biochemistry 36: 8640-8648. http://dx.doi.org/10.1021/bi970367p
16 Friesen. MC; Benke. G; Del Monaco. A; Dennekamp. M; Fritschi. L; de Klerk. N; Moving. JL; Macfarlane. E;
17 Sim. MR. (2009). Relationship between cardiopulmonary mortality and cancer risk and
18 quantitative exposure to polycyclic aromatic hydrocarbons, fluorides, and dust in two prebake
19 aluminum smelters. Cancer Causes Control 20: 905-916. http://dx.doi.org/10.1007/sl0552-009-
20 9329-8
21 Friesen, MC; Demers, PA; Spinelli, JJ; Eisen, EA; Lorenzi, MF; Le, ND. (2010). Chronic and acute effects of
22 coal tar pitch exposure and cardiopulmonary mortality among aluminum smelter workers. Am J
23 Epidemiol 172: 790-799. http://dx.doi.org/10.1093/aie/kwq208
24 Friesen, MC; Macnab, YC; Marion, SA; Demers, PA; Davies, HW; Teschke, K. (2006). Mixed models and
25 empirical bayes estimation for retrospective exposure assessment of dust exposures in Canadian
26 sawmills. Ann Occup Hyg 50: 281-288. http://dx.doi.org/10.1093/annhyg/mei076
27 Gale, CR; O'Callaghan, FJ; Bredow, M; Martyn, CM; Team, ALSoPaCS. (2006). The influence of head
28 growth in fetal life, infancy, and childhood on intelligence at the ages of 4 and 8 years. Pediatrics
29 118: 1486-1492. http://dx.doi.org/10.1542/peds.2005-2629
30 Galvan, N; Page, TJ; Czuprynski, CJ; Jefcoate, CR. (2006). Benzo(a)pyrene and 7,12-
31 dimethylbenz(a)anthrecene differentially affect bone marrow cells of the lymphoid and myeloid
32 lineages. Toxicol Appl Pharmacol 213: 105-116. http://dx.doi.Org/10.1016/i.taap.2005.09.018
33 Gao. M; Li. Y; Sun. Y; Long. J; Kong. Y; Yang. S; Wang. Y. (2011a). A common carcinogen benzo[a]pyrene
34 causes p53 overexpression in mouse cervix via DNA damage. Mutat Res 724: 69-75.
35 http://dx.doi.0rg/10.1016/i.mrgentox.2011.06.008
36 Gao. M; Li. Y; Sun. Y; Shah. W; Yang. S; Wang. Y; Long. J. (2011b). Benzo[a]pyrene exposure increases
37 toxic biomarkers and morphological disorders in mouse cervix. Basic & Clinical Pharmacology &
38 Toxicology Online Pharmacology Online 109: 398-406. http://dx.doi.org/lO.llll/j. 1742-
39 7843.2011.00755.x
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 Gao. M; Long. J; Li. Y; Shah. W; Fu. L; Liu. J; Wang. Y. (2010). Mitochondrial decay is involved in BaP-
2 induced cervical damage. Free Radic Biol Med 49: 1735-1745.
3 http://dx.doi.0rg/10.1016/i.freeradbiomed.2010.09.003
4 Garcon, G; Gosset, P; Garry, S; Marez, T; Hannothiaux, MH; Shirali, P. (2001a). Pulmonary induction of
5 proinflammatory mediators following the rat exposure to benzo(a)pyrene-coated onto Fe2O3
6 particles. Toxicol Lett 121: 107-117.
7 Garcon, G; Zerimech, F; Hannothiaux, M; Gosset, P; Martin, A; Marez, T; Shirali, P. (2001b). Antioxidant
8 defense disruption by polycyclic aromatic hydrocarbons-coated onto Fe(2)O(3) particles in
9 human lung cells (A549). Toxicology 166: 129-137.
10 Gaspari, L; Chang, SS; Santella, RM; Garte, S; Pedotti, P; Taioli, E. (2003). Polycyclic aromatic
11 hydrocarbon-DNA adducts in human sperm as a marker of DNA damage and infertility. Mutat
12 Res 535: 155-160.
13 Geacintov. NE; Cosman. M; Hingerty. BE; Amin. S; Brovde. S; Patel. DJ. (1997). NMR solution structures
14 of stereoisometric covalent polycyclic aromatic carcinogen-DNA adduct: principles, patterns,
15 and diversity [Review]. Chem Res Toxicol 10: 111-146. http://dx.doi.org/10.1021/tx9601418
16 Gehlhaus. MW; Gift. JS; Hogan. KA; Kopylev. L; Schlosser. PM; Kadry. AR. (2011). Approaches to cancer
17 assessment in EPA's Integrated Risk Information System [Review]. Toxicol Appl Pharmacol 254:
18 170-180. http://dx.doi.0rg/10.1016/i.taap.2010.10.019
19 Centner. NJ; Weber. LP. (2011). Intranasal benzo[a]pyrene alters circadian blood pressure patterns and
20 causes lung inflammation in rats. Arch Toxicol 85: 337-346. http://dx.doi.org/10.1007/s00204-
21 010-0589-6
22 Gibbs, GW. (1985). Mortality of aluminum reduction plant workers, 1950 through 1977. J Occup Environ
23 Med 27: 761-770.
24 Gibbs, GW; Armstrong, B; Sevigny, M. (2007). Mortality and cancer experience of Quebec aluminum
25 reduction plant workers. Part 2: mortality of three cohorts hired on or before January 1, 1951. J
26 Occup Environ Med 49: 1105-1123. http://dx.doi.org/10.1097/JOM.Ob013e318157d34a
27 Gibbs, GW; Horowitz, I. (1979). Lung cancer mortality in aluminum reduction plant workers. J Occup
28 Med 21: 347-353.
29 Gibbs, GW; Sevigny, M. (2007a). Mortality and cancer experience of Quebec aluminum reduction plant
30 workers, part 4: cancer incidence. J Occup Environ Med 49: 1351-1366.
31 http://dx.doi.org/10.1097/JOM.Ob013e318156ecbc
32 Gibbs. GW; Sevigny. M. (2007b). Mortality and cancer experience of Quebec aluminum reduction plant
33 workers. Part 3: monitoring the mortality of workers first employed after January 1, 1950. J
34 Occup Environ Med 49: 1269-1287. http://dx.doi.org/10.1097/JOM.Ob013e3181593da8
35 Ginsberg. G; Hattis. D; Sonawane. B. (2004). Incorporating pharmacokinetic differences between
36 children and adults in assessing children's risks to environmental toxicants [Review]. Toxicol
37 Appl Pharmacol 198: 164-183. http://dx.doi.Org/10.1016/i.taap.2003.10.010
38 Godschalk. R; Curfs. D; Bartsch. H; Van Schooten. FJ; Nair. J. (2003). Benzo[a]pyrene enhances lipid
39 peroxidation induced DNA damage in aorta of apolipoprotein E knockout mice. Free Radic Res
40 37: 1299-1305.
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 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 Grant, G; Roe, FJ. (1963). The effect of phenanthrene on tumour induction by 3,4-benzopyrene
11 administered to newly born mice. Br J Cancer 17: 261-265.
12 Grice. HC; Clayson. DB; Flamm. WG; Ito. N; Kroes. R; Newberne. PM; Scheuplein. R. (1986). Panel
13 discussion: Possible mechanisms of BHA carcinogenicity from a consideration of its chemical and
14 biological properties. Food Chem Toxicol 24: 1235-1242.
15 Grimmer. G; Brune. H; Deutsch-Wenzel. R; Dettbarn. G; Misfeld. J. (1984). Contribution of polycyclic
16 aromatic hydrocarbons to the carcinogenic impact of gasoline engine exhaust condensate
17 evaluated by implantation into the lungs of rats. J Natl Cancer Inst 72: 733-739.
18 Grimmer. G; Brune. H; Deutsch-Wenzel. R; Naujack. KW; Misfeld. J; Timm. J. (1983). On the contribution
19 of polycyclic aromatic hydrocarbons to the carcinogenic impact of automobile exhaust
20 condensate evaluated by local application onto mouse skin. Cancer Lett 21: 105-113.
21 Grova, N; Schroeder, H; Farinelle, S; Prodhomme, E; Valley, A; Muller, CP. (2008). Sub-acute
22 administration of benzo[a]pyrene (B[a]P) reduces anxiety-related behaviour in adult mice and
23 modulates regional expression of N-methyl-D-aspartate (NMDA) receptors genes in relevant
24 brain regions. Chemosphere 73: S295-S302.
25 http://dx.doi.0rg/10.1016/i.chemosphere.2007.12.037
26 Grova, N; Valley, A; Turner, JD; Morel, A; Muller, CP; Schroeder, H. (2007). Modulation of behavior and
27 NMDA-R1 gene mRNA expression in adult female mice after sub-acute administration of
28 benzo(a)pyrene. Neurotoxicology 28: 630-636. http://dx.doi.Org/10.1016/i.neuro.2007.01.010
29 Gustavsson, P; Jakobsson, R; Johansson, H; Lewin, F; Norell, S; Rutkvist, LE. (1998). Occupational
30 exposures and squamous cell carcinoma of the oral cavity, pharynx, larynx, and oesophagus: A
31 case-control study in Sweden. Occup Environ Med 55: 393-400.
32 Gustavsson. P; Plato. N; Lidstrom. EB; Hogstedt. C. (1990). Lung cancer and exposure to diesel exhaust
33 among bus garage workers. Scand J Work Environ Health 16: 348-354.
34 Habs. M; Jahn. SA; Schmahl. D. (1984). Carcinogenic activity of condensate from coloquint seeds
35 (Citrullus colocynthis) after chronic epicutaneous administration to mice. J Cancer Res Clin Oncol
36 108: 154-156. http://dx.doi.org/10.1007/BF00390988
37 Habs. M; Schma'hl. D; Misfeld. J. (1980). Local carcinogenicity of some environmentally relevant
38 polycyclic aromatic hydrocarbons after lifelong topical application to mouse skin. Arch
39 Geschwulstforsch 50: 266-274.
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 Hackman. P; Hou. SM; Nyberg. F; Pershagen. G; Lambert. B. (2000). Mutational spectra at the
2 hypoxanthine-guanine phosphoribosyltransferase (HPRT) locus in T-lymphocytes of nonsmoking
3 and smoking lung cancer patients. Mutat Res 468: 45-61.
4 Hainaut, P; Pfeifer, GP. (2001). Patterns of p53 G->T transversions in lung cancers reflect the primary
5 mutagenic signature of DNA-damage by tobacco smoke. Carcinogenesis 22: 367-374.
6 Hakim, FT; Memon, SA; Cepeda, R; Jones, EC; Chow, CK; Kasten-Sportes, C; Odom, J; Vance, BA;
7 Christensen, BL; Mackall, CL; Gress, RE. (2005). Age-dependent incidence, time course, and
8 consequences of thymic renewal in adults. J Clin Invest 115: 930-939.
9 http://dx.doi.org/10.1172/JCI22492
10 Hales, DB. (2002). Testicular macrophage modulation of Leydig cell steroidogenesis [Review]. J Reprod
11 Immunol57:3-18.
12 Hammond. EC; Selikoff. IJ; Lawther. PL; Seidman. H. (1976). Inhalation of benzpyrene and cancer in man.
13 Ann N Y Acad Sci 271: 116-124. http://dx.doi.Org/10.llll/i.1749-6632.1976.tb23100.x
14 Hanahan. D; Weinberg. RA. (2011). Hallmarks of cancer: The next generation [Review]. Cell 144: 646-
15 674. http://dx.doi.0rg/10.1016/i.cell.2011.02.013
16 Hansen. ES. (1989). Cancer incidence in an occupational cohort exposed to bitumen fumes. Scand J Work
17 Environ Health 15: 101-105.
18 Hansen. ES. (1991). Mortality of mastic asphalt workers. Scand J Work Environ Health 17: 20-24.
19 http://dx.doi.org/10.5271/siweh.1739
20 Hardin. JA; Hinoshita. F; Sherr. PH. (1992). Mechanisms by which benzo[a]pyrene, an environmental
21 carcinogen, suppresses B cell lymphopoiesis. Toxicol Appl Pharmacol 117: 155-164.
22 Harvey, RG. (1996). Mechanisms of carcinogenesis of polycyclic aromatic hydrocarbons. Polycycl Aromat
23 Compd 9: 1-23. http://dx.doi.org/10.1080/10406639608031196
24 Hattis, D; Goble, R; Russ, A; Chu, M; Ericson, J. (2004). Age-related differences in susceptibility to
25 carcinogenesis: A quantitative analysis of empirical animal bioassay data. Environ Health
26 Perspect 112: 1152-1158. http://dx.doi.org/10.1289/ehp.6871
27 Henry, MC; Port, CD; Bates, RR; Kaufman, DG. (1973). Respiratory tract tumors in hamsters induced by
28 benzo(a)pyrene. Cancer Res 33: 1585-1592.
29 Higginbotham, S; Ramakrishna, NVS; Johansson, SL; Rogan, EG; Cavalieri, EL. (1993). Tumor-initiating
30 activity and carcinogenicity of dibenzo[a,l]pyrene versus 7,12-dimethylbenz[a]anthracene and
31 benzo[a]pyrene at low doses in mouse skin. Carcinogenesis 14: 875-878.
32 http://dx.doi.0rg/10.1093/carcin/14.5.875
33 Hoffmann. D; Rathkamp. G; Nesnow. S; Wynder. EL. (1972). Fluoranthenes: Quantitative determination
34 in cigarette smoke, formation by pyrolysis, and tumor-initiating activity. J Natl Cancer Inst 49:
35 1165-1175.
36 Holladay. SD; Smialowicz. RJ. (2000). Development of the murine and human immune system:
37 Differential effects of immunotoxicants depend on time of exposure [Review]. Environ Health
38 Perspect 108: 463-473.
39 Holladay. SD; Smith. BJ. (1994). Fetal hematopoietic alterations after maternal exposure to
40 benzo[a]pyrene: A cytometric evaluation. J Toxicol Environ Health 42: 259-273.
41 http://dx.doi.org/10.1080/15287399409531878
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 Holladay. SD; Smith. BJ. (1995). Benzo[a]pyrene-induced alterations in total immune cell number and
2 cell-surface antigen expression in the thymus, spleen and bone marrow of B6C3F1 mice. Vet
3 Hum Toxicol 37: 99-104.
4 Homburger, F; Hsueh, SS; Kerr, CS; Russfield, AB. (1972). Inherited susceptibility of inbred strains of
5 Syrian hamsters to induction of subcutaneous sarcomas and mammary and gastrointestinal
6 carcinomas by subcutaneous and gastric administration of polynuclear hydrocarbons. Cancer
7 Res 32: 360-366.
8 Hood, DB; Nayyar, T; Ramesh, A; Greenwood, M; Inyang, F. (2000). Modulation in the developmental
9 expression profile of Spl subsequent to transplacental exposure of fetal rats to desorbed
10 benzo[a]pyrene following maternal inhalation. Inhal Toxicol 12: 511-535.
11 http://dx.doi.org/10.1080/089583700402897
12 Hsu. PC: Chen. IY: Pan. CH: Wu. KY: Pan. MH: Chen. JR: Chen. CJ: Chang-Chien. GP: Hsu. CH: Liu. CS: Wu.
13 MT. (2006). Sperm DNA damage correlates with polycyclic aromatic hydrocarbons biomarker in
14 coke-oven workers. Int Arch Occup Environ Health 79: 349-356.
15 http://dx.doi.org/10.1007/s00420-005-0066-3
16 IARC (International Agency for Research on Cancer). (1973). Certain polycyclic aromatic hydrocarbons
17 and heterocyclic compounds. Lyon, France.
18 IARC (International Agency for Research on Cancer). (1983). Polynuclear aromatic hydrocarbons, Part 1:
19 Chemical, environmental and experimental data. Lyon, France.
20 http://monographs.iarc.fr/ENG/Monographs/vol32/volume32.pdf
21 IARC (International Agency for Research on Cancer). (2002). Section 3. Carcinogenicity studies in
22 experimental animals [IARC Monograph] (pp. 973-1003). Lyon, France.
23 http://monographs.iarc.fr/ENG/Monographs/vol83/mono83-6C.pdf
24 IARC (International Agency for Research on Cancer). (2003). Predictive value of rodent forestomach and
25 gastric neuroendocrine tumours in evaluating carcinogenic risks to humans: Views and expert
26 opinions of an IARC working group, Lyon, 29 November -1 December 1999. (IARC Technical
27 Publication No. 39). Lyon, France.
28 http://monographs.iarc.fr/ENG/Publications/techrep39/IARCrep39.pdf
29 IARC (International Agency for Research on Cancer). (2004). Tobacco smoke and involuntary smoking
30 [IARC Monograph] (pp. 1-1438). (RISKLINE/2004080004). Lyon, France.
31 http://monographs.iarc.fr/ENG/Monographs/vol83/mono83.pdf
32 IARC (International Agency for Research on Cancer). (2010). Some non-heterocyclic polycyclic aromatic
33 hydrocarbons and some related exposures [IARC Monograph] (pp. 1-853). Lyon, France.
34 http://monographs.iarc.fr/ENG/Monographs/vol92/mono92.pdf
35 Ickstadt. K: Schafer. M: Fritsch. A: Schwender. H: Abel. J: Bolt. HM: Bruning. T: Ko. YD: Vetter. H: Harth.
36 \A (2008). Statistical methods for detecting genetic interactions: a head and neck squamous-cell
37 cancer study. J Toxicol Environ Health A 71: 803-815.
38 http://dx.doi.org/10.1080/15287390801985745
39 Invang. F; Ramesh. A; Kopsombut. P; Niaz. MS; Hood. DB; Nyanda. AM; Archibong. AE. (2003). Disruption
40 of testicular steroidogenesis and epididymal function by inhaled benzo(a)pyrene. Reprod Toxicol
41 17:527-537.
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 I PCS (International Programme on Chemical Safety). (1998). Selected non-heterocyclic polycyclic
2 aromatic hydrocarbons [WHO EHC]. In Environmental Health Criteria (pp. CHF 174).
3 (CIS/00/00657). Geneva, Switzerland: World Health Organization.
4 http://www.inchem.org/documents/ehc/ehc/ehc202.htm
5 Jeffrey, AM; Jennette, KW; Blobstein, SH; Weinstein, IB; Beland, FA; Harvey, RG; Kasal, H; Miura, I;
6 Nakanishi, K. (1976). Letter: Benzo[a]pyrene-nucleic acid derivative found in vivo: structure of a
7 benzo[a]pyrenetetrahydrodiol epoxide-guanosine adduct. J Am Chem Soc 98: 5714-5715.
8 Jensen, TK; Henriksen, TB; Hjollund, NH; Scheike, T; Kolstad, H; Giwercman, A; Ernst, E; Bonde, JP;
9 Skakkebaek, NE; Olsen, J. (1998). Adult and prenatal exposures to tobacco smoke as risk
10 indicators of fertility among 430 Danish couples. Am J Epidemiol 148: 992-997.
11 Jerina, DM; Chadha, A; Cheh, AM; Schurdak, ME; Wood, AW; Sayer, JM. (1991). Covalent bonding of
12 bay-region diol epoxides to nucleic acids [Review]. Adv Exp Med Biol 283: 533-553.
13 Jiang. H; Gelhaus. SL; Mangal. D; Harvey. RG; Blair. IA; Penning. TM. (2007). Metabolism of
14 benzo[a]pyrene in human bronchoalveolar H358 cells using liquid chromatography-mass
15 spectrometry. Chem Res Toxicol 20: 1331-1341. http://dx.doi.org/10.1021/tx700107z
16 Jiang. H; Shen. YM; Quinn. AM; Penning. TM. (2005). Competing roles of cytochrome P450 1A1/1B1 and
17 aldo-keto reductase 1A1 in the metabolic activation of (+/-)-7,8-Dihydroxy-7,8-dihydro-
18 benzo[a]pyrene in human bronchoalveolar cell extracts. Chem Res Toxicol 18: 365-374.
19 http://dx.doi.org/10.1021/tx0497245
20 Jules. GE; Pratap. S; Ramesh. A; Hood. DB. (2012). In utero exposure to benzo(a)pyrene predisposes
21 offspring to cardiovascular dysfunction in later-life. Toxicology 295: 56-67.
22 http://dx.doi.0rg/10.1016/i.tox.2012.01.017
23 Kao, J; Patterson, FK; Hall, J. (1985). Skin penetration and metabolism of topically applied chemicals in
24 six mammalian species, including man: an in vitro study with benzo[a]pyrene and testosterone.
25 Toxicol Appl Pharmacol 81: 502-516.
26 Kapitulnik, J; Wislocki, PG; Levin, W; Yagi, H; Jerina, DM; Conney, AH. (1978). Tumorigenicity studies with
27 diol-epoxides of benzo(a)pyrene which indicate that "(+/-)"-trans-7"beta",8"alpha"-dihydroxy-
28 9"alpha" 10"alpha"-epoxy-7,8,9,10-tetrahydrobenzo(a)pyrene is an ultimate carcinogen in
29 newborn mice. Cancer Res 38: 354-358.
30 Kappes, U; Schliemann-Willers, S; Bankova, L; Heinemann, C; Fischer, TW; Ziemer, M; Schubert, H;
31 Norgauer. J; Fluhr. JW; Eisner. P. (2004). The quality of human skin xenografts on SCID mice: A
32 noninvasive bioengineering approach. Br J Dermatol 151: 971-976.
33 http://dx.doi.0rg/10.llll/i.1365-2133.2004.06191.x
34 Karakava. A; Ates. I; Yucesoy. B. (2004). Effects of occupational polycyclic aromatic hydrocarbon
35 exposure on T-lymphocyte functions and natural killer cell activity in asphalt and coke oven
36 workers. Hum Exp Toxicol 23: 317-322.
37 Karakava. A; Yucesoy. B; Turhan. A; Erdem. O; Burgaz. S; Karakava. AE. (1999). Investigation of some
38 immunological functions in a group of asphalt workers exposed to polycyclic aromatic
39 hydrocarbons. Toxicology 135: 43-47.
40 Karlehagen. S; Andersen. A; Ohlson. CG. (1992). Cancer incidence among creosote-exposed workers.
41 Scand J Work Environ Health 18: 26-29. http://dx.doi.org/10.5271/siweh.1612
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 Kee. K; Flores. M; Cedars. Ml; Reijo Pera. RA. (2010). Human primordial germ cell formation is
2 diminished by exposure to environmental toxicants acting through the AHR signaling pathway.
3 Toxicol Sci 117: 218-224. http://dx.doi.org/10.1093/toxsci/kfql79
4 Keohavong, P; Lan, Q; Gao, WM; DeMarini, DM; Mass, MJ; Li, XM; Roop, BC; Weissfeld, J; Tian, D;
5 Mumford, JL. (2003). K-ras mutations in lung carcinomas from nonsmoking women exposed to
6 unvented coal smoke in China. Lung Cancer 41: 21-27. http://dx.doi.org/10.1016/S0169-
7 5002(03)00125-9
8 Ketkar, M; Reznik, G; Schneider, P; Mohr, U. (1978). Investigations on the carcinogenic burden by air
9 pollution in man Intratracheal instillation studies with benzo(a)pyrene in bovine serum albumin
10 in Syrian hamsters. Cancer Lett 4: 235-239.
11 Klemme, JC; Mukhtar, H; Elmets, CA. (1987). Induction of contact hypersensitivity to
12 dimethylbenz(a)anthracene and benzo(a)pyrene in C3H/HeN mice. Cancer Res 47: 6074-6078.
13 Knafla. A; Petrovic. S; Richardson. M; Campbell. J; Rowat. C. (2011). Development and application of a
14 skin cancer slope factor for exposures to benzo[a]pyrene in soil. Regul Toxicol Pharmacol 59:
15 101-110. http://dx.doi.0rg/10.1016/i.vrtph.2010.09.011
16 Knuckles. ME; Inyang. F; Ramesh. A. (2001). Acute and subchronic oral toxicities of benzo[a]pyrene in F-
17 344 rats. Toxicol Sci 61: 382-388.
18 Kolb. B; Whishaw. IQ. (1989). Plasticity in the neocortex: mechanisms underlying recovery from early
19 brain damage. Prog Neurobiol 32: 235-276.
20 Kong. LY; Luster. Ml; Dixon. D; O'Grady. J; Rosenthal. GJ. (1994). Inhibition of lung immunity after
21 intratracheal instillation of benzo(a)pyrene. Am J Respir Crit Care Med 150: 1123-1129.
22 Koreeda, M; Moore, PD; Wislocki, PG; Levin, W; Yagi, H; Jerina, DM. (1978). Binding of benzo[a]pyrene
23 7,8-diol-9,10-epoxides to DNA, RNA, and protein of mouse skin occurs with high
24 stereoselectivity. Science 199: 778-781.
25 Kristensen, P; Eilertsen, E; Einarsdottir, E; Haugen, A; Skaug, V; Ovrebo, S. (1995). Fertility in mice after
26 prenatal exposure to benzo[a]pyrene and inorganic lead. Environ Health Perspect 103: 588-590.
27 http://dx.doi.org/10.1289/ehp.95103588
28 Kroese, ED; Muller, JJA; Mohn, GR; Dortant, PM; Wester, PW. (2001). Tumorigenic effects in Wistar rats
29 orally administered benzo[a]pyrene for two years (gavage studies): Implications for human
30 cancer risks associated with oral exposure to polycyclic aromatic hydrocarbons. (658603 010).
31 Bilthoven, The Netherlands: National Institute for Public Health and the Environment (RIVM).
32 http://www.rivm.nl/bibliotheek/rapporten/658603010.pdf
33 Kummer. V; Maskova. J; Zraly. Z; Matiasovic. J; Faldyna. M. (2007). Effect of postnatal exposure to
34 benzo[a]pyrene on the uterus of immature rats. Exp Toxicol Pathol 59: 69-76.
35 http://dx.doi.0rg/10.1016/i.etp.2007.02.007
36 Kummer. V; Maskova. J; Zraly. Z; Neca. J; Simeckova. P; Vondracek. J; Machala. M. (2008). Estrogenic
37 activity of environmental polycyclic aromatic hydrocarbons in uterus of immature Wistar rats.
38 Toxicol Lett 180: 212-221. http://dx.doi.Org/10.1016/i.toxlet.2008.06.862
39 Kuper. CF; Beems. RB; Bruiintjes. JP; Schuurman. HJ; Vos. JG. (1992). Normal development, growth, and
40 aging of the thymus. In U Mohr; DL Dungworth; CC Capen (Eds.), Pathobiology of the aging rat,
41 volume 1 (pp. 25-49). Washington, DC: ILSI press.
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 Kuper. CF; De Heer. E; Van Loveren. H; Vos. JG. (2002). Immune system. In MA Wallig (Ed.), Handbook of
2 toxicologic pathology (2nd ed., pp. 585-644). San Diego, CA: Academic Press.
3 Landreth, KS. (2002). Critical windows in development of the rodent immune system [Review]. Hum Exp
4 Toxicol 21: 493-498.
5 Landreth, KS; Dodson, SVM. (2005). Development of the rodent immune system. In SD Holladay (Ed.),
6 Developmental immunotoxicology (pp. 3-19). Boca Raton, FL: CRC Press.
7 http://dx.doi.org/10.1201/9781420024036.ptl
8 Lavoie, EJ; Amin, S; Hecht, SS; Furuya, K; Hoffmann, D. (1982). Tumour initiating activity of dihydrodiols
9 of benzo[b]fluoranthene, benzo[j]fluoranthene, and benzo[k]fluoranthene. Carcinogenesis 3:
10 49-52. http://dx.doi.0rg/10.1093/carcin/3.l.49
11 Lavoie, EJ; Braley, J; Rice, JE; Rivenson, A. (1987). Tumorigenic activity of non-alternant polynuclear
12 aromatic hydrocarbons in newborn mice. Cancer Lett 34: 15-20.
13 Lavoie. EJ; Cai. ZW; Meschter. CL; Wevand. EH. (1994). Tumorigenic activity of fluoranthene, 2-
14 methylfluoranthene and 3-methylfluoranthene in newborn CD-I mice. Carcinogenesis 15: 2131-
15 2135. http://dx.doi.org/10.1093/carcin/15.10.2131
16 Lavoue. J; Gerin. M; Cote. J; Lapointe. R. (2007). Mortality and cancer experience of Quebec aluminum
17 reduction plant workers. Part I: The reduction plants and coal tar pitch volatile (CTPV) exposure
18 assessment. J Occup Environ Med 49: 997-1008.
19 http://dx.doi.org/10.1097/JOM.Ob013e3181484cf3
20 Levin. W; Wood. AW; Wislocki. PG; Kapitulnik. J; Yagi. H; Jerina. DM; Conney. AH. (1977). Carcinogenicity
21 of benzo-ring derivatives of benzo(a)pyrene on mouse skin. Cancer Res 37: 3356-3361.
22 Li, D; Firozi, PF; Wang, LE; Bosken, CH; Spitz, MR; Hong, WK; Wei, Q. (2001). Sensitivity to DNA damage
23 induced by benzo(a)pyrene diol epoxide and risk of lung cancer: a case-control analysis. Cancer
24 Res 61: 1445-1450.
25 Li, Z; Chadalapaka, G; Ramesh, A; Khoshbouei, H; Maguire, M; Safe, S; Rhoades, RE; Clark, R; Jules, G;
26 Mccallister, M; Aschner, M; Hood, DB. (2012). PAH particles perturb prenatal processes and
27 phenotypes: protection from deficits in object discrimination afforded by dampening of brain
28 oxidoreductase following in utero exposure to inhaled benzo(a)pyrene. Toxicol Sci 125: 233-247.
29 http://dx.doi.org/10.1093/toxsci/kfr261
30 Liu, M; Grant, SG; Macina, OT; Klopman, G; Rosenkranz, HS. (1997). Structural and mechanistic bases for
31 the induction of mitotic chromosomal loss and duplication (malsegregation1) in the yeast
32 Saccharomyces cerevisiae: Relevance to human Carcinogenesis and developmental toxicology.
33 Mutat Res-Fundam Mol Mech Mutagen 374: 209-231. http://dx.doi.org/10.1016/S0027-
34 5107(96)00236-9
35 Liu. S; Abdelrahim. M; Khan. S; Ariazi. E; Jordan. VC; Safe. S. (2006). Aryl hydrocarbon receptor agonists
36 directly activate estrogen receptor alpha in MCF-7 breast cancer cells. Biol Chem 387: 1209-
37 1213. http://dx.doi.org/10.1515/BC.2006.149
38 Liu. SH; Wang. JH; Chuu. JJ; Lin-Shiau. SY. (2002). Alterations of motor nerve functions in animals
39 exposed to motorcycle exhaust. J Toxicol Environ Health A 65: 803-812.
40 http://dx.doi.org/10.1080/00984100290071144
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 Liu. Z; Muehlbauer. KR; Schmeiser. HH; Hergenhahn. M; Belharazem. D; Hollstein. MC. (2005). p53
2 mutations in benzo(a)pyrene-exposed human p53 knock-in murine fibroblasts correlate with
3 p53 mutations in human lung tumors. Cancer Res 65: 2583-2587.
4 http://dx.doi.org/10.1158/0008-5472.CAN-04-3675
5 Luebke, RW; Chen, DH; Dietert, R; Yang, Y; King, M; Luster, Ml. (2006). The comparative immunotoxicity
6 of five selected compounds following developmental or adult exposure [Review]. J Toxicol
7 Environ Health B Crit Rev 9: 1-26. http://dx.doi.org/10.1080/15287390500194326
8 Lummus, ZL; Henningsen, G. (1995). Modulation of T-cell ontogeny by transplacental benzo(a)pyrene.
9 International Journal of Immunopharmacology 17: 339-350.
10 Luster, Ml; Portier, C; Pait, DG; White, KL, Jr; Gennings, C; Munson, AE; Rosenthal, GJ. (1992). Risk
11 assessment in immunotoxicology: I: Sensitivity and predictability of immune tests. Fundam Appl
12 Toxicol 18: 200-210. http://dx.doi.org/10.1016/0272-0590(92)90047-1
13 Lyte. M; Bick. PH. (1985). Differential immunotoxic effects of the environmental chemical
14 benzo[a]pyrene in young and aged mice. Mech Ageing Dev 30: 333-341.
15 http://dx.doi.org/10.1016/0047-6374(85)90122-8
16 Ma. Q; Lu. AY. (2007). CYP1A induction and human risk assessment: an evolving tale of in vitro and in
17 vivo studies [Review]. Drug Metab Dispos 35: 1009-1016.
18 http://dx.doi.org/10.1124/dmd.107.015826
19 Mackall. CL; Fleisher. TA; Brown. MR; Andrich. MP; Chen. CC; Feuerstein. IM; Horowitz. ME; Magrath. IT;
20 Shad. AT; Steinberg. SM. (1995). Age, thymopoiesis, and CD4+ T-lymphocyte regeneration after
21 intensive chemotherapy. N Engl J Med 332: 143-149.
22 http://dx.doi.org/10.1056/NEJM199501193320303
23 Mackenzie, KM; Angevine, DM. (1981). Infertility in mice exposed in utero to benzo(a)pyrene. Biol
24 Reprod 24: 183-192.
25 Mahler, BJ; Van Metre, PC; Bashara, TJ; Wilson, JT; Johns, DA. (2005). Parking lot sealcoat: an
26 unrecognized source of urban polycyclic aromatic hydrocarbons. Environ Sci Technol 39: 5560-
27 5566. http://dx.doi.org/10.1021/es0501565
28 Malaveille, C; Kuroki, T; Sims, P; Grover, PL; Bartsch, H. (1977). Mutagenicity of isomeric diol-epoxides of
29 benzo[a]pyrene and benz[a]anthracene in S. typhimurium TA98 and TA100 and in V79 Chinese
30 hamster cells. Mutat Res 44: 313-326.
31 Mancini. R; Romano. G; Sgambato. A; Flamini. G; Giovagnoli. MR; Boninsegna. A; Carraro. C; Vecchione.
32 A; Cittadini. A. (1999). Polycyclic aromatic hydrocarbon-DNA adducts in cervical smears of
33 smokers and nonsmokers. Gynecol Oncol 75: 68-71. http://dx.doi.org/10.1006/gyno.1999.5525
34 Marshall. CJ; Vousden. KH; Phillips. DH. (1984). Activation of c-Ha-ras-1 proto-oncogene by in vitro
35 modification with a chemical carcinogen, benzo(a)pyrene diol-epoxide. Nature 310: 586-589.
36 http://dx.doi.org/10.1038/310586aO
37 Mass. MJ; Jeffers. AJ; Ross. JA; Nelson. G; Galati. AJ; Stoner. GD; Nesnow. S. (1993). Ki-ras oncogene
38 mutations in tumors and DNA adducts formed by benz[j]aceanthrylene and benzo[a]pyrene in
39 the lungs of strain A/J mice. Mol Carcinog 8: 186-192.
40 http://dx.doi.org/10.1002/mc.2940080309
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 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
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 children living in Upper Silesia, Poland. Mutagenesis 21: 295-304.
2 http://dx.doi.org/10.1093/mutage/gel037
3 Miller, KP; Ramos, KS. (2001). Impact of cellular metabolism on the biological effects of benzo[a]pyrene
4 and related hydrocarbons [Review]. Drug Metab Rev 33: 1-35. http://dx.doi.org/10.1081/DMR-
5 100000138
6 Miller, MM; Plowchalk, PR; Weitzman, GA; London, SIM; Mattison, DR. (1992). The effect of
7 benzo(a)pyrene on murine ovarian and corpora lutea volumes. Am J Obstet Gynecol 166: 1535-
8 1541.
9 Mohamed. e; Song. WH; Oh. SA; Park. YJ; You. YA; Lee. S; Choi. JY; Kim. YJ; Jo. I; Pang. MG. (2010). The
10 transgenerational impact of benzo(a)pyrene on murine male fertility. Hum Reprod 25: 2427-
11 2433. http://dx.doi.org/10.1093/humrep/deq205
12 Moody. RP; Chu. I. (1995). Dermal exposure to environmental contaminants in the Great Lakes [Review].
13 Environ Health Perspect 103 Suppl 9: 103-114.
14 Moody. RP; Joncas. J; Richardson. M; Chu. I. (2007). Contaminated soils (I): In vitro dermal absorption of
15 benzo[a]pyrene in human skin. J Toxicol Environ Health A 70: 1858-1865.
16 http://dx.doi.org/10.1080/15287390701459296
17 Moolgavkar. SH; Knudson. AG. Jr. (1981). Mutation and cancer: A model for human carcinogenesis. J
18 Natl Cancer Inst 66: 1037-1052.
19 Moolgavkar. SH; Venzon. DJ. (1979). Two-event models for carcinogenesis: Incidence curves for
20 childhood and adult tumors. Math Biosci 47: 55-77. http://dx.doi.org/10.1016/0025-
21 5564(79)90005-1
22 Moulin, JJ; Wild, P; Mur, JM; Lafontaine, M; Lefer, M; Mercier-Gallay, M; Villemot, P; Whebi, V; Coulon,
23 JP. (1989). Risk of lung, larynx, pharynx and buccal cavity cancers among carbon electrode
24 manufacturing workers. Scand J Work Environ Health 15: 30-37.
25 http://dx.doi.org/10.5271/siweh.1885
26 Munson, AE; White, KL; Lysy, HH. (1985). Effects of Subchronic Fourteen Day Exposure to Benzopyrene
27 in B6C3F1 Female Mice on Host Resistance. Submitted under TSCA Section FYI.
28 Munson, AE; White, KU. (1983). Effects of subchronic 14-day exposure to benzopyrene in b6c3fl male
29 and female mice on specific immunological parameters. American Petroleum Institute.
30 N'Diaye, M; Le Ferrec, E; Lagadic-Gossmann, D; Corre, S; Gilot, D; Lecureur, V; Monteiro, P; Rauch, C;
31 Galibert. MD; Fardel. O. (2006). Aryl hydrocarbon receptor- and calcium-dependent induction of
32 the chemokine CCL1 by the environmental contaminant benzo[a]pyrene. J Biol Chem 281:
33 19906-19915. http://dx.doi.org/10.1074/ibc.M601192200
34 Nanez. A; Ramos. IN; Ramos. KS. (2011). A mutant Ahr allele protects the embryonic kidney from
35 hydrocarbon-induced deficits in fetal programming. Environ Health Perspect 119: 1745-1753.
36 http://dx.doi.org/10.1289/ehp.1103692
37 Naslund. I; Rubio. CA; Auer. GU. (1987). Nuclear DNA changes during pathogenesis of squamous
38 carcinoma of the cervix in 3,4-benzopyrene-treated mice. Anal Quant Cytol Histol 9: 411-418.
39 Naufal. Z: Zhiwen. L: Zhu. L: Zhou. GD: Mcdonald. T: He. LY: Mitchell. L: Ren. A: Zhu. H: Finnell. R:
40 Donnelly. KC. (2010). Biomarkers of exposure to combustion by-products in a human population
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 in Shanxi, China. J Expo Sci Environ Epidemiol 20: 310-319.
2 http://dx.doi.org/10.1038/ies.2009.19
3 Neal, J; Rigdon, RH. (1967). Gastric tumors in mice fed benzo(a)pyrene: A quantitative study. Tex Rep
4 Biol Med 25: 553-557.
5 Neal, MS; Mulligan Tuttle, AM; Casper, RF; Lagunov, A; Foster, WG. (2010). Aryl hydrocarbon receptor
6 antagonists attenuate the deleterious effects of benzo[a]pyrene on isolated rat follicle
7 development. Reprod Biomed Online 21: 100-108.
8 http://dx.doi.0rg/10.1016/i.rbmo.2010.03.025
9 Neal, MS; Zhu, J; Foster, WG. (2008). Quantification of benzo[a]pyrene and other PAHs in the serum and
10 follicular fluid of smokers versus non-smokers. Reprod Toxicol 25: 100-106.
11 http://dx.doi.0rg/10.1016/i.reprotox.2007.10.012
12 Neal. MS; Zhu. J; Holloway. AC; Foster. WG. (2007). Follicle growth is inhibited by benzo-[a]-pyrene, at
13 concentrations representative of human exposure, in an isolated rat follicle culture assay. Hum
14 Reprod 22: 961-967. http://dx.doi.org/10.1093/humrep/del487
15 Nebert. DW; Puga. A; Vasiliou. V. (1993). Role of the Ah receptor and the dioxin-inducible [Ah] gene
16 battery in toxicity, cancer, and signal transduction [Review]. Ann N Y Acad Sci 685: 624-640.
17 http://dx.doi.0rg/10.llll/i.1749-6632.1993.tb35928.x
18 Nesnow. S; Mass. MJ; Ross. JA; Galati. AJ; Lambert. GR; Gennings. C; Carter. WH; Stoner. GD. (1998a).
19 Lung tumorigenic interactions in strain A/J mice of five environmental polycyclic aromatic
20 hydrocarbons. Environ Health Perspect 106 Suppl 6: 1337-1346.
21 Nesnow, S; Ross, JA; Mass, MJ; Stoner, GD. (1998b). Mechanistic relationships between DNA adducts,
22 oncogene mutations, and lungtumorigenesis in strain A mice. Exp Lung Res 24: 395-405.
23 Nesnow, S; Ross, JA; Stoner, GD; Mass, MJ. (1995). Mechanistic linkage between DNA adducts,
24 mutations in oncogenes and tumorigenesis of carcinogenic environmental polycyclic aromatic
25 hydrocarbons in strain A/J mice. Toxicology 105: 403-413. http://dx.doi.org/10.1016/0300-
26 483X(95)03238-B
27 Nesnow, S; Ross, JA; Stoner, GD; Mass, MJ. (1996). Tumorigenesis of Carcinogenic Environmental
28 Polycyclic Aromatic Hydrocarbons in Strain A/J Mice: Linkage to DNA Adducts and Mutations in
29 Oncogenes. Polycycl Aromat Compd 10: 259-266.
30 http://dx.doi.org/10.1080/10406639608034705
31 Nesnow. S; Triplett. LL; Slaga. TJ. (1983). Mouse skin tumor initiation-promotion and complete
32 carcinogenesis bioassays: mechanisms and biological activities of emission samples. Environ
33 Health Perspect 47: 255-268.
34 Nettesheim. P; Griesemer. RA; Martin. DH; Caton. JE. Jr. (1977). Induction of preneoplastic and
35 neoplastic lesions in grafted rat tracheas continuously exposed to benzo(a)pyrene. Cancer Res
36 37: 1272-1278.
37 Newbold. RF; Brookes. P. (1976). Exceptional mutagenicity of a benzo(a)pyrene diol epoxide in cultured
38 mammalian cells. Nature 261: 52-54. http://dx.doi.org/10.1038/261052aO
39 Nikonova. TV. (1977). [Transplacental effect of benz(a)pyrene and pyrene]. Biull Eksp Biol Med 84: 88-
40 91.
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 Niu. Q; Zhang. H; Li. X; Li. M. (2010). Benzo[a]pyrene-induced neurobehavioral function and
2 neurotransmitter alterations in coke oven workers. Occup Environ Med 67: 444-448.
3 http://dx.doi.org/10.1136/oem.2009.047969
4 OECD (Organisation for Economic Co-operation and Development). (2009). Series on testing and
5 assesment: Number 39. Guidance document on acute inhalation toxicity testing [OECD SIDS].
6 (ENV/JM/MONO(2009)28; JT03268144). Paris, Fr.
7 http://search.oecd. org/officialdocuments/displavdocumentpdf/?cote=env/im/mono%282009%
8 2928&doclanguage=en
9 Okev, AB; Riddick, PS; Harper, PA. (1994). Molecular biology of the aromatic hydrocarbon (dioxin)
10 receptor [Review]. Trends Pharmacol Sci 15: 226-232. http://dx.doi.org/10.1016/0165-
11 6147(94)90316-6
12 Olsen. AK; Andreassen. A; Singh. R; Wiger. R; Duale. N; Farmer. PB; Brunborg. G. (2010). Environmental
13 exposure of the mouse germ line: DNA adducts in spermatozoa and formation of de novo
14 mutations during spermatogenesis. PLoS ONE 5: el!349.
15 http://dx.doi.org/10.1371/iournal.pone.0011349
16 Olsson. A; Kromhout. H; Agostini. M; Hansen. J; Lassen. CF; Johansen. C; Kjaerheim. K; Langard. S;
17 Stucker. I; Ahrens. W; Behrens. T; Lindbohm. ML; Heikkila. P; Heederik. D; Portengen. L; Shaham.
18 J; Ferro. G; de Vocht. F; Burstyn. I; Boffetta. P. (2010). A case-control study of lung cancer nested
19 in a cohort of European asphalt workers. Environ Health Perspect 118: 1418-1424.
20 http://dx.doi.org/10.1289/ehp.0901800
21 Park, JH; Gelhaus, S; Vedantam, S; Oliva, AL; Batra, A; Blair, IA; Troxel, AB; Field, J; Penning, TM. (2008).
22 The pattern of p53 mutations caused by PAH o-quinones is driven by 8-oxo-dGuo formation
23 while the spectrum of mutations is determined by biological selection for dominance. Chem Res
24 Toxicol 21: 1039-1049. http://dx.doi.org/10.1021/tx700404a
25 Park, JH; Troxel, AB; Harvey, RG; Penning, TM. (2006). Polycyclic aromatic hydrocarbon (PAH) o-
26 quinones produced by the aldo-keto-reductases (AKRs) generate abasic sites, oxidized
27 pyrimidines, and 8-oxo-dGuo via reactive oxygen species. Chem Res Toxicol 19: 719-728.
28 http://dx.doi.org/10.1021/tx0600245
29 Pate Capps, N; Stewart, A; Burns, C. (2009). The interplay between secondhand cigarette smoke,
30 genetics, and cervical cancer: a review of the literature [Review]. Biol Res Nurs 10: 392-399.
31 http://dx.doi.org/10.1177/1099800408330849
32 Pavanello. S; Favretto. D; Brugnone. F; Mastrangelo. G; Dal Pra. G; Clonfero. E. (1999).
33 HPLC/fluorescence determination of anti-BPDE-DNA adducts in mononuclear white blood cells
34 from PAH-exposed humans. Carcinogenesis 20: 431-435.
35 Pavanello. S; Pulliero. A; Siwinska. E; Mielzynska. D; Clonfero. E. (2005). Reduced nucleotide excision
36 repair and GSTMl-null genotypes influence anti-B[a]PDE-DNA adduct levels in mononuclear
37 white blood cells of highly PAH-exposed coke oven workers. Carcinogenesis 26: 169-175.
38 http://dx.doi.org/10.1093/carcin/bgh303
39 Pavanello. S; Siwinska. E; Mielzynska. D; Clonfero. E. (2004). GSTM1 null genotype as a risk factor for
40 anti-BPDE-DNA adduct formation in mononuclear white blood cells of coke-oven workers. Mutat
41 Res 558: 53-62. http://dx.doi.Org/10.1016/i.mrgentox.2003.10.019
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 Penit. C; Vasseur. F. (1989). Cell proliferation and differentiation in the fetal and early postnatal mouse
2 thymus. J Immunol 142: 3369-3377.
3 Penning, TM; Burczynski, ME; Hung, CF; Mccoull, KD; Palackal, NT; Tsuruda, LS. (1999). Dihydrodiol
4 dehydrogenases and polycyclic aromatic hydrocarbon activation: generation of reactive and
5 redox active o-quinones [Review]. Chem Res Toxicol 12: 1-18.
6 http://dx.doi.org/10.1021/tx980143n
7 Penning, TM; Ohnishi, ST; Ohnishi, T; Harvey, RG. (1996). Generation of reactive oxygen species during
8 the enzymatic oxidation of polycyclic aromatic hydrocarbon trans-dihydrodiols catalyzed by
9 dihydrodiol dehydrogenase. Chem Res Toxicol 9: 84-92. http://dx.doi.org/10.1021/tx950055s
10 Perera, F; Li, TY; Lin, C; Tang, D. (2012a). Effects of prenatal polycyclic aromatic hydrocarbon exposure
11 and environmental tobacco smoke on child IQ in a Chinese cohort. Environ Res 114: 40-46.
12 http://dx.doi.0rg/10.1016/i.envres.2011.12.011
13 Perera. FP; Li. Z; Whvatt. R; Hoepner. L; Wang. S; Camann. D; Rauh. V. (2009). Prenatal airborne
14 polycyclic aromatic hydrocarbon exposure and child IQ at age 5 years. Pediatrics 124: e!95-
15 e202. http://dx.doi.org/10.1542/peds.2008-3506
16 Perera. FP; Rauh. V; Whvatt. RM; Tang. D; Tsai. WY; Bernert. JT; Tu. YH; Andrews. H; Barr. DB; Camann.
17 DE; Diaz. D; Dietrich. J; Reyes. A; Kinney. PL. (2005a). A summary of recent findings on birth
18 outcomes and developmental effects of prenatal ETS, PAH, and pesticide exposures [Review].
19 Neurotoxicology 26: 573-587. http://dx.doi.Org/10.1016/i.neuro.2004.07.007
20 Perera. FP; Rauh. V; Whvatt. RM; Tsai. WY; Bernert. JT; Tu. YH; Andrews. H; Ramirez. J; Qu. L; Tang. D.
21 (2004). Molecular evidence of an interaction between prenatal environmental exposure and
22 birth outcomes in a multiethnic population. Environ Health Perspect 112: 626-630.
23 Perera, FP; Tang, D; Rauh, V; Lester, K; Tsai, WY; Tu, YH; Weiss, L; Hoepner, L; King, J; Del Priore, G;
24 Lederman, SA. (2005b). Relationships among polycyclic aromatic hydrocarbon-DNA adducts,
25 proximity to the World Trade Center, and effects on fetal growth. Environ Health Perspect 113:
26 1062-1067.
27 Perera, FP; Tang, D; Wang, S; Vishnevetskv, J; Zhang, B; Diaz, D; Camann, D; Rauh, V. (2012b). Prenatal
28 Polycyclic Aromatic Hydrocarbon (PAH) Exposure and Child Behavior at Age 6-7 Years. Environ
29 Health Perspect 120: 921-926. http://dx.doi.org/10.1289/ehp.1104315
30 Perera, FP; Weinstein, IB. (2000). Molecular epidemiology: Recent advances and future directions
31 [Review]. Carcinogenesis 21: 517-524. http://dx.doi.Org/10.1093/carcin/21.3.517
32 Perrin. J; Tassistro. V; Mandon. M; Grillo. JM; Botta. A; Sari-Minodier. I. (2011a). Tobacco consumption
33 and benzo(a)pyrene-diol-epoxide-DNA adducts in spermatozoa: in smokers, swim-up procedure
34 selects spermatozoa with decreased DNA damage. Fertil Steril 95: 2013-2017.
35 http://dx.doi.0rg/10.1016/i.fertnstert.2011.02.021
36 Perrin. J; Tassistro. V; Paulmyer-Lacroix. O; Courbiere. B; Botta. A; Sari-Minodier. I. (2011b). In smokers.
37 swim-up and discontinuous gradient centrifugation recover spermatozoa with equally lower
38 amounts of DNA damage than spermatozoa obtained from neat semen. Fertil Steril 95: 2680-
39 2682. http://dx.doi.0rg/10.1016/i.fertnstert.2011.04.092
40 Petrie. HT. (2002). Role of thymic organ structure and stromal composition in steady-state postnatal T-
41 cell production [Review]. Immunol Rev 189: 8-19.
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 Pfeifer. GP; Denissenko. MF; Olivier. M; Tretyakova. N; Hecht. SS; Hainaut. P. (2002). Tobacco smoke
2 carcinogens, DNA damage and p53 mutations in smoking-associated cancers [Review].
3 Oncogene 21: 7435-7451. http://dx.doi.org/10.1038/si.onc.1205803
4 Pfeifer, GP; Hainaut, P. (2003). On the origin of G --> T transversions in lung cancer [Review]. Mutat Res
5 526: 39-43. http://dx.doi.org/10.1016/S0027-5107(03)00013-7
6 Pfeiffer, CA; Allen, E. (1948). Attempts to produce cancer in rhesus monkeys with carcinogenic
7 hydrocarbons and estrogens. Cancer Res 8: 97-127.
8 Pfeiffer, EH. (1977). Oncogenic interaction of carcinogenic and non-carcinogenic polycyclic aromatic
9 hydrocarbons in mice. In Air pollution and cancer in man. Lyon, France: IARC.
10 Phillips, PH. (2002). Smoking-related DNA and protein adducts in human tissues [Review].
11 Carcinogenesis 23: 1979-2004.
12 Pittelkow. MR; Perry. HO; Muller. SA; Maughan. WZ; O'Brien. PC. (1981). Skin cancer in patients with
13 psoriasis treated with coal tar: A 25-year follow-up study. Arch Dermatol 117: 465-468.
14 Poel. WE. (1959). Effect of carcinogenic dosage and duration of exposure on skin-tumor induction in
15 mice. J Natl Cancer Inst 22: 19-43.
16 Poel. WE. (1960). Skin as a test site for the bioassay of carcinogens and carcinogen precursors [IARC
17 Monograph]. In IARC Monographs on The Evaluation of Carcinogenic Risk of Chemicals To Man
18 (pp. 611-631). Poel, WE.
19 Postlind. H; Vu. TP; Tukey. RH; Quattrochi. LC. (1993). Response of human CYPl-luciferase plasmids to
20 2,3,7,8-tetrachlorodibenzo-p-dioxin and polycyclic aromatic hydrocarbons. Toxicol Appl
21 Pharmacol 118: 255-262. http://dx.doi.org/10.1006/taap.1993.1031
22 Pru, JK; Kaneko-Tarui, T; Jurisicova, A; Kashiwagi, A; Selesniemi, K; Tilly, JL. (2009). Induction of
23 proapoptotic gene expression and recruitment of p53 herald ovarian follicle loss caused by
24 polycyclic aromatic hydrocarbons. Reproductive Sciences 16: 347-356.
25 http://dx.doi.org/10.1177/1933719108327596
26 Puisieux, A; Lim, S; Groopman, J; Ozturk, M. (1991). Selective targeting of p53 gene mutational hotspots
27 in human cancers by etiologically defined carcinogens. Cancer Res 51: 6185-6189.
28 Pukkala, E. (1995). Cancer risk by social class and occupation : a survey of 109,000 cancer cases among
29 Finns of working age. In J Wahrendorf (Ed.). Basel, Switzerland: Karger.
30 OJu, C; Cheng, S; Xia, Y; Peng, B; Tang, Q; Tu, B. (2011). Effects of subchronic benzo(a)pyrene exposure
31 on neurotransmitter receptor gene expression in the rat hippocampus related with spatial
32 learning and memory change. Toxicology 289: 83-90.
33 http://dx.doi.0rg/10.1016/i.tox.2011.07.012
34 Ramesh. A; Inyang. F; Lunstra. DP; Niaz. MS; Kopsombut. P; Jones. KM; Hood. DB; Hills. ER; Archibong.
35 AE. (2008). Alteration of fertility endpoints in adult male F-344 rats by subchronic exposure to
36 inhaled benzo(a)pyrene. Exp Toxicol Pathol 60: 269-280.
37 http://dx.doi.0rg/10.1016/i.etp.2008.02.010
38 Ramesh. A; Walker. SA; Hood. DB; Guillen. MD; Schneider. K; Weyand. EH. (2004). Bioavailability and risk
39 assessment of orally ingested polycyclic aromatic hydrocarbons [Review]. Int J Toxicol 23: 301-
40 333. http://dx.doi.org/10.1080/10915810490517063
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 Ramos. KS; Moorthy. B. (2005). Bioactivation of polycyclic aromatic hydrocarbon carcinogens within the
2 vascular wall: implications for human atherogenesis [Review]. Drug Metab Rev 37: 595-610.
3 http://dx.doi.org/10.1080/03602530500251253
4 Rask-Nielsen, R. (1950). On the susceptibility of the thymus, lung, subcutaneous and mammary tissues in
5 strain street mice to direct application of small doses of four different carcinogenic
6 hydrocarbons. Br J Cancer 4: 108-116.
7 Raveh, D; Slaga, TJ; Huberman, E. (1982). Cell-mediated mutagenesis and tumor-initiating activity of the
8 ubiquitous polycyclic hydrocarbon, cyclopenta[c,d]pyrene. Carcinogenesis 3: 763-766.
9 http://dx.doi.Org/10.1093/carcin/3.7.763
10 Raychoudhury, SS; Kubinski, D. (2003). Polycyclic aromatic hydrocarbon-induced cytotoxicity in cultured
11 rat Sertoli cells involves differential apoptotic response. Environ Health Perspect 111: 33-38.
12 Revel. A; Raanani. H; Younglai. E; Xu. J; Han. R; Savouret. JF; Casper. RF. (2001). Resveratrol, a natural
13 aryl hydrocarbon receptor antagonist, protects sperm from DNA damage and apoptosis caused
14 by benzo(a)pyrene. Reprod Toxicol 15: 479-486.
15 Rice. JE; Makowski. GS; Hosted. TJ; Lavoie. EJ. (1985). Methylene-bridged bay region chrysene and
16 phenanthrene derivatives and their keto-analogs: mutagenicity in Salmonella typhimurium and
17 tumor-initiating activity on mouse skin. Cancer Lett 27: 199-206.
18 http://dx.doi.org/10.1016/0304-3835(84)90119-8
19 Robinson. M; Laurie. RD; Bull. RJ; Stober. JA. (1987). Carcinogenic effects in A/J mice of particulate of a
20 coal tar paint used in potable water systems. Cancer Lett 34: 49-54.
21 http://dx.doi.org/10.1016/0304-3835(87)90072-3
22 Robles. R; Morita. Y; Mann. KK; Perez. Gl; Yang. S; Matikainen. T; Sherr. DH; Tilly. JL (2000). The aryl
23 hydrocarbon receptor, a basic helix-loop-helix transcription factor of the PAS gene family, is
24 required for normal ovarian germ cell dynamics in the mouse. Endocrinology 141: 450-453.
25 Rodriguez, JW; Kirlin, WG; Wirsiy, YG; Matheravidathu, S; Hodge, TW; Urso, P. (1999). Maternal
26 exposure to benzo[a]pyrene alters development of T lymphocytes in offspring.
27 Immunopharmacol Immunotoxicol 21: 379-396. http://dx.doi.org/10.3109/08923979909052769
28 Roe, FJ; Peto, R; Kearns, F; Bishop, D. (1970). The mechanism of carcinogenesis by the neutral fraction of
29 cigarette smoke condensate. Br J Cancer 24: 788-806.
30 Roe, FJ; Waters, MA. (1967). Induction of hepatoma in mice by carcinogens of the polycyclic
31 hydrocarbon type. Nature 214: 299-300.
32 Roelofzen. JH; Aben. KK; Oldenhof. UT; Coenraads. PJ; Alkemade. HA; van de Kerkhof. PC; van der Valk.
33 PG; Kiemeney. LA. (2010). No increased risk of cancer after coal tar treatment in patients with
34 psoriasis or eczema. J Invest Dermatol 130: 953-961. http://dx.doi.org/10.1038/iid.2009.389
35 Rogan. EG; Devanesan. PD; Ramakrishna. NV; Higginbotham. S; Padmavathi. NS; Chapman. K; Cavalieri.
36 EL; Jeong. H; Jankowiak. R; Small. GJ. (1993). Identification and quantitation of benzo[a]pyrene-
37 DNA adducts formed in mouse skin. Chem Res Toxicol 6: 356-363.
38 http://dx.doi.org/10.1021/tx00033a017
39 Rojas. M; Alexandrov. K; Cascorbi. I; Brockmoller. J; Likhachev. A; Pozharisski. K; Bouvier. G; Auburtin. G;
40 Mayer. L; Kopp-Schneider. A; Roots. I; Bartsch. H. (1998). High benzo[a]pyrene diol-epoxide DNA
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 adduct levels in lung and blood cells from individuals with combined CYP1A1 Mspl/Msp-
2 GSTM1*0/*0 genotypes. Pharmacogenetics 8: 109-118.
3 Rojas, M; Cascorbi, I; Alexandrov, K; Kriek, E; Auburtin, G; Mayer, L; Kopp-Schneider, A; Roots, I; Bartsch,
4 hL (2000). Modulation of benzo[a]pyrene diolepoxide-DNA adduct levels in human white blood
5 cells by CYP1A1, GSTM1 and GSTT1 polymorphism. Carcinogenesis 21: 35-41.
6 http://dx.doi.0rg/10.1093/carcin/21.l.35
7 Rojas, M; Marie, B; Vignaud, JM; Martinet, N; Siat, J; Grosdidier, G; Cascorbi, I; Alexandrov, K. (2004).
8 High DNA damage by benzo[a]pyrene 7,8-diol-9,10-epoxide in bronchial epithelial cells from
9 patients with lung cancer: comparison with lung parenchyma. Cancer Lett 207: 157-163.
10 http://dx.doi.0rg/10.1016/i.canlet.2003.ll.016
11 Rubes, J; Rybar, R; Prinosilova, P; Veznik, Z; Chvatalova, I; Solansky, I; Sram, RJ. (2010). Genetic
12 polymorphisms influence the susceptibility of men to sperm DNA damage associated with
13 exposure to air pollution. Mutat Res 683: 9-15.
14 http://dx.doi.0rg/10.1016/i.mrfmmm.2009.09.010
15 Ruchirawat. M; Navasumrit. P; Settachan. D. (2010). Exposure to benzene in various susceptible
16 populations: co-exposures to 1,3-butadiene and PAHs and implications for carcinogenic risk.
17 Chem Biol Interact 184: 67-76. http://dx.doi.Org/10.1016/i.cbi.2009.12.026
18 Sadeu. JC; Foster. WG. (2011). Cigarette smoke condensate exposure delays follicular development and
19 function in a stage-dependent manner. Fertil Steril 95: 2410-2417.
20 http://dx.doi.0rg/10.1016/i.fertnstert.2011.03.072
21 Saffiotti, U; Montesano, R; Sellakumar, AR; Kaufman, DG. (1972). Respiratory tract carcinogenesis
22 induced in hamsters by different dose levels of benzo[a]pyrene and ferric oxide. J Natl Cancer
23 Inst 49: 1199-1204.
24 Sakkas, D; Alvarez, JG. (2010). Sperm DNA fragmentation: mechanisms of origin, impact on reproductive
25 outcome, and analysis [Review]. Fertil Steril 93: 1027-1036.
26 http://dx.doi.0rg/10.1016/i.fertnstert.2009.10.046
27 Saunders, CR; Das, SK; Ramesh, A; Shockley, DC; Mukherjee, S. (2006). Benzo(a)pyrene-induced acute
28 neurotoxicity in the F-344 rat: role of oxidative stress. J Appl Toxicol 26: 427-438.
29 http://dx.doi.org/10.1002/iat.1157
30 Saunders, CR; Ramesh, A; Shockley, DC. (2002). Modulation of neurotoxic behavior in F-344 rats by
31 temporal disposition of benzo(a)pyrene. Toxicol Lett 129: 33-45.
32 Saunders. CR; Shockley. DC; Knuckles. ME. (2001). Behavioral effects induced by acute exposure to
33 benzo(a)pyrene in F-344 rats. Neurotox Res 3: 557-579. http://dx.doi.org/10.1007/BF03033211
34 Schmahl. D; Schmidt. KG; Habs. M. (1977). Syncarcinogenic action of polycyclic hydrocarbons in
35 automobile exhaust gas condensates. In Air pollution and cancer in man. Lyon, France: IARC.
36 Schmidt. JV; Su. GH; Reddy. JK; Simon. MC; Bradfield. CA. (1996). Characterization of a murine Ahr null
37 allele: involvement of the Ah receptor in hepatic growth and development. PNAS 93: 6731-6736.
38 Schmidt. KG; Schma'hl. D; Misfeld. J. (1973). Investigations on the carcinogenic burden by air pollution in
39 man. VI. Experimental investigations to determine a dose-response relationship and to estimate
40 a threshold dose of benzo(a)pyrene in the skin of two different mouse strains. Zentralbl
41 Bakteriol, Parasitenkd, Infektionskrankh Hyg, Abt 1: Orig, Reihe B 158: 62-68.
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 Schnizlein. CT; Munson. AE; Rhoades. RA. (1987). Immunomodulation of local and systemic immunity
2 after subchronic pulmonary exposure of mice to benzo(a)pyrene. International Journal of
3 Immunopharmacology 9: 99-106.
4 Schonland, SO; Zimmer, JK; Lopez-Benitez, CM; Widmann, T; Ramin, KD; Goronzy, JJ; Weyand, CM.
5 (2003). Homeostatic control of T-cell generation in neonates. Blood 102: 1428-1434.
6 http://dx.doi.org/10.1182/blood-2002-ll-3591
7 Secretan, B; Straif, K; Baan, R; Grosse, Y; El Ghissassi, F; Bouvard, V; Benbrahim-Tallaa, L; Guha, N;
8 Freeman, C; Galichet, L; Cogliano, V. (2009). A review of human carcinogens-Part E: tobacco,
9 areca nut, alcohol, coal smoke, and salted fish. Lancet Oncol 10: 1033-1034.
10 Seli, E; Gardner, DK; Schoolcraft, WB; Moffatt, O; Sakkas, D. (2004). Extent of nuclear DNA damage in
11 ejaculated spermatozoa impacts on blastocyst development after in vitro fertilization. Fertil
12 Steril 82: 378-383. http://dx.doi.Org/10.1016/i.fertnstert.2003.12.039
13 Shamsuddin. AK; Gan. R. (1988). Immunocytochemical localization of benzo(a)pyrene-DNA adducts in
14 human tissue. Hum Pathol 19: 309-315. http://dx.doi.org/10.1016/S0046-8177(88)80524-0
15 Sharovskava. J; Kobliakova. I; Solomatina. N; Kobliakov. V. (2006). Effect of some carcinogenic and non-
16 carcinogenic polycyclic aromatic hydrocarbons on gap junction intercellular communication in
17 hepatoma cell cultures. Eur J Cell Biol 85: 387-397. http://dx.doi.Org/10.1016/i.eicb.2005.ll.006
18 Shen. H: Spitz. MR: Qiao. Y: Guo. Z: Wang. LE: Bosken. CH: Amos. Cl: Wei. Q. (2003). Smoking, DNA
19 repair capacity and risk of nonsmall cell lung cancer. Int J Cancer 107: 84-88.
20 http://dx.doi.org/10.1002/iic.11346
21 Shen, YM; Troxel, AB; Vedantam, S; Penning, TM; Field, J. (2006). Comparison of p53 mutations induced
22 by PAH o-quinones with those caused by anti-benzo[a]pyrene diol epoxide in vitro: role of
23 reactive oxygen and biological selection. Chem Res Toxicol 19: 1441-1450.
24 http://dx.doi.org/10.1021/tx0601206
25 Shimizu, Y; Nakatsuru, Y; Ichinose, M; Takahashi, Y; Kume, H; Mimura, J; Fujii-Kuriyama, Y; Ishikawa, T.
26 (2000). Benzo[a]pyrene carcinogenicity is lost in mice lacking the aryl hydrocarbon receptor.
27 PNAS 97: 779-782. http://dx.doi.Org/10.1073/pnas.97.2.779
28 Silverstone, AE; Cantor, H; Goldstein, G; Baltimore, D. (1976). Terminal deoxynucleotidyl transferase is
29 found in prothymocytes. J Exp Med 144: 543-548.
30 Sivak, A; Niemeier, R; Lynch, D; Beltis, K; Simon, S; Salomon, R; Latta, R; Belinky, B; Menzies, K; Lunsford,
31 A; Cooper. C; Ross. A; Bruner. R. (1997). Skin carcinogenicity of condensed asphalt roofing fumes
32 and their fractions following dermal application to mice. Cancer Lett 117: 113-123.
33 http://dx.doi.org/10.1016/S0304-3835(97)00214-0
34 Slaga. TJ; Gleason. GL; Mills. G; Ewald. L; Fu. PP; Lee. HM; Harvey. RG. (1980). Comparison of the skin
35 tumor-initiating activities of dihydrodiols and diol-epoxides of various polycyclic aromatic
36 hydrocarbons. Cancer Res 40: 1981-1984.
37 Slaga. TJ; Huberman. E; Selkirk. JK; Harvey. RG; Bracken. WM. (1978). Carcinogenicity and mutagenicity
38 of benz(a)anthracene diols and diol-epoxides. Cancer Res 38: 1699-1704.
39 Scares. SR; Melo. MA. (2008). Cigarette smoking and reproductive function [Review]. Curr Opin Obstet
40 Gynecol 20: 281-291. http://dx.doi.org/10.1097/GCO.Ob013e3282fc9cle
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 Spinelli. JJ; Band. PR; Svirchev. LM; Gallagher. RP. (1991). Mortality and cancer incidence in aluminum
2 reduction plant workers. J Occup Med 33: 1150-1155.
3 Spinelli, JJ; Demers, PA; Le, IMP; Friesen, MD; Lorenzi, MF; Fang, R; Gallagher, RP. (2006). Cancer risk in
4 aluminum reduction plant workers (Canada). Cancer Causes Control 17: 939-948.
5 http://dx.doi.org/10.1007/sl0552-006-0031-9
6 Steiner, PE. (1955). Carcinogenicity of multiple chemicals simultaneously administered. Cancer Res 15:
7 632-635.
8 Stephanou, P; Konstandi, M; Pappas, P; Marselos, M. (1998). Alterations in central monoaminergic
9 neurotransmission induced by polycyclic aromatic hydrocarbons in rats. European Journal of
10 Drug Metabolism and Pharmacokinetics 23: 475-481. http://dx.doi.org/10.1007/BF03189998
11 Stern, RS; Liebman, EJ; Vakeva, L. (1998). Oral psoralen and ultraviolet-A light (PUVA) treatment of
12 psoriasis and persistent risk of nonmelanoma skin cancer. PUVA Follow-up Study. J Natl Cancer
13 Inst 90: 1278-1284.
14 Stern. RS; Zierler. S; Parrish. JA. (1980). Skin carcinoma in patients with psoriasis treated with topical tar
15 and artificial ultraviolet radiation. Lancet 1: 732-735.
16 Straif. K; Baan. R; Grosse. Y; Secretan. B; El Ghissassi. F; Cogliano. V. (2005). Carcinogenicity of polycyclic
17 aromatic hydrocarbons. Lancet Oncol 6: 931-932. http://dx.doi.org/10.1016/S1470-
18 2045(05)70458-7
19 Sugiyama. T. (1973). Chromosomal aberrations and carcinogenesis by various benz[a]anthracene
20 derivatives. Cancer Sci 64: 637-639.
21 Suter, M; Abramovici, A; Showalter, L; Hu, M; Shope, CD; Varner, M; Aagaard-Tillery, K. (2010). In utero
22 tobacco exposure epigenetically modifies placental CYP1A1 expression. Metabolism 59: 1481-
23 1490. http://dx.doi.0rg/10.1016/i.metabol.2010.01.013
24 Swartz, WJ; Mattison, DR. (1985). Benzo(a)pyrene inhibits ovulation in C57BL/6N mice. Anat Rec 212:
25 268-276. http://dx.doi.org/10.1002/ar.1092120308
26 Szczeklik, A; Szczeklik, J; Galuszka, Z; Musial, J; Kolarzyk, E; Targosz, D. (1994). Humoral
27 immunosuppression in men exposed to polycyclic aromatic hydrocarbons and related
28 carcinogens in polluted environments. Environ Health Perspect 102: 302-304.
29 Szymanska, K; Matos, E; Hung, RJ; Wunsch-Filho, V; Eluf-Neto, J; Menezes, A; Daudt, AW; Brennan, P;
30 Boffetta, P. (2010). Drinking of mate and the risk of cancers of the upper aerodigestive tract in
31 Latin America: a case-control study. Cancer Causes Control 21: 1799-1806.
32 http://dx.doi.org/10.1007/sl0552-010-9606-6
33 Takizawa. K; Yagi. H; Jerina. DM; Mattison. DR. (1984). Murine strain differences in ovotoxicity following
34 intraovarian injection with benzo(a)pyrene, (+)-(7R,8S)-oxide, (-)-(7R,8R)-dihydrodiol, or (+)-
35 (7R,8S)-diol-(9S,10R)-epoxide-2. Cancer Res 44: 2571-2576.
36 Talaska. G; Ginsburg. D; Ladow. K; Puga. A; Dalton. T; Warshawsky. D. (2006). Impact of Cypla2 or Ahr
37 gene knockout in mice: implications for biomonitoring studies. Toxicol Lett 162: 246-249.
38 http://dx.doi.0rg/10.1016/i.toxlet.2005.09.020
39 Talaska. G; Jaeger. M; Reilman. R; Collins. T; Warshawsky. D. (1996). Chronic, topical exposure to
40 benzo[a]pyrene induces relatively high steady-state levels of DNA adducts in target tissues and
41 alters kinetics of adduct loss. PNAS 93: 7789-7793.
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 Tamaki. A; Hayashi. H; Nakajima. H; Takii. T; Katagiri. D; Miyazawa. K; Hirose. K; Onozaki. K. (2004).
2 Polycyclic aromatic hydrocarbon increases mRNA level for interleukin 1 beta in human
3 fibroblast-like synoviocyte line via aryl hydrocarbon receptor. Biol Pharm Bull 27: 407-410.
4 Tang. D; Li. TY; Liu. JJ; Chen. YH; Qu. L; Perera. F. (2006). PAH-DNA adducts in cord blood and fetal and
5 child development in a Chinese cohort. Environ Health Perspect 114: 1297-1300.
6 http://dx.doi.org/10.1289/ehp.8939
7 Tang, D; Warburton, D; Tannenbaum, SR; Skipper, P; Santella, RM; Cereijido, GS; Crawford, FG; Perera,
8 FP. (1999). Molecular and genetic damage from environmental tobacco smoke in young
9 children. Cancer Epidemiol Biomarkers Prev 8: 427-431.
10 Tang. PL; Li. TY; Liu. JJ; Zhou. ZJ; Yuan. T; Chen. YH; Rauh. VA; Xie. J; Perera. F. (2008). Effects of prenatal
11 exposure to coal-burning pollutants on children's development in China. Environ Health
12 Perspect 116: 674-679. http://dx.doi.org/10.1289/ehp.10471
13 Tang. Q; Xia. Y; Cheng. S; Tu. B. (2011). Modulation of behavior and glutamate receptor mRNA
14 expression in rats after sub-chronic administration of benzo(a)pyrene. Biomed Environ Sci 24:
15 408-414. http://dx.doi.org/10.3967/0895-3988.2011.04.012
16 Temple. L; Kawabata. TT; Munson. AE; White. KL. (1993). Comparison of ELISA and Plaque-Forming Cell
17 Assays for Measuring the Humoral Immune Response to SRBC in Rats and Mice Treated with
18 Benzo[a]pyrene or Cyclophosphamide. Toxicol Sci 21: 412-419.
19 http://dx.doi.0rg/10.1093/toxsci/21.4.412
20 Thirman. MJ; Albrecht. JH; Krueger. MA; Erickson. RR; Cherwitz. PL; Park. SS; Gelboin. HV; Holtzman. JL
21 (1994). Induction of cytochrome CYPIA1 and formation of toxic metabolites of benzo[a]pyrene
22 by rat aorta: a possible role in atherogenesis. PNAS 91: 5397-5401.
23 Thyssen, J; Althoff, J; Kimmerle, G; Mohr, U. (1981). Inhalation studies with benzo[a]pyrene in Syrian
24 golden hamsters. J Natl Cancer Inst 66: 575-577.
25 Topping, DC; Martin, DH; Nettesheim, P. (1981). Determination of cocarcinogenic activity of
26 benzo[e]pyrene for respiratory tract mucosa. Cancer Lett 11: 315-321.
27 http://dx.doi.org/10.1016/0304-3835(81)90097-5
28 Tornqvist, S; Norell, S; Ahlbom, A; Knave, B. (1986). Cancer in the electric power industry. Br J Ind Med
29 43: 212-213.
30 Triolo, AJ; Aponte, GE; Herr, PL. (1977). Induction of aryl hydrocarbon hydroxylase and forestomach
31 tumors ben benzo(a)pyrene. Cancer Res 37: 3018-3021.
32 U.S. EPA (U.S. Environmental Protection Agency). (1988). Recommendations for and documentation of
33 biological values for use in risk assessment [EPA Report]. (EPA/600/6-87/008). Cincinnati, OH.
34 http://cfpub.epa. gov/ncea/cfm/recordisplay.cfm?deid=34855
35 U.S. EPA (U.S. Environmental Protection Agency). (1990a). Development of a dose-response model for
36 inhaled B(a)P [EPA Report]. (EPA-600/R-92-135).
37 http://www.ntis.gov/search/product.aspx?ABBR=PB93161016
38 U.S. EPA (U.S. Environmental Protection Agency). (1990b). Ingestion dose-response model to
39 benzo(a)pyrene. (EPA/600/R-92/045).
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). (1991a). Dose-response analysis of ingested
2 Benzo(a)pyrene (CAS No. 50-32-8) [EPA Report]. (EPA600R92045).
3 http://www.ntis.gov/search/product.aspx?ABBR=PB93167484
4 U.S. EPA (U.S. Environmental Protection Agency). (1991b). Drinking water criteria document for
5 polycyclic aromatic hydrocarbons (PAHs). (ECAOCIND010).
6 http://www.ntis.gov/search/product.aspx?ABBR=PB92173459
7 U.S. EPA (U.S. Environmental Protection Agency). (1991c). Guidelines for developmental toxicity risk
8 assessment [EPA Report]. (EPA/600/FR-91/001). Washington, DC: U.S. Environmental Protection
9 Agency, Risk Assessment Forum, http://www.epa.gov/raf/publications/guidelines-dev-toxicity-
10 risk-assessment.htm
11 U.S. EPA (U.S. Environmental Protection Agency). (1992). Draft report: A cross-species scaling factor for
12 carcinogen risk assessment based on equivalence of mg/kg(3/4)/day. Fed Reg 57: 24151-24173.
13 U.S. EPA (U.S. Environmental Protection Agency). (1994). Methods for derivation of inhalation reference
14 concentrations and application of inhalation dosimetry [EPA Report]. (EPA/600/8-90/066F).
15 Research Triangle Park, NC. http://cfpub.epa.gov/ncea/cfm/recordisplav.cfm?deid=71993
16 U.S. EPA (U.S. Environmental Protection Agency). (2000). Science policy council handbook: Risk
17 characterization [EPA Report]. (EPA 100-B-00-002). Washington, D.C.
18 http://www.epa.gov/osa/spc/pdfs/rchandbk.pdf
19 U.S. EPA (U.S. Environmental Protection Agency). (2002). A review of the reference dose and reference
20 concentration processes [EPA Report]. (EPA/630/P-02/002F). Washington, DC: Risk Assessment
21 Forum, U.S. Environmental Protection Agency.
22 http://cfpub.epa.gov/ncea/cfm/recordisplay.cfm?deid=51717
23 U.S. EPA (U.S. Environmental Protection Agency). (2004). Risk Assessment Guidance for Superfund
24 (RAGS), Volume I: Human health evaluation manual, (part E: Supplemental guidance for dermal
25 risk assessment): Final [EPA Report]. (EPA/540/R/99/005). Washington, DC.
26 http://www.epa.gov/oswer/riskassessment/ragse/index.htm
27 U.S. EPA (U.S. Environmental Protection Agency). (2005a). Guidelines for carcinogen risk assessment
28 [EPA Report]. (EPA/630/P-03/001F). Washington, DC. http://www.epa.gov/cancerguidelines/
29 U.S. EPA (U.S. Environmental Protection Agency). (2005b). Supplemental guidance for assessing
30 susceptibility from early-life exposure to carcinogens [EPA Report] (pp. 1125-1133). (EPA/630/R-
31 03/003F). Washington. DC. http://www.epa.gov/cancerguidelines/guidelines-carcinogen-
32 supplement.htm
33 U.S. EPA (U.S. Environmental Protection Agency). (2010). Revised draft: Multistage Weibull time-to-
34 tumor model in EPA's benchmark dose software (BMDS): Methodology description. Washington,
35 DC.
36 U.S. EPA (U.S. Environmental Protection Agency). (2011). Recommended use of body weight 3/4 as the
37 default method in derivation of the oral reference dose [EPA Report]. (EPA/100/R11/0001).
38 Washington, DC. http://www.epa.gov/raf/publications/interspecies-extrapolation.htm
39 U.S. EPA (U.S. Environmental Protection Agency). (2012a). Benchmark Dose Software (BMDS). Version
40 2.2. Retrieved from http://www.epa.gov/NCEA/bmds/index.html
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 U.S. EPA (U.S. Environmental Protection Agency). (2012b). Benchmark dose technical guidance.
2 (EPA/100/R-12/001). Washington, DC.
3 http://www.epa.gov/raf/publications/pdfs/benchmark dose guidance.pdf
4 Urano, K; Katakai, Y; Tokuda, Y; Uevama, Y; Nomura, T; Yamamoto, S. (1995). Failure of genotoxic
5 carcinogens to produce tumors in human skin xenografts transplanted to SCID mice.
6 Carcinogenesis 16: 2223-2226. http://dx.doi.Org/10.1093/carcin/16.9.2223
7 Urso, P; Gengozian, N. (1980). Depressed humoral immunity and increased tumor incidence in mice
8 following in utero exposure to benzo[alpha]pyrene. J Toxicol Environ Health 6: 569-576.
9 http://dx.doi.org/10.1080/15287398009529874
10 Urso, P; Gengozian, N. (1982). Alterations in the humoral immune response and tumor frequencies in
11 mice exposed to benzo[a]pyrene and X-rays before or after birth. J Toxicol Environ Health 10:
12 817-835. http://dx.doi.org/10.1080/15287398209530297
13 Urso. P; Gengozian. N. (1984). Subnormal expression of cell-mediated and humoral immune responses
14 in progeny disposed toward a high incidence of tumors after in utero exposure to
15 benzo[a]pyrene. J Toxicol Environ Health 14: 569-584.
16 http://dx.doi.org/10.1080/15287398409530606
17 Urso. P; Johnson. RA. (1987). Early changes in T lymphocytes and subsets of mouse progeny defective as
18 adults in controlling growth of a syngeneic tumor after in utero insult with benzo(a)pyrene.
19 Immunopharmacology 14: 1-10.
20 Urso. P; Johnson. RA. (1988). Quantitative and functional change in T cells of primiparous mice following
21 injection of benzo(a)pyrene at the second trimester of pregnancy. Immunopharmacol
22 Immunotoxicol 10: 195-217. http://dx.doi.org/10.3109/08923978809014333
23 Urso, P; Kramer, MK. (2008). The role of adherent cells in the immunosuppressed state of mouse
24 progeny transplacentally exposed to benzo(alpha)pyrene. In Vitro Cell Dev Biol Anim 44: 273-
25 279. http://dx.doi.org/10.1007/sll626-008-9112-2
26 Urso, P; Ryan, MC; Bennett, JS. (1988). Changes in peripheral blood cells in mice after injection with
27 benzo-a-pyrene during pregnancy. Immunopharmacol Immunotoxicol 10: 179-194.
28 http://dx.doi.org/10.3109/08923978809014332
29 Urso, P; Zhang, W; Cobb, JR. (1992). Immunological consequences from exposure to benzo(a)pyrene
30 during pregnancy. Scand J Immunol Suppl 11: 203-206.
31 van den Berg. FA; Baken. KA; Vermeulen. JP; Gremmer. ER; van Steeg. H; van Loveren. H. (2005). Use of
32 the local lymph node assay in assessment of immune function. Toxicology 211: 107-114.
33 http://dx.doi.0rg/10.1016/i.tox.2005.03.003
34 van Lipzig. MM; Vermeulen. IMP; Gusinu. R; Legler. J; Frank. H; Seidel. A; Meerman. JH. (2005). Formation
35 of estrogenic metabolites of benzo[a]pyrene and chrysene by cytochrome P450 activity and
36 their combined and supra-maximal estrogenic activity. Environ Toxicol Pharmacol 19: 41-55.
37 http://dx.doi.0rg/10.1016/i.etap.2004.03.010
38 Varghese. SK; Gangamma. S. (2009). Particle deposition in human respiratory system: deposition of
39 concentrated hygroscopic aerosols. Inhal Toxicol 21: 619-630.
40 http://dx.doi.org/10.1080/08958370802380792
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 Vesselinovitch. SD; Kyriazis. AP; Mihailovich. N; Rao. KV. (1975). Conditions modifying development of
2 tumors in mice at various sites by benzo(a)pyrene. Cancer Res 35: 2948-2953.
3 Vesselinovitch, SD; Rao, KV; Mihailovich, N. (1979). Neoplastic response of mouse tissues during
4 perinatal age periods and its significance in chemical carcinogenesis [Review]. Natl Cane Inst
5 Monogr239-250.
6 Vineis, P; Anttila, S; Benhamou, S; Spinola, M; Hirvonen, A; Kiyohara, C; Garte, SJ; Puntoni, R; Rannug, A;
7 Strange, RC; Taioli, E. (2007). Evidence of gene gene interactions in lung carcinogenesis in a large
8 pooled analysis [Review]. Carcinogenesis 28: 1902-1905.
9 http://dx.doi.org/10.1093/carcin/bgm039
10 Vineis, P; Perera, F. (2007). Molecular epidemiology and biomarkers in etiologic cancer research: the
11 new in light of the old [Review]. Cancer Epidemiol Biomarkers Prev 16: 1954-1965.
12 http://dx.doi.org/10.1158/1055-9965.EPI-07-0457
13 Vondracek. J; Kozubfk. A; Machala. M. (2002). Modulation of estrogen receptor-dependent reporter
14 construct activation and GO/Gl-S-phase transition by polycyclic aromatic hydrocarbons in
15 human breast carcinoma MCF-7 cells. Toxicol Sci 70: 193-201.
16 Wang. F; He. Y; Guo. H; Li. J; Yang. Y; Wu. Z; Zheng. H; Wu. T. (2010). Genetic variants of nucleotide
17 excision repair genes are associated with DNA damage in coke oven workers. Cancer Epidemiol
18 Biomarkers Prev 19: 211-218. http://dx.doi.org/10.1158/1055-9965.EPI-09-0270
19 Wattenberg. LW. (1974). Inhibition of carcinogenic and toxic effects of polycyclic hydrocarbons by
20 several sulfur-containing compounds. J Natl Cancer Inst 52: 1583-1587.
21 Waylen, AL; Metwally, M; Jones, GL; Wilkinson, AJ; Ledger, WL (2009). Effects of cigarette smoking upon
22 clinical outcomes of assisted reproduction: a meta-analysis [Review]. Hum Reprod Update 15:
23 31-44. http://dx.doi.org/10.1093/humupd/dmn046
24 Wei, Q; Cheng, L; Amos, Cl; Wang, LE; Guo, Z; Hong, WK; Spitz, MR. (2000). Repair of tobacco
25 carcinogen-induced DNA adducts and lung cancer risk: a molecular epidemiologic study. J Natl
26 Cancer Inst 92: 1764-1772.
27 Wei. SJ; Chang. RL; Merkler. KA; Gwynne. M; Cui. XX; Murthy. B; Huang. MT; Xie. JG; Lu. YP; Lou. YR;
28 Jerina, DM; Conney, AH. (1999). Dose-dependent mutation profile in the c-Ha-ras proto-
29 oncogene of skin tumors in mice initiated with benzo[a]pyrene. Carcinogenesis 20: 1689-1696.
30 Weinberg, CR; Wilcox, AJ; Baird, DP. (1989). Reduced fecundability in women with prenatal exposure to
31 cigarette smoking. Am J Epidemiol 129: 1072-1078.
32 Wester. PW; Kroes. R. (1988). Forestomach carcinogens: pathology and relevance to man [Review].
33 Toxicol Pathol 16: 165-171.
34 Wester. RC; Maibach. HI; Bucks. DA; Sedik. L; Melendres. J; Liao. C; Dizio. S. (1990). Percutaneous
35 absorption of [14C]DDT and [14C]benzo[a]pyrene from soil. Fundam Appl Toxicol 15: 510-516.
36 Weyand. EH; Chen. YC; Wu. Y; Koganti. A; Dunsford. HA; Rodriguez. LV. (1995). Differences in the
37 tumorigenic activity of a pure hydrocarbon and a complex mixture following ingestion:
38 Benzo[a]pyrene vs manufactured gas plant residue. Chem Res Toxicol 8: 949-954.
39 http://dx.doi.org/10.1021/tx00049a008
40 Wevand. EH; He. ZM; Ghodrati. F; Wu. Y; Marshall. MV; Lavoie. EJ. (1992). Effect of fluorine substitution
41 on benzo[j]fluoranthene genotoxicity. Chem Biol Interact 84: 37-53.
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 Weyand. EH; Lavoie. EJ. (1988). Comparison of PAH DNA adduct formation and tumor initiating activity
2 in newborn mice. Proc Am Assoc Cancer Res 29: 98.
3 White, KL; Lvsv, HH; Holsapple, MP. (1985). Immunosuppression by polycyclic aromatic hydrocarbons: a
4 structure-activity relationship in B6C3F1 and DBA/2 mice. Immunopharmacology 9: 155-164.
5 WHO (World Health Organization). (2012). Guidance for Immunotoxicity risk assessment for chemicals.
6 http://www.who.int/entity/ipcs/methods/.../guidance immunotoxicity.pdf
7 Whyatt, RM; Santella, RM; Jedrychowski, W; Garte, SJ; Bell, DA; Ottman, R; Gladek-Yarborough, A;
8 Cosma, G; Young, TL; Cooper, TB; Randall, MC; Manchester, DK; Perera, FP. (1998). Relationship
9 between ambient air pollution and DNA damage in Polish mothers and newborns. Environ
10 Health Perspect 106: 821-826.
11 Wilson, JS; Holland, LM. (1988). Periodic response difference in mouse epidermis chronically exposed to
12 crude-oils or BaP: Males vs. females. Toxicology 50: 83-94. http://dx.doi.org/10.1016/0300-
13 483X(88)90123-0
14 Winker. N; Tuschl. H; Kovac. R; Weber. E. (1997). Immunological investigations in a group of workers
15 exposed to various levels of polycyclic aromatic hydrocarbons. J Appl Toxicol 17: 23-29.
16 Wislocki. PG: Bagan. ES: Lu. AYH: Doolev. KL: Fu. PP: Han-Hsu. H: Beland. FA: Kadlubar. FF. (1986).
17 Tumorigenicity of nitrated derivatives of pyrene, benz[a]anthracene, chrysene, and
18 benzo[a]pyrene in the newborn mouse assay. Carcinogenesis 7: 1317-1322.
19 http://dx.doi.0rg/10.1093/carcin/7.8.1317
20 Wolff. RK: Griffith. WC: Henderson. RF: Hahn. FF: Harkema. J. R.: Rebar. AH: Eidson. AF: Mcclellan. RO.
21 (1989). Effects of repeated inhalation exposures to 1-nitropyrene, benzo(a)pyrene, Ga2O3
22 particles, and SO2 alone and in combinations on particle clearance, bronchoalveolar lavage fluid
23 composition, and histopathology. J Toxicol Environ Health 27: 123-138.
24 http://dx.doi.org/10.1080/15287398909531283
25 Wood, AW; Levin, W; Chang, RL; Huang, MT; Ryan, DE; Thomas, PE; Lehr, RE; Kumar, S; Koreeda, M;
26 Akagi, H; Ittah, Y; Dansette, P; Yagi, H; Jerina, DM; Conney, AH. (1980). Mutagenicity and tumor-
27 initiating activity of cyclopenta(c,d)pyrene and structurally related compounds. Cancer Res 40:
28 642-649.
29 Wormley, DP; Chirwa, S; Nayyar, T; Wu, J; Johnson, S; Brown, LA; Harris, E; Hood, DB. (2004). Inhaled
30 benzo(a)pyrene impairs long-term potentiation in the Fl generation rat dentate gyrus. Cell Mol
31 Biol (Noisy-le-grand) 50: 715-721.
32 Wu. J; Hou. H; Ritz. B; Chen. Y. (2010). Exposure to polycyclic aromatic hydrocarbons and missed
33 abortion in early pregnancy in a Chinese population. Sci Total Environ 408: 2312-2318.
34 http://dx.doi.0rg/10.1016/i.scitotenv.2010.02.028
35 Wu. J; Ramesh. A; Navyar. T; Hood. DB. (2003a). Assessment of metabolites and AhR and CYP1A1 mRNA
36 expression subsequent to prenatal exposure to inhaled benzo(a)pyrene. Int J Dev Neurosci 21:
37 333-346.
38 Wu. MT: Pan. CH: Huang. YL: Tsai. PJ: Chen. CJ: Wu. TN. (2003b). Urinary excretion of 8-hydroxy-2-
39 deoxyguanosine and 1-hydroxypyrene in coke-oven workers. Environ Mol Mutagen 42: 98-105.
40 http://dx.doi.org/10.1002/em.10176
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 Wynder. EL; Fritz. L; Furth. N. (1957). Effect of concentration of benzopyrene in skin carcinogenesis. J
2 Natl Cancer Inst 19: 361-370.
3 Wynder, EL; Hoffmann, D. (1959). A study of tobacco carcinogenesis. VII. The role of higher polycyclic
4 hydrocarbons. Cancer 12: 1079-1086. http://dx.doi.org/10.1002/1097-
5 0142(195911/12)12:6<1079::AID-CNCR2820120604>3.0.CO;2-I
6 Xia. Y; Cheng. S; He. J; Liu. X; Tang. Y; Yuan. H; He. L; Lu. T; Tu. B; Wang. Y. (2011). Effects of subchronic
7 exposure to benzo[a]pyrene (B[a]P) on learning and memory, and neurotransmitters in male
8 Sprague-Dawley rat. Neurotoxicology 32: 188-198.
9 http://dx.doi.0rg/10.1016/i.neuro.2010.12.015
10 Xia, Y; Zhu, P; Han, Y; Lu, C; Wang, S; Gu, A; Fu, G; Zhao, R; Song, L; Wang, X. (2009). Urinary metabolites
11 of polycyclic aromatic hydrocarbons in relation to idiopathic male infertility. Hum Reprod 24:
12 1067-1074. http://dx.doi.org/10.1093/humrep/depQ06
13 Xu. C: Chen. J: Qiu. Z: Zhao. Q: Luo. J: Yang. L: Zeng. H: Huang. Y: Zhang. L: Cao. J: Shu. W. (2010).
14 Ovotoxicity and PPAR-mediated aromatase downregulation in female Sprague-Dawley rats
15 following combined oral exposure to benzo[a]pyrene and di-(2-ethylhexyl) phthalate. Toxicol
16 Lett 199: 323-332. http://dx.doi.Org/10.1016/i.toxlet.2010.09.015
17 Xu. D; Penning. TM; Blair. IA; Harvey. RG. (2009). Synthesis of phenol and quinone metabolites of
18 benzo[a]pyrene, a carcinogenic component of tobacco smoke implicated in lung cancer. J Org
19 Chem 74: 597-604. http://dx.doi.org/10.1021/io801864m
20 Xu. GB; Yu. CP. (1985). Theoretical lung deposition of hygroscopic NaCI aerosols. Aerosol Sci Technol 4:
21 455-461. http://dx.doi.org/10.1080/02786828508959070
22 Xu, Z; Brown, LM; Pan, GW; Liu, TF; Gao, GS; Stone, BJ; Cao, RM; Guan, DX; Sheng, JH; Yan, ZS; Dosemeci,
23 M; Fraumeni, JF; Blot, WJ. (1996). Cancer risks among iron and steel workers in Anshan, China,
24 Part II: Case-control studies of lung and stomach cancer. Am J Ind Med 30: 7-15.
25 http://dx.doi.org/10.1002/(SICI)1097-0274(199607)30:l<7::AID-AJIM2>3.0.CO;2-#
26 Xue, W; Warshawsky, D. (2005). Metabolic activation of polycyclic and heterocyclic aromatic
27 hydrocarbons and DNA damage: a review [Review]. Toxicol Appl Pharmacol 206: 73-93.
28 http://dx.doi.0rg/10.1016/i.taap.2004.ll.006
29 Ye, X; Skjaerven, R; Basso, O; Baird, DP; Eggesbo, M; Cupul Uicab, LA; Haug, K; Longnecker, MP. (2010).
30 In utero exposure to tobacco smoke and subsequent reduced fertility in females. Hum Reprod
31 25: 2901-2906. http://dx.doi.org/10.1093/humrep/deq235
32 Yu. D; Berlin. JA; Penning. TM; Field. J. (2002). Reactive oxygen species generated by PAH o-quinones
33 cause change-in-function mutations in p53. Chem Res Toxicol 15: 832-842.
34 Zandi-Nejad. K; Luyckx. VA; Brenner. BM. (2006). Adult hypertension and kidney disease - The role of
35 fetal programming [Review]. Hypertension 47: 502-508.
36 http://dx.doi.org/10.1161/01.HYP.0000198544.09909.la
37 Zenzes. MT; Puy. LA; Bielecki. R. (1998). Immunodetection of benzo[a]pyrene adducts in ovarian cells of
38 women exposed to cigarette smoke. Mol Hum Reprod 4: 159-165.
39 http://dx.doi.0rg/10.1093/molehr/4.2.159
This document is a draft for review purposes only and does not constitute Agency policy.
R-33 DRAFT—DO NOT CITE OR QUOTE
-------
Toxicological Review ofBenzo[a]pyrene
1 Zenzes. MT; Puy. LA; Bielecki. R; Reed. TE. (1999). Detection of benzo[a]pyrene diol epoxide-DNA
2 adducts in embryos from smoking couples: evidence for transmission by spermatozoa. Mol Hum
3 Reprod 5: 125-131.
4 Zhang, HM; Nie, JS; Li, X; Niu, Q. (2012). Characteristic analysis of peripheral blood mononuclear cell
5 apoptosis in coke oven workers. J Occup Health 54: 44-50. http://dx.doi.org/10.1539/ioh.ll-
6 0155-OA
7 Zhang, Y; Ramos, KS. (1997). The induction of proliferative vascular smooth muscle cell phenotypes by
8 benzo[a]pyrene does not involve mutational activation of ras genes. Mutat Res 373: 285-292.
9 http://dx.doi.org/10.1016/50027-5107(96)00213-8
10 Zheng, SJ; Tian, HJ; Cao, J; Gao, YQ. (2010). Exposure to di(n-butyl)phthalate and benzo(a)pyrene alters
11 IL-1(3 secretion and subset expression of testicular macrophages, resulting in decreased
12 testosterone production in rats. Toxicol Appl Pharmacol 248: 28-37.
13 http://dx.doi.0rg/10.1016/i.taap.2010.07.008
14
15
This document is a draft for review purposes only and does not constitute Agency policy.
R-34 DRAFT—DO NOT CITE OR QUOTE
------- |