DRAFT - DO NOT CITE OR QUOTE                              EPA/635/R-os/oi2A
                                                    www.epa.gov/iris
f/EPA
      DEVELOPMENT OF A RELATIVE
  POTENCY FACTOR (RPF) APPROACH
       FOR POLYCYCLIC AROMATIC
     HYDROCARBON (PAH) MIXTURES
            In Support of Summary Information on the
            Integrated Risk Information System (IRIS)
                         February 2010
                             NOTICE


This document is an External Review draft. This information is distributed solely for the
purpose of pre-dissemination peer review under applicable information quality guidelines. It has
not been formally disseminated by EPA. It does not represent and should not be construed to
represent any Agency determination or policy. It is being circulated for review of its technical
accuracy and science policy implications.
                   U.S. Environmental Protection Agency
                          Washington, DC

-------
                                    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 or recommendation for use.
                                                   DRAFT - DO NOT CITE OR QUOTE

-------
                                  EXECUTIVE SUMMARY


       The U.S. Environmental Protection Agency's (U.S. EPA's) Integrated Risk Information
System (IRIS) Program is releasing for scientific review a relative potency factor (RPF)
approach for polycyclic aromatic hydrocarbon (PAH) mixtures as one approach for assessing
cancer risk from exposure to PAH mixtures.  The RPF analysis under review is not a
reassessment of individual PAH carcinogenicity, but rather provides a cancer risk estimate for
PAH mixtures by summing doses of component PAHs after scaling the doses (with RPFs)
relative to the potency of an index PAH (i.e., benzo[a]pyrene). The cancer risk is then estimated
using the dose-response curve for the index PAH. RPFs for seven individual PAHs were
developed in the U.S. EPA (1993) Provisional Guidance for Quantitative Risk Assessment of
PAHs (Provisional Guidance} and are utilized extensively within U.S. EPA program offices and
other regulatory agencies. The RPF analysis provided in the current report includes more recent
data and an analysis of both  tumorigenicity and genotoxicity data for PAHs.
       The Supplemental Guidance for Conducting Health Risk Assessment of Chemical
Mixtures (U.S. EPA, 2000) indicates that approaches based on whole mixtures are preferred to
component approaches, such as the RPF approach. Risk assessment approaches based on
toxicity evaluations of whole mixtures inherently address specific interactions among PAHs and
account for the toxicity of unidentified components of PAH mixtures. They also do not require
assumptions regarding the toxicity of individual components (e.g., dose additivity or response
additivity). While whole mixture assessment is preferred, there are challenges associated with
using these approaches.  There are very few toxicity data available for whole PAH mixtures and,
in most cases, chemical analyses of the composition of mixtures are limited.  In addition, PAH-
containing mixtures tend to be very complex; the composition of these mixtures appears to vary
across  sources releasing these mixtures to the environment and in various environmental media
in which they occur. For these reasons, a whole mixtures approach may not always be
practicable for risk assessment purposes. This report provides recommendations for
development of the RPF approach for PAH mixtures health risk assessment and includes:

    (1) A rationale for recommending an RPF approach (Chapter 2);

    (2) A summary of previous approaches for developing the RPF approach for PAHs
       (Chapter 3);

    (3) An evaluation of the  carcinogenicity of individual PAHs (Chapter 4);

    (4) Methods for dose-response assessment and individual study RPF calculation (Chapter 5);

    (5) Selection of PAHs for inclusion in the RPF approach (Chapter 6);


                                       iii           DRAFT - DO NOT CITE OR QUOTE

-------
   (6) Derivation of RPFs for selected PAHs (Chapter 7); and

   (7) Characterization of strengths, weaknesses, and uncertainties associated with the RPF
       approach to PAH cancer risk assessment (Chapter 8).

       The RPF approach involves two key assumptions related to the application of a dose-
additivity model:  (1) a imilar toxicological action of PAH components in the mixture; and
(2) interactions among PAH mixture components do not occur at low levels of exposure typically
encountered in the environment. Mechanistic studies indicate that the mutagenic and tumor-
initiating activity of carcinogenic PAHs requires metabolic activation to reactive intermediates
(e.g., dihydrodiol epoxides, quinones, radical cations), which covalently modify
deoxyribonucleic acid (DNA) targets resulting in mutation, and that tumor promotion and
progression phases may involve parent compound binding to the Ah receptor (AhR) and
subsequent alterations of gene expression or a cell proliferation response to metabolite
cytotoxicity (see Section 2.4, Similarities in Mode of Carcinogenic Action for PAHs, and
Figure 2-3, Overview of the proposed key events in the mode of action for PAH
carcinogenicity). As such, there is evidence that an assumption of a similar toxicological action
is reasonable; however, the carcinogenic process for individual PAHs is likely related to some
unique combination of multiple molecular events resulting from the formation of several reactive
species. The second assumption of no interactions at low levels of exposure is also reasonable,
but cannot be conclusively demonstrated in experimental systems (see  Section 2.8, Dose
Additivity of PAHs in Combined Exposures). Use of the RPF approach assumes that doses of
component chemicals that act in a similar manner can be added together, after scaling the
potencies relative to the index chemical.  The assumptions of toxicological  similarity and no
interaction effects at low environmental exposure levels that are inherent in the dose-additivity
model  are generally supported by the experimental data for PAHs (see  Sections 2.4 and 2.7).
       Several approaches have been used previously for the determination of RPFs for PAHs
(see Chapter 3). In the published literature, RPF values were proposed in at least one analysis
for a total of 27 PAHs (see Table 3-1). Because these approaches generally relied on similar
bioassay data and modeling methods, the resulting RPF values are considered comparable for
most PAHs across analyses.
       There is a large PAH database on carcinogenicity in animal bioassays, genotoxicity in
various test systems, and bioactivation to tumorigenic and/or genotoxic metabolic intermediates.
The RPF analysis presented here includes only unsubstituted PAHs with three or more fused
aromatic rings containing only carbon and hydrogen atoms, because these are the most widely
studied members of the PAH chemical class. The study types that were considered most useful
for RPF derivation were rodent carcinogenicity bioassays (all routes) in which one or more PAH
was tested at the same time as benzo[a]pyrene.  In addition, in vivo and in vitro data for cancer-
related endpoints in which one or more PAH and benzo[a]pyrene was tested simultaneously were

                                       iv           DRAFT - DO NOT CITE OR QUOTE

-------
obtained, including studies on the formation of DNA adducts, mutagenicity, chromosomal
aberrations, sister chromatid exchange frequency, aneuploidy, DNA damage/repair/
recombination, unscheduled DNA synthesis, and cell transformation. Although it would be
possible to calculate RPFs from studies where a PAH and benzo[a]pyrene were tested by the
same laboratory using the same test system but at different times, this approach was not
considered because it could introduce differences in the dose-response information that are
unrelated to the chemical (e.g., variability associated with laboratory environment conditions,
animal handling, food supply, etc.). Thus, studies in which benzo[a]pyrene was not tested
simultaneously with another PAH were not considered in the RPF calculations.
       Studies of AhR binding/activation were not considered for use in deriving RPFs because
there does not appear to be a clear relationship between affinity for the AhR and carcinogenic
potency.  For example, highly mutagenic fjord-region PAHs are potent carcinogens despite
exhibiting lower AhR affinity (reviewed by Bostrom et al., 2002). Likewise, some PAHs that
strongly activate the AhR, such as benzo[k]fluoranthene (Machala et al., 2001), are only weakly
carcinogenic. In addition, some studies have demonstrated the formation of DNA adducts in the
liver of AhR knock-out mice following intraperitoneal  or oral exposure to benzo[a]pyrene
(Sagredo et al., 2006; Uno et al., 2006; Kondraganti et  al., 2003), indicating that Ah
responsiveness is not strictly required for metabolic activation and genotoxicity.  These findings
suggest that there may be alternative (i.e., non-AhR-mediated) mechanisms of benzo[a]pyrene
activation in the mouse liver, and that AhR affinity would not be a good predictor of
carcinogenic potency. Also, several studies indicate that AhR-mediated CYP1 Al induction
potency does not correlate well with carcinogenic potency.  These studies compared CYP1 Al
induction potency for several PAHs using assays to measure ethoxyresorufm O-deethylase
(EROD) activity, CYP1 Al protein,  and messenger ribonucleic acid (mRNA) levels, or chemical-
activated luciferase reporter gene expression (Bosveld  et al., 2002; Machala et al., 2001; Bols et
al., 1999; Till et al.,  1999; Willett et al., 1997).
       Several study types were excluded from the database because they did not provide
carcinogenicity or cancer-related endpoint information for individual PAHs.  These include
biomarker studies measuring DNA adducts in humans, studies of PAH metabolism, and studies
of PAH mixtures. Although these studies contain important information on human exposure to
PAH mixtures and the mode of action for PAH toxicity, they generally do not contain dose-
response information that would be useful for calculation of RPF estimates.
       A database of primary literature relevant to the  RPF approach for PAHs was developed by
performing a comprehensive review of the scientific literature dating from the 1950s through
2009 on the carcinogenicity and genotoxicity of PAHs. The search identified over 900 individual
publications for a target list of 74 PAHs (see Table  2-1) that have been identified in
environmental media or for which toxicological data are available.  Review of these publications
                                        v           DRAFT - DO NOT CITE OR QUOTE

-------
resulted in the identification of more than 600 papers that included carcinogenicity or cancer-
related endpoint data on at least one PAH and benzo[a]pyrene tested at the same time.
       References in the PAH database were sorted into the following major categories: cancer
bioassays, in vivo studies of cancer-related endpoints, and in vitro studies of cancer-related
endpoints.  These categories were further sorted by route (for bioassays) or by endpoint (for
cancer-related endpoints).  Each study was reviewed, and critical study details were extracted
into tables for each individual endpoint (see Chapter 4). The tables also include an initial
determination of whether the data from each study meet selection criteria for use in the RPF
analysis. Studies with data on selected PAHs  and benzo[a]pyrene were considered for RPF
determination, even if a particular PAH has not been classified by U.S. EPA or International
Agency for Research on Cancer (IARC)  as a carcinogen. Studies were included in the analysis if
the following selection criteria were met:

       •  Benzo[a]pyrene was tested simultaneously with another PAH;

       •  A statistically increased incidence of tumors was observed with benzo[a]pyrene
          administration, compared with control incidence;

       •  Benzo[a]pyrene produced a statistically significant change in a  cancer-related
          endpoint finding;

       •  Quantitative results were presented;

       •  The carcinogenic response observed in either the benzo[a]pyrene- or other PAH-
          treated animals at the lowest dose level was not saturated (i.e., tumor incidence at the
          lowest dose was <90%), with the exception of tumor multiplicity findings; and

       •  There were no study quality concerns or potential confounding  factors that precluded
          use (e.g., no concurrent control, different vehicles, strains, etc. were used for the
          tested PAH and benzo[a]pyrene; use of cocarcinogenic vehicle; PAHs of questionable
          purity; unexplained mortality in treated or control animals).

       If the above criteria were met, studies were selected for use in the analysis  regardless of
whether positive or nonpositive results were reported.  Studies with positive findings were used
for calculation of RPFs. Studies with nonpositive findings were used in a weight of evidence
evaluation to select PAHs for inclusion in the RPF approach (see Section 6.1).
       Dose-response data were extracted from studies with positive findings that met selection
criteria. For studies that reported results graphically, individual data points were extracted using
digitizing software. In all, over 300 data sets were extracted, reflecting dose-response data from
at least one study for 51 of the 74 PAHs included in the analysis. All of the extracted data are
presented in Appendix C of this report.
                                        vi           DRAFT - DO NOT CITE OR QUOTE

-------
       While tumor multiplicity data from tumor bioassays are not generally used to estimate
cancer potency., these data were included in the dose-response assessment in order to determine
whether they could serve as a reliable measure of relative cancer potency.  Several bioassays
reported data on both tumor incidence and tumor number, providing information that was later
used to compare relative potencies estimated from these two endpoints. Statistical analyses were
performed on tumor bioassay data to determine whether the tumor incidence or multiplicity
observed at a particular dose represented a statistically significant increase over  controls. If
statistical analyses were not described in the original report, incidence data were analyzed using
Fisher's exact test and the Cochran-Armitage trend test. Positive findings were indicated by a
significant (p < 0.05) difference for at least one dose group by comparison to control (in Fisher's
exact or an equivalent test) or a significant dose-response trend (Cochran-Armitage or
equivalent) for multidose studies.  For tumor bioassay data reported as tumor count, a t-test was
conducted (when variance data were available) to determine whether the count was significantly
different from control (p < 0.05). The results of the statistical analyses are shown with the  dose-
response data in Appendix C.  Statistical analyses of the cancer-related endpoint data were  not
conducted; the study author's conclusions as to response (positive or nonpositive) was used.
       Chapter 5  describes the methods used for both the dose-response assessment and the RPF
calculation in detail.  The general equation for estimating an RPF was the ratio of the slope of the
dose-response curve for the subject PAH to the slope of the dose-response curve for
benzo[a]pyrene. For bioassay data, tumor incidences were modeled using the multistage model
within the U.S. EPA Benchmark Dose (BMD) Software (Version  1.3.2). For cancer-related
endpoint data in quantal form, this model was also used; for continuous data (either tumor
multiplicity or cancer-related endpoint data), the simplest continuous model (linear) within the
software was applied. Whenever the data allowed, benchmark response (BMR) values of 10%
for quantal data and 1 standard deviation (SD) from the control value for continuous data were
used to calculate the slope by linear extrapolation to the origin for consistency across data sets.
Alternative BMR values were used in select instances, as described in  Section 5.3. For data sets
that included only a single dose, or those for which no model fit was achieved with the selected
models, a point estimate RPF1 was calculated. As Table G-2 indicates, final RPFs for five
compounds (benz[a]anthracene, benz[b,c]aceanthrylene, benz[j]aceanthrylene,
dibenzo[a,h]pyrene, and naphtho[2,3-e]pyrene) are based exclusively on point estimates; the
remaining 19 PAHs had at least one dataset that could be modeled (see Appendix G).
       The RPFs calculated  from individual studies for each PAH were used in  a weight of
evidence evaluation to select PAHs for inclusion in the RPF approach (see Chapter 6) and in the
derivation of a final RPF for each compound (Chapter 7).  The selection of PAHs to be included
JFor the purpose of this report, the term "point estimate RPF" is used to describe an RPF calculated from a single
point on the dose-response curve for both the PAH of interest and benzo[a]pyrene. This term distinguishes the RPF
from one calculated using a BMD modeling result from multidose data.

                                        vii           DRAFT - DO NOT CITE OR QUOTE

-------
in the RPF approach began with an evaluation of whether the available data were adequate to
assess the carcinogenicity of each compound.  At least one RPF value was calculated for each of
51 PAHs. For 16 of these compounds, only a single RPF value derived from an in vitro cancer-
related endpoint (primarily mutagenicity assays) was available (see Table 6-1).  Due to the
limited data available for these 16 compounds, no further evaluation of these PAHs was
conducted, and they were not selected for inclusion in the RPF approach.
       For the remaining 35 PAHs, a weight of evidence evaluation (see Figure 6-1) was
conducted to assess the evidence that each PAH could induce a carcinogenic response. For the
purposes of this analysis, PAHs were assumed to be carcinogenic due to toxicological similarity
to the indicator compound, benzo[a]pyrene. The weight of evidence approach was developed to
determine whether the available information for each PAH was adequate for inclusion in the RPF
approach. If the data were not considered adequate, then the PAH was excluded. In vivo tumor
bioassays that included benzo[a]pyrene were given the greatest weight in assessing the
carcinogenicity of a given PAH; data from other bioassays and cancer-related endpoint studies
were used to supplement the weight of evidence when the bioassay data that included
benzo[a]pyrene were conflicting or nonpositive.  Structural alerts for PAH carcinogenicity or
mutagenicity (as defined in Section 2.5 as the presence of a classic bay or fjord region in a PAH
containing at least four benzene rings) were noted in the evaluation for each PAH, but were not
used explicitly in the weight of evidence evaluation.
       The weight of evidence evaluation (Chapter 6) indicated that the available data were
adequate to determine that 24 of the 35 PAHs were carcinogenic, that 3 PAHs (anthracene,
phenanthrene, and pyrene) were not carcinogenic, and that data were inadequate to evaluate the
carcinogenicity for  8 PAHs. The eight PAHs with inadequate data were excluded from the RPF
approach. For the three PAHs for which there were sufficient data to conclude that they were not
carcinogenic (i.e., robust nonpositive tumor bioassay data and cancer-related endpoint data), a
final RPF of zero was recommended. While there  is little quantitative difference between
selecting a final RPF of zero for a given PAH and excluding that PAH from the RPF approach,
this is an important distinction for uncertainty analysis. There is substantial uncertainty in the
risk associated with PAHs that are excluded from the RPF approach due to inadequate data;
these compounds could be of low or high potency.  However, for PAHs with an RPF of zero,
there is evidence to suggest that these compounds are not carcinogenic, and the uncertainty
associated with the  cancer risk for these compounds is markedly reduced.
       For each of the remaining 24 compounds, a final nonzero RPF was derived. A number of
options were considered for deriving an RPF from  among the numerous values calculated for
each individual PAH.  These options included: prioritizing bioassay RPFs from different
exposure routes based  on environmentally relevant routes; prioritizing bioassay RPFs based on
target organs considered relevant to human susceptibility to PAH carcinogenesis; prioritizing
RPFs based on quality of the underlying study; prioritizing cancer-related endpoints by their

                                       viii          DRAFT - DO NOT CITE OR QUOTE

-------
correlation with bioassay potency (i.e., ability to predict bioassay potency); and aggregating
RPFs across all bioassays, across all cancer-related endpoints, or across all endpoints. In the
end, it was concluded that the available data did not provide a clear scientific basis for
prioritizing RPFs except for a preference for bioassay data over cancer-related endpoints. As a
consequence, final RPFs were derived from bioassay data for any PAH that had at least one RPF
based on a bioassay.
       For each carcinogenic PAH with bioassay data, the average RPF was calculated from
bioassays with positive results. For those PAHs that did not have an estimated RPF based on a
bioassay, but for which the weight of evidence evaluation indicated a carcinogenic response
(e.g., dibenz[a,c]anthracene), the final RPF was calculated from all cancer-related endpoint
studies with  positive results.  In both cases, nonpositive results were not included in the
calculation.  The final RPF for each PAH was reported to  one significant figure.  The range of
RPF values was also reported. Presenting the RPFs in this manner provides an average and
maximum estimate for each PAH that has  data from multiple studies.
       Several options were considered for the determination of final RPFs (e.g., arithmetic
mean, geometric mean, weighted average, maximum, or order of magnitude estimates).  The
arithmetic mean and range were chosen as a simple approach to describing the calculated RPF
values available for each PAH. Other estimates were not considered appropriate due to the
limited number of RPF values calculated for most PAHs and the variability in the RPF estimates.
Most PAHs (18/24, 73%) had <3 calculated RPF values and the range of RPF values was greater
than an order of magnitude for several compounds (7/24 PAHs). The variability in RPF
estimates is likely due to differences in study design parameters (e.g., route, species/strain,
exposure duration, exposure during sensitive time periods, initiation versus promotion and
complete carcinogenesis protocols, tumor incidence versus multiplicity reporting) and dose-
response methods (modeled versus point estimates). Calculation of a weighted average was not
possible because there is no clear scientific rationale for choosing among study types or tumor
data outcomes. Providing order of magnitude estimates, as has been previously done for
estimating RPFs for PAHs, was not considered to be superior to calculating simple means.
Including the range in the estimated RPFs  was considered to be informative to the user for
characterizing uncertainty.
       Once a final RPF was derived for a given PAH, the resulting value was assigned a
relative confidence rating of high, medium, or low confidence. The relative confidence rating
characterized the nature of the database upon which the final RPF was based.  Confidence
rankings were based on the robustness of the database. For final RPFs based on tumor bioassay
data, confidence ratings considered both the available tumor bioassays and the availability of
supporting data for cancer-related endpoints.  The most important factors that were considered
included the availability of in vivo data and whether multiple exposure routes were represented.
Other database characteristics that  were considered included the availability of more than one in

                                        ix           DRAFT - DO NOT CITE OR QUOTE

-------
vivo study, and whether effects were evident in more than one sex or species. Very low relative
confidence was reserved for final RPFs based on cancer-related endpoint data only (e.g.,
dibenz[a,c]anthracene). An RPF of zero was only applied if the data implied high or medium
relative confidence.
       Table 1 shows the average RPFs based on tumor bioassay data with their associated range
and relative confidence ratings, and an overview of the tumor bioassay database (total number of
studies, exposure routes tested, species tested, and sexes tested) for each PAH. Table 2 shows
the average RPF for dibenz[a,c]anthracene, the only RPF based on cancer-related endpoint data,
with its associated range, relative confidence rating, and an overview of the database for this
compound.
                                                    DRAFT - DO NOT CITE OR QUOTE

-------
       Table 1. PAHs with final RTFs based on tumor bioassay data
PAH
Anthanthrene
Anthracene
B enz [a] anthracene
Benz[b,c]aceanthrylene, 11H-
Benzo [bjfluoranthene
Benzo [c]fluorene
Benz [e] aceanthry lene
Benzo [g,h,i]pery lene
Benz [j ] aceanthry lene
Benzo [j]fluoranthene
Benzo [kjfluoranthene
Benz [1] aceanthry lene
Chrysene
Cyclopenta[c,d]pyrene
Cyclopenta[d,e,f]chrysene, 4H-
Dibenzo [a,e]fluoranthene
Dibenzo [a,e]py rene
Dibenz [a,h] anthracene
Dibenzo [a,h]pyrene
Dibenzo [a,i]py rene
Dibenzo [a,l]py rene
Fluoranthene
Indeno [ 1,2,3 -c,d]pyrene
Naphtho[2,3-e]pyrene
Phenanthrene
Pyrene
Average
RPF
0.4
0
0.2
0.05
0.8
20
0.8
0.009
60
0.3
0.03
5
0.1
0.4
0.3
0.9
0.4
10
0.9
0.6
30
0.08
0.07
0.3
0
0
Range of
RPFs
0.2-0.5
0
0.02-0.4
0.05
0.1-2
1-50
0.6-0.9
0.009
60
0.01-1
0.03-0.03
4-7
0.04-0.2
0.07-1
0.2-0.5
0.7-1
0.3-0.4
l^K)
0.9
0.5-0.7
10-40
0.009-0.2
0.07
0.3
0
0
Relative
confidence
Medium
Medium3
Medium
Low
High
Medium
Low
Low
Low
High
Medium
Low
High
Medium
Low
Low
Low
High
Low
Low
Medium
Low
Low
Low
High
Medium
Number of
datasets
2
1 (nonpositive)
3
1
5
2
2
1
1
5
2
2
7
5
2
2
2
3
1
2
3
5
1
1
3 (nonpositive)
7 (nonpositive)
Exposure routes tested
Dermal, lung implantation
Dermal
Dermal, intraperitoneal
Dermal
Dermal, intraperitoneal, lung implantation
Oral, intraperitoneal
Dermal
Lung implantation
Intraperitoneal
Dermal, intraperitoneal, lung implantation
Dermal, lung implantation
Dermal
Dermal, intraperitoneal, lung implantation
Dermal, intraperitoneal
Dermal
Dermal
Dermal
Dermal, intraperitoneal, lung implantation
Dermal
Dermal
Dermal, intraperitoneal
Intraperitoneal
Lung implantation
Dermal
Dermal, intraperitoneal, lung implantation
Dermal, intraperitoneal
Species
tested
Mouse, rat
Mouse
Mouse
Mouse
Mouse, rat
Mouse
Mouse
Rat
Mouse
Mouse, rat
Mouse, rat
Mouse
Mouse, rat
Mouse
Mouse
Mouse
Mouse
Mouse, rat
Mouse
Mouse
Mouse
Mouse
Rat
Mouse
Mouse, rat
Mouse
Sexes tested
Female
Female
Female, male
Female
Female, male
Female
Female, male
Female
Male
Female, male
Female
Female, male
Female, male
Female, male
Female
Female
Female
Female, male
Female
Female
Female, male
Female, male
Female
Female
Female, male
Female, male
""Reflects availability of data from anthracene exposure via another exposure route in a study that did not include benzo[a]pyrene.
                                                           XI
DRAFT - DO NOT CITE OR QUOTE

-------
       Table 2. PAHs with final RTFs based on cancer-related endpoint data
       (no tumor bioassay data available)
PAH
Dibenz[a,c]anthracene









Average
RPF
4









Range of
RPFs
0.04-50









Relative
confidence
Very low









Types of studies
Total =14 studies
One in vivo DNA adduct
Six in vitro bacterial
mutagenicity
One in vitro mammalian
mutagenicity
One in vitro morphological/
malignant transformation
Three in vitro DNA damage
Two in vitro DNA adducts
Multiple dose studies
Total = 6 studies
Four in vitro bacterial
mutagenicity
One in vitro DNA
damage
One in vitro DNA
adduct



       The cancer risk for a PAH mixture of concern is determined by multiplying the
benzo[a]pyrene equivalent dose or concentration by the benzo[a]pyrene cancer toxicity value
(e.g., oral slope factor). Benzo[a]pyrene equivalents are calculated by multiplying the
concentration (or dose) of a particular PAH component in the mixture by its RPF. The proposed
RPF approach considers each of the bioassay types used for RPF derivation to be equivalent for
the purpose of determining relative potency to benzo[a]pyrene.
       According to the Supplemental Guidance for Assessing Susceptibility from Early-Life
Exposure to Carcinogens (U.S. EPA, 2005b), benzo[a]pyrene is carcinogenic by a mutagenic
mode of action. A common mutagenic mode of action for other carcinogenic PAHs is
hypothesized based on information available for the indicator chemical, benzo[a]pyrene (U.S.
EPA, 2005b). When assessing PAH cancer risks for lifestages under 16 years of age, or for
lifetime exposures that include early-life exposures, the RPF values should be applied with
specific exposure information  to the benzo[a]pyrene cancer risk estimates including adjustment
for early-life susceptibility, through the application of age-dependent adjustment factors
(ADAFs).
       A description of uncertainties and limitations is crucial to interpretation of the RPF
approach for PAH mixtures risk assessment (see Chapter 8).  Many of the general uncertainties
related to chemical-specific risk assessment are also applicable to the proposed RPF approach for
PAHs (e.g., appropriateness of animal models, low-dose and interspecies extrapolation,
variability within the human population). Use of a component-based approach for mixtures risk
assessment leads to additional  uncertainties related to adequate characterization of the mixture
and the potential interactions that may occur between individual components within the mixture
(i.e., PAHs and other chemicals).  The RPF approach is limited by the small number of PAHs for
which there are analytical chemistry and toxicology data, and thus may result in underestimation
of actual cancer risks from complex PAH mixtures. There are uncertainties and limitations
related to the size  and nature of the PAH database, the human relevance of animal data,
                                       xn
DRAFT - DO NOT CITE OR QUOTE

-------
assumptions regarding mode of action and dose additivity, and cross-route extrapolation.
Specific uncertainties that are related to dose-response assessment (i.e., calculation of RPFs) and
the selection of single RPF values for each PAH are also discussed in Chapter 8.
       In summary, the current analysis represents a significant improvement upon the previous
component-based approaches for PAH mixtures risk assessment.  One of the most important
improvements is the consideration of data from a comprehensive review of the scientific
literature dating from the 1950s through 2008 on the carcinogenicity and genotoxicity of PAHs.
The search identified over 900 individual publications for a target list of 74 PAHs that have been
identified in environmental media and for which toxicological data are available. Review of
these publications resulted in the identification of more than 600 papers that included
carcinogenicity or cancer-related endpoint data on at least one PAH and benzo[a]pyrene tested at
the same time. Dose-response data were extracted, and RPFs from individual studies were
calculated from over 300 data sets representing 51 individual PAHs. For 35 compounds, a
weight of evidence evaluation was conducted to select PAHs for inclusion in the RPF approach;
data were inadequate to conduct such an evaluation for the remaining 16 compounds. A final
RPF was derived for each PAH based on tumor bioassay data (if available) or cancer-related
endpoint data (if no tumor bioassay RPFs were available).  Final RPFs were derived for
27 PAHs, significantly increasing the number of PAHs that can be addressed through this
approach.  Each RPF was assigned a relative confidence rating reflecting the nature of the tumor
bioassay or cancer-related endpoint database that was used to derive the final RPF for that PAH.
                                       xiii           DRAFT - DO NOT CITE OR QUOTE

-------
                               CONTENTS
LIST OF TABLES	xvii
LIST OF FIGURES	xxi
LIST OF ABBREVIATIONS AND ACRONYMS	xxiv
AUTHORS, CONTRIBUTORS, AND REVIEWERS	xxvi

1. BACKGROUND FOR THE DEVELOPMENT OF RELATIVE POTENCY FACTOR
  APPROACH FOR PAH MIXTURES HEALTH ASSESSMENT	1

2. RATIONALE FOR RECOMMENDING AN RPF APPROACH	2
 2.1. PAHs AS A CHEMICAL CLASS	4
 2.2. THE TOXICOLOGICAL DATABASE FOR PAHs	20
 2.3. BENZO[A]PYRENE AS AN INDEX CHEMICAL	21
 2.4. SIMILARITIES IN MODE OF CARCINOGENIC ACTION FOR PAHs	23
 2.5. STRUCTURAL ALERTS FORPAH CARCINOGENESIS	34
 2.6. SIMILARITIES IN RELATIVE POTENCY ACROSS ENDPOINTS	35
 2.7. SIMILARITIES IN RELATIVE POTENCY ESTIMATES ACROSS SPECIES
     AND EXPOSURE ROUTES	37
 2.8. DOSE ADDITIVITY OF PAHs IN COMBINED EXPOSURES	38

3. DISCUS SIGN OF PREVIOUSLY PUBLISHED RPF APPRO ACHES	44
 3.1. PREVIOUS EFFORTS TO VALIDATE THE RPF APPROACH	52

4. EVALUATION OF THE CARCINOGENICITY OF INDIVIDUAL PAHs	55
 4.1. DATABASE OF STUDIES ON PAH CARCINOGENICITY AND CANCER-
     RELATED ENDPOINTS	55
 4.2. STUDIES IN HUMANS	57
 4.3. STUDIES IN ANIMALS	57
  4.3.1. In Vivo Cancer Bioassays in Animals	86
     4.3.1.1. Dermal Exposure	86
     4.3.1.2. Intraperitoneal Exposure	89
     4.3.1.3. Subcutaneous Injection Exposure	90
     4.3.1.4. Oral Exposure	91
     4.3.1.5. Other Routes	92
  4.3.2. In Vivo Studies of Cancer-Related Endpoints	93
     4.3.2.1. DNAAdducts	93
     4.3.2.2. Clastogenicity or Sister Chromatid Exchange Frequency	95
     4.3.2.3. In Vivo Mutagenicity	96
  4.3.3. In Vitro Studies of Cancer-Related Endpoints	97
     4.3.3.1. Bacterial Mutageni city	97
     4.3.3.2. Mammalian Mutageni city	98
     4.3.3.3. Morphological/Malignant Cell Transformation	99
     4.3.3.4. DNAAdducts	100
     4.3.3.5. DNADamage/Repair	101
     4.3.3.6. Clastogenicity or Sister Chromatid Exchange Frequency	102
 4.4. SUMMARY OF INFORMATION AVAILABLE TO DEVELOP RPFs FOR
     INDIVIDUAL PAHs	103
                                xiv        DRAFT - DO NOT CITE OR QUOTE

-------
5. METHODS FOR DOSE-RESPONSE ASSESSMENT AND RPF CALCULATION	104
 5.1. CHOICE OF DOSE-RESPONSE DATA	104
  5.1.1. Dose-Response Data for Tumor Bioassays	104
  5.1.2. Dose-Response Data for Cancer-Related Endpoint Studies	105
 5.2. OVERALL FORM OF RPF ESTIMATE	106
 5.3. RPF CALCULATION FOR MULTIDOSE DATASETS	106
 5.4. RPF CALCULATION FOR SINGLE DOSE DATASETS	108
 5.5. DOSE CONVERSION FOR RPF CALCULATION	109
 5.6. SPECIAL CONSIDERATIONS FOR RPF CALCULATION USING TUMOR
     BIOASSAYDATA	110
 5.7. SPECIAL CONSIDERATIONS FOR RPF CALCULATION USING CANCER-
     RELATED ENDPOINT DATA	Ill

6. SELECTION OF PAHs FOR INCLUSION IN RELATIVE POTENCY APPROACH	113
 6.1. METHOD FOR SELECTING PAHs FOR INCLUSION IN RELATIVE POTENCY
     APPROACH	114
 6.2. WEIGHT OF EVIDENCE EVALUATION FOR 35 INDIVIDUAL PAHs	117

7. DERIVATION OF FINAL RPFs FOR SELECTED PAHs	190
 7.1. METHODS FOR DERIVING FINAL RPFs	190
 7.2. CONFIDENCE RATINGS FOR FINAL RPFs	194
 7.3. APPLICATION OF RPFs FOR ASSESSING CANCER RISKS FROM EXPOSURE
     TO PAH MIXTURES	196
 7.4. SUSCEPTIBILITY FROM EARLY LIFE EXPOSURE TO CARCINOGENS	196

8. UNCERTAINTIES AND LIMITATIONS ASSOCIATED WITH THE RPF APPROACH
  	198
 8.1. DOSE-RESPONSE ASSESSMENT FOR INDIVIDUAL PAHs	199
 8.2. SELECTION OF PAHs FOR INCLUSION IN RPF APPROACH	201
 8.3. DERIVATION OF A FINAL RPF FOR EACH PAH	203
 8.4. USE OF ANIMAL DATA TO PREDICT HUMAN CANCER RISK FOR PAHs	208
 8.5. ASSUMPTIONS OF A COMMON MODE OF ACTION AND DOSE ADDITIVITY 210
 8.6. EXTRAPOLATION OF RPFs ACROSS EXPOSURE ROUTES	211

9. REFERENCES	218

APPENDIX A. SECONDARY SOURCES REVIEWED FOR IDENTIFICATION OF
            PRIMARY LITERATURE	A-l

APPENDIX B. BIBLIOGRAPHY OF STUDIES WITHOUT BENZO[A]PYRENE AS
            A REFERENCE COMPOUND	B-l
  B.I. BIBLIOGRAPHY OF BIOASSAYS WITHOUT BENZO[A]PYRENE	B-4
  B.2. BIBLIOGRAPHY OF STUDIES ON CANCER-RELATED ENDPOINTS
      WITHOUT BENZO[A]PYRENE	B-10

APPENDIX C. DOSE-RESPONSE DATA FOR POTENCY CALCULATIONS	C-l
                              xv        DRAFT - DO NOT CITE OR QUOTE

-------
APPENDIX D. BENCHMARK DOSE MODELING OUTPUTS	D-l
  D.I. DERMAL BIO AS SAYS	D-l
  D.2. INTRAPERITONEAL BIOASSAYS	D-43
  D.3. LUNG IMPLANTATION BIO AS SAYS	D-82
  D.5. BACTERIAL MUTAGENICITY	D-l 17
  D.6. MAMMALIAN MUTAGENICITY	D-126
  D.7. MALIGNANT TRANSFORMATION	D-154
  D.8. IN VITRO DNA DAMAGE	D-186

APPENDIX E. CALCULATION OF RPFs	E-l

APPENDIX F. EXAMPLE CALCULATION OF RPF DETECTION LIMIT	F-l

APPENDIX G: EVALUATION OF ALTERNATIVES FOR RANKING RPFs	G-1
  G.I. OPTIONS FOR RANKING TUMOR BIO AS SAY RPFs	G-l
  G.2. RANKING NONBIO AS SAY DAT A	G-7
                              xvi        DRAFT - DO NOT CITE OR QUOTE

-------
                                  LIST OF TABLES
1.  PAHs with final RPFs based on tumor bioassay data	xi

2.  PAHs with final RPFs based on cancer-related endpoint data (no tumor bioassay data
   available)	xii

2-1.  PAHs evaluated in the RPF analysis	5

2-2.  Studies of binary mixtures of PAHs and tumorigenicity	40

3-1.  Comparison among various relative potency estimates for PAHs from the published
     literature and regulatory agencies (1984-2004)	45

4-1.  Study summaries:  dermal bioassays of benzo[a]pyrene and at least one other PAH	59

4-2.  Study summaries:  intraperitoneal bioassays of benzo[a]pyrene and at least one other
     PAH	63

4-3.  Study summaries:  subcutaneous bioassays of benzo[a]pyrene and at least one other
     PAH	65

4-4.  Study summaries:  oral bioassays of benzo[a]pyrene and at least one other PAH	66

4-5.  Study summaries:  other route bioassays of benzo[a]pyrene and at least one other
     PAH	67

4-6.  Study summaries:  in vivo DNA adducts with benzo[a]pyrene and at least one other
     PAH	68

4-7.  Study summaries:  in vivo clastogenicity or sister chromatid exchange with
     benzo[a]pyrene and at least one other PAH	70

4-8.  Study summaries:  in vivo mutagenicity with benzo[a]pyrene and at least one other
     PAH	72

4-9.  Study summaries:  in vitro bacterial mutagenicity with benzo[a]pyrene and at least one
     other PAH	73

4-10. Study summaries: in vitro mammalian mutagenicity assays with benzo[a]pyrene
      and at least one other PAH	76

4-11. Study summaries: in vitro morphological/malignant cell transformation with
      benzo[a]pyrene and at least one other PAH	79

4-12. Study summaries: in vitro DNA adducts with benzo[a]pyrene and at least one
      other PAH	81
                                      xvii          DRAFT - DO NOT CITE OR QUOTE

-------
4-13.  Study summaries:  in vitro DNA damage, repair, or synthesis with benzo[a]pyrene
      and at least one other PAH	82

4-14.  Study summaries:  in vitro clastogenicity or sister chromatid exchange with
      benzo[a]pyrene and at least one other PAH	84

5-1. Comparison between molar and mass-based RPF	110

6-1. PAHs with only one RPF from a single in vitro cancer-related endpoint study and
    excluded from RPF approach	113

6-2. Results of weight of evidence evaluation for 27 PAHs selected for inclusion in the
    RPF approach	118

7-1. Final RPFs based on tumor bioassay data	193

7-2. Final RPFs based on cancer-related endpoint data (no tumor bioassay data available)	194

7-3. Relative confidence ratings for RPFs	195

7-4. Sample calculation of estimated cancer risk for benz[a]anthracene with the
    application of ADAFs	197

8-1. Results of simple linear regression of log-transformed average tumor bioassay
    RPF versus log average genotoxicity RPF	206

8-2. PAHs with RPFs of varying relative confidence	207

8-3. Comparisons among average tumor bioassay RPF values by exposure route and
    target organ	213

B-l. Bioassays with and without benzo[a]pyrene by PAH	B-2

C-l. Dermal bioassays:  dose-response information for incidence data	C-2

C-2. Dermal bioassays:  dose-response information for tumor multiplicity	C-8

C-3. Intraperitoneal bioassays: dose-response information for incidence data	C-l3

C-4. Intraperitoneal bioassays: dose-response information for tumor multiplicity	C-23

C-5. Lung implantation bioassays: dose-response information for incidence data	C-27

C-6. Oral bioassays: dose-response information for incidence data	C-31

C-7. Oral bioassays: dose-response information for tumor multiplicity	C-31

C-8. In vitro bacterial mutagenicity:  data use	C-32
                                      xviii         DRAFT - DO NOT CITE OR QUOTE

-------
C-9. In vitro bacterial mutagenicity:  dose-response data	C-35




C-10. In vitro mammalian mutagenicity:  data use	C-53




C-ll. In vitro mammalian mutagenicity:  dose-response data	C-56




C-12. In vitro malignant/morphological cell transformation:  data use	C-64




C-13. In vitro malignant/morphological cell transformation:  dose-response data	C-66




C-14. In vitro DNA adducts: data use	C-74




C-15. In vitro DNA adducts: dose-response data	C-75




C-16. In vitro DNA damage: data use	C-78




C-ll. In vitro DNA damage: dose-response data	C-79




C-18. In vitro clastogenicity: data use	C-84




C-19. In vitro clastogenicity: dose-response data	C-85




C-20. In vivo DNA adducts: data use	C-86




C-21. In vivo DNA adducts: dose-response data	C-87




C-22. In vivo clastogenicity: data use	C-95




C-23. In vivo clastogenicity: dose-response data	C-96




E-l. Dermal bioassays: RPF calculations for incidence data	E-2




E-2. Dermal bioassays: RPF calculations for multiplicity data	E-5




E-3. Intraperitoneal bioassays:  RPF  calculations for incidence data	E-6




E-4. Intraperitoneal bioassays:  RPF  calculations for multiplicity data	E-8




E-5. Lung implantation bioassays: RPF calculations (incidence data)	E-9




E-6. Oral bioassays: RPF calculations (incidence and multiplicity data)	E-10




E-7. In vivo DNA adducts:  RPF calculations	E-ll




E-8. In vivo clastogenicity or sister chromatid exchange:  RPF calculation	E-14




E-9. In vitro bacterial mutagenicity:  RPF calculations	E-16




E-10. In vitro mammalian mutagenicity:  RPF calculations	E-22






                                        xix          DRAFT - DO NOT CITE OR QUOTE

-------
E-ll. In vitro morphological/malignant transformation: RPF calculation	E-25

E-12. In vitro DNAadducts:  RPF calculations	E-27

E-13. In vitro DNA damage:  RPF calculations	E-29

E-14. In vitro clastogenicity or sister chromatid exchange:  RPF calculations	E-31

F-l.  Example data for calculation of RPF detection limit	F-l

G-l.  Comparisons among average nonzero tumor bioassay-based RPF values by
     exposure route	G-3

G-2.  Comparisons among average nonzero tumor bioassay-based RPF values by
     calculation method	G-6

G-3.  Results of simple linear regression of log-transformed average genotoxicity RPF
     versus log average tumor bioassay RPF	G-8
                                      xx           DRAFT - DO NOT CITE OR QUOTE

-------
 1                                    LIST OF FIGURES
 2
 3
 4   2-1.  Structural features of PAHs	20
 5
 6   2-2.  Metabolic pathways for benzo[a]pyrene	24
 7
 8   2-3.  Overview of the proposed key events in the mode of action for PAH carcinogenicity	26
 9
10   2-4.  Structures of the four stereoisomeric adduct moieties, anti-[EaP]-N2-dG., derived from
11        the trans- or cis- covalent binding of (+)-a«ft'-BaP diol epoxide or (-)-awft'-BaP diol
12        epoxide to dG residues in DNA	27
13
14   2-5.  Depurinating adducts of benzo[a]pyrene formed by one-electron oxidation	28
15
16   2-6.  Spectrum of DNA adducts anticipated with PAH o-quinones	29
17
18   2-7.  Interaction of PAHs with the AhR - regulation of genes related to induction of
19        metabolism  and cell differentiation and proliferation	31
20
21   6-1.  Weight of evidence analysis of for selection of PAHs to be included in the RPF
22        approach	115
23
24   6-2.  2,3-Acepyrene (ACEP) RPFs	120
25
26   6-3.  Anthanthrene (AA) RPFs	122
27
28   6-4.  Anthracene  (AC) RPFs	124
29
30   6-5.  Benz[a]anthracene (BaA) RPFs	126
31
32   6-6.  1 lH-Benz[b,c]aceanthrylene (BbcAC) RPFs	128
33
34   6-7.  Benzo[b]fluoranthene (BbF) RPFs	130
35
36   6-8.  1 lH-Benzo[b]fluorene (BbFE) RPFs	132
37
38   6-9.  Benzo[c]fluorene  (BcFE) RPFs	134
39
40   6-10. Benz[e]aceanthrylene (BeAC) RPFs	136
41
42   6-11. Benzo[e]pyrene (BeP) RPFs	138
43
44   6-12. Benzo[g,h,i]fluoranthene (BghiF) RPFs	140
45
46   6-13. Benzo[g,h,i]perylene (BghiP) RPFs	142
47
48   6-14. Benz[j]aceanthrylene(BjAC)RPFs	144
49


                                           xxi          DRAFT - DO NOT CITE OR QUOTE

-------
 1   6-15. Benzo[j]fluoranthene(BjF)RPFs	146
 2
 3   6-16. Benzo[k]fluoranthene (BkF) RPFs	148
 4
 5   6-17. Benz[l]aceanthrylene (B1AC) RPFs	150
 6
 7   6-18. Chrysene (CH) RPFs	152
 8
 9   6-19. Coronene (CO) RPFs	154
10
11   6-20. Cyclopenta[c,d]pyrene (CPcdP) RPFs	156
12
13   6-21. Cyclopenta[d,e,f]chrysene (CPdefC) RPFs	158
14
15   6-22. Dibenz[a,c]anthracene (DBacA) RPFs	160
16
17   6-23. Dibenzo[a,e]fluoranthene (DBaeF) RPFs	162
18
19   6-24. Dibenzo[a,e]pyrene (DBaeP) RPFs	164
20
21   6-25. Dibenz[a,h]anthracene (DBahA) RPFs	166
22
23   6-26. Dibenzo[a,h]pyrene (DBahP) RPFs	168
24
25   6-27. Dibenzo[a,i]pyrene (DbaiP) RPFs	170
26
27   6-28. Dibenzo[a,l]pyrene (DBalP) RPFs	173
28
29   6-29. Fluoranthene (FA) RPFs	175
30
31   6-30. Fluorene (FE) RPFs	177
32
33   6-31. Indeno[l,2,3-c,d]pyrene(IP)RPFs	179
34
35   6-32. Naphtho[2,3-e]pyrene (N23eP) RPFs	181
36
37   6-33. Perylene (Pery) RPFs	183
38
39   6-34. Phenanthrene (PH) RPFs	185
40
41   6-35. Pyrene (Pyr) RPFs	187
42
43   6-36. Triphenylene (Tphen) RPFs	189
44
45   8-1. Correlation between incidence and multiplicity RPFs	204
46
47   G-l. Average bioassay RPF versus average in vivo DNA adduct RPF	G-9
48
49   G-2. Average bioassay RPF versus average in vivo nonbioassay RPF	G-10
                                          xxii         DRAFT - DO NOT CITE OR QUOTE

-------
1
2   G-3. Average bioassay RPF versus average nonbioassay RPF	G-ll
o
J
4   G-4. Average bioassay RPF versus average in vitro nonbioassay RPF	G-12
5
6
                                        xxiii         DRAFT - DO NOT CITE OR QUOTE

-------
   LIST OF ABBREVIATIONS AND ACRONYMS
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46


ADAF
AEL
Ah
AhR
ATSDR
AUC
BMD
BMR
CASRN
CCRIS
CHO
CYP
dG
DMSO
DNA
DSSTOX
EOPP
EROD
HPRT
IARC
IRIS
MGP
MN-PCE
mRNA
MVK
NTP
OEHHA
PAC
PAH
PCB
PCR
PEF
QSAR
RNA
RPF
RTD
SD
TK
TIDAL
TEF
TK
TPA
TSCATS
age-dependent adjustment factor
acceptable exposure level
aryl hydrocarbon
Ah receptor
Agency for Toxic Substances and Disease Registry
area under the curve
benchmark dose
benchmark response
Chemical Abstract Service Registry Number
Chemical Carcinogenesis Research Information System
Chinese hamster ovary
cytochrome P450
deoxyguanosine
dimethyl sulfoxide
deoxyribonucleic acid
Distributed Structure-Searchable Toxicity
estimated order of potential potency
ethoxyresorufin O-deethylase
hypoxanthine-guanine phosphoribosyl transferase gene
International Agency for Research on Cancer
Integrated Risk Information System
manufactured gas plant
micronuleated polychromatic erythrocyte
messenger ribonucleic acid
Moolgavkar-Venson-Knudsen two-stage model
National Toxicology Program
Office of Environmental Health Hazard Assessment, California EPA
polycyclic aromatic compound
polycyclic aromatic hydrocarbon
polychlorinated biphenyl
polymerase chain reaction
potency equivalency factor
quantitative structure activity relationship
ribonucleic acid
relative potency factor
relative tumor dose
standard deviation
thymidine kinase locus
time-integrated DNA adduct level
toxicity equivalency factor
thymidine kinase
12-O-tetra-decanoylphorbol-13-acetate
Toxic Substances Control Act Test Submissions
                     xxiv
DRAFT - DO NOT CITE OR QUOTE

-------
1   U.S. EPA        U.S. Environmental Protection Agency
2   WHO           World Health Organization
o
J
4    Abbreviations for PAH chemical names are provided in Table 2-1.
5
                                        xxv          DRAFT - DO NOT CITE OR QUOTE

-------
 1                     AUTHORS, CONTRIBUTORS, AND REVIEWERS
 2
 3   PROJECT CO-MANAGERS
 4
 5   Lynn Flowers, Ph.D., DABT
 6   National Center for Environmental Assessment
 7   Office of Research and Development
 8   U.S. Environmental Protection Agency
 9   Washington, DC
10
11   Martin Gehlhaus, III
12   National Center for Environmental Assessment, IRIS Program
13   Office of Research and Development
14   U.S. Environmental Protection Agency
15   Washington, DC
16
17   AUTHORS
18
19   Lynn Flowers, Ph.D., DABT
20   National Center for Environmental Assessment
21   Office of Research and Development
22   U.S. Environmental Protection Agency
23   Washington, DC
24
25   Martin Gehlhaus, III
26   National Center for Environmental Assessment, IRIS Program
27   Office of Research and Development
28   U.S. Environmental Protection Agency
29   Washington, DC
30
31   Karen Hogan
32   National Center for Environmental Assessment, IRIS Program
33   Office of Research and Development
34   U.S. Environmental Protection Agency
35   Washington, DC
36
37   Channa Keshava, Ph.D.
38   National Center for Environmental Assessment, IRIS Program
39   Office of Research and Development
40   U.S. Environmental Protection Agency
41   Washington, DC
42
43   Glenn Rice, Ph.D.
44   National Center for Environmental Assessment
45   Office of Research and Development
46   U.S. Environmental Protection Agency
47   Cincinnati, OH
48
                                          xxvi          DRAFT - DO NOT CITE OR QUOTE

-------
 1   Jamie Strong, Ph.D.
 2   National Center for Environmental Assessment, IRIS Program
 3   Office of Research and Development
 4   U.S. Environmental Protection Agency
 5   Washington, DC
 6
 7   Linda Teuschler, Ph.D.
 8   National Center for Environmental Assessment
 9   Office of Research and Development
10   U.S. Environmental Protection Agency
11   Cincinnati, OH
12
13   Stephen Nesnow, Ph.D.
14   Environmental Carcinogenesis Division
15   National Health and Environmental Effects Research Laboratory
16   Office of Research and Development
17   Research Triangle Park, NC
18
19   Chao Chen, Ph.D.
20   National Center for Environmental Assessment
21   Office of Research and Development
22   Washington, DC
23
24   Heather Carlson-Lynch, S.M.
25   Syracuse Research Corporation, Inc.
26   Syracuse, NY
27
28   Julie Stickney, Ph.D., DABT
29   Syracuse Research Corporation, Inc.
30   Syracuse, NY
31
32   Peter R. McClure, Ph.D., DABT
33   Syracuse Research Corporation, Inc.
34   Syracuse, NY
35
36   Amber B acorn
37   Syracuse Research Corporation, Inc.
38   Syracuse, NY
39
40
                                          xxvii          DRAFT - DO NOT CITE OR QUOTE

-------
 1        1.  BACKGROUND FOR THE DEVELOPMENT OF A RELATIVE POTENCY
 2           FACTOR APPROACH FOR PAH MIXTURES HEALTH ASSESSMENT
 o
 4          This analysis focuses on the relative potency factor (RPF) approach that is based on
 5   component PAHs in PAH mixtures. U.S. EPA held a peer consultation workshop to outline some
 6   of the important issues related to approaches for PAH mixtures risk assessment. These issues are
 7   discussed in Peer Consultation Workshop on Approaches to Poly cyclic Aromatic Hydrocarbon
 8   (PAH) Health Assessment (U.S. EPA, 2002) and the accompanying discussion document.  Health
 9   assessments for 15 unsubstituted, nonheterocyclic poly cyclic aromatic hydrocarbons (PAHs)
10   with three or more rings are currently entered on EPA's IRIS database. Benzo[a]pyrene is the
11   only PAH for which there are robust animal dose-response data for the oral, dermal, and
12   inhalation routes.
13          In  1993, U.S. EPA published the Provisional Guidance for Quantitative Risk Assessment
14   of PAHs (Provisional Guidance}.  The Provisional Guidance recommended estimated orders of
15   potential potency (EOPP) for individual PAHs that could be used in a component-based
16   approach to PAH mixtures risk assessment.  The Provisional Guidance recommended EOPPs for
17   seven PAHs categorized as Group B2 (probable human carcinogens) under the 1986 U.S. EPA
18   Cancer Guidelines: benzo[a]pyrene, benz[a]anthracene, benzo[b]fluoranthene, benzo[k]fluor-
19   anthene, chrysene, dibenz[a,h]anthracene, and indeno[l,2,3-c,d]pyrene (U.S. EPA,  1993).  The
20   current analysis extends the 1993 Provisional Guidance and provides recommendations for
21   further development of this approach to PAH mixtures risk assessment.  The assessment includes
22   the following:
23
24      (1) A rationale for recommending an order of potency, or RPF, approach;
25
26      (2) A summary of previous approaches for developing the RPF approach for PAHs;
27
28      (3) Identification of individual carcinogenic PAHs that could be included in the RPF
29          approach;
30
31      (4) Identification of potential index chemicals;
32
33      (5) Presentation of the available literature for in vivo carcinogenicity and both in vivo  and in
34          vitro cancer-related endpoint assays for individual PAHs;
35
36      (6) Development of a recommendation for the RPF approach for PAH mixtures; and
37
38      (7) Characterization of strengths, weaknesses, and uncertainties associated with the
39          recommended approaches.
40
41
42

                                            1           DRAFT - DO NOT CITE OR QUOTE

-------
 1               2.  RATIONALE FOR RECOMMENDING AN RTF APPROACH
 2
 3
 4          PAHs are a concern as human health hazards, because many PAHs are demonstrated
 5   tumorigenic agents in animal bioassays and are active in in vivo or in vitro tests for genotoxicity
 6   or deoxyribonucleic acid (DNA) damage. PAHs do not occur in the  environment as isolated
 7   entities; they primarily occur in complex mixtures generated from the combustion or pyrolysis of
 8   substances containing carbon and hydrogen.  Several complex mixtures of PAHs have been
 9   classified as possibly carcinogenic, probably  carcinogenic, or carcinogenic to humans (Straif et
10   al., 2005; U.S. EPA, 2002; Bostrom et al., 2002; WHO, 1998; ATSDR, 1995; IARC, 1985,
11   1984a, b, 1983).
12          In accordance with U.S. EPA (2000, 1986) guidance for health  risk assessment of
13   chemical mixtures, assessment of the cancer risk from long-term human exposure to a particular
14   PAH mixture would best be conducted with quantitative information on the dose-response
15   relationship  for cancer from chronic exposure to the mixture of concern. When data for the
16   mixture of concern are not available, U.S. EPA (2000, 1986) guidance  recommends using
17   toxicity data on a "sufficiently similar" mixture. However, quantitative cancer dose-response
18   information  exists only for a few complex mixtures generated from the combustion or pyrolysis
19   of organic matter; for example, tobacco smoke, coke oven emissions, and emissions from roofing
20   tar pots (see Bostrom et al., 2002; Albert et al.,  1983). U.S. EPA's IRIS database currently
21   includes assessments for only three PAH-containing mixtures:  coke  oven emissions, creosote,
22   and diesel emissions. The availability of oral carcinogenicity bioassays of manufactured gas
23   plant (MGP) residue (Weyand et al., 1995) and coal tar preparations  (Gulp et al., 1998; Gaylor et
24   al., 1998) has expanded the PAH mixture cancer database.
25          Component-based approaches, involving an analysis of the toxicity of components of the
26   mixture, are recommended when appropriate toxicity data on a complex mixture of concern, or
27   on a "sufficiently similar" mixture, are unavailable (U.S. EPA, 2000, 1986).  Component-based
28   approaches involving dose addition (such as the RPF approach) are recommended when
29   components in the mixture are judged to act in a lexicologically similar manner. In the RPF
30   approach,  doses of component chemicals that act in a lexicologically similar manner are added
31   together, after scaling the doses relative to the potency of an index chemical  (U.S. EPA, 2000,
32   1986). Then, using the dose-response curve of the index chemical, the response to the total
33   equivalent dose in the mixture is estimated. The index compound is  typically the best-studied
34   member of the class with the largest body of  available data describing exposure and health
35   effects. The index chemical should have a quantitative dose-response assessment of acceptable
36   scientific quality and must have (or be expected to have) similar toxic effects to the rest of the
37   members of the class.
                                                         DRAFT - DO NOT CITE OR QUOTE

-------
 1          For exposure situations in which dose-response data for the PAH mixture or a sufficiently
 2   similar mixture are not available (e.g., the source of the PAH contamination may be mixed or
 3   unknown), there are at least three practical advantages of an RPF approach that uses
 4   benzo[a]pyrene as the index PAH:
 5
 6       (1) Benzo[a]pyrene is routinely assayed and detected in environmental media contaminated
 7          with PAH mixtures;
 8
 9       (2) Benzo[a]pyrene is the only PAH for which robust cancer dose-response data involving
10          chronic exposures are available; and
11
12       (3) There is a large database of studies in which the potency of benzo[a]pyrene is compared
13          with the potency of other PAHs in various assays.
14
15          The database includes animal tumorigenicity2 assays involving dermal or parenteral
16   administration, and in vivo and in vitro assays of cancer-related endpoints (e.g., various
17   genotoxic endpoints).  Thus, RPFs for a number of PAHs can be derived.
18          The RPF approach involves two key assumptions related to the application of a dose-
19   additivity model: (1) the assumption of similar toxicological action; and (2) the assumption that
20   interactions among PAH mixture components do not occur at low levels of exposure typically
21   encountered in the environment.
22          Mechanistic studies indicate that the mutagenic and tumor-initiating activity of most
23   carcinogenic PAHs requires metabolic activation to reactive intermediates (e.g., stereospecific
24   dihydrodiol epoxides). For several PAHs (e.g., benzo[a]pyrene, dibenz[a,h]anthracene,
25   dibenzo[a,l]pyrene), there is evidence that DNA damage associated with metabolism can lead to
26   mutations in cancer-related genes. Tumor promotion and progression by PAHs may involve
27   parent compound binding to the aryl hydrocarbon (Ah) receptor and subsequent alterations of
28   gene expression, as well as by cell proliferation in response to cytotoxic effects from metabolites
29   (see Section 2.4, Similarities in Mode of Carcinogenic Action for PAHs).  As such, there is
30   evidence that an assumption of similar toxicological action is reasonable; however, the
31   carcinogenic process for individual PAHs is likely to be related to  some unique combination of
32   multiple molecular events resulting from the formation of several reactive species.  The second
33   assumption of no interactions at low levels of exposure is also reasonable, but has not been
34   conclusively demonstrated in experimental systems (see  Section 2.8, Dose Additivity of PAHs in
35   Combined Exposures).
36          Key limitations to the RPF approach, relative to whole mixture approaches, are:
37   (1) RPFs have been derived for a limited  number of PAHs; and (2) cancer risks from non-PAH
38   components, unidentified PAHs, and heterocyclic and substituted PAHs in PAH mixtures are not
     throughout this report, the term "tumorigenicity" is used to describe the production of either benign or malignant
     tumors.

                                             3           DRAFT - DO NOT CITE OR QUOTE

-------
 1   estimated. The first of these limitations is being addressed, to the degree allowable by available
 2   data, by the derivation of RPFs for numerous PAHs as discussed in Chapters 4 through 7 of this
 3   report. If non-PAH carcinogenic components are identified and quantified in the complex
 4   mixture of concern and appropriate dose-response data are available, the second limitation can be
 5   addressed by adding the cancer risk from PAH components estimated by the RPF approach to
 6   cancer risks estimated for the non-PAH carcinogenic components of the mixture.  Previous
 7   efforts to validate the  RPF approach using data for PAH mixtures are discussed in Section 3.1.
 8   These validation efforts compared the cancer risk of a PAH mixture measured experimentally
 9   with the cancer risk that was predicted using the RPF method but were limited by the small
10   number of compounds for which RPFs and analytical data were available (Muller et al., 1997;
11   McClure, 1996; Goldstein et al., 1994; Clement Associates, 1990, 1988; Krewski et al., 1989).
12   Validation of the updated approach presented here would be of value, either using previous data
13   on PAH mixtures (human and animal) or using new data collected with the main purpose of
14   evaluating the validity of the approach.
15
16   2.1.  PAHs AS A CHEMICAL CLASS
17          The PAH chemical class has been variously defined to include organic compounds
18   containing either two  or more, or three or more, fused rings made up of carbon and hydrogen
19   atoms (i.e., unsubstituted parent PAHs and their alkyl-substituted derivatives) (WHO, 1998).
20   Most PAHs are high-melting, high-boiling point, lipophilic compounds, predominately generated
21   from the incomplete combustion or pyrolysis of organic matter. The PAH chemical class
22   includes alkylated PAHs (e.g., 1,4-dimethylphenanthrene  and 5-methylchrysene), but not
23   heterocyclic compounds containing N, S, or O or PAHs substituted with N-, S-, or O-containing
24   groups; these are included in a larger chemical class, often referred to as polycyclic aromatic
25   compounds (PACs) (WHO, 1998). The number of chemicals that comprise the PAHs class is
26   unknown; however, there are thought to be hundreds of individual PAHs present as components
27   of complex mixtures (WHO, 1998). The analysis presented here is limited in focus to include
28   only unsubstituted PAHs with three or more fused aromatic rings containing only carbon and
29   hydrogen atoms, because these are the most widely studied members of the PAH chemical class.
30   Naphthalene is a widely studied two-ring PAH compound; however, a separate toxicological
31   review and carcinogenicity assessment is being developed by the IRIS Program for this
32   compound and it is not included in this RPF approach.  The list of PAH compounds that were
33   considered for inclusion in this analysis is presented in Table 2-1 along with the Chemical
34   Abstracts Service Registry Numbers (CASRNs) and the abbreviations that are utilized in tables
35   throughout the report.
                                                        DRAFT - DO NOT CITE OR QUOTE

-------
       Table 2-1. PAHs evaluated in the RTF analysis
             PAH
       (common synonyms)
 CASRN
Abbreviation
Structure
Molecular
 weight
 (g/mol)
Benzo[a]pyrene
          BaP
                                        252.31
Aceanthrylene
202-03-09
ACEA
                   202.26
Acenaphthene
 83-32-9
AN
                   154.21
Acenaphthylene
208-96-8
ANL
                   152.20
Acephenanthrylene
201-06-9
APA
                   202.26
                                                      DRAFT - DO NOT CITE OR QUOTE

-------
       Table 2-1.  PAHs evaluated in the RTF analysis
              PAH
       (common synonyms)
 CASRN
Abbreviation
Structure
Molecular
 weight
 (g/mol)
Acepyrene, 2,3-
25732-74-5
ACEP
                    228.29
Anthanthrene
 191-26-4
AA
                    276.34
Anthracene
 120-12-7
AC
                    178.23
Benzacenaphthylene
76774-50-0
BAN
                    202.26
B enz [a] anthracene
  56-55-3
BaA
                    228.29
Benzo [a]fluoranthene
 203-33-8
BaF
                    252.32
                                                       DRAFT - DO NOT CITE OR QUOTE

-------
       Table 2-1.  PAHs evaluated in the RTF analysis
              PAH
       (common synonyms)
CASRN
Abbreviation
Structure
Molecular
 weight
 (g/mol)
Benzo[a]fluorene
238-84-6
BaFE
                    216.28
Benzo[a]perylene
191-85-5
BaPery
                    302.38
Benz[b,c]aceanthrylene, 11H-
202-94-8
BbcAC
                    240.30
Benz [b] anthracene
(naphthacene)
 92-24-0
BbA
                    228.29
Benzo[b]chrysene
214-17-5
BbC
                    278.35
                                          7            DRAFT - DO NOT CITE OR QUOTE

-------
       Table 2-1. PAHs evaluated in the RTF analysis
              PAH
       (common synonyms)
CASRN
Abbreviation
Structure
Molecular
 weight
 (g/mol)
Benzo [b]fluoranthene
205-99-2
BbF
                   252.32
Benzo [b]fluorene, 11H
243-17-4
BbFE
                   216.28
Benzo [b]perylene
197-70-6
BbPeiy
                   302.38
Benzo[c]chrysene
194-69-4
BcC
                   278.35
Benzo [c]fluorene
205-12-9
BcFE
                   216.28
                                          8            DRAFT - DO NOT CITE OR QUOTE

-------
       Table 2-1.  PAHs evaluated in the RTF analysis
              PAH
       (common synonyms)
CASRN
Abbreviation
Structure
Molecular
 weight
 (g/mol)
Benzo [c]phenanthrene
195-19-7
BcPH
                    228.29
Benz [e] aceanthry lene
199-54-2
BeAC
                    252.32
Benzo [e]pyrene
192-97-2
BeP
                    252.32
Benzo [g,h,i]fluoranthene
203-12-3
BghiF
                    226.28
Benzo [g,h,i]pery lene
191-24-2
BghiP
                    276.34
                                                       DRAFT - DO NOT CITE OR QUOTE

-------
       Table 2-1.  PAHs evaluated in the RTF analysis
              PAH
       (common synonyms)
 CASRN
Abbreviation
Structure
Molecular
 weight
 (g/mol)
Benzo[g]chrysene
 196-78-1
BgC
                    278.35
Benz |] ] aceanthry lene
 202-33-5
BjAC
                    252.32
Benzo[j]fluoranthene
 205-82-3
BjF
                    252.32
Benzo fkjfluoranthene
 207-08-9
BkF
                    252.32
Benz [1] aceanthry lene
 211-91-6
B1AC
                    252.32
Benzophenanthrene
65777-08-4
BPH
                    228.29
                                          10
                      DRAFT - DO NOT CITE OR QUOTE

-------
       Table 2-1.  PAHs evaluated in the RTF analysis
              PAH
       (common synonyms)
  CASRN
Abbreviation
Structure
Molecular
 weight
 (g/mol)
Chrysene
 218-01-9
CH
                    228.29
Coronene
  191-07-1
CO
                    300.36
Cyclopent[h,i]aceanthrylene
131581-33-4
CPhiACEA
                    226.28
Cyclopenta[c,d]pyrene
27208-37-3
CPcdP
                    226.28
Cyclopenta[d,e,f]chrysene, 4H-
 202-98-2
CPdefC
                    240.30
Cyclopenta[d,e,f]phenanthrene
 203-64-5
CPdefPH
                    190.24
                                          11           DRAFT - DO NOT CITE OR QUOTE

-------
       Table 2-1.  PAHs evaluated in the RTF analysis
              PAH
       (common synonyms)
  CASRN
Abbreviation
Structure
Molecular
 weight
 (g/mol)
Cyclopenta[h,i]acephenanthrylene
114959-37-4
CPhiAPA
                    226.28
Cyclopentaphenanthrene
 219-08-9
CPPH
                    216.28
Cyclopenteno-l,2-benzanthracene, 5,6-
 7099-43-6
CPBA
                    268.36
Dibenz[a,c]anthracene
(benzotriphenylene)
 215-58-7
DBacA
                    278.35
Dibenzo[a,c]fluorene, 13H-
 201-65-0
DBacFE
                    266.34
                                          12            DRAFT - DO NOT CITE OR QUOTE

-------
       Table 2-1.  PAHs evaluated in the RTF analysis
              PAH
       (common synonyms)
 CASRN
Abbreviation
Structure
Molecular
 weight
 (g/mol)
Dibenzo [a,e]fluoranthene
5385-75-1
DBaeF
                    302.38
Dibenzo [a,e]py rene
 192-65-4
DBaeP
                    302.38
Dibenzo [a,f]fluoranthene
(indeno [1,2,3 -fg] naphthacene)
203-11-2
DBafF
                    302.38
Dibenzo [a,g]fluorene, 13H-
207-83-0
DBagFE
                    266.34
Dibenz[a,h]anthracene
 53-70-3
DBahA
                    278.35
                                          13            DRAFT - DO NOT CITE OR QUOTE

-------
       Table 2-1.  PAHs evaluated in the RTF analysis
              PAH
       (common synonyms)
 CASRN
Abbreviation
Structure
Molecular
 weight
 (g/mol)
Dibenzo [a,h]pyrene
 189-64-0
DBahP
                    302.38
Dibenzo [a,i]py rene
 189-55-9
DBaiP
                    302.38
Dibenzo [a,l]py rene
 191-30-0
DBalP
                    302.38
Dibenzo [b,e]fluoranthene
2997-45-7
DBbeF
                    302.38
Dibenzo [e,l]pyrene
(dibenzo[fg,op]naphthacene)
 192-51-8
DBelP
                    302.38
                                          14            DRAFT - DO NOT CITE OR QUOTE

-------
       Table 2-1. PAHs evaluated in the RTF analysis
              PAH
       (common synonyms)
  CASRN
Abbreviation
Structure
Molecular
 weight
 (g/mol)
Dibenzo [h,rst]pentaphene
  192-47-2
DBhrstPent
                   352.43
Dibenz \j ,mno] acephenanthry lene
153043-82-4
DBjmnoAPH
                   276.34
Dibenz [k,mno]acephenanthrylene
153043-81-3
DBkmnoAPH
                   276.34
Dihydroaceanthrylene, 1,2-
 641-48-5
DACEA
                   204.27
Fluoranthene
 206-44-0
FA
                   202.26
                                         15            DRAFT - DO NOT CITE OR QUOTE

-------
       Table 2-1. PAHs evaluated in the RTF analysis
              PAH
       (common synonyms)
  CASRN
Abbreviation
Structure
Molecular
 weight
 (g/mol)
Fluorene
  86-73-7
FE
                    166.22
Indeno [ 1,2,3 -c,d]fluoranthene
  193-43-1
IF
                   276.34
Indeno [ 1,2,3 -c,d]pyrene
  193-39-5
IP
                   276.34
Naphth[l ,2,3 -mno]acephenanthrylene
113779-16-1
N123mnoAPH
                   276.34
                                         16            DRAFT - DO NOT CITE OR QUOTE

-------
       Table 2-1. PAHs evaluated in the RTF analysis
              PAH
       (common synonyms)
  CASRN
Abbreviation
Structure
Molecular
 weight
 (g/mol)
Naphtho[ 1,2-b]fluoranthene
111189-32-3
N12bF
                    302.38
Naphtho[2, l-a]fluoranthene
 203-20-3
N21aF
                    302.38
Naphtho[2,3-a]pyrene
(naphtho [2,1,8-qra]naphthacene)
  196-42-9
N23aP
                    302.38
Naphtho[2,3-e]pyrene
(dibenzo [de,qr] naphthacene)
  193-09-9
N23eP
                    302.38
                                           17           DRAFT - DO NOT CITE OR QUOTE

-------
       Table 2-1.  PAHs evaluated in the RTF analysis
              PAH
       (common synonyms)
CASRN
Abbreviation
Structure
Molecular
 weight
 (g/mol)
Pentacene
135-48-8
PCE
                    278.35
Pentaphene
(dibenzphenanthrene, 2,3:6,7-)
222-93-5
Pent
                    278.35
Perylene
198-55-0
Pery
                    252.32
Phenanthrene
 85-01-8
PH
                    178.23
Picene
213-46-7
Pic
                    278.35
                                          18           DRAFT - DO NOT CITE OR QUOTE

-------
Table 2-1. PAHs evaluated in the RTF analysis


1
PAH
(common synonyms)
Pyrene
Tribenzofluoranthene 3,4-10,11-12,13-


Triphenylene
CASRN
129-00-0
13579-05-0


217-59-4
Abbreviation
Pyr
TBF


Tphen
Structure
^




CC
[

l""
%X^

/
Vv
^
^




\
=<
x^^t/^^
~^^^



!3
)
Molecular
weight
(g/mol)
202.26
352.43


228.29

2 Unsubstituted PAHs have been further classified into alternant and nonalternant
3 compounds. Alternant PAHs are those compounds composed solely of fused benzene rings,
4 while nonalternant PAHs contain both benzene and five carbon rings. Among alternant PAHs,
5 important structural features related to enhanced mutagenicity and carcinogenicity include the
6 presence of at least four rings (Bostrom et al., 2002). Common structural features of PAH
7 compounds are illustrated in Figure 2-1 .




                            19
DRAFT - DO NOT CITE OR QUOTE

-------
                    Benzo[a]pyrene                   Pyrene
                            Examples of Alternant PAHs
                        Fluorene
                                             Fluoranthene
                           Examples of Nonalternant PAHs
                                               Fjord-region
                        Chrysene
                                                  Benzo[c]phenanthrene
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
                     Bay-region and Fjord-regions of PAHs

       Figure 2-1. Structural features of PAHs.
2.2.  THE TOXICOLOGICAL DATABASE FOR PAHs
       Over the last 30- to 50-years, a large PAH database has been generated including studies
of carcinogenicity in animal bioassays, genotoxicity in various test systems, and metabolism
(bioactivation) to tumorigenic and/or genotoxic intermediates.  Carcinogenicity and genotoxicity
data are sufficient to classify a number of individual PAHs as possibly carcinogenic to humans
(WHO, 1998; U.S. EPA, 1993; IARC, 1989,  1986, 1985, 1984a, b, 1983).  Other PAHs have
been tested for tumorigenicity and/or genotoxicity, but either nonpositive or equivocal results
were obtained; for many PAHs, positive results were only observed in genotoxicity assays (e.g.,
pyrene). Many studies have been performed to provide further understanding about the
carcinogenic mode of action of PAHs (see Bostrom et al., 2002; WHO, 1998; ATSDR, 1995).
Therefore, the PAH database contains studies that evaluate:
                                            20
                                                    DRAFT - DO NOT CITE OR QUOTE

-------
 1          •  Metabolism to reactive intermediates;
 2          •  Characterization of PAH-DNA adducts;
 3          •  Mutagenicity of PAHs in bacterial and mammalian cells;
 4          •  Mutation spectra in identified oncogene and tumor suppressor genes;
 5          •  Clastogenic effects;
 6          •  Cell transformation; and
 7          •  Initiation and promotion of carcinogenicity (complete carcinogenesis).
 8
 9          A limitation to the database is the lack of data from long-term oral or inhalation cancer
10   studies for most individual PAH compounds.  The only PAH for which there are robust animal
11   dose-response data is benzo[a]pyrene (Kroese et al., 2001; Gulp et al., 1998, 1996a, b; Thyssen
12   etal., 1981, 1980; Rigdon et al., 1969; Rigdon and Neal, 1969, 1966; Neal and Rigdon, 1967).
13   Furthermore,  most of the toxicological data available for PAHs relate to cancer or genotoxicity.
14   Available information on the systemic, noncarcinogenic effects of PAHs is limited, although
15   immunological, neurotoxic, and developmental effects have been noted in animal studies and
16   some human studies (for earlier reviews, see WHO, 1998; ATSDR, 1995). As a result, the
17   relative potency methodology described here is applied only to cancer risk assessment for PAHs.
18
19   2.3.  BENZO[A]PYRENE AS AN INDEX CHEMICAL
20          Because long-term animal studies are not available for many individual PAHs, it is
21   necessary to choose an appropriate index chemical for comparison of relative carcinogenic
22   potency. The index compound is typically the best-studied member of the class, with the largest
23   body of available data describing exposure and health effects. The index chemical should have a
24   quantitative dose-response assessment of acceptable scientific quality and must have (or be
25   expected to have) similar toxic effects to the rest of the members of the class.
26          Although the PAH composition of complex mixtures varies, benzo[a]pyrene is
27   considered to be present in significant amounts in certain occupational environments and urban
28   settings (WHO, 1998; Petry et al., 1996; ATSDR, 1995). Benzo[a]pyrene is one of the most
29   potent of the carcinogenic PAHs and has, therefore, been proposed to contribute significantly to
30   the carcinogenicity of a PAH mixture, even when present in low concentrations (Petry et al.,
31   1996). Benzo[a]pyrene is also the best-studied PAH compound, with carcinogenicity bioassay
32   data available for several routes of exposure and a considerable number of studies on
33   carcinogenic mode of action.  Benzo[a]pyrene has been characterized as reasonably anticipated
34   to be a human carcinogen (NTP, 2005) or carcinogenic to humans (Straif, 2005).
35          The laboratory animal database for benzo[a]pyrene is robust.  Benzo[a]pyrene has been
36   shown to induce tumors at the site of administration and at distal sites in numerous studies.
37   Dose-response data for tumors are available for the oral, inhalation, and dermal routes of
38   administration in multiple species. There are methodological limitiations associated with the

                                             21           DRAFT - DO NOT CITE OR QUOTE

-------
 1   inhalation data (Thyssen et al., 1981), although positive findings in intratracheal instillation
 2   studies support the observed positive response. Dermal exposure studies with several strains of
 3   mice also provide data on dose-related tumor incidences (Albert et al., 1991; Warshawsky and
 4   Barkley, 1987; Habs et al., 1984, 1980; Nesnow et al., 1983; Wynder et al., 1957).
 5          The animal carcinogenicity database for benzo[a]pyrene includes several well-conducted
 6   oral cancer bioassays. Kroese et al. (2001) conducted a well-designed gavage study of
 7   benzo[a]pyrene carcinogenicity and found that benzo[a]pyrene induced tumors at multiple sites
 8   in rats of both sexes, specifically in the liver, forestomach, auditory canal, and oral cavity. In
 9   another well-conducted study, using Ah-responsive B6C3Fi female mice exposed to
10   benzo[a]pyrene in the diet (Beland and Gulp, 1998; Gulp et al., 1998), only portal-of-entry
11   tumors were found, including papillomas and/or carcinomas of the forestomach, esophagus,
12   tongue, and larynx. Earlier, a number of related studies were conducted to evaluate the
13   carcinogenicity of benzo[a]pyrene in feed in Ah-responsive white Swiss mice (Rigdon and Neal,
14   1969, 1966; Neal and Rigdon, 1967). These studies were not conducted using standard, modern
15   toxicological methods and have several limitations, including inconsistent dosing protocols;
16   varying ages  of the animals; use of benzene as a solvent; small numbers of animals; and
17   evaluation of only a limited number of tissues. These studies do, however, provide useful dose-
18   response information on benzo[a]pyrene carcinogenicity. Following oral administration via
19   feeding of benzo[a]pyrene, site-of-contact tumors (both papillomas and carcinomas) were
20   induced in the forestomach, esophagus, and larynx of mice (Gulp et al., 1998; Neal and Rigdon,
21   1967) and rats (Brune et al., 1981). The results following inhalation,  dermal, or oral exposure
22   are further supported by numerous mechanistic studies or assays using infant mice, susceptible
23   transgenic strains, or Ah-receptor knockout mice.
24          Benzo[a]pyrene is a complete carcinogen and likely acts by initiating tumors through
25   direct DNA damage as well as by promoting tumor growth. Benzo[a]pyrene has been shown to
26   be mutagenic in multiple assay systems.  Several modes of carcinogenic action are possible.
27   These include:
28
29          (1) Alteration of pathways regulating cell proliferation and survival (Tannheimer et al.,
30              1998);
31
32          (2) Inhibition of intracellular communication (Sharovskaia et al., 2003; Blaha et al.,
33             2002; Rummel et al., 1999);
34
35          (3) Altered intracellular Ca2+ signaling (Tannheimer et al., 1998);
36
37          (4) Modulation of cell survival, cell proliferation, and altered growth via generation of
38             oxidative stress and activation of oxidant stress signaling (Burdick et al., 2003; Miller
39             and Ramos, 2001);
40
41          (5) Altered apoptosis processes (Chen et al., 2003);

                                             22           DRAFT - DO NOT CITE OR QUOTE

-------
 1
 2          (6) Dysregulation of normal circulating hormone levels or activity affecting
 3             tumorigenesis in reproductive tissues (Safe and Wormke, 2003; Archibong et al.,
 4             2002) or the central nervous system (Dasgupta and Lahiri, 1992);
 5
 6          (7) Disruption of cell cycle kinetics in breast cancer cells (Jeffy et al., 2002, 2000); and
 7
 8          (8) Disruption of DNA repair through alteration of ribonucleic acid (RNA) polymerase
 9             activity (Shah and Bhattacharya, 1989).
10
11          Oral (dietary) carcinogenicity bioassays are available that compare MGP residue
12   (Weyand et al., 1995) or coal tar preparations (Culp et al., 1998; Gaylor et al., 1998) with
13   benzo[a]pyrene. In both cases, there were significant differences in the target organ distribution
14   of tumors between benzo[a]pyrene and complex mixtures of PAHs.  Following dietary
15   administration, benzo[a]pyrene-induced tumors were observed primarily at the point of contact
16   (i.e., the forestomach), while MGP residue and coal tar produced tumors in the lung, liver,
17   forestomach, skin, and other organs.  Tissue-specific differences in metabolic activation and
18   DNA binding of PAHs may contribute to the observed differences in target organ sensitivity
19   (Weyand and Wu, 1995; Culp and Beland, 1994). However, a dietary study in A/J mice
20   (Weyand et al., 2004) showed that benzo[a]pyrene could induce significant increases in the
21   incidences of lung adenomas and forestomach carcinomas.  Further,  a gavage study in rats
22   (Kroese et al.,  2001) demonstrated that oral exposure to benzo[a]pyrene could induce tumors in
23   the liver and auditory canal; no lung tumors were observed. The latter two studies indicate that,
24   contrary to the conclusions of earlier studies, benzo[a]pyrene can induce tumors at distal sites.
25          In summary, benzo[a]pyrene is the most appropriate compound to use as an index
26   chemical for carcinogenic PAHs. It is well-studied, with a robust database of both bioassay data
27   and mode of action information.  Benzo[a]pyrene is a complete carcinogen with both initiating
28   and promoting properties, is among the most potent PAH carcinogens, and is prevalent in many
29   complex environmental mixtures. No alternative index chemical was identified from the list of
30   target PAHs.
31
32   2.4.  SIMILARITIES IN MODE OF CARCINOGENIC ACTION FOR PAHs
33          Toxicological similarity of chemicals is the basis for the assumption of dose additivity
34   that underlies the RPF approach  (U.S. EPA, 1990).  The carcinogenic mode of action for PAHs
35   has been extensively reviewed (Ramesh, 2004; CCME, 2003; Bostrom et al.,  2002; Larsen and
36   Larsen, 1998; WHO,  1998; Muller et al., 1997; Sjogren et al., 1996;  ATSDR, 1995; Malcolm
37   and Dobson, 1994; U.S. EPA, 1990).  Key events that have been associated with PAH
38   carcinogenicity include:
39
                                            23           DRAFT - DO NOT CITE OR QUOTE

-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
   Oxidative metabolism to reactive intermediates that covalently bind to DNA, RNA,
   and proteins (benzo[a]pyrene metabolism is illustrated in Figure 2-2);

   Formation of DNA adducts;

   Tumor initiation due to mutations in cancer-related genes (e.g., tumor suppressor
   genes or oncogenes); and

   Tumor promotion related to cytotoxicity and formation of reactive oxygen species,
   and/or Ah receptor (AhR) affinity and upregulation of genes related to
   biotransformation, growth, and differentiation.
                                                     BaP 1,6-hydroquinont     BaP 1,6-semtquinone     BaP 1.6-quinone
Reprinted from Impact of cellular metabolism on the biological effects of benzo[a]pyrene
and related hydrocarbons, 2001 by Miller, KP; Ramos, KS; with permission of Taylor &
Francis.

Source: Miller and Ramos (2001).

Figure 2-2. Metabolic pathways for benzo[a]pyrene.
                                             24
                                              DRAFT - DO NOT CITE OR QUOTE

-------
 1           Formation of reactive intermediates andDNA adducts. Each of the key events identified
 2    above is affected by the chemical structure of the individual PAH. At least three distinct
 3    molecular mechanisms have been proposed to explain the tumor initiation process of PAHs (Xu
 4    et al., 2009; Jiang et al., 2007, 2005; Xue and Warshawsky, 2005; Bolton et al., 2000; Penning et
 5    al.,  1999; Harvey, 1996; Cavalieri and Rogan, 1995).  These modes of action include the
 6    formation of diol epoxides, radical cations, and o-quinones (Figure 2-3).  Diol epoxide formation
 7    leads to stable and unstable DNA adducts, mainly at guanine and adenine, which can lead to
 8    mutations in proto-oncogenes and tumor-suppressor genes. Radical  cation formation may lead to
 9    the  generation of unstable adducts at guanine and adenine, leading to apurinic sites and mutation
10    in HRAS.  o-Quinone formation could lead to stable and unstable DNA adducts and generation of
11    reactive oxygen species, inducing mutations in RP53. The evidence supporting the role of these
12    reactive metabolites in tumor initiation includes a characterization of the  specific DNA adducts
13    arising from PAH metabolism and observations of mutagenesis resulting from direct exposure.
14    Figure 2-3 illustrates the proposed key steps in the mode of action for PAH carcinogenesis.
15    These include the interaction of reactive metabolites with DNA to form adducts, induction of
16    depurination, transversion mutations (e.g., GC^-TA or AT^-TA), and oxidative damage to
17    DNA, and tumor promotion mediated by AhR-mediated effects on gene regulation.
18
                                            25           DRAFT - DO NOT CITE OR QUOTE

-------
                              Key Events in the Mode of Action for PAH Carcinogenicity
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
                                                                                Progression
                                                                   Upregulation of
                                                                    les related to
                                                                  biotransformation,
                                                                    growth, and
                                                                   differentiation
       Figure 2-3. Overview of the proposed key events in the mode of action for
       PAH carcinogenicity.
       The formation of diol epoxides is a proposed key step in the most established mode of
action for PAH-induced carcinogenicity. Extensive studies of the metabolism of carcinogenic
PAHs suggest that bay- and fjord-region diol epoxides are some of the ultimate reactive
metabolites of PAHs (Jerina et al., 1978; Jerina and Lehr, 1977). These metabolites are
generally formed through cytochrome P450 (CYP) oxidation to form epoxides and epoxide
hydrolase cleavage resulting in diol formation.  CYP1 Al  appears to be the primary isozyme
involved in diol epoxide formation; however, other isozymes may also contribute to PAH
metabolism (i.e., CYPIA2, CYP1B1, CYP3A4) (Bostrom et al., 2002; ATSDR, 1995). Non-
alternant PAHs, composed of fused benzenoid and five-membered rings, may be metabolized
through other pathways resulting in the formation of reactive intermediates that bind to DNA.
Classic bay- and fjord-region diol epoxides may be formed from these compounds; however,
epoxide formation at cyclopenta-ring structures has also been demonstrated to result in DNA
adduct formation (Bostrom et al., 2002).
       Many studies have been performed to evaluate the formation of DNA adducts following
in vivo or in vitro exposure to PAHs. Diol epoxide metabolites interact preferentially with the
exocyclic amino groups of deoxyguanine and deoxyadenine (Geacintov et al., 1997; Jerina et al.,
                                            26
                                                    DRAFT - DO NOT CITE OR QUOTE

-------
 1    1991). Adducts may give rise to mutations, unless these adducts are removed by DNA repair
 2    processes prior to replication.  The stereochemical nature of the diol epoxide metabolite (i.e.,
 3    anti- versus syn-diol epoxides) affects the number and type of adducts and mutation that occurs.
 4    Figure 2-4 presents the structures of four stereoisomeric adducts arising from the interaction of
 5    benzo[a]pyrene diol epoxide metabolites with the deoxyguanosine (dG) residues in DNA
 6    (Geacintov et al., 1997).  Transversion mutations (e.g., GC—>TA or AT—>TA) are the most
 7    common type of mutation found in mammalian cells following diol epoxide exposure (Bostrom
 8    et al., 2002).
              R    HO,,,
                   HO
                 10S (+)-trans-anti-[BaP]-N2-dG
                                                 R
                                                     HO
                                                            OH
                                                       (-)-trans-anti-[BaP]-N2-dG
10
11
12
13
14
15
16
17
18
19
              R
                   HO
                          OH
                 10R (+)-c/s-an//-[BaP]-A/2-dG
       Source:  Geacintov et al. (1997).
                                                                 OH
                                                   10S (-)-c/s-a/7f/-[BaP]-AP-dG
       Figure 2-4. Structures of the four stereoisomeric adduct moieties,
       anti-\Ba¥]-N2-AG, derived from the trans- or cis- covalent binding of
       (+)-a«ft'-BaP diol epoxide or (-)-a«ft'-BaP diol epoxide to dG residues in DNA.

       Radical cation formation involves a one-electron oxidation that produces electrophilic
radical cation intermediates (Cavalieri and Rogan, 1995, 1992). Oxidation of this type can occur
by CYP or peroxidase enzymes (i.e., horseradish peroxidase, prostaglandin H synthetase).
Radical cations can be further metabolized to phenols and quinones (Cavalieri et al., 1988a), or
                                            27
                                                    DRAFT - DO NOT CITE OR QUOTE

-------
 1
 2
 o
 3
 4
 5
 6
 7
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
they can form unstable adducts with DNA that ultimately result in depurination (Cavalieri et al.,
2005, 1993; Rogan et al., 1993). Radical cations have been shown to play a major role in
formation of DNA adducts for several carcinogenic PAHs (e.g., 7,12-dimethylbenzanthracene,
benzo[a]pyrene, dibenzo[a,l]pyrene).  The predominant depurinating adducts occur at the
N-3 and N-7 positions of adenine and the C-8 and N-7 positions of guanine (Cavalieri and
Rogan,  1995; Li et al., 1995).  Figure 2-5 illustrates three depurinating adducts of
benzo[a]pyrene formed by one-electron oxidation. Abasic sites resulting from base depurination
undergo error-prone excision repair to induce mutations. In the case of dibenzo[a,l]pyrene-
treated mouse skin, repair error from abasic sites resulted in H-ras oncogene mutations that
underwent rapid clonal expansion and regression (Chakravarti et al., 2000). H-ras mutations in
mouse skin papillomas also corresponded to adenine and guanine depurinating adducts resulting
from exposure to dibenzo[a,l]pyrene, 7,12-dimethyl-benz[a]anthracene, benzo[a]pyrene, and
benzo[a]pyrene-7,8-dihydrodiol (Chakravarti et al., 2008).
             o
              BaP-6-C8-guanine
BaP-6-N7-guanine
BaP-6-N7-adenine
       Reprinted from Central role of radical cations in metabolic activation of poly cyclic
       aromatic hydrocarbons, 1995 by Cavalieri, EL; Rogan, EG; with permission of Taylor &
       Francis.

       Source: Cavalieri and Rogan (1995).

       Figure 2-5. Depurinating adducts of benzo[a]pyrene formed by one-electron
       oxidation.

       o-Quinone metabolites of PAHs are formed by enzymatic dehydrogenation of
dihydrodiols (Bolton et al., 2000; Penning et al., 1999; Harvey, 1996; ATSDR, 1995).
Dihydrodiol dehydrogenase enzymes are members of the a-keto reductase gene superfamily.
o-Quinone metabolites are potent cytotoxins,  are weakly mutagenic,  and are capable of
producing a broad spectrum of DNA damage.  These metabolites can interact directly with DNA
and can also result in production of reactive oxygen species (i.e., hydroxyl and superoxide
radicals) that may produce further cytotoxicity and DNA damage. The DNA damage caused by
                                             28
                                                    DRAFT - DO NOT CITE OR QUOTE

-------
 1

 2
 o
 3

 4

 5

 6

 7
 9
10
11
12

13

14

15

16

17
o-quinones may include the formation of stable adducts (Balu et al., 2006), N-7 depurinating

adducts (McCoull et al., 1999), oxidative base damage (i.e., 8-oxo-2'-dG or 8-oxo-dG) (Park et
al., 2006, 2005), and strand scission (Flowers-Geary et al., 1997).  The reactive oxygen species

generated by the o-quinone of benzo[a]pyrene and other PAH o-quinones have been shown to

induce mutation in the p53 tumor suppressor gene (Park et al., 2008; Shen et al., 2006; Yu et al.,

2002). Figure 2-6 illustrates the  spectrum of DNA adducts associated with PAH o-quinones.
                        o-Quinone adducts
                                                                      ROS modifications
               stable adducts
               translesional synthesis
               G to T transversions
                               depurinating adducts
                               apurinic sites
                               G to T transversions
8'-oxo-dG
base pair mismatch
G to T transversions
                                                                                 base propenals
       Source: Bolton et al. (2000).

       Figure 2-6. Spectrum of DNA adducts anticipated with PAH o-quinones.


       The cytotoxicity of o-quinone metabolites may also contribute to tumor promotion via

inflammatory responses leading to cell proliferation (Burdick et al., 2003).

       Genotoxicity andmutagenicity. The genotoxicity and mutagenicity of PAHs have been

demonstrated in various bacterial and mammalian assays (see Section 4.3.2 below) (reviewed in

WHO, 1998; ATSDR, 1995). Mutagenesis of PAHs in the Ames assay (Salmonella
                                              29
                                                     DRAFT - DO NOT CITE OR QUOTE

-------
 1    typhimuriuni) as well as other bacterial assays requires the presence of a mammalian metabolic
 2    enzyme system. In most cases, this is supplied by postmitochondrial supernatant (S9) from the
 3    liver of rodents treated with an enzyme inducer. Mammalian cell mutagenesis in Chinese
 4    hamster V79 cells and mouse lymphoma L5178Y cells also requires metabolic activation in the
 5    form of a rodent S9 mix or co-cultivation with metabolically active rodent cells (i.e., cell-
 6    mediated assay). Several studies have noted a correlation between mutagenic potency and tumor
 7    initiation potency in the two-stage dermal carcinogenicity assay for multiple PAH compounds
 8    (LaVoie et al., 1985,  1979; Raveh et al., 1982).
 9           Tumor promotion and the AhR. The ability of certain PAHs to act as tumor promoters as
10    well as initiators may increase their carcinogenic potency (Andrews et al., 1978). The
11    promotional effects of PAHs appear to be related to AhR affinity and the upregulation of genes
12    related to growth and differentiation (Bostrom et al., 2002). Figure 2-7 illustrates the function of
13    the AhR and depicts the genes regulated by this receptor as belonging to two major functional
14    groups (i.e., induction of metabolism or regulation cell differentiation and proliferation). PAHs
15    bind to the cytosolic AhR in complex  with heat shock protein 90.  The ligand-bound receptor is
16    then transported to nucleus in complex with the AhR nuclear translocator protein.  The AhR
17    complex interacts with AhR elements  of DNA to increase the transcription of proteins associated
18    with induction of metabolism and regulation of cell differentiation and proliferation.
19
                                            30           DRAFT - DO NOT CITE OR QUOTE

-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
         *
         PAH
                          -»tAHR|Hsp90
                                 [Hsp90
AHR

Hsp90
Hsp90
Hsp90
Hsp90

ARNT

                                              Enhanced
                                              specific
                                              mRNA
                                              production




m
St..
|AHR
ARNT
AHREDNA


                                      Increased
                                      synthesis of
                                      PAH metabolizing
                                      enzymes
                                                    Increased
                                                    synthesis of
                                                    proteins that
                                                    regulate cell
                                                    differentiation and
                                                    proliferation
       Reprinted from Molecular biology of the aromatic hydrocarbon (dioxin) receptor, 1994
       by Okey, AB; et al. with permission of Elsevier.
       Source:  Okey etal. (1994).
       Figure 2-7. Interaction of PAHs with the AhR - regulation of genes related
       to induction of metabolism and cell differentiation and proliferation.
       Tumor promotion and cytotoxicity.  PAHs are metabolized to o-quinones, which are
cytotoxic and can generate reactive oxygen species (Bolton et al., 2000; Penning, 1999).  PAH
o-quinones reduce the viability and survival of rat and human hepatoma cells (Flowers-Geary et
al., 1996, 1993). Inflammatory responses to cytotoxicity may contribute to the tumor promotion
process.  For example, benzo[a]pyrene quinones (1,6-, 3,6-, and 6,12-benzo[a]pyrene-quinone)
generated reactive oxygen species and increased cell proliferation by enhancing the epidermal
growth factor receptor pathway in cultured breast epithelial cells (Burdick et al., 2003). Dermal
exposure of mice to dibenzo[a,l]pyrene and dimethyl-benz[a]anthracene resulted in an
inflammatory response that was correlated with epidermal hyperplasia and skin tumor promotion
(Casale et al., 2000, 1997). The extent of epidermal hyperplasia was correlated with the cytokine
mRNA response in lymph nodes and skin of treated mice (Casale et al., 2000).
       Genetic  targets and tumor formation. DNA adducts and oncogenes/tumor suppressor
gene mutations  have been demonstrated in tumor tissue from humans and laboratory animals.
                                            31
                                                    DRAFT - DO NOT CITE OR QUOTE

-------
 1   DeMarini et al. (2001) demonstrated mutations in the p53 tumor suppressor gene and the K-ras
 2   oncogene in the lung tumors of nonsmokers, whose tumors were associated with exposure to
 3   smoky coal. Lung tumors were obtained from 24 nonsmoking women from China (age 30-
 4   63 years, mean age 48.5 ± 8.8 years) who used smoky coal in their homes without chimneys.
 5   Bronchioloalveolar adenocarcinoma and acinar adenocarcinoma were observed in 54 and 46% of
 6   the women studied, respectively. The observed mutations in lung tumors were primarily G—>T
 7   transversions at either K-ras or p53.  The mutation hotspots in the lung tumors that were
 8   examined corresponded with hot spots for PAH adducts (codon 154), cigarette smoke associated
 9   mutations (codon 249), and both of these events together (codon 273).  The mutation spectrum
10   was described as unique and consistent with exposure to PAHs in the absence of cigarette smoke.
11          Mutations in the K-ras, H-ras, and p53 genes were assessed in forestomach tumors
12   (n = 31) of mice fed benzo[a]pyrene in the diet (0, 5, 25, or 100 ppm) for 2 years (Gulp et al.,
13   2000). Sixty-eight percent of 31 forestomach tumors analyzed had K-ras mutations, which were
14   G—>T or C transversions in codon 12 or 13.  H-ras (codon 13) and p53 mutations characterized
15   as G—>T or C transversions were also each found in 10% of forestomach tumors.
16   [32P]-postlabeling of forestomach DNA  of benzo[a]pyrene-treated mice revealed one maj or
17   adduct characterized as dG N2 BPDE. In mice exposed to benzo[a]pyrene at several
18   concentrations in the diet for 4 weeks (5, 25, and 100 ppm), there was an approximate linear
19   relationship between the daily dose of benzo[a]pyrene (in units of jig/day) and the concentration
20   of dG-N2-BPDE-DNA adducts in the forestomach (Gulp et al., 2000, 1996a).  In contrast, the
21   tumor dose-response data in mice exposed for 2 years showed a sharp increase in incidence
22   between the 5-ppm group (6% of mice had forestomach tumors) and the 25-ppm group (78% had
23   forestomach tumors) (Gulp et al., 1996a). The appearance of increased levels of BPDE-DNA
24   adducts in the target tissue at 28 days is  temporally consistent with the contribution of these
25   adducts to the initiation of forestomach tumors at 25 and 100  ppm benzo[a]pyrene in the diet.
26   However, the absence of a sharp increase in the BPDE-DNA relationship between 5 and 25 ppm
27   benzo[a]pyrene is consistent with the possible contributions of mutagenic modes of action other
28   than the diol epoxide pathway (i.e., formation of depurinated DNA adducts from the radical
29   cation or aldo-keto-reductase pathways and reactive oxygen species DNA damage from the aldo-
30   keto-reductase pathway).
31          A series of experiments designed to evaluate the mechanistic relationship between PAH
32   DNA adducts, oncogene mutations, and lung tumorigenesis were performed in the A/J mouse
33   lung model (Nesnow et al., 1998a, 1996, 1995; Mass  et al., 1993). Tumorigenic potency in the
34   lung of A/J mice varied over 2 orders of magnitude following a single intraperitoneal injection of
35   seven PAHs of varying structure (benzo[a]pyrene, benzo[b]fluoranthene, benz[j]aceanthrylene,
36   dibenz[a,h]anthracene, dibenzo[a,l]pyrene, cyclopenta[c,d]pyrene, and 5-methylchrysene).
37   When considering the non-alkylated PAHs, the number of lung adenomas per mouse was highest
                                            32          DRAFT - DO NOT CITE OR QUOTE

-------
 1   for benz[j]aceanthrylene and cyclopenta[c,d]pyrene, each of which contain a pentacyclic ring
 2   feature. The major DNA adducts identified in the mouse lung included:
 o
 4          (1) Bay region diol epoxide adducts for benzo[a]pyrene, dibenz[a,h]anthracene, and
 5             5-methylcholanthrene;
 6
 7          (2) Phenolic diol epoxide adducts for benzo[b]fluoranthene;
 8
 9          (3) Cyclopenta-ring adducts for cyclopenta[c,d]pyrene and benz[j]aceanthrylene;
10
11          (4) Bisdihydrodiol epoxide adducts for dibenz[a,h]anthracene; and
12
13          (5) Fjord-region diol epoxide adducts for dibenzo[a,l]pyrene (Nesnow et al., 1998a,
14             1996, 1995; Mass et al., 1993).
15
16          Guanine adducts were most common for all PAHs; however, adenine adducts were also
17   demonstrated for dibenzo[a,l]pyrene and benz[j]aceanthrylene.  Quantitative analysis of DNA
18   adducts by [32P]-postlabeling illustrates the importance of measuring DNA adduct levels over
19   time.  A time-integrated DNA adduct level (TIDAL) was linearly related to the dose of a
20   particular PAH. The relationship of TIDAL level to tumor formation was similar for PAHs that
21   produce different types of adducts and different mutations in the Ki-ras oncogene.  This suggests
22   that the probability of tumor formation for these PAHs may be related to the extent of overall
23   DNA damage and repair rather than the formation of a specific adduct at specific sites.
24          The DNA sequence analysis of Ki-ras mutations in lung adenomas at codons 12 and 61
25   was generally consistent with the DNA adduct data in that PAHs that produced guanine adducts
26   also produced Ki-ras guanine mutations (Nesnow et al.,  1998a, 1996, 1995; Mass et al., 1993).
27   Cyclopenta[c,d]pyrene, benz[j]aceanthrylene, and 5-methylchrysene produced large numbers of
28   adenomas per mouse (>90) and also produced a large proportion of tumors with CGT mutations
29   at Ki-ras codon 12.  Cyclopenta-ring adduct formation by cyclopenta[c,d]pyrene and
30   benz[j]aceanthrylene was correlated with the formation of GGT—>CGT mutations at Ki-ras
31   codon 12. The primary mutation type for benzo[a]pyrene, benzo[b]fluoranthene, and
32   dibenzo[a,l]pyrene was the GGT^TGT mutation, which is associated with the formation of diol
33   epoxide guanine adducts. Dibenz[a,h]anthracene did not induce mutations in Ki-ras codons 12
34   or 61; however, diol epoxide guanine adducts and lung adenomas in A/J mice were observed.
35   This suggests that a different genetic target may be  involved in carcinogenicity of this
36   compound.
37          H-ras mutations were studied in skin papillomas of SENCAR mice  resulting from dermal
38   initiation by benzo[a]pyrene or benzo[a]pyrene-7,8-dihydrodiol (400 nmol) followed by
39   12-O-tetra-decanoylphorbol-13-acetate (TPA) promotion (Chakravarti et al., 2008). Polymerase
40   chain reaction (PCR) amplification of the H-ras gene and sequencing revealed that codon 13
                                            33           DRAFT - DO NOT CITE OR QUOTE

-------
 1   (GGC to GTC) and codon 61 (CAA to CTA) mutations in papillomas corresponded to the
 2   relative levels of depurinating adducts of guanine and adenine, despite the formation of
 3   significant amounts of stable DNA adducts.
 4          Other studies also suggest that multiple genetic targets may be involved in PAH
 5   mutagenicity and carcinogenicity (Conney et al., 2001; Smith et al., 2000). Smith et al. (2000)
 6   indicated that diol epoxide adducts and mutations were observed in the p53 tumor suppressor
 7   gene following in vitro exposure of cultured human bronchial epithelial cells to metabolites of
 8   benzo[a]pyrene, chrysene, benzo[c]phenanthrene, and benzo[g]chrysene. PAH adducts and
 9   corresponding mutations preferentially formed at lung mutational hot spots (codons 154, 157,
10   158, 245, 248, and 273), suggesting that PAHs may contribute to the mutation spectrum
11   observed in human lung cancer.  Conney et al. (2001) provided evidence that dose-dependent
12   differences may exist for the mutation spectra seen in PAH-induced tumors. Skin papillomas
13   induced by benzo[a]pyrene in female mice were examined for mutations in the c-Ha-ras proto-
14   oncogene. The major difference between high- and low-dose groups was mutations at exon 2 of
15   the c-Ha-ras gene, with the proportion of AT base pair mutations higher in the low-dose group.
16   Dose-dependent changes in the mutation profile were also evident in Chinese hamster V79 cells
17   exposed to the diol epoxides of benzo[a]pyrene and benzo[c]phenanthrene (i.e., the proportion of
18   AT mutations decreased with increasing concentration).
19          In conclusion, there is evidence that an assumption of a similar toxicological action is
20   reasonable for PAHs; however, the carcinogenic process for individual PAHs is likely to be
21   related to some unique combination of multiple molecular events resulting from formation of
22   several reactive species. For these reasons, the use of an RPF approach to estimate cancer risk
23   associated with PAH exposure is considered appropriate.  A common mutagenic mode of action
24   for carcinogenic PAHs is hypothesized based on information available for the indicator
25   chemical, benzo[a]pyrene (U.S. EPA, 2005b). The uncertainties and limitations related to the
26   mode of action assumption for PAH-induced cancer are further discussed in Section 8.5.
27
28   2.5. STRUCTURAL ALERTS FOR PAH CARCINOGENESIS
29          The carcinogenic activity of PAH compounds is influenced by specific structural
30   features. For example, alternant PAHs having four or more benzene rings exhibit greater
31   carcinogenic potency than PAHs with two or three benzene rings (Bostrom et al., 2002). The
32   carcinogenic activity  of PAHs is also related to the specific arrangement of the benzene rings.
33   As described in Section 2.4, PAHs that form bay- and fjord-region diol or dihydrodiol epoxides
34   are more potent carcinogens compared with linear PAHs that lack this structural feature
35   (Bostrum et  al., 2002).  These metabolites are resistant to detoxification due to stereochemical
36   effects and, consequently, are more likely to be mutagenic and cause cancer (Buterin et al., 2000;
37   Chang et al., 1981; Buening et al., 1979; MacLeod et al.,  1979; Flesher et al., 1976).
38   Dihydrodiol epoxides formed at other positions on the PAH molecule (i.e., not the bay- or fjord-

                                            34           DRAFT - DO NOT CITE OR QUOTE

-------
 1   regions) are more accessible to glutathione transferase detoxification and are less potent
 2   mutagens and carcinogens (MacLeod et al., 1979; Flesher et al., 1976). Nonalternant PAHs
 3   containing fused benzenoid and five-membered rings, can also be metabolized to bay- and fjord-
 4   region diol epoxides (Bostrum et al., 2002); however, epoxide formation at the cyclopenta- ring
 5   structure may also contribute to carcinogenicity (Bostrum et al., 2002; Nyholm et al., 1996).
 6          PAHs with at least four rings and a classic bay- or fjord-region (formed entirely by
 7   benzene rings; see Figure 2-1) may be characterized as containing structural alerts for
 8   carcinogenesis. However, this structural characterization is likely to be overly simplistic and
 9   other features may be important to carcinogenesis. Recent studies have applied quantitative
10   structure activity relationship  (QSAR) methods to evaluate the relationship between specific
11   PAH structural features and mechanistic events related to carcinogenesis (Bruce et al., 2008;
12   Vijayalakshmi et al., 2008).
13
14   2.6.  SIMILARITIES IN RELATIVE POTENCY ACROSS ENDPOINTS
15          Studies that have evaluated the association between cancer-related endpoints and
16   tumorigencity of PAHs are briefly summarized below.
17          Several studies have been performed that compare the bacterial or mammalian cell
18   mutagenicity of various PAHs with tumor initiating activity or complete carcinogenesis
19   (Blackburn et al., 1996; LaVoie et al., 1985, 1981, 1979; Raveh et al., 1982; Andrews et al.,
20   1978). In general, mutagenicity appears to correlate best with tumor initiation. Complete
21   carcinogenicity is not well-predicted by positive findings in short-term mutagenicity assays.
22   Andrews et al. (1978) tested 24 PAHs for bacterial mutagenicity in the Ames test and compared
23   these findings to evidence of carcinogenicity (parent and metabolites) from previously published
24   studies. Positive findings of both mutagenicity and carcinogenicity were only reported for 14 of
25   the 24 PAHs evaluated. Eight of the 10 remaining PAHs were found to be mutagenic in the
26   Ames assay, but were not carcinogenic in animal studies. LaVoie et al. (1979) compared the
27   mutagenicity, tumor-initiating activity, and complete carcinogenicity of several series of
28   structurally related PAHs.  Tumor-initiating activity was found to correspond with complete
29   carcinogenicity.  Quantitation of mutagenicity in the Ames assay for structurally related PAHs
30   failed to provide a reliable indication of tumor-initiating activity or complete carcinogenicity. In
31   addition, mutagenicity results  could not be used to predict which PAHs would be
32   noncarcinogenic.  Many PAHs were active mutagens, but were not shown to be carcinogenic.
33   Studies using methylated derivatives of anthracene demonstrated a correlation between
34   mutagenicity of specific metabolites and tumor initiating activity in mouse skin (LaVoie et al.,
35   1985). Raveh et al.  (1982) reported that the mutagenic response to PAHs in Chinese hamster
36   V79 cells was similar to the skin tumor initiating activity observed in SENCAR mice.
37   Benzo[a]pyrene was demonstrated to be a more potent mutagen and skin tumor initiator than
38   cyclopenta[c,d]pyrene.

                                             3 5          DRAFT - DO NOT CITE OR QUOTE

-------
 1          Blackburn et al. (1996) compared the predictive power of a mutagenicity test (the
 2   Modified Ames Test, which uses enhanced extraction techniques and greater levels of S9 to
 3   improve performance when oils are tested) and DNA adduct formation (measured by
 4   P32-postlabelling) to predict the dermal carcinogenicity of 120 PAH-containing oils.  The
 5   Modified Ames Test provided greater accuracy in predicting carcinogenicity (96%). In addition,
 6   the mutagenicity index estimated from this test correlated strongly (r2 > 0.83) with PAH content
 7   of the oils.  The DNA adduct assay predicted carcinogenicity correctly with about 73% accuracy;
 8   however, the study authors indicated that the lower predictability may have resulted from the use
 9   of adduct data that were collected while the assay was still undergoing development.
10          Sjogren et al. (1996) performed a multivariate analysis of data for 29 PAHs to evaluate
11   the relevance of different biological assays to the carcinogenic properties of PAHs. This analysis
12   considered carcinogenicity (International Agency for Research on Cancer [IARC] weight of
13   evidence and QSAR predictions), bacterial mutagenicity, inhibition or enhancement of bacterial
14   mutagenicity, AhR affinity, and enzyme induction. Bacterial mutagenicity data were poorly
15   correlated with observed and predicted cancer data, while AhR affinity variables were
16   statistically relevant to describe these data.
17          Other studies suggest that the relationship between affinity for the AhR and carcinogenic
18   potency is unclear. For example, highly mutagenic fjord-region PAHs are potent carcinogens
19   despite exhibiting lower AhR affinity (reviewed by Bostrom et al., 2002). Likewise, some PAHs
20   that strongly activate the AhR, such as benzo[k]fluoranthene (Machala et al., 2001), are only
21   weakly carcinogenic. In addition,  some studies have demonstrated the formation of DNA
22   adducts in the liver of AhR knock-out mice following intraperitoneal or oral exposure to
23   benzo[a]pyrene (Sagredo et al., 2006; Uno et al., 2006; Kondraganti et al., 2003), indicating that
24   Ah responsiveness is not strictly required for metabolic activation and genotoxicity. These
25   findings suggest that there may be alternative (i.e., non-AhR mediated) mechanisms of
26   benzo[a]pyrene activation in the mouse liver, and that AhR affinity would not be  a good
27   predictor of carcinogenic potency.
28          AhR-mediated CYP1 Al induction by PAHs is considered to contribute to tumorigenesis
29   by increasing the production of DNA-reactive metabolites (Ayrton et al.,  1990). However,
30   CYP1 Al induction potency alone does not appear to correlate well with carcinogenic potency of
31   PAHs. Ethoxyresorufm O-deethylase (EROD) activity was evaluated as a measure of CYP1 Al
32   induction in rat hepatocytes (Bosveld et al., 2002; Till et al., 1999; Willett et al., 1997) and trout
33   liver cells (Bols et al., 1999).  Till et al. (1999) additionally measured levels of CYP1A1 protein
34   and mRNA. Machala et al. (2001) measured PAH activation of the AhR  using a chemical -
35   activated luciferase reporter gene assay. Comparable results were observed across studies, and
36   benzo[k]fluoranthene was consistently demonstrated to be the most potent inducer of CYP1 Al.
37   Chrysene, benzo[b]fluoranthene, dibenz[a,h]anthracene, and indeno[l,2,3-c,d]pyrene were also
38   demonstrated to be more potent inducers of CYP1 Al than benzo[a]pyrene.  However, most of

                                             36            DRAFT - DO NOT CITE OR QUOTE

-------
 1   these PAH compounds (except dibenz[a,h]anthracene) are considerably less potent as
 2   carcinogens in animal bioassays.
 3          Ross et al. (1995) evaluated the relationship between TIDAL values for DNA adduct
 4   formation and lung adenoma formation in A/J mice.  The TIDAL value versus tumor relationship
 5   was similar for five different PAHs, suggesting a correlation between adduct levels and tumor
 6   formation (regression analysis was not performed). As described above, the relationship of
 7   TIDAL level to tumor formation was similar for PAHs that produce different types of adducts
 8   and different mutations in the Ki-ras oncogene, suggesting that the probability of tumor
 9   formation may be related to the extent of overall DNA damage and repair (Nesnow et al., 1998a,
10   1996, 1995; Mass et al., 1993).
11          To summarize, various cancer-related endpoints have been associated with PAH
12   carcinogenicity.  Tumor initiation ability was shown to correspond well with complete
13   carcinogenicity, while some studies suggested that bacterial mutagenesis assays of individual
14   PAHs were not highly correlated with tumor formation (Sjogren et al., 1996; Lavoie et al.,  1979).
15   DNA adduct formation corresponded with lung adenoma formation in A/J mice for several
16   PAHs (Sjogren et al., 1996; Ross et al., 1995; LaVoie et al., 1979).  The development of RPFs in
17   this analysis considered both tumorigenicity and cancer-related endpoints (e.g., mutagenicity,
18   clastogenicity, morphological transformation).  Studies of AhR binding/activation were not
19   considered for use in deriving RPFs because there does not appear to be a clear relationship
20   between affinity for the AhR and carcinogenic potency of PAHs.
21
22   2.7. SIMILARITIES IN RELATIVE POTENCY  ESTIMATES ACROSS SPECIES AND
23   EXPOSURE ROUTES
24          Available studies suggest that the potency of individual PAHs is generally consistent
25   across species and study protocols.  The consistency of potency estimates based on in vivo
26   tumorigenicity studies conducted using different study protocols and exposure routes in varying
27   species/strains of test animals is summarized below.
28          Nisbet and LaGoy (1992) and Clement Associates (1988) reported that RPFs for PAHs
29   are reasonably consistent across different study protocols using varying species/strains of
30   laboratory animals. RPF estimates were calculated in multiple test systems including mouse skin
31   complete carcinogenesis studies, mouse skin tumor initiation studies,  studies in rat lung
32   (implantation), other rat studies (intrapulmonary injection, subcutaneous injection), and newborn
33   mouse studies (intraperitoneal  injection).  The RPF estimates for specific PAHs calculated from
34   different assay systems varied  by less than an order of magnitude. The relative potency of
35   individual PAHs to benzo[a]pyrene was also shown to be very similar when based on data in
36   different strains of mice using  different mouse tumor initiation models (Slaga and Fisher, 1983).
37   Muller et al. (1997) compared  the relative potency of benzo[a]pyrene and 3-methylcholanthrene
38   from data generated in three species (rat, mouse, and hamster). Similar RPF values (i.e., within a

                                            37           DRAFT - DO NOT CITE OR QUOTE

-------
 1   factor of 2) were derived for oral exposures in mice, rats, and hamsters. In their comparison
 2   across different exposure routes (oral, respiratory, and dermal), Muller et al. (1997) reported
 3   similar relative potencies for benzo[a]pyrene and 3-methylcholanthrene (within a factor of 2) for
 4   data from rats exposed via oral and respiratory routes, and for mice exposed via oral and dermal
 5   routes. The relative potency for respiratory exposure in mice was an order of magnitude lower
 6   than relative potencies for the other two exposure routes.
 7          Schneider et al. (2002) performed a more recent evaluation of the impact of exposure
 8   route on the determination of RPFs. Potency ratios were calculated for several carcinogenicity
 9   bioassays by dividing the carcinogenic potency of a PAH mixture by the carcinogenic potency of
10   benzo[a]pyrene as a single substance.  The potency ratios were observed to vary by exposure
11   route and target organ. For  example, potency ratios associated with forestomach tumors from
12   oral exposure ranged from 0.7 to 1.2 (i.e., the  potencies of the PAH mixtures and benzo[a]pyrene
13   to induce forestomach tumors were approximately equal). This suggested that these tumors may
14   be attributable to the benzo[a]pyrene content of the mixture. Potency ratios for skin tumor
15   production from dermal exposure ranged from 2 to 11, whereas RPFs calculated for lung tumors
16   from oral exposure, pulmonary implantation, or inhalation were greater than 20.  These results
17   suggested that the benzo[a]pyrene content of PAH mixtures may be only slightly responsible for
18   lung and dermal carcinogenicity. Schneider et al. (2002) suggested that RPF estimates should be
19   derived separately for oral, dermal, and inhalation exposure using studies with the relevant
20   exposure pathway.
21          To summarize, there is some consistency  within the in vivo carcinogenicity  database for
22   relative potency estimates derived from different species and strains exposed by  various routes,
23   although this is an area for which further research is needed.  However, Schneider et al. (2002)
24   have cautioned that potency ratios appear to cluster by exposure route and target organ and have
25   suggested that route-specific RPFs be  developed. There is also some concern regarding the use
26   of benzo[a]pyrene as an index chemical to estimate lung cancer from PAH mixtures, considering
27   that the lung is relatively insensitive to benzo[a]pyrene-induced tumorigenicity following oral
28   exposure (Gaylor et al., 1998). Section 8.6 provides a comparison of RPF values calculated  in
29   this report, using bioassay data from different exposure routes and study designs. RPF values
30   were comparable across most  exposure routes, with the exception of the newborn mouse
31   intraperitoneal injection studies.
32
33   2.8. DOSE ADDITIVITY OF PAHs IN COMBINED EXPOSURES
34          Use of the RPF approach assumes that doses of component chemicals that act in a similar
35   manner can be added together, after scaling the potencies relative to the index chemical, and that
36   interaction effects do not occur at low environmental exposure levels (U.S. EPA, 2000, 1986).
37   The level of confidence in the RPF approach is increased if dose additivity can be demonstrated
38   experimentally, even with simple mixtures. For PAHs, the assumption of dose additivity at low

                                            3 8          DRAFT - DO NOT CITE OR QUOTE

-------
 1    exposures cannot be confirmed or refuted based on the available experimental data.  It appears
 2    that interactions may occur at higher doses of complex PAH mixtures (see below).
 3           The complexity of potential interactions for tumorigenesis of binary mixtures of PAHs is
 4    illustrated in Table 2-2. The nature of the interaction varies with the PAHs evaluated and the
 5    study conditions (e.g., vehicle used, dose selection, study method).  Many studies were designed
 6    to evaluate the combined administration of a known carcinogen with either a weak carcinogen or
 7    a noncarcinogenic PAH. The true nature of the interaction (i.e.,  additive, synergistic, or
 8    antagonistic) can be difficult to determine in studies wherein the tumorigenic response is not
 9    measured for both PAHs given alone and in combination. These studies can distinguish between
10    an enhanced or cocarcinogenic response and an inhibitory response, but a further classification
11    cannot be made.  The interactions described as cocarcinogenic in Table 2-2 may be either
12    additive or synergistic in nature.
13
                                             39          DRAFT - DO NOT CITE OR QUOTE

-------
            Table 2-2.  Studies of binary mixtures of PAHs and tumorigenicity
Reference
Cavalieri et al., 1983
DiGiovanni et al., 1982
Falketal., 1964
Laviketal., 1942
Pfeiffer, 1973
Slagaetal., 1979
Steiner, 1955; Steiner
and Falk, 1951
Van Duuren and
Goldschmidt, 1976;
Goldschmidtetal.,
1973
Van Duuren etal.,
1973
Warshawsky etal.,
1993
Endpoint
Mouse skin
carcinogenicity
Skin tumor initiation in
mice
Sarcoma induction in
mice by subcutaneous
injection
Mouse skin tumors
Sarcoma induction in
mice by subcutaneous
injection
Skin tumor initiation in
mice
Sarcoma induction in
mice by subcutaneous
injection
Mouse skin
carcinogenicity
Mouse skin
carcinogenicity
Mouse skin
carcinogenicity
Findings
BaP and CPcdP given together resulted in a
synergistic effect at low and intermediate doses;
three- to sevenfold increase in relative risk at
intermediate dose of both BaP and CPcdP as
compared to the sum of the relative risk for the
same dose of each PAH given alone.
BeP increased BaP tumor initiation (30%f),
inhibited tumor initiation by DMBA (84% J.) and
DBahA (48%J.) and produced no change in
combination with 3-MC; DBacA inhibited tumor
initiation by DMBA (92%|), DBahA (39%|), and
3-MC (61%J.) and produced no change in
combination with BaP.
PH inhibited tumor response of DBahA in ethyl
laurate vehicle (approximately 30%J,, estimated
from graph); tumor response was enhanced in
triethylene glycol vehicle (approximately 50%| to
100% tumor-bearing animals, estimated from
graph).
3-MC and BaP, DBahA, or BaA essentially
additive.
BaP and DBahA less than additive; tumor response
for combined treatment was within 10% of DBahA
response.
BeP, Pyr, or FA increased skin tumor initiation by
BaP (30, 35, and 23%|, respectively); BeP, Pyr, or
FA decreased skin tumor initiation by DMBA (84,
50, and 34%|, respectively).
DBahA and 3-MC in combination roughly additive;
BaA and CH in combination resulted in a
synergistic effect (9%| above additive response);
BaA and DBahA in combination resulted in
inhibition (48%J, below additive response).
BeP, BghiP, Pyr, or FA and BaP increased tumors
over BaP alone (>50% increase in incidence, also
t multiplicity); no tumors were observed for PAHs
without BaP.
BaP and BghiP had cocarcinogenic effect (23%|
over BaP response alone).
Nontumorigenic dose of BaP increased tumor
incidence produced by CH (16%|), AC (8%|), and
FA (8%t).
Net effect
S
Co, I
Co, I
A
I
Co, I
A, S, and I
S
Co
S
1
2
o
5
4
5
3-MC = 3-methylchloanthrene; A = additive; Co = cocarcinogenic (enhanced tumorigenicity, study design does not
allow for determination of A or S); DMBA = 7,12-dimethyl-benz[a]anthracene; I = inhibitory; S = synergistic

       Slooff et al. (1989) reviewed the available data addressing the carcinogenicity of
individual PAHs and in combination.  It was concluded that a generally additive effect was
observed following administration of more than two different PAHs in weight ratios similar to
those occurring in ambient air or in various emissions. Combinations of only two PAHs
                                             40
                                                      DRAFT - DO NOT CITE OR QUOTE

-------
 1   produced either additive, synergistic, or inhibitory effects. The complexity of the interaction
 2   among single PAH compounds is thought to be related to effects on metabolic enzyme systems
 3   including induction processes and competitive inhibition.  The generally additive response noted
 4   for a more complex mixture may reflect the balance between inhibitory and synergistic
 5   processes.
 6          Additivity has been observed in carcinogenicity studies of complex mixtures of PAHs.
 7   Schmahl et al. (1977) evaluated the production of skin tumors following combined dermal
 8   treatment with 11 PAHs found as constituents of automobile exhaust. Tumor findings were
 9   presented separately for two groups of PAHs. High potency carcinogens (Group 1) included
10   benzo[a]pyrene, dibenz[a,h]anthracene, benz[a]anthracene, and benzo[b]fluoranthene. Lower
11   potency PAHs (Group 2) included anthracene, benzo[e]pyrene, benzo[g,h,i]perylene, chrysene,
12   fluoranthene, phenanthrene, and pyrene. Chronic dermal exposure to PAHs in both groups
13   resulted in an additive response when compared to the tumor response for each group alone.
14          Nesnow et al. (1998b) evaluated lung tumor formation in A/J mice following combined
15   administration of five carcinogenic PAH compounds (benzo[a]pyrene, benzo[b]fluoranthene,
16   dibenz[a,h]anthracene, 5-methylchrysene, and cyclopenta[c,d]pyrene). High and low doses were
17   selected for each PAH in this study based on toxicity, survival, range of response, and predicted
18   tumor yield. The ratio of PAH doses was designed to simulate PAH ratios found in
19   environmental air and emissions samples. PAHs were administered to mice in a 25 factorial
20   study design yielding 32 dose groups (combination of five PAHs at high and low doses).  The
21   formation of lung adenomas was evaluated 8 months following intraperitoneal injection of PAH
22   mixtures. A response surface model was used to evaluate specific interactions among PAHs.
23   The results of the study indicated that greater-than-additive effects were seen at low doses, while
24   less-than-additive effects were observed at high  doses. However, the magnitude of the
25   interactions was relatively small (twofold), suggesting that potential interactions  are limited in
26   extent.
27          Dermal application of binary mixtures of PAHs has also been shown to produce additive,
28   synergistic, and inhibitory effects on DNA binding in mouse skin (Hughes and Phillips, 1993,
29   1990). Hermann (1981) demonstrated that many PAHs could both enhance and inhibit the
30   bacterial mutagenicity of benzo[a]pyrene depending on the relative concentrations in the binary
31   mixture. Binary mixtures of benzo[a]pyrene and benzo[e]pyrene produced a synergistic
32   response in the TA98 strain of S. typhimurium (which detects frameshift mutations) and
33   antagonistic and additive effects in strain TA100 (which detects a broad spectrum of mutations)
34   depending on the concentration (Hass et al., 1981). Binary mixtures of PAHs have also been
35   shown to produce antagonistic or less-than-additive effects in the Ames assay of bacterial
36   mutagenicity  (Barrai et al., 1992;  Salamone et al.,  1979a). Vaca et al. (1992) demonstrated an
37   additive effect for sister chromatid exchange induction by combined administration of
                                            41           DRAFT - DO NOT CITE OR QUOTE

-------
 1   benzo[a]pyrene and fluoranthene in human peripheral lymphocytes cocultured with
 2   polychlorinated biphenyl-induced rodent liver cells.
 3          The effects of binary PAH mixtures on gene expression, DNA adduct formation,
 4   apoptosis, and cell cycle are additive compared to the effects of the individual compounds in
 5   human hepatoma cells (HepG2) (Staal et al., 2007). Equimolar and equitoxic mixtures of
 6   benzo[a]pyrene with either dibenzo[a,l]pyrene, dibenz[a,h]anthracene, benzo[b]fluoranthene,
 7   fluoranthene, or 1-methylphenanthrene were studied.  PAH mixtures showed an additive effect
 8   on apoptosis and on cell cycle blockage. The effects of binary mixtures of PAHs on gene
 9   expression were generally additive or slightly  antagonistic.
10          Additivity has also been observed for the mutagenicity of PAHs administered as a
11   complex mixture (Bostrom et al., 1998; Kaden et al., 1979). Kaden et al. (1979) evaluated the
12   bacterial mutagenicity of the PAH fraction of kerosene soot using resistance to 8-azaguanine as a
13   genetic marker for forward mutation in S. typhimurium. Approximately half of the PAHs tested
14   (34 of 70) produced a significant increase in the mutant fraction in this assay system. The
15   mutagenicity of the complex soot mixture was demonstrated to be approximately equal to the
16   additive mutagenicity of the individual components.  Bostrom et al. (1998) reported additivity in
17   the Ames test of bacterial mutagenesis (i.e., reversion to histidine independence) for a mixture of
18   four PAHs (benzo[a]pyrene, benz[a]anthracene, fluorene, and pyrene) using four different strains
19   of S. typhimurium.
20          Mechanistic studies have suggested that the outcome of the interaction between two
21   PAHs in a binary mixture is dependent on changes in metabolism. PAHs can act as both
22   inducers and competitive inhibitors of the CYP enzymes that are responsible for generation of
23   reactive metabolites.  Benzo[e]pyrene has been shown to alter the oxidative metabolism of
24   benzo[a]pyrene, which may be related to the cocarcinogenic effect seen in skin tumor initiation
25   studies (Baird et al., 1984). Alterations in the types and amounts of benzo[a]pyrene metabolites
26   suggest that benzo[e]pyrene-induced changes may be isozyme specific (Smolarek and Baird,
27   1984). An increase in the formation of benzo[a]pyrene DNA  adducts has also been
28   demonstrated for coadministration  of benzo[e]pyrene in SENCAR mouse skin (Smolarek et al.,
29   1987). Fluoranthene and pyrene have been shown to increase the formation of benzo[a]pyrene-
30   DNA adducts in mouse skin following a combined treatment (Rice et al., 1988, 1984).
31   Enhancement of the metabolism of benzo[a]pyrene to diol epoxide metabolites and subsequent
32   DNA binding may explain the increased carcinogenic effect in this case. Phenanthrene did not
33   increase the formation of benzo[a]pyrene-DNA adducts and was not shown to be cocarcinogenic
34   following combined administration with benzo[a]pyrene in this study.  Cherng et al. (2001)
35   demonstrated that benzo[g,h,i]perylene increased the formation of benzo[a]pyrene adducts in
36   hepatoma cells (HepG2) by enhancing benzo[a]pyrene induction of CYP1A1. Benzo[g,h,i]-
37   perylene increased the nuclear accumulation of the AhR and/or the activation of the AhR to a
38   DNA-binding form (Cherng et al., 2001). Benzo[k]fluoranthene altered the metabolic profile of

                                            42           DRAFT - DO NOT CITE OR QUOTE

-------
 1    benz[a]anthracene by increasing the activity of CYP1A1 (Schmoldt et al., 1981).  The bacterial
 2    mutagenicity of benz[a]anthracene was enhanced by use of a rodent liver S9 that was obtained
 3    from animals previously exposed to other PAHs (Norpoth et al., 1984).  Coadministration of
 4    benzo[a]pyrene and benz[a]anthracene to hamster embryo cell cultures resulted in decreases in
 5    the metabolism of benzo[a]pyrene, the level of DNA binding, and the mutation frequency in
 6    hamster V79 cells (Smolarek et al., 1986).
 7           In summary, combined administration of binary mixtures of PAHs can result in several
 8    types of joint action (i.e., additive,  synergistic, or antagonistic).  The nature of the joint action
 9    appears to be dependent on the characteristics of the individual PAHs, related changes in
10    metabolism and possibly the test species/strain.  PAHs can act as both inducers and competitive
11    inhibitors of the CYP enzymes that are responsible for generation of reactive metabolites.
12    Additivity has been observed for some complex  mixtures of PAHs, suggesting a balance in the
13    relative metabolism of individual PAHs. For the purposes of this analysis, an assumption is
14    made that the combination of individual PAHs results in additive effects.  Additional research is
15    needed to characterize the validity  of this assumption.
16
                                            43           DRAFT - DO NOT CITE OR QUOTE

-------
1            3. DISCUSSION OF PREVIOUSLY PUBLISHED RPF APPROACHES
2
3
4          There are multiple analyses available for the derivation of relative potency estimates for
5   individual PAHs. All of these analyses utilize benzo[a]pyrene as the index chemical. Table 3-1
6   compares relative cancer potency values for PAHs presented by several authors. A review of the
7   derivation of these relative potency values follows.
                                          44          DRAFT - DO NOT CITE OR QUOTE

-------
Table 3-1. Comparison among various relative potency estimates for PAHs from the published literature and
regulatory agencies (1984-2004)
PAH
Acenaphthene
Acenaphthylene
Anthanthrene
Anthracene
Benzo[a]pyrene
Benz[a]anthracene
Benzo [b] fluoranthene
Benzo[c]phenanthrene
Benzo[e]pyrene
Benzo[g,h,i]perylene
Benzo [j ] fluoranthene
Benzo [k] fluoranthene
Chrysene
Coronene
Cyclopenta[c,d]
pyrene
Dibenz[a,h]
anthracene
Dibenz[a,c]anthracene
Dibenzo[a,e]pyrene
Dibenzo[a,h]pyrene
Dibenzo[a,i]pyrene
Dibenzo[a,l]pyrene
Fluoranthene
Fluorene
Indenof 1,2,3-
c,d]pyrene
Perylene
Abbr
AN
ANL
AA
AC
BaP
BaA
BbF
BcPH
BeP
BghiP
BjF
BkF
CH
CO
CPcdP
DBahA
DBacA
DBaeP
DBahP
DBaiP
DBalP
FA
FE
IP
Pery
U.S.
EPA
(1993)




1
0.1
0.1




0.01
0.001


1







0.1

Chu
and
Chen
(1984)




1
0.013
0.08




0.004
0.001


0.69







0.017

Clement
(1988)


0.32

1
0.145
0.14

0.004
0.022
0.061
0.066
0.0044

0.023
1.11







0.232

Clement
(1990)


0.316

1

0.1228

0.007
0.0212
0.0523
0.0523











0.278

Rugen et
al. (1989)




1
0.004-
0.006
0.0235



0.0763




0.599







0.00599

Slooffet
al. (1989)



0
1
0-0.04



0.01-0.03

0.03-0.09
0.05-0.89








0-0.06

0-0.08

Kroese
etal.
(2001)



0
1
0.1



0.03

<0.1
0.1-0.03








0.01

0.1

Nisbet
and
LaGoy
(1992)
0.001
0.001

0.01
1
0.1
0.1


0.01

0.1
0.01


5





0.001
0.001
0.1

Malcolm
and
Dobson
(1994)
0.001
0.001

0.01
1
0.1
0.1

0.01
0.01
0.1
0.1
0.01
0.001
0.1
1
0.1




0.001
0.001
0.1
0.001
Meek
etal.
(1994)




1

0.06



0.05
0.04











0.12

Muller
etal.
(1997)


0.28

1
0.014
0.11
0.023
0
0.012
0.045
0.037
0.026

0.012
0.89


1.2
1.1



0.067

Larsen and
Larsen
(1998)


0.3
0.0005
1
0.005
0.1
0.023
0.002
0.02
0.05
0.05
0.03

0.02
1.1

0.2
1
0.1
1
0.05

0.1

Collins
etal.
(1998)




1
0.1
0.1



0.1
0.1
0.01




1
10
10
10


0.1

Cali-
fornia
EPA
(2004)






0.62



0.52

0.17





11
12





                                         45
DRAFT - DO NOT CITE OR QUOTE

-------
      Table 3-1. Comparison among various relative potency estimates for PAHs from the published literature and
      regulatory agencies (1984-2004)



PAH
Phenanthrene
Pyrene



Abbr
PH
Pyr

U.S.
EPA
(1993)


Chu
and
Chen
(1984)




Clement
(1988)

0.081


Clement
(1990)




Rugen et
al. (1989)




Slooffet
al. (1989)
0.01


Kroese
etal.
(2001)
<0.01

Nisbet
and
LaGoy
(1992)
0.001
0.001
Malcolm
and
Dobson
(1994)
0.001
0.001

Meek
etal.
(1994)



Muller
etal.
(1997)
0.00064
0

Larsen and
Larsen
(1998)
0.0005
0.001

Collins
etal.
(1998)


Cali-
fornia
EPA
(2004)


Abbr = abbreviation
                                                 46
DRAFT - DO NOT CITE OR QUOTE

-------
 1          U.S. EPA (1993) presented RPFs (termed EOPPs) for seven PAHs (benzo[a]pyrene,
 2   benz[a]anthracene, benzo[b]fluoranthene, benzo[k]fluoranthene, chrysene,
 3   dibenz[a,h]anthracene, indeno[l,2,3-c,d]pyrene) as Provisional Guidance for the risk evaluation
 4   of PAHs.  On the IRIS database, the current entries for all seven of these compounds contain a
 5   cancer weight of evidence classification of Group B2 (probable human carcinogen, based on
 6   sufficient evidence of carcinogenicity in animals) (www.epa.gov/iris).  U.S. EPA (1993)
 7   indicated that the data for PAHs did not meet the criteria for the development of toxicity
 8   equivalency factors (TEFs). In particular, the existing database was limited primarily to studies
 9   of metabolism, genotoxicity, and cancer, and the assumptions of the dose-additivity model (i.e.,
10   toxicological  similarity and no interactions at low concentrations) were not proven or refuted.
11   The EOPP terminology was used because this approach was limited to skin painting data and
12   was based on benzo[a]pyrene exposure from a single (oral) pathway (for the derivation of the
13   slope factor).  This analysis considered only a small subset of PAHs routinely measured in PAH
14   mixtures at hazardous waste sites.  The EOPP values were based on previous evaluations
15   conducted by Chu and Chen (1984) and Clement Associates (1988) and were calculated for
16   various test systems (i.e., mouse skin carcinogenesis, subcutaneous injection in mice,
17   intrapulmonary administration to rats, tumor initiation on mouse skin, and intraperitoneal
18   injection in newborn mice) (Clement Associates, 1988). Various statistical methods for
19   combining data sets were considered; however, final EOPP values were based  on a single test
20   system (skin painting) and were rounded to the closest order of magnitude. The EOPPs were
21   recommended for the oral exposure route only, because the quantitative dose-response
22   assessment for benzo[a]pyrene was from an oral carcinogenicity bioassay (i.e., an oral cancer
23   slope factor).  This recommendation was, however, complicated by the fact that the EOPPs were
24   derived from comparisons based on dermal exposure.
25          Chu and Chen (1984) presented RPF values for the seven PAH compounds described in
26   the Provisional Guidance described above (U.S. EPA, 1993) (see Table 3-1). These values were
27   calculated using mouse skin painting data only.  Tumor incidence data were modeled using the
28   linearized multistage  model and the resulting EDio and ql* (upper confidence limit of the linear
29   slope) were presented for target PAHs and benzo[a]pyrene. The RPFs listed in Table 3-1
30   represent the ratio of the ql * value for a PAH compound to the ql * value for benzo[a]pyrene
31   (i.e., ql*PAH-ql*Bap).
32          Clement Associates (1988) identified 11 published studies that concurrently compared
33   the carcinogenicity of benzo[a]pyrene with one or more other PAHs, and used  the data to derive
34   relative cancer potencies for 13 PAHs, including benzo[a]pyrene.  Test protocols used in this
35   analysis included mouse skin complete carcinogenesis, initiation-promotion on mouse skin,
36   subcutaneous injection into mice, lung implantation in rats, and intraperitoneal injection into
37   newborn mice. Tumor incidence data were fit to a simplified version of the Moolgavkar-
38   Venson-Knudsen (MVK) two-stage model and to the linearized multistage model to obtain low-

                                            47           DRAFT - DO NOT  CITE OR QUOTE

-------
 1   dose cancer potency values (transition rates and low-dose slope factors, respectively). Most of
 2   the estimates were derived using data for multiple exposure levels and controls, but some were
 3   based on a single exposure level and a control. RPFs were calculated as the ratio of the
 4   estimated transition rate or slope factor for a particular PAH to the corresponding values for
 5   benzo[a]pyrene from the same study. Clement Associates (1988) selected representative RPFs
 6   for each of the studied PAHs based on evaluations of the quality of the studies from which the
 7   estimates were obtained.
 8          Clement Associates (1990) also derived relative cancer potencies for eight PAHs based
 9   on tumor incidence data from rat lung implantation data only (Deutsch-Wenzel,  1983). The data
10   were restricted to a single group of studies using a defined experimental  protocol in order to
11   address issues of questionable data quality associated with other studies. Data quality concerns
12   cited for  other studies include variation in survival, saturation of the carcinogenic effect,
13   outmoded pathological classification, and inadequate controls.  The RPF values based on rat lung
14   implantation data were comparable to those originally derived by Clement Associates (1988)
15   (see Table 3-1).
16          Rugen et al. (1989) proposed a relative potency  approach to establish acceptable
17   exposure levels (AELs) for six carcinogenic PAHs in drinking water (listed in Table 3-1).  These
18   authors reviewed mouse skin painting studies in which the cancer potency of benzo[a]pyrene
19   was compared with those of other PAHs (Bingham and Falk, 1969; Wynder and Hoffmann,
20   1961, 1959a, b). The following relationship was used to calculate conversion factors to derive
21   AELs for these PAHs from the AEL for benzo[a]pyrene:  relative tumor dose (RTD) =
22   (di/ni)/(d2/n2); where di and ni represented a dosage level and associated tumor incidence after a
23   given exposure  duration to a certain PAH, PAHi, and d2 and n2  represented similar quantities  for
24   exposure to the index PAH, benzo[a]pyrene, for the same exposure duration. The AEL for a
25   particular PAH was then derived with the following relationship: AEL(PAHi) = AEL(benzo[a]Pyrene) x
26   RTD(pAHi)- In this approach, RTDs for PAHs more potent than benzo[a]pyrene were less
27   than 1 and RTDs for PAHs less potent than benzo[a]pyrene were greater than 1.  The reciprocal
28   of the RTDs derived by Rugen et al. (1989) were comparable to the RPFs presented by other
29   authors and are presented as such in Table 3-1.
30          The Netherlands (RIVM) proposed RPF values for 10 PAHs (naphthalene, anthracene,
31   phenanthrene, fluoranthene, chrysene, benz[a]anthracene, benzo[k]fluoranthene, benzo[a]pyrene,
32   benzo[g,h,i]perylene, and indeno[l,2,3-c,d]pyrene) (Slooff et al., 1989).  RPFs were calculated
33   as a ratio of ED50 values that were  calculated using a simple linear model. For dermal studies in
34   which the latency period was determined, the tumor incidence was divided by latency and
35   concentration, and the values were averaged for the different concentrations. Kroese et al.
36   (2001) provided an update of the RPF values calculated by Slooff et al. (1989) by incorporating
37   more recent evaluations conducted by other authors (Larsen and Larsen, 1998; Nesnow et al.,
                                            48          DRAFT - DO NOT CITE OR QUOTE

-------
 1   1998b; Muller, 1997; Nisbet and LaGoy, 1992). The RPF values for chrysene and fluoranthene
 2   were decreased, while other values remained similar to those originally proposed (see Table 3-1).
 3          Nisbet and LaGoy (1992) proposed toxicity equivalence factors for 17 PAHs commonly
 4   found at hazardous waste sites. These authors reviewed published studies in which the
 5   tumorigenic potencies of one or more PAHs were compared with benzo[a]pyrene (essentially the
 6   same as those reviewed by Clement Associates, 1988) and rounded, to an order of magnitude, the
 7   estimates presented by Clement Associates (1988) for seven carcinogenic PAHs (dibenz[a,h]-
 8   anthracene, benz[a]anthracene, benzo[b]fluoranthene, benzo[k]fluoranthene, indeno[l,2,3-c,d]-
 9   pyrene, benzo[g,h,i]perylene, and chrysene) (see Table 3-1). Nisbet and LaGoy (1992) argued
10   that the rounded estimates more accurately reflected the uncertainty in the estimates than the
11   values presented by Clement Associates (1988). Nisbet and LaGoy (1992) stated that Clement
12   Associates (1988) proposed a TEF of 0.32 for anthracene (CASRN 120-12-7), but examination
13   of the original report  shows that Clement Associates (1988) proposed this value for anthanthrene
14   (CASRN 191-26-4) and did not propose a value for anthracene. Nisbet and LaGoy (1992)
15   assigned a value of 0.01 to anthracene.  In addition, Nisbet and LaGoy (1992) arbitrarily
16   assigned TEFs of 0.001 to eight other PAHs for which adequate evidence of carcinogenicity in
17   animals was not available (acenaphthene, acenaphthylene, fluoranthene, fluorene,  2-methyl-
18   naphthalene, naphthalene, phenanthrene, and pyrene). In defense of this assignment, the
19   argument was made that some of these PAHs have been shown to have some, albeit limited,
20   evidence for carcinogenic or genotoxic activity in some studies (e.g., phenanthrene and
21   naphthalene3). The RPF value proposed for dibenz[a,h]anthracene was substantially higher than
22   that proposed by Clement Associates (1988). Nisbet and LaGoy (1992) indicate that their
23   analysis of the dose-response data suggests that an RPF value of 5 is more appropriate for
24   environmental exposures where the chemically-related tumor incidence rate would be
25   approximately <25%.
26          Malcolm and  Dobson (1994) used RPFs for 23 PAHs to calculate environmental
27   assessment levels for atmospheric PAHs (sponsored by the Great Britain  Department of the
28   Environment). The RPFs were derived from previously reported review papers (Nisbet and
29   LaGoy, 1992; Rugen et al.,  1989; Clement Associates, 1988; Chu and Chen, 1984), as well as the
30   primary literature describing pulmonary implant, skin painting, subcutaneous injection, and
31   mouse skin DNA binding studies. No information was provided regarding the methodology used
32   to derive RPFs from specific experimental studies. The proposed RPF values for individual
33   PAHs were the highest values reported in the literature.  Many of the RPF values are similar to
34   those reported by Nisbet and LaGoy (1992).  RPFs were additionally reported for
35   benzo[e]pyrene, coronene, cyclopenta[c,d]pyrene, dibenz[a,c]anthracene, and perylene.  The
36   benzo[e]pyrene and cyclopenta[c,d]pyrene RPFs were apparently calculated directly from mouse
     3It should be noted that a recent bioassay for naphthalene has shown increased incidence of nasal tumors in exposed
     rats (NTP, 2000).
                                            49           DRAFT - DO NOT CITE OR QUOTE

-------
 1   skin painting studies (Habs et al., 1980; Hoffmann and Wynder, 1966; Wynder and Hoffmann,
 2   1959a, b). Coronene and perylene were arbitrarily assigned RPF values of 0.001 given the IARC
 3   and U.S. EPA designation as "not classifiable as to human carcinogen!city" (similar approach to
 4   Nisbet and LaGoy, 1992). Dibenz[a,c]anthracene was assigned an RPF value of 0.1 based on the
 5   IARC designation of "possibly carcinogenic to humans."
 6          Health Canada (Meek et al.,  1994) proposed RPFs for five PAHs (benzo[a]pyrene,
 7   benzo[b]fluoranthene,  benzo[k]fluoranthene, benzo[j]fluoranthene, and indeno[l,2,3-cd]pyrene)
 8   based on the results of multistage modeling of incidence data in Osborne-Mendel rats treated by
 9   lung implantation (Deutsch-Wenzel et al., 1983). Values were based on a comparison of the
10   doses that caused a 5% increase in tumor incidence (ED0s). RPFs were calculated as the ratio of
11   the ED05 for benzo[a]pyrene to the ED0s for a specific PAH compound.
12          The Ontario Ministry of Environment and Energy (Muller et al., 1997) proposed RPF
13   values for 209 PAHs using data from dermal studies in mouse skin or rat lung bioassays. Most
14   of these PAHs were alkylated PAHs, PAH metabolites, or heterocyclic PAH compounds. The
15   17 unsubstituted PAHs that were evaluated in this analysis are listed in Table 3-1. Muller et al.
16   (1997) derived a standard time of observation in order to account for varying study duration
17   across experiments. Several dose-response models were considered for the evaluation of tumor
18   incidence  and multiplicity, and linear regression was selected as the preferable method.
19   Tumorigenic potency (i.e., the slope of incidence/mg) was determined separately for each data
20   set based on the following order of preference regarding study type: tumor initiation in
21   CD-I  mice, tumor initiation in SENCAR mice, rat lung implantation, and complete
22   carcinogenicity in C57BL mice.  RPFs were determined as the ratio of PAH potency to the
23   potency of benzo[a]pyrene.  RPF values derived by Muller et al. (1997) were comparable to
24   values estimated by other authors.
25          Larsen and Larsen (1998) estimated RPFs for 23 PAHs based on a compilation of
26   available carcinogenicity data in animals using oral, pulmonary, and skin application of PAHs.
27   The authors indicated that these values represent an entirely subjective estimate of relative
28   potency; however, further detail regarding the derivation of RPF estimates was not provided.
29          Collins et al. (1998) developed RPFs (termed potency equivalency factors [PEFs]) for
30   21 PAHs;  10 of these were either methyl- or nitro-substituted or heterocyclic PAHs. A hierarchy
31   of data types was utilized to provide an order of preference for data utilization in calculating
32   RPFs.  Because the analysis focused on PAHs as air contaminants, tumor data from inhalation
33   studies were preferred (although none were found), followed by intratracheal or intrapulmonary
34   instillation, oral administration, skin-painting, and subcutaneous or intraperitoneal injection.
35   Genotoxicity and structure activity data were considered the least-preferred data type for
36   calculation of RPFs. Collins et al. (1998) noted that a wide range of PEFs were observed for
37   individual chemicals using different types of data (e.g., mutagenicity versus tumor data). The
38   basis for the derivation of individual RPF values was presented in a California EPA (2002)

                                            50           DRAFT - DO NOT CITE OR QUOTE

-------
 1   technical support document.  RPF values for benz[a]anthracene, benzo[b]fluoranthene,
 2   benzo[j]fluoranthene, benzo[k]fluoranthene, indeno[l,2,3-c,d]pyrene, and chrysene were similar
 3   to those described by Clement Associates (1988).  Additional RPFs for dibenzo[a,e]pyrene,
 4   dibenzo[a,h]pyrene, dibenzo[a,i]pyrene, and dibenzo[a,l]pyrene were calculated using mouse
 5   skin and rat mammary gland data (Cavalieri et al., 1991, 1989). A cancer slope factor was
 6   directly calculated for dibenz[a,h]anthracene using the tumor incidence data from a drinking
 7   water study (Snell and Stewart, 1962).  The relative potency of dibenz[a,h]anthracene was
 8   estimated to be 0.1, when compared to the oral potency for benzo[a]pyrene.
 9          Revised California EPA RPFs were recently developed for benzo[b]fluoranthene,
10   benzo[j]fluoranthene, chrysene, dibenzo[a,h]pyrene, and dibenzo[a,i]pyrene (California EPA,
11   2004). Cancer potency estimates were derived from lung adenoma data in newborn mice treated
12   by intraperitoneal injection. Potency estimates represented the upper 95% confidence limit on
13   the linear term of the multistage model fit for the newborn mouse dose-response data. Because
14   benzo[a]pyrene was demonstrated to be  75 times more toxic in newborn mouse intraperitoneal
15   assays than in adult oral studies, oral equivalent potencies for individual PAHs were derived by
16   adjusting the  cancer potency downward  by a factor of 75.  The RPFs listed in Table 3-1 were
17   calculated as  the ratio of the oral equivalent potency for a PAH to the oral potency estimate for
18   benzo[a]pyrene. This methodology resulted in a significant increase in RPF values for
19   benzo[b]fluoranthene, benzo[j]fluoranthene, and chrysene when compared with other
20   approaches.
21          In summary, several approaches  are available  for the determination of RPFs for PAHs.
22   RPF values are proposed in at least one study for a total of 27 PAHs (see Table 3-1).  Because
23   these approaches generally rely on similar bioassay data and modeling methods, the resulting
24   RPF values are fairly comparable for most PAHs across studies.  Reports by Larsen and  Larsen
25   (1998) and Malcolm and Dobbs (1994) did not provide sufficient information on the
26   methodology used to calculate RPF  estimates and  are therefore more uncertain. Variable RPF
27   estimates were reported for benz[a]anthracene, chrysene,  and indeno[l,2,3-c,d]pyrene. RPF
28   values were also highly variable for dibenzo[a,e]pyrene, dibenzo[a,h]pyrene, dibenzo[a,i]pyrene,
29   and dibenzo[a,l]pyrene; however, these were only presented in a few recent studies. As
30   indicated above, the recent California EPA (2004) approach to estimating RPFs provides
31   considerably  higher RPF values for benzo[b]fluoranthene, benzo[j]fluoranthene, and  chrysene,
32   compared with other approaches.
33          U.S. EPA is reevaluating the RPF approach for PAHs in this analysis due to the evolution
34   of the state of the science and increased understanding of PAH toxicology. A great deal  of
35   scientific research on PAHs has been conducted since the 1993 Provisional Guidance was
36   developed. Toxicological data are available for  a larger number of PAHs  and cancer-related
37   endpoints.  However, the  database for PAHs still does not meet the criteria for the derivation of
38   TEFs.  U.S. EPA (2000) defines TEFs as special types of RPFs that are derived when there are

                                             51           DRAFT - DO NOT CITE OR QUOTE

-------
 1   abundant data supporting a specific mode of action that is pertinent to all health endpoints. RPFs
 2   may be derived when the mode of action is less certain or is known for only a subset of all health
 3   endpoints.  The major differences in the use of TEFs and RPFs is that TEFs are applied to all
 4   health endpoints, exposure routes, and exposure durations (U.S. EPA, 2000), while RPFs may be
 5   limited to specific endpoints, routes, or durations.  In the case of PAHs, there are inadequate data
 6   to identify a specific mode of action that is applicable across all health endpoints. Most of the
 7   available toxicological data are limited to cancer endpoints and there are few data on the
 8   potential mode(s) of action for other effects.  As a result, the more generalized RPF  approach is
 9   considered appropriate for PAHs.
10
11   3.1.  PREVIOUS EFFORTS TO VALIDATE THE RPF APPROACH
12          Several  studies have attempted to validate the RPF approach by comparing the cancer
13   risk of a PAH mixture measured experimentally with the cancer risk that was predicted using the
14   RPF method (Muller et al., 1997; McClure, 1996;  Goldstein et al., 1994; Clement Associates,
15   1990, 1988; Krewski et al., 1989). These studies provide semi-quantitative information on the
16   overall uncertainty in using a component-based approach. Consistent findings were not reported
17   across these studies. Some studies suggested that the RPF approach would closely predict the
18   cancer risks associated with PAH mixtures, while  others  indicated that cancer risks might be
19   over- or underestimated.
20          Clement Associates (1988) evaluated the usefulness of selected RPFs to predict the tumor
21   incidence observed in a mouse skin painting assay. Schmahl et al. (1977) exposed groups of
22   mice to multiple doses of benzo[a]pyrene alone or to one of two defined mixtures of PAHs. The
23   first of these mixtures was comprised of benzo[a]pyrene, dibenz[a,h]anthracene,
24   benz[a]anthracene, and benzo[b]fluoranthene.  The second mixture contained seven PAHs:
25   phenanthrene, anthracene, fluoranthene, pyrene, chrysene, benzo[e]pyrene, and
26   benzo[g,h,i]perylene. The predicted tumor incidences for the animals treated with the mixtures
27   were calculated from benzo[a]pyrene equivalents of the mixture and dose-response modeling of
28   the Schmahl et al. (1977) data for benzo[a]pyrene  alone.  Predicted tumor incidences for the first
29   mixture were comparable to observed tumor incidences, while predicted values were greater than
30   the observed values for the second mixture.
31          Clement Associates (1990) examined the utility of a relative potency approach, in which
32   relative cancer potency estimates of eight PAHs were used, to predict the cancer potencies of
33   each of four complex mixtures containing  many PAHs and other substances: gasoline engine
34   exhaust condensate, flue-gas condensate from coal-fired  residential furnaces, diesel  engine
35   exhaust condensate, and sidestream smoke condensate of cigarettes. Relative cancer potencies
36   (compared to benzo[a]pyrene) for each of the four complex mixtures were calculated using a
37   simplified version of the MVK two-stage model and tumor incidence data from a series of
38   published rat lung implantation studies that examined the carcinogenicity of each complex

                                            52           DRAFT - DO NOT CITE OR QUOTE

-------
 1   mixture, various subfractions of the mixtures, and benzo[a]pyrene (Grimmer et al., 1988,
 2   1987a, b, 1984). Lung implantation data (Deutsch-Wenzel, 1983) were used to calculate RPFs
 3   for benzo[b]fluoranthene, benzo[e]pyrene, benzo[j]fluoranthene, benzo[k]fluoranthene,
 4   indeno[l,2,3-c,d]pyrene, anthanthrene, benzo[g,h,i]perylene, and benzo[a]pyrene. The sum of
 5   the benzo[a]pyrene exposure equivalents for the eight PAHs (i.e., the sum of the products of the
 6   relative cancer potencies of the eight PAHs multiplied by their concentrations in the respective
 7   complex mixtures) accounted for only minor fractions of the total carcinogenicity of each of the
 8   four complex mixtures. When the assumption was made that each of the eight PAHs was as
 9   potent as benzo[a]pyrene, the sum of the benzo[a]pyrene equivalents still accounted for only
10   minor fractions of the carcinogenicity  of each mixture. Clement Associates (1990) concluded
11   that the cancer risk associated with a complex PAH mixture could not be estimated reliably from
12   measurements of a few indicator components, and further speculated that the underestimation
13   occurred because complex mixtures that occur in the environment contain many PAHs that have
14   not been studied in cancer tests, but may be  carcinogenic. In addition, complex PAH mixtures
15   found in the environment contain other potential carcinogens including substituted and
16   heterocyclic PAHs and non-PAH components.
17          Krewski et al. (1989) compared the observed tumor response rate for two PAH mixtures
18   in mice with the tumor response predicted using the RPFs for  13 individual PAHs; chemical
19   characterization of the mixture was not provided. With the exception of the highest dose, the
20   predicted tumor response for mixture 1 was  similar to the observed response.  For mixture 2, the
21   predicted tumor response value was higher than the observed response.
22          Goldstein et al. (1994) compared the experimental carcinogenicity of a MGP residue to
23   the predicted cancer risk using the Nisbet and LaGoy (1992) RPF scheme. The RPF method
24   underestimated the carcinogenicity of the mixture. The lack of correspondence was suggested to
25   be related to the presence of unidentified carcinogens in the mixture or possible synergistic
26   interactions between PAHs.
27          McClure et al. (1996) compared the tumor response predicted using U.S. EPA's 1993
28   provisional values (i.e., EOPPs) to the observed response reported in studies of mice exposed to
29   synthetic and complex mixtures of PAHs. The results of this analysis were mixed. EOPP values
30   closely predicted the mouse tumor response to subcutaneous or dermal application of synthetic
31   mixtures containing relatively potent carcinogens, while overestimating the response to synthetic
32   mixtures containing only relatively weak carcinogens (similar to findings of Clement Associates,
33   1988). Mouse skin tumor initiation with several coal liquids was closely predicted by the EOPP
34   approach; however, this method underestimated the tumor response from lung implantation of
35   coal furnace emission condensate and  its PAH-containing neutral fraction.
36          The validation analyses that were performed by Muller et al. (1997) consisted of
37   component versus whole mixture risk comparisons using data for smoky coal and coke oven
38   emissions.  The human lung cancer risks that were estimated using the RPF approach were

                                            53           DRAFT - DO NOT CITE OR  QUOTE

-------
1   compared to the whole mixture cancer risk derived from epidemiology studies. The relative
2   content of PAHs (compared to benzo[a]pyrene) in the mixture was determined analytically (for
3   smoky coal and coke oven emissions) or was estimated as a standard mixture assumed to
4   represent an average PAH profile. The RPF method produced PAH cancer risk estimates that
5   were significantly lower than the risk estimates derived from epidemiology studies.
6
                                          54           DRAFT - DO NOT CITE OR QUOTE

-------
 1          4.  EVALUATION OF THE CARCINOGENICITY OF INDIVIDUAL PAHs
 2
 3
 4   4.1.  DATABASE OF STUDIES ON PAH CARCINOGENICITY AND CANCER-
 5   RELATED ENDPOINTS
 6          A database of primary literature relevant to the RPF approach for PAHs was developed.
 7   This was accomplished through the following means:
 8
 9          •  Definition of the study types that were considered relevant to relative potency
10             development;
11
12          •  Review of reference lists from review articles and other secondary sources;
13
14          •  Identification of selected PAHs to be included in search of open literature;
15
16          •  Performance of targeted searches of open literature on selected PAHs; and
17
18          •  Population of the database with references and meaningful keywords.
19
20          The study types that were considered most useful for RPF derivation were rodent
21   carcinogenicity bioassays (all routes) in which one or more PAH was tested at the same time as
22   benzo[a]pyrene.  In addition, in vivo and in vitro data for cancer-related endpoints (in which one
23   or more PAH and benzo[a]pyrene was tested simultaneously) were obtained, including studies
24   on the formation of DNA adducts, mutagenicity, chromosomal aberrations, aneuploidy, DNA
25   damage/repair/recombination, unscheduled DNA  synthesis, and  cell transformation.  Although it
26   would be possible to calculate RPFs from studies where a PAH and benzo[a]pyrene were tested
27   by the same laboratory using the same test system but at different times, this approach was not
28   considered because it could introduce differences in the dose-response information that are
29   unrelated to the chemical (e.g., variability associated with laboratory environment conditions,
30   animal handling, food supply).  Thus, studies in which benzo[a]pyrene was not tested
31   simultaneously with another PAH were not considered  for use in calculating RPFs.  Studies that
32   did not include benzo[a]pyrene were, however, considered useful for evaluating the weight of
33   evidence for selecting PAHs to be included in the RPF  approach.
34          Several study types were initially excluded from the database because they did not
35   provide carcinogenicity or cancer-related endpoint information for individual  PAHs. These
36   include biomarker studies measuring DNA adducts in humans, studies of PAH metabolism, and
37   studies of PAH mixtures. Although these studies  contain important information on human
38   exposure to PAH mixtures and the mode of action for PAH toxicity, they generally do not
39   contain dose-response information that would be useful for calculation of RPF estimates. In
40   addition to the primary bioassay and cancer-related endpoint studies described above, the RPF
                                            55           DRAFT - DO NOT CITE OR QUOTE

-------
 1   database also includes information on PAH mode of carcinogenic action, interactions among
 2   PAHs in mixtures, and the influence of exposure route on carcinogenic action of PAHs.
 3          Primary studies were identified through the review of available secondary sources and
 4   review articles, supplemented by a targeted literature search. A complete list of the secondary
 5   sources that were reviewed is contained in Appendix A.  A literature search strategy was
 6   developed by first constructing a list of the individual PAHs to be included. The list of PAHs
 7   was restricted to unsubstituted PAHs with three or more fused aromatic rings containing only
 8   carbon and hydrogen atoms, because these are the most widely studied members of the PAH
 9   chemical class. Heterocyclic PACs or PAHs with substituted groups (e.g.,  alkyl, hydroxyl,
10   sulfhydryl, amino, or nitro groups) were not included.  An initial search yielded a list of PAHs
11   for which toxicological data are available. Individual PAHs were then chosen for the literature
12   search because they were known to have toxicological information relevant to cancer, and in
13   most cases, their presence in environmental sources of PAH exposure was known.  Using these
14   criteria and excluding benzo[a]pyrene, 74 PAHs were identified from primary and secondary
15   sources (see Table 2-1 in Chapter 2).
16          A search of the open literature was conducted in the MEDLINE (PubMed) database for
17   the 74 PAHs that were identified. This database encompasses many of the  studies that would
18   also be found in TOXLINE and CANCERLIT (the latter is no longer available as a separate
19   database). MEDLINE was searched by CASRN in conjunction with cancer and cancer-related
20   endpoint keywords.  The search was not limited by publication date to ensure that all relevant
21   studies were identified.  A few compounds did not show any result when searched by CASRN.
22   For these PAHs, an additional search by name was conducted.  Search results, including
23   MEDLINE keywords, were downloaded directly into the working RPF database.
24          In addition to MEDLINE, computer searches of the following databases and websites
25   were conducted:  IARC, World Health Organization (WHO), Agency for Toxic Substances and
26   Disease Registry (ATSDR), Health Canada, the National Toxicology Program (NTP), California
27   EPA's Office of Environmental Health Hazard Assessment (OEHHA), the  Substance Registry
28   System, the Chemical Carcinogenesis Research Information System (CCRIS), the Toxic
29   Substance Control Act Test Submission (TSCATS) database, and the Distributed Structure-
30   Searchable Toxicity (DSSTOX) database.
31          Primary and secondary studies were entered in the RPF database and relevant keywords
32   (identifying study type, whether benzo[a]pyrene was included, route of administration, target
33   organ, etc.) were identified for each study.  The list of keywords was developed in order to
34   facilitate database searching for references on a specific topic.  Quality assurance procedures
35   were employed to ensure that database references were properly keyword-coded for retrieval.
36
                                            56          DRAFT - DO NOT CITE OR QUOTE

-------
 1   4.2.  STUDIES IN HUMANS
 2          Numerous studies have evaluated cancer outcomes in PAH-exposed individuals
 3   (reviewed in Bostrom et al., 2002; WHO, 1998; ATSDR,  1995; IARC, 1987, 1983, 1973).
 4   However, since these exposures were to complex mixtures containing multiple PAH
 5   carcinogens, they did not provide adequate data to evaluate the human carcinogenicity of
 6   individual PAH compounds. Epidemiology studies have focused on occupational exposure to
 7   PAH mixtures.  Emissions from coke production, coal gasification, aluminum production, iron
 8   and steel founding, coal tars, coal tar pitches, and soot have produced lung cancer in humans
 9   (Bostrom et al., 2002). Skin and scrotal cancers have resulted from exposure to coal tar, coal tar
10   pitches, nonrefmed mineral oils, shale oils, and soot (Larsen and Larsen, 1998; WHO, 1998;
11   ATSDR, 1995).  Occupational studies clearly demonstrate exposure-response relationships for
12   PAH mixtures; however, quantitative estimates of risk are limited primarily to lung cancer in
13   coke oven workers (Bostrom et al., 2002; Larsen and Larsen, 1998; ATSDR, 1995).
14          Biomonitoring of exposure to PAHs includes measurement of DNA and protein adducts
15   and measurement of urinary metabolites of PAHs, studies on genetic polymorphisms of CYP450
16   and other enzymes, and changes in PAH metabolism (Bostrom et al.,  2002; Larsen and Larsen,
17   1998; ATSDR,  1995).  While these studies demonstrate the degree of exposure to PAHs from
18   various settings, quantitative dose-response data for humans exposed to individual PAHs are not
19   available. Cancer-related endpoint studies that were performed using human cell lines are
20   presented with similar assays in other mammalian species in Section 4.3.
21
22   4.3.  STUDIES IN ANIMALS
23          The database of studies investigating cancer or cancer-related endpoints in animals
24   exposed to PAHs is extensive. For the purpose of developing relative potency estimates, only
25   those studies that included at least one selected PAH and benzo[a]pyrene as a reference
26   compound were reviewed.  Studies were excluded if PAH potency comparisons were not
27   conducted in the same laboratory in concurrent experiments.  Studies without benzo[a]pyrene are
28   listed in two separate bibliographies in Appendix B. Table B-l shows PAHs that were assayed
29   with or without benzo[a]pyrene.  Table B-l shows that 32 of the 74 PAHs were assayed with
30   benzo[a]pyrene; an additional 14 PAHs were not tested in the same study as benzo[a]pyrene.
31   The remaining 28 PAHs either have only cancer-related endpoint data, or have neither bioassays
32   nor cancer-related endpoint data. Bioassays without benzo[a]pyrene were considered in the
33   weight of evidence evaluation for individual PAHs (Section 6.1). Studies that provided only
34   information on PAH mixtures or PAH metabolites were not reviewed or summarized for this
35   analysis.
36          References in the database were sorted by keyword into the following major categories:
37   cancer bioassays, in vivo studies of cancer-related endpoints, and in vitro studies of cancer-
38   related endpoints. These categories were further divided by route (for bioassays) or by endpoint

                                            57           DRAFT - DO NOT CITE OR QUOTE

-------
 1   (for cancer-related endpoints). Each study was reviewed, and critical study details were
 2   extracted into tables (Tables 4-1 through 4-14) for each individual endpoint.  Studies with data
 3   on selected PAHs and benzo[a]pyrene were used, even if a particular PAH has not been
 4   evaluated by U.S. EPA or IARC for carcinogenicity. Studies were included in the analysis if the
 5   following selection criteria were met:
 6
 7          •  Benzo[a]pyrene was tested simultaneously with another PAH;
 8
 9          •  A statistically increased incidence of tumors was observed with benzo[a]pyrene
10             administration;
11
12          •  Benzo[a]pyrene produced a statistically significant change in a cancer-related
13             endpoint finding;
14
15          •  Quantitative results were presented;
16
17          •  The carcinogenic response observed in either the benzo[a]pyrene- or other PAH-
18             treated animals at the lowest dose level was not saturated (i.e., tumor incidence at the
19             lowest dose was <90%); and
20
21          •  There were no study quality concerns or potential confounding factors that precluded
22             use (e.g., no concurrent control, different vehicles, strains, etc. were used for the
23             tested PAH and benzo[a]pyrene; use of cocarcinogenic vehicle; PAHs of questionable
24             purity; unexplained mortality in treated or control animals).
25
                                             58           DRAFT - DO NOT CITE OR QUOTE

-------
Table 4-1.  Study summaries: dermal bioassays of benzo[a]pyrene and at least one other PAH
Record
number
Reference
Mouse"
strain
Exposure
Follow up
Vehicle
Promoter
Tumor type
Positive
result
Nonpositive
result
Meets selection
criteria?
Comments
Complete carcinogenicity studies
480
600
22390
19320
22400
13640
13650
620
17660
610
19760
18570
Bingham and Falk,
1969
Habsetal., 1980
Wynder and
Hoffmann, 1959a
LaVoie et al., 1979
Wynder and
Hoffmann, 1959b
Cavalieri et al., 1983
Cavalieri et al.,
1981b
Hoffmann and
Wynder, 1966
Cavalieri et al., 1977
Higginbotham et al.,
1993
Masuda and
Kagawa, 1972
Hechtetal., 1974
CH3/He
NMRI
Swiss
HA/ICR
Swiss
albino
Swiss
Swiss
Swiss
Ha/ICR/
Mil Swiss
Swiss
Swiss
Ha/ICR/
Mil Swiss
Ha/ICR/
Mil Swiss
3 times/wk
2 times/wk
(4 times for
CO) for life
3 times/wk
3 times/wk
3 times/wk
2 times/wk
for 48 wk
2 times/wk
for 30 wk
3 times/wk
for 12 mo
2 times/wk
for 30 wk
2 times/wk
3 times/wk
for
60 applica-
tions
3 times/wk
for 17 mo
50 wk
Until
moribund or
dead
6-14 mo
Unspecified
10-22 mo
Until 2 cm
tumor or
61 wk
Until 2 cm
tumor,
moribund, or
57 wk
Up to 1 5 mo
Until
moribund,
dead, or
after 70 wk
40 wk
7 mo
72 wk
Toluene
or n-do-
decane
Acetone
(DMSO
for CO)
Cyclo-
hexane
Acetone
Acetone
Acetone
Acetone
Dioxane
Acetone
Acetone
Dioxane
Acetone
None
None
None
None
None
None
None
None
None
None
None
None
Malignant
and benign
Papilloma,
carcinoma,
sarcoma
Papilloma,
carcinoma
Unspecified
Papilloma,
carcinoma
Papilloma,
adenoma,
carcinoma
Primarily
squamous
cell
carcinoma
Papillomas
Papilloma,
kerato-
acanthoma,
carcinoma
Papilloma,
carcinoma
Unspecified
Unspecified
BaA
BbF
BbF, BjF
CH, BbF,
BjF,
DBaeP,
DBahP,
DBaiP
CH,
DBahA,
DBaiP
CPcdP
CPcdP
DBaeP,
DBahP,
DBaiP,
DBaeF
DBahP,
AA
DBaiP
DBaiP
CH

BkF, BjF, CPcdP,
CO, IP
BghiF, BkF
AC, Pyr, BghiF,
BkF, AA, BeP,
DBelP, IP,
BghiP, N23eP
AC, BeP, Pyr, FA

ACEP

BaA



No
Yes
No
No
No
Yes
Yes
Yes
Yes
No
No
No
BaP administered in different vehicle.
n-Dodecane cocarcinogenic with BaA.
No concurrent untreated, toluene, or
n-dodecane control.

Deaths prior to first tumor appearance.
No concurrent control.
Reiterates data published elsewhere.
Deaths prior to first tumor appearance.
Not clear if BaP administered
simultaneously. No concurrent
control.
Reports both incidence and
multiplicity.
Tumor incidence not useable because
BaP tumor incidence was 100%.
Tumor multiplicity data available for
dose-response assessment.
Paper in German. Paper reports
compound as DBaiP; LaCassagne et
al. (1968) state that it is actually
DBaeF. DBahP incidence >90% at
lowest dose.
DBahP incidence >90% at lowest
dose.
No tumors with BaP.
No concurrent untreated or vehicle
control; lowest dose DBaiP gave
100% incidence.
BaP dose not reported.
                                         59
DRAFT - DO NOT CITE OR QUOTE

-------
Table 4-1.  Study summaries: dermal bioassays of benzo[a]pyrene and at least one other PAH
Record
number
21310
23310
23840
Reference
Shubiketal., 1960
Pfeiffer and Allen,
1948
Barry et al., 1935
Mouse"
strain
Syrian
golden
hamster
Rhesus
monkey
Un-
specified
Exposure
2 times/wk
for 10 wk
various
2 times/wk
Follow up
75 wk
Various
1-2+ yr
Vehicle
Mineral
oil
Sesame
oil
Benzene
Promoter
None
None
None
Tumor type
None
Various
Epithelioma,
papilloma
Positive
result

Multiple
Multiple
Nonpositive
result
DBahA, BaA


Meets selection
criteria?
No
No
No
Comments
Small number of animals (5/sex/dose).
Sequential exposure to multiple
compounds; no concurrent untreated
control.
Test compounds from various sources
gave differing results; purity may be
suspect; use of benzene vehicle
confounds tumorigenicity results; no
benzene or untreated control.
Initiation studies
24800
21410
630
16310
10200
18570
22500
Nesnow et al., 1984
Slaga et al., 1978
LaVoie et al., 1982
Weyand et al., 1992
El-Bayoumy et al.,
1982
Hechtetal., 1974
Van Duuren et al.,
1966
SENCAR
CD-I
CrliCD-
1[ICR]
BR
Crl:CD-l
CrliCD-
1[ICR]
BR
Ha/ICR/
Mil Swiss
ICR/HA
Single
Single
10 subdoses
every other d
5 or
10 applica-
tions given
every other d
10 subdoses
every other d
10 subdoses
every other d
Single
31 wk
27 wk
Unspecified
Until
promotion
complete
Unspecified
Until
promotion
complete
63 wk
Acetone
Acetone
Acetone
Acetone
Acetone
Acetone
Acetone
TPA 2 ng
2 times/wk
for 30 wk
TPA 10 ug
2 times/wk
for 26 wk
TPA 2.5 ug
3 times/wk
for 20 wk
TPA 2.5 ug
3 times/wk
for 20 wk
TPA 2.5 ug
3 times/wk
for 25 wk
TPA 2.5 ug
3 times/wk
for 20 wk
Croton resin,
25 Ug
3 times/wk
Papilloma
Papilloma
Primarily
squamous
cell
papilloma
Unspecified
Primarily
squamous
cell
papilloma
Unspecified
Papilloma,
carcinoma
BeAC,
B1AC
BaA
BbF, BjF,
BkF
BjF
CH
CH
CH, BbF




Pery, Pyr

BghiF
Yes
Yes
Yes
Yes
Yes
Yes
No
Reports both incidence and
multiplicity.
Tumor incidence data not useable
because BaP gave 93% tumor
incidence. Tumor multiplicity data
available for dose-response
assessment.
Reports both incidence and
multiplicity.
Tumor incidence data not useable
because BaP gave 100% tumor
incidence. Tumor multiplicity data
available for dose-response
assessment. DNA adducts,
mutagenicity also evaluated.
Tumor incidence data not useable
because single dose CH gave 100%
tumor incidence; BaP gave 90% tumor
incidence. Tumor multiplicity data
available for dose-response
assessment.
Reports both incidence and
multiplicity.
BaP gave 100% tumor incidence.
Corollary data with acetone only as
promotion agent not included.
                                         60
DRAFT - DO NOT CITE OR QUOTE

-------
Table 4-1.  Study summaries: dermal bioassays of benzo[a]pyrene and at least one other PAH
Record
number
24300
19320
21420
15640
620
610
13660
19360
13650
20830
Reference
Rice et al., 1985
LaVoie et al., 1979
Slaga, et al., 1980
Raveh et al., 1982
Hoffmann and
Wynder, 1966
Higginbotham et al.,
1993
Cavalieri et al., 1991
LaVoie et al., 1985
Cavalieri et al.,
1981b
Roe, 1962
Mouse"
strain
CD-I
HA/ICR
Swiss
albino
SENCAR
SENCAR
Ha/ICR/
Mil Swiss
SENCAR
SENCAR
Crl:CD/l
(ICR)BR
CD-I
Albino
Exposure
10 subdoses
every other d
10 subdoses
every other d
Single
Single
Single
Single
Single
10 subdoses
every other d
10 subdoses
every other d
Single
Follow up
Until
promotion
complete
Until
promotion
complete
15 wk
25 wk
6 mo
27 wk
16 wk and
27 wk (two
experiments)
Unspecified
57 wk
Until
promotion
complete
Vehicle
Acetone
Acetone
or
dioxane
Acetone
Un-
specified
Dioxane
Acetone
Acetone
Acetone
Acetone
Acetone
Promoter
TPA
0.0025%
3 times/wk
for 20 wk
TPA 2.5 ug
3 times/wk
for 20 wk or
croton oil
2.5%
3 times/wk
TPA 2 ug
2 times/wk
TPA 2 ng
2 times/wk
for 25 wk
Croton oil
TPA
2.6 nmol,
2 times/wk
TPA
3. 24 nmol
2 times/wk
for 1 1 wk
TPA 2.5 ug
3 times/wk
for 20 wk
TPA
0.017umol
2 times/wk
for 40 wk
Croton oil
once/wk for
20 wk
Tumor type
Unspecified
Unspecified
Papilloma
Papilloma
Papillomas
Papillomas,
few
carcinomas
Primarily
papilloma
Unspecified
Papilloma
Papilloma
Positive
result
CH,
CPdefC
CH,
DBaeP,
DBahP,
DBaiP,
N23eP
CH,
DBahA,
CPcdP
DBaeF,
DBaeP,
DBahP,
DBaiP,
N23eP
DBaiP
DBaiP

CPcdP

Nonpositive
result

FA, AA, DBelP,
BghiP, IP
BeP, DBacA

IP, AA, BghiP,
DBelP


AC
ACEP
PH
Meets selection
criteria?
Yes
No
Yes
Yes
Yes
No
Yes
Yes
Yes
No
Comments
Tumor incidence data not useable
because all compounds gave >90%
tumor incidence. Tumor multiplicity
data available for dose-response
assessment.
Reiterates data published elsewhere.
Not clear if BaP done simultaneously
but protocol, vehicle, and follow-up
are the same. Reports both incidence
and multiplicity.
Reports both incidence and
multiplicity.
Paper reports compound as DBaiP;
LaCassagne et al. (1968) state that it is
actually DBaeF.
No tumors with BaP.
Tumor incidence data not useable
because lowest dose DBaiP gave
>90% tumor incidence. Tumor
multiplicity data from both
experiments available for dose-
response assessment.

Reports both incidence and
multiplicity.
BaP not simultaneous.
                                         61
DRAFT - DO NOT CITE OR QUOTE

-------
       Table 4-1. Study summaries:  dermal bioassays of benzo[a]pyrene and at least one other PAH
Record
number
16440
17450
18680
19420
13660
15700
Reference
Woodetal., 1980
Bruneetal., 1978
Hoffmann et al.,
1972
LaVoieetal., 1981
Cavalieri et al., 1991
Rice et al., 1988
Mouse"
strain
CD-I
NMRI
Ha/ICR/
Mil Swiss
HA/ICR
Swiss
albino
SENCAR
CD-I
Exposure
Single
Unspecified
10 subdoses
every other d
10 subdoses
every other d
Single
10 subdoses
every other d
Follow up
27 wk
Unspecified
Until
promotion
complete
Unspecified
27 wk
24 wk
Vehicle
Acetone
Un-
specified
Acetone
Acetone
Acetone
Acetone
Promoter
TPA 16 nmol
2 times/wk
for 26 weeks
TPA
Croton oil
2.5% for
20 wk
TPA 2.5 ug
3 times/wk
for 20 wk
None
TPA 2.5 ug
3 times/wk
for 20 wk
Tumor type
Unspecified
Unspecified
Unspecified
Unspecified
Primarily
papilloma
Unspecified
Positive
result




DBalP
CH,
BbcAC,
CPdefC
Nonpositive
result
Pyr, CPcdP
AC
FA
PH


Meets selection
criteria?
Yes
No
Yes
Yes
Yes
Yes
Comments

Study design not reported. Results
reported qualitatively.


Initiating dose only; no promoter.
Tumor incidence data not useable
because lowest dose DBalP gave
>90% tumor incidence. Tumor
multiplicity data available for dose-
response assessment.
Not clear if BaP done simultaneously
for all PAHs.
Cocarcinogenicity studies
18700
21430
21840
21850
21920
Horton and
Christian, 1974
Slaga et al., 1979
Van Duuren and
Goldschmidt, 1976
Van Duuren et al.,
1973
Warshawsky et al.,
1993
C3H
CD-I
ICR/Ha
Swiss
ICR/HA
C3H/HEJ
2 times/wk
for 80 wk
Single
3 times/wk
3 times/wk
for 52 wk
2 times/wk
82 wk
30 wk
368 or 440 d
52 wk
Until lesion
developed or
104 wk
n-Do-
decane/
decalin
mixture
Acetone
Acetone
Acetone
Toluene
or n-do-
decane
None
TPA 10 ug
2 times/wk
for 30 wk
None
None
None
Carcinoma,
papilloma
Papilloma
Papilloma
None
Unspecified
DBacA,
Pyr
BeP



CH, FA, Tphen,
Pery,

Pyr, BghiP, BeP,
FA
Pyr, BghiP, BeP
AC, CH, Pyr, FA,
PH
No
No
Yes
No
No
Not clear if BaP done simultaneously.
Experiments with decalin
(noncarcinogen) and 50/50 decalin/
dodecane mix (cocarcinogenic). No
data for BaP in 50/50 mix. No vehicle
control in decalin.
No concurrent control. Study aimed at
exploring interactions; not clear if BaP
done simultaneously.

Qualitative results reported.
No tumors with BaP.
"Except where noted, all studies were conducted in mice.




DMSO = dimethyl sulfoxide
                                                      62
DRAFT - DO NOT CITE OR QUOTE

-------
Table 4-2. Study summaries:  intraperitoneal bioassays of benzo[a]pyrene and at least one other PAH
Record
number
Reference
Mouse strain"
Exposure
Follow
up
Vehicle
Target
organ(s)
Tumor type(s)
Positive
result
Non-
positive
result
Meets selection
criteria?
Comments
Newborn mouse studies
13610
17560
640
7510
22040
22510
Busby et al.,
1984
Busby et al.,
1989
LaVoie et al.,
1987
LaVoie et al.,
1994
Weyand and
LaVoie, 1988
Wislocki et al.,
1986
Swiss-
Webster
BLU:Ha
(ICR)
Swiss-
Webster
BLU:Ha
(ICR)
CD-I
CD-I
CD-I
CD-I
1st, 8th, 15thd
1st, 8th, 15thd
1st, 8th, 15thd
1st, 8th, 15thd
1st, 8th, 15thd
1st, 8th, 15thd
26 wk
26 wk
52 wk
12 mo
Not
reported
12 mo
DMSO
DMSO
DMSO
DMSO
DMSO
DMSO
Lung
Lung
Lung,
liver
Lung,
liver
Lung,
liver
Lung,
liver,
lymphatic
system
Adenoma,
adenocarcinoma
Adenoma,
adenocarcinoma
Adenoma,
hepatoma
Foci, adenoma,
carcinoma
Unspecified
Adenoma,
carcinoma,
lymphoma
FA
FA
BbF, BjF
FA
Not
reported
CH, BaA

Pyr, CH
BkF, IP


Pyr
Yes
Yes
Yes
Yes
No
Yes
Tumor incidence data not useable
because lowest dose BaP gave >90%
tumor incidence. Tumor multiplicity
data available for dose-response
assessment.
Reports both incidence and multiplicity.

Reports both incidence and multiplicity.
Abstract only; dose-response information
not included.
Reports both incidence and multiplicity.
Studies in A/J mice
11190
23960 and
23450
22670
24590
Mass et al., 1993
Nesnow et al.,
1998a, 1995
Nesnow et al.,
1996
Nesnow et al.,
1998b
A/J
A/J
A/J
A/J
Single
Single
Single
Single
8 mo
8 mo
8 mo
8 mo
Tri-
caprylin
Tri-
caprylin
Tri-
caprylin
Tri-
caprylin
Lung
Lung
Lung
Lung
Adenoma,
carcinoma
Adenoma
Adenoma
Adenoma
BjAC
BbF,
DBahA,
CPcdP
BbF,
DBahA,
CPcdP
CPcdP,
BbF,
DBahA,
BjAC,
DBalP




No
No
No
Yes
Reiterates data reported elsewhere
(Record 24590).
Reiterates data reported elsewhere
(Record 24590).
(Reiterates data reported elsewhere
(Record 24590).)
Raw data (both incidence and
multiplicity) obtained courtesy of S.
Nesnow.
                                          63
DRAFT - DO NOT CITE OR QUOTE

-------
       Table 4-2. Study summaries: intraperitoneal bioassays of benzo[a]pyrene and at least one other PAH

Record
number
20920
24801

Reference
Rossetal., 1995
Weyand et al.,
2004

Mouse strain"
A/J
A/J

Exposure
Single
Single

Follow
up
240 d
260 d

Vehicle
Tri-
caprylin
Tri-
caprylin

Target
organ(s)
Lung
Lung

Tumor type(s)
Adenoma
Adenoma

Positive
result
BbF,
DBahA,
CPcdP
BcFE
Non-
positive
result
Pyr


Meets selection
criteria?
No
Yes

Comments
Reiterates data reported elsewhere
(Record 24590).

"All studies were conducted in mice.
                                                   64
DRAFT - DO NOT CITE OR QUOTE

-------
Table 4-3.  Study summaries: subcutaneous bioassays of benzo[a]pyrene and at least one other PAH
Record
number
23840
220

18350
23200



660


23310

24290

24310
21560

Reference
Barry et al., 1935
Bryan and
Shimkin, 1943

Grant and Roe,
1963
Homburger et al.,
1972



Pfeiffer, 1977


Pfeiffer and Allen,
1948

Rask-Nielson,
1950

Roe and Waters,
1967
Steiner, 1955

Species
Mouse
Mouse

Mouse
Hamster



Mouse


Monkey

Mouse

Mouse
Mouse

Strain
Unspeci-
fied
C3H

Albino
Various



NMRI


Rhesus

Street

Swiss
albino
C57BL

Exposure
site
Unspecified
Right axilla

Neck
Groin



Neck


Various

Thymus,
lung,
mammary
area

Not
specified
Interscapular

Exposure
Single
Single

1st d after
birth
Single



Single


Various

Single

Istd after
birth
Single

Follow
up
1-2+ yr
until
20mm
tumor
52-62 wk
52 wk



114wk


variable

30 mo

50-60 wk
22-28 mo

Vehicle
Lard
Tricaprylin

Aqueous
gelatin
Tricaprylin



Tricaprylin


Sesame oil

Paraffin

Aqueous
gelatin
Tricaprylin

Target
organ(s)
Injection
site
Injection
site

Lung
Injection
site; lung



Injection
site


Various

Various

Liver
Injection
site

Tumor
type(s)
Sarcoma
Unspecified

Adenoma
Various



Sarcoma


Various

Various

Hepatoma
Sarcoma

Positive
result
Multiple
DBahA


BaA



DBahA


Multiple

DBahA

PH
DBahA,
BaA, CH

Nonpositive
result



PH












AC, PH

Meets selection
criteria?
No
No

Yes
No



No


No

No

No
No

Comments
Test compounds from
various sources gave
differing results; purity
may be suspect; no
untreated control.
No concurrent untreated
control.


Study aimed at
evaluating strain
specificity of
tumorigenicity. BaA
results equivocal. Not
clear if BaP treatment
simultaneous. "Aged"
mice used as controls;
aged mice allowed to
live 16 weeks longer.
Less than 10% of
100 control mice alive
at 1 14 wk; control data
not provided.
Sequential exposure to
multiple compounds; no
concurrent untreated
control.
Number of control and
exposed varies by
tumor type reported;
BaP nontumorigenic;
DBahA results
equivocal; results
unclear.
Study methodology and
results not detailed; PH
results equivocal.
No concurrent untreated
control; study aimed at
evaluating interactions.
                                         65
DRAFT - DO NOT CITE OR QUOTE

-------
Table 4-4.  Study summaries: oral bioassays of benzo[a]pyrene and at least one other PAH
Record
number
17280
23880
24801
Reference
Biancifiori and Caschera,
1962
Huggins and Yang, 1962
Weyand et al., 2004
Species
Mouse
Rat
Mouse
Strain
BALB/c
Sprague-
Dawley
A/J
Exposure
route
Gavage
Gavage
Diet
Exposure
2 times/wk,
15 wk
Single
Daily,
260 d
Follow up
Variable;
50-60 wk
Not
reported
260 d
Target
organ(s)
Mammary
gland
Mammary
gland
Lung
Tumor
type(s)
Carcinomas
and
sarcomas
Unspecified
Adenoma
Positive
result
DBahA

BcFE
Non-
positive
result

BaA, PH

Meets selection
criteria?
No
No
Yes
Comments
Tumors observed after DBahA only
in pseudopregnant mice, not virgin
mice.
Untreated control information not
included.

                                         66
DRAFT - DO NOT CITE OR QUOTE

-------
Table 4-5. Study summaries: other route bioassays of benzo[a]pyrene and at least one other PAH
Record
number
21750
17620
13660
21620
20280
17940
22000
21500
23910
Reference
Topping et al., 1981
Cavalieri et al.,
1988b
Cavalieri et al., 1991
Sugiyama, 1973
Pataki and Huggins,
1969
Deutsch-Wenzel et
al., 1983
Wenzel-Hartung et
al., 1990
Soltetal., 1987
Nikonova, 1977
Species
Rat
Rat
Rat
Rat
Rat
Rat
Rat
Hamster
Mouse
Strain
F344
Sprague-
Dawley
Sprague-
Dawley
Long-
Evans
Sprague-
Dawley
Osborne-
Mendel
Osborne-
Mendel
Syrian
golden
A
Exposure route
Implantation in
transplanted
tracheas
Intramammillary
Intramammillary
Intramuscular
Intravenous
Lung implantation
Lung implantation
Painting buccal
pouch
Subcutaneous (FO)
and transplacental
(Fl)
Exposure
Release
from pellet
Single
Single
Single
3 doses 3 d
apart
Release
from pellet
Release
from pellet
2 times/wk
for 20 wk
GD 18 or
19
Follow up
28 mo
20 wk
Until 2 cm
tumor or
24 wk
9 mo
98 d
Until
moribund
or dead
Until
moribund
or dead
Up to
44 wk
lyr
Vehicle
Beeswax
pellet
None
Trioctanoin
Sesame oil
Lipid
emulsion
Beeswax/
trioctanoin
Beeswax/
trioctanoin
Paraffin oil
Sunflower
oil
Target
organ(s)
Tracheal
epithelium
Mammary
Mammary,
other
Injection
site
Mammary
Lung
Lung
Buccal
pouch
Lung,
mammary,
liver,
injection
site
Tumor type(s)
Carcinoma,
sarcoma
Adeno-
carcinoma,
adenofibroma,
fibrosarcoma
Adeno-
carcinoma,
adenofibroma,
fibrosarcoma,
squamous cell
carcinoma
Sarcoma
Unspecified
Carcinoma,
sarcoma
Carcinoma
Carcinoma
Adenoma
Positive
result


DBalP


BbF,
BjF,
BkF,
IP, AA,
BghiP
CH,
DBahA


Non-
positive
result
BeP
DBahA,
BaA

BaA
BaA
BeP
PH
BaA
Pyr
Meets
selection
criteria?
No
No
No
No
No
Yes
Yes
No
No
Comments
Interaction
information
included.
Control data
from untreated
mammary
glands of same
rats.
DBalP
produced
tumors in all
animals at the
lowest dose.
BaP gave 100%
tumor
incidence.
No control
group.


Fewer than
20 animals per
group; negative
result.
Transplacental
exposure not
quantified.
                                         67
DRAFT - DO NOT CITE OR QUOTE

-------
Table 4-6.  Study summaries: in vivo DNA adducts with benzo[a]pyrene and at least one other PAH
Record
number
6210
17420
17630
18810
18790
10900
13200
11190
8010
22670
23960
24590
22810

20650
20920
16310
22040
Reference
Arifetal., 1997
Brookes and
Lawley, 1964
Cavalieri et al.,
1981a
Hughes and
Phillips, 1990
Hughes and
Phillips, 1991
Koganti et al.,
2000
Li et al., 2002
Mass et al.,
1993
Nesnow et al.,
1993b
Nesnow et al.,
1996
Nesnow et al.,
1995
Nesnow et al.,
1998a
Phillips et al
1979
Reddy etal.,
1984
Ross et al.,
1995
Weyand etal.,
1992
Weyand and
LaVoie, 1988
Route of
administration
Intramammillary
Dermal
Dermal
Dermal
Dermal
Oral-diet
Gavage or oral-
diet
Intraperitoneal
Intraperitoneal
Intraperitoneal
Intraperitoneal
Intraperitoneal
Dermal

Dermal
Intraperitoneal
Dermal
Intraperitoneal
Exposure
frequency
Single dose
Single dose
Single dose
Single dose
Single dose
14 d
1 time/d for 1—
4 d; diet 14 d
Single dose
Single dose
Single dose
Single dose
Single dose
Single dose

4 doses (0, 6,
30, 54 hr)
Single dose
Single dose
Postnatal d 1,
8, 15
Hours between
dosing and
sacrifice
48
various to —12 d
4,24
0.5, 1,2,4,7,21,
84 d
24
not stated

24, 48, 72
1, 3, 7, 14, 28, 56 d
7d
7d
various to 2 1 d
19 24 48 72 96
120, 144
24
0, 1, 3, 5, 7, 14, 21 d
24
24
Tissue analyzed
Mammary
epithelium, lung
Skin
Skin
Skin, lung
Skin
Lung
Mammary gland
and liver; lung
Lung
Lung, liver,
peripheral blood
lymphocytes
Lung
Lung
Lung
Skin

Skin
Lung
Skin
Lung, liver
Method of
analysis
[32P] postlabeling
[3H] prelabeling
[3H] or [14C]
prelabeling
[32P] postlabeling
[32P] postlabeling
[32P] postlabeling
[32P] postlabeling
[32P] postlabeling
[32P] postlabeling
[32P] postlabeling
[32P] postlabeling
[32P] postlabeling
[3H]-Prelabeling

[32P] postlabeling
[32P] postlabeling
[32P] postlabeling
[32P] postlabeling
PAHs evaluated"
DBalP
DBacA, DBahA
CPcdP, ACEP
DBalP, DBaeP,
DBahP, DBaiP
DBaeP
BcFE, BaFE, BbFE
BcFE
BjAC
BbF
BbF, DBahA,
CPcdP
BbF, DBahA,
CPcdP
BbF, CPcdP,
DBahA, DBalP
BaA DBacA
DBahA
AC, BaA, BghiP,
BeP, CH, DBacA,
DBahA, Pery, Pyr
BbF, CPcdP,
DBahA
BjF
BbF, BjF, BkF
Meets selection
criteria?
Yes
No
Yes
Yes
No
No
No
Yes
Yes
No
No
Yes
Yes

No
No
No
No
Comments

Data on individual compounds not
reported.

24-hr experiment with DBaeP and
DBalP; 84-d experiment with all.
No quantitative information; abstract
only.
Not quantified.
Not quantified; BaP administered by
gavage, BcFE admin in diet.

Peaks differ temporally; study also
correlates number of adducts in
organs.
Not quantified.
Not quantified.
Used data from Ross et al., 1995 (ref
20920) to calculate slope.


Semiquantitative data only.
Reiterates data published elsewhere
(Record 24590).
Not quantified.
No quantitative data; abstract only.
                                         68
DRAFT - DO NOT CITE OR QUOTE

-------
       Table 4-6. Study summaries:  in vivo DNA adducts with benzo[a]pyrene and at least one other PAH
Record
number
24801
24790
Reference
Weyand et al.,
2004
Kligerman et
al., 2002
Route of
administration
Oral-diet or
intraperitoneal
Intraperitoneal
and oral
Exposure
frequency
14 d diet;
single dose
intraperitoneal
Single dose
Hours between
dosing and
sacrifice
24
7d
Tissue analyzed
Lung,
forestomach
Peripheral blood
lymphocytes
Method of
analysis
[32P] postlabeling
[32P] postlabeling
PAHs evaluated"
BcFE
BaA, BbF, CH
Meets selection
criteria?
Yes
Yes
Comments

Data in both rats and mice.
"Positive findings were reported for all PAHs evaluated.
                                                    69
DRAFT - DO NOT CITE OR QUOTE

-------
Table 4-7. Study summaries: in vivo clastogenicity or sister chromatid exchange with benzo[a]pyrene and at
least one other PAH

Record
number
24740
14270
17190




19030

24720

24790


20200
20230

20950

21050

21770


Reference
Allen et al., 1999
He and Baker,
1991
Bayer, 1978




Katzetal., 1981

Kligerman et al.,
1986

Kligerman et al.,
2002


Oshiro et al.,
1992
Paikaetal., 1981

Roszinsky-
Kocher et al.,
1979
Salamone et al.,
1981

Tsuchimoto and
Matter, 1981


Species
Mice
Mice
Hamsters




Mice

Mice

Mice and
rats


Mice
Mice

Hamsters

Mice

Mice


Strain
A/Jor
p53 +/+,
+/-, and
HRA/Skh
hairless
Chinese




B6C3Fi/
BR

C57BL6

CD-I
Swiss
mice; CD
rats


CD-I
CBA/J

Chinese

B6C3Fi

CD-I


Route of
administration
Intraperitoneal
Dermal
Intraperitoneal




Intraperitoneal

Gavage

Oral and
intraperitoneal


Peroral
Intraperitoneal

Intraperitoneal

Intraperitoneal

Intraperitoneal


Vehicle
Tricaprylin
Acetone
Tricaprylin




DMSO

Corn oil

Sunflower
seed oil


Polyethylene
glycol
DMSO

Tricapryline

Not
specified

DMSO


Exposure
Single
Single
Single




At 0 and
24 hr

Single

Single


1 time/d,
4d
single

2 doses
24 hr
apart
2 doses
24 hr
apart
2 doses
24 hr
apart
Hours between
dosing and
sacrifice
48 or 72 hr
24 hr
24 hr for
aberrations; 30 hr
for micronuclei



various; 24, 30,
48, 72 hr after last
dose
23.5-25 hr

7d


24 hr after 2nd
and 4th treatment
16-20 hr

24 hr after 2nd
treatment

24, 48, 72 hr after
2nd treatment

6 hr after 2nd
treatment


Tissue
analyzed
Bone
marrow or
peripheral
blood
Keratino-
cytes
Bone
marrow




Bone
marrow

Peripheral
blood

Whole
blood or
mono-
nuclear
leukocytes

Peripheral
blood
Bone
marrow

Bone
marrow

Bone
marrow

Bone
marrow

Clasto-
genic
end point
Micro-
nuclei
Micro-
nuclei
Gaps,
breaks,
micro-
nuclei,
sister
chromatid
exchanges
micro-
nuclei

Sister
chromatid
exchanges
Sister
chromatid
exchange,
micro-
nuclei

Micro-
nuclei
Sister
chromatid
exchanges
Sister
chromatid
exchanges,
aberrations
Micro-
nuclei

Micro-
nuclei


Positive
results
DBalP
CH
PH (high
dose
only)





B1AC

BaA,
BbF, CH





PH, CH,
DBahA,
BaA,
BbF, BeP




Non-
positive
results

Pyr





DBaiP,
AC,
BghiP, Pyr





Pyr, AC
Pyr

AC

AC, Pyr

Pyr

Meets
selection
criteria?
Yes
Yes
Yes




No

Yes

Yes


No
No

Yes

Yes

Yes


Comments







No quantitative data.



All positive for sister
chromatid exchange
via intraperitoneal
administration;
mixed results for oral
administration.
No quantitative data;
published as abstract.
No quantitative data.

Positive results for
sister chromatid
exchanges, not
aberrations.




                                           70
DRAFT - DO NOT CITE OR QUOTE

-------
Table 4-7. Study summaries: in vivo clastogenicity or sister chromatid exchange with benzo[a]pyrene and at
least one other PAH
Record
number
21390
21620
Reference
Sirianni and
Huang, 1978
Sugiyama, 1973
Species
Mice
Rats
Strain
C3H/SI
Long-
Evans
Route of
administration
V79 cells in dif-
fusion chamber
implanted in
peritoneal
cavity of mice
Intravenous
Vehicle

Lipid
emulsion
Exposure

Single
Hours between
dosing and
sacrifice

12, 24 hr
Tissue
analyzed
Chinese
hamster
V79 cells
Bone
marrow
Clasto-
genic
end point
Sister
chromatid
exchanges
Gaps,
breaks
Positive
results


Non-
positive
results
AC, Pyr,
Pery
BaA
Meets
selection
criteria?
Yes
Yes
Comments


                                           71
DRAFT - DO NOT CITE OR QUOTE

-------
Table 4-8. Study summaries: in vivo mutagenicity with benzo[a]pyrene and at least one other PAH
Record
number
18130
13980
11190
23960
22670
24590
21370
21830
22450
Reference
Fahmy and Fahmy,
1980
Frolich and Wurgler,
1990
Mass et al., 1993
Nesnowetal., 1995
Nesnow et al., 1996
Nesnowetal., 1998b
Simmon et al., 1979
Valencia and
Houtchens, 1981
Zijlstra and Vogel,
1984
Species/strain
Drosophila
melanogaster
D. melanogaster
A/3 mice
A/J mice
A/J mice
A/J mice
Swiss Webster
mice
D. melanogaster
D. melanogaster
Route of
administration
Suspension in
media
Suspension in
media
Intraperitoneal
Intraperitoneal
Intraperitoneal
Intraperitoneal
PAHs
intramuscular or
peroral;
microorganisms
intraperitoneal
Filter feeding
Abdominal
injection
Exposure
frequency/follow up
48-72 hr
48-72 hr
3 d/8 mo
Single injection/
8 mo
Single injection/
8 mo
Single injection/
8 mo
Single injection/4 hr
48-72 hr
Not applicable
Mutagenic end point
Somatic mutation; eye color
mosaicism
Somatic mutation and
recombination test; wing
spots
Mutations in codon 12 of
the Ki-ras oncogene; PCR
and DNA sequencing of
lung tumor DNA
Mutations in codon 12 of
the Ki-ras oncogene; PCR
and DNA sequencing of
lung tumor DNA
Mutations in codon 12 of
the Ki-ras oncogene; PCR
and DNA sequencing of
lung tumor DNA
Mutations in codons 12 and
61 of the Ki-ras oncogene;
PCR and DNA sequencing
of lung tumor DNA
Intraperitoneal host
mediated assay;
mutagenicity in S.
typhimurium and
Saccharomyes cerevisiae of
recovered microorganisms
Sex-linked recessive lethal
test
Sex-linked recessive lethal
test; 2-3 translocation and
ring-X loss
Positive
result


BjAC
BbF,
DBahA,
CPcdP
BbF,
DBahA,
CPcdP
BbF,
DBahA,
CPcdP,
BjAC,
DBalP



Non-
positive
result
BaA
BaA




AC, BaA,
BeP, CH,
PH
Pyr
BaA
Meets
selection
criteria?
Yes
No
No
No
No
No
No
No
No
Comments

Inconsistent results for BaA; significant
effects only seen with cross-breeding of
strains selected for enhanced metabolic
activity (not standard strains).
Quantitative dose-response data were
not available. Different mutation
sequences observed; GGT— >TGT for
BaP and GG1WCGT for BjAC;
mutation sequence for BjAC may
correlate with cyclopenta-adduct
formation.
Quantitative dose-response data were
not available. GGT— >TGT mutations
for BaP and BbF; GG1WCGT for
CPcdP; no mutations seen for DBahA.
Quantitative dose-response data were
not available. GGT— >TGT mutations
for BaP and BbF; GGT^CGT for
CPcdP; no mutations seen for DBahA.
Quantitative dose-response data were
not available. Mutations in codon 12,
GGT^TGT for BaP, BbF, and DBalP;
GGT^CGT for CPcdP and BjAC; no
mutations seen for DBahA; GTT
mutations seen for all other PAHs. Only
DBalP caused mutations in codon 61.
Assay was not considered sensitive
enough for detecting carcinogens.
Results were negative for BaP.
Results were negative for BaP.
                                          72
DRAFT - DO NOT CITE OR QUOTE

-------
Table 4-9. Study summaries: in vitro bacterial mutagenicity with benzo[a]pyrene and at least one other PAH
Record
number
17030
23830
23660
17380
9560
17590
17630
9620
24030
13860
18030
18050
18180
24080
14080
14170
14260
Reference
Andrews et al.,
1978
Baker et al., 1980
Bartsch et al., 1980
Bosetal., 1988
Carver et al., 1985
Carver et al., 1986
Cavalieri et al.,
1981a
Chang etal., 2002
De Flora et al., 1984
Devanesan etal.,
1990
Dunkeletal., 1984
Eisenstadt and
Gold, 1978
Florin et al., 1980
Gibson et al., 1978
Gold and
Eisenstadt, 1980
Outline et al., 1982
Hassetal., 1981
Salmonella strain(s)
TA100, TA1527,
TA1538
TA100
TA100, TA1535,
TA98
TA98, TA100
TA98, TA100
TA100
TM677
TA100
TA1535, TA1537,
TA1538, TA98,
TA100
TA100, TA98
TA1535, TA1537,
TA1538, TA98,
TA100
TA1537, TA100
TA98, TA100
TA1535, TA1537,
TA1538, TA98
TA100
TA98, TA100
TA98, TA100
Activation system
Ar S9 and others
Guinea pig MC S9
and others
Rat MC S9
Rat Ar S9
S9
Ar rat and Ar hamster
S9
ArS9
Rat Ar S9
Rat AR S9
Rat Ar S9
Rat, mouse, hamster
ArS9
Rat Ar S9
Rat Ar and MC S9
Nonenzymatic
(gamma radiation)
Rat MC S9
Rat Ar S9 compare to
PGS from ram
seminal vesicles
Rat Ar S9
Positive result
AA, DBahA, DBajA, DBacA, BghiP,
BeP
DBaiP, BaA, DBacA, DBahA
BaA
PH, Pyr
Pery
BaA, BghiF, Pery
CPcdP, ACEP, Pyr
BghiF, BcPH
BaA, Pery, BeP
DBaeP, DBalP
BaA, BeP, PH, Pyr
CPcdP
BaA, CH, Pery, CO
BaA, BghiP, CH, FE, Pyr
CPcdP
BaA, CH

Nonpositive
result








AC

AC


DBahA, AC,
Pic, Tphen


BeP
Meets
selection
criteria?
Yes
Yes
Yes
Yes
No
Yes
Yes
Yes
Yes
No
No
Yes
Yes
Yes
Yes
No
Yes
Comments
TA100 results include BaP.


Qualitative data for other PAHs (no BaP);
quantitative data with BaP comparison for
PHandPyrinTAlOO.
The response varied at different
concentrations of S9; BaP was more potent
at low S9 while Pery was more potent at
high S9.
Qualitative data also presented for other
PAHs. S9 concentration varied;
400 uL/plate optimal.
BaP data from previous publication used.
Dose-response data not provided for Pyr.


No concurrent control.
Dose-response data not provided.


AN, PH also tested; toxicity interfered with
mutagenicity testing.
BaP and CPcdP maximal responses
occurred at different S9 levels.
BaP tested in TA98, BaA and CH tested in
TA100.

                                           73
DRAFT - DO NOT CITE OR QUOTE

-------
Table 4-9. Study summaries: in vitro bacterial mutagenicity with benzo[a]pyrene and at least one other PAH
Record
number
18650
10670
19000
24680
19320
19360
23650
15170
20220
20530
20450
20490
20560
20880
21000
21040
Reference
Hermann, 1981
Johnsen et al., 1997
Kadenetal., 1979
Lafleuretal., 1993
LaVoie et al., 1979
LaVoie et al., 1985
McCann et al., 1975
Norpothetal., 1984
Pahlman and
Pelkonen, 1987
Penman et al., 1980
Phillipson and
loannides, 1989
Poncelet et al., 1978
Probst etal., 1981
Rosenkranz and
Poirier, 1979
Sakaietal., 1985
Salamone et al.,
1979a
Salmonella strain(s)
TA98
TA98
TM677
TM677
TA98, TA100
TA98, TA100
TA1535, TA1537,
TA98, TA100
TA100
TA100
TM677
TA100
TA1530, TA1535,
TA1537, TA1538,
TA98, TA100
TA1530, TA1535,
TA1537, TA1538,
TA98, TA100
TA1530, TA1535
TA97, TA98, TA100
TA1535, TA1537,
TA1538, TA98,
TA100
Activation system
Rat Ar S9
Rat control or PB S9
Rat Ar or PB S9
ArPMS
Rat Ar S9
Rat Ar S9
Rat Ar S9
Rat and mouse S9;
induction by Clophen
ASOand 18PAHs
S9 from control, MC,
or TCDD treated rats
and mice
Rat Ar or PB S9
S9 isolated from
mouse, hamster, rat,
pig, and human
S9 (origin unknown)
Rat Ar S9
Uninduced rat S9
Rat Ar S9
Rat Ar S9
Positive result
BbA, BaA, CH, FA, Tphen, BeP,
DBacA, DBahA, BbF, Pery, DBalP,
DBaiP, AA, CO
BjAC, B1AC
AN, ANL, Pyr, BbFE, CPcdP, BaA, CH,
Tphen, FA, BeP, Pery, BghiP, AA,
DBacA, DBahA, DBbeF
CPcdP, APA, ACEA, CPhiAPA,
CPhiACEA
BeP, Pery

DBaiP, BeP, DBacA, DBahA, CH, BaA
BaA
BaA, CH, Tphen, DBacA, DBahA
Pery, CPcdP, DBacA
BaA, DBaiP, DBahA
CO, Tphen, FA, BghiP
BbA, DBacA

FE (equiv.), AC, PH, FA, CH, Pyr, BeP,
Pery, BghiP, CO
BaA, BeP (equiv.), BghiP, DBaiP, BPH,
CH, CO, DBacA, PCE
Nonpositive
result
AC, PH, FE,
Pyr, BbFE

FE, AC, PH, Pic,
CO


AC
Pyr, AC, PH, FE

AN, AC, PH,
FE, Pyr, BeP,
Pery, PCE


BbF
AC, DBahA,
PH, Pyr, DBaiP
AC, BaA, BeP,
CH, PH

AC, BaFE,
BbFE, FA, Pery,
Pyr
Meets
selection
criteria?
Yes
Yes
Yes
Yes
Yes
Yes
Yes
No
Yes
No
Yes
No
No
Yes
Yes
No
Comments


Mutagenic activity relative to BaP reported.

Several other PAHs were evaluated, but not
concurrent with BaP.


S9 composition was different for BaA and
BaP; result cannot be compared.

No concurrent control values were reported.

Qualitative data reported in published
abstract.
Data reported as minimum mutagenic
concentration (nmol/mL).


Increase in spontaneous mutation rate was
indicated, but dose data were not provided.
                                           74
DRAFT - DO NOT CITE OR QUOTE

-------
        Table 4-9.  Study summaries: in vitro bacterial mutagenicity with benzo[a]pyrene and at least one other PAH
Record
number
13260
11860
21360
21640
16180
16440
Reference
Salamone et al.,
1979b
Sangaiah et al.,
1983
Simmon, 1979a
Teranishi et al.,
1975
Utesch et al., 1987
Wood et al., 1980
Salmonella strain(s)
TA98, TA100
TA1535, TA1537,
TA1538, TA98,
TA100
TA1535, TA1536,
TA1537, TA1538,
TA98, TA100
TA1535, TA1536,
TA1537, TA1538
TA100
TA98, TA100
Activation system
Rat Ar S9
Rat Ar S9
Rat Ar S9
S9 from rats treated
with PB and MC or
DBahA
Intact or
homogenized
hepatocytes from Ar
treated rats
Rat Ar S9 and
purified MFO
enzymes system
Positive result
DBaiP
BjAC
BaA, BeP
DBaiP, DBaeP
BaA
CPcdP
Nonpositive
result


AC, CH, PH
DBahA, BaA,
BeP


Meets
selection
criteria?
No
Yes
Yes
Yes
Yes
Yes
Comments
Dose-response data were not completely
reported; maximal response information
(dose and number of revertants) was
presented in text; BaP max response at
different S9 than DBaiP.
Dose-response data for BaP was presented
for TA98 only.




Ar = Arochlor 1254-treated; MC = 3-methylcholanthrene-treated; PB = phenobarbital-treated; PMS = postmitochondrial supernatant
                                                           75
DRAFT - DO NOT CITE OR QUOTE

-------
Table 4-10. Study summaries: in vitro mammalian mutagenicity assays with benzo[a]pyrene and at least one
other PAH
Record
number
16900
16920
16930
16940
16910
13440
17140
24670
18260
14250
18750
18740
24120
18990
24720
Reference
Allen-Hoffmann and
Rheinwald, 1984
Amacher and Paillet,
1982
Amacher and Paillet,
1983
Amacher and Turner,
1980
Amacher et al., 1980
Bairdetal., 1984
Barfknecht et al., 1982
Durant et al., 1999
Gehlyetal., 1982
Hassetal., 1982
Huberman, 1975
Huberman and Sachs,
1976
Huberman and Sachs,
1974
Jotz and Mitchell,
1981
Kligerman et al., 1986
Cell type
Human epidermal
keratinocyte
Mouse lymphoma cells
(L5178Y)
Mouse lymphoma cells
(L5178Y)
Mouse lymphoma cells
(L5178Y)
Mouse lymphoma cells
(L5178Y)
V79 Chinese hamster cells
TK6 human lymphoblast cells
HlAlv2 human
lymphoblastoid cells
C3H/10T1/2 clone 8 mouse
fibroblast cells
V79 Chinese hamster cells
V79 Chinese hamster cells
V79 Chinese hamster cells
V79 Chinese hamster cells
Mouse lymphoma cells
(L5178Y)
Mouse lymphoma cells
(L5178Y)
Metabolic activation
None
Syrian golden hamster S9 mix or
cocultivated hamster hepatocytes
Cocultivated rat hepatocytes
S9 from eight rodent species or
strain; one rat strain induced by Ar
Rat Ar and noninduced S9
Hamster embryo cells
Rat Ar S9
Transfected with cyplal cDNA
None
Hamster embryo cells
Hamster cells
Hamster embryo cells
Hamster embryo cells
Rat Ar S9
Rat Ar S9
Mutagenesis assay
6-Thioguanine resistance (HPRT)
Trifluorothymidine resistance
(thymidine kinase locus [TK])
Trifluorothymidine resistance (TK)
Trifluorothymidine resistance (TK)
Trifluorothymidine resistance (TK)
6-Thioguanine resistance (HPRT)
Trifluorothymidine resistance (TK)
Trifluorothymidine resistance (TK)
Ouabain resistance (HPRT)
Ouabain and 6-thioguanine
resistance (HPRT)
8-Azaguanine resistance (HPRT)
Ouabain and 8-azaguanine
resistance (HPRT)
8-Azaguanine resistance (HPRT)
Trifluorothymidine resistance (TK)
Trifluorothymidine resistance (TK)
Positive
result

BaA

AC, BaA
BaA

FA, BaA,
CH, Tphen,
CPcdP
BaPery,
BbPery,
DBaeF,
DBafF,
DBahP,
DBaiP,
DBelP,
N23aP,
N23eP

DBaiP,
DBahP

DBacA,
DBahA
(both weak)

Pyr
B1AC
Non-
positive
result
BaA

BaA

AC, Pyr
BeP
PH, AC,
ACEP
DBjlF,
N12bF
BeP

BaA, Pyr
Pyr, PH,
CH, BaA
BaA


Meets selection
criteria?
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Comments



AC data not
useable; BaP
not
simultaneous.











                                          76
DRAFT - DO NOT CITE OR QUOTE

-------
Table 4-10. Study summaries: in vitro mammalian mutagenicity assays with benzo[a]pyrene and at least one
other PAH
Record
number
19180
24680
24170
7550
19870
20040
11450
15630
15640
21410
21720
21730
Reference
Krahn and
Heidelberger, 1977
Lafleur et al., 1993
Langenbach et al.,
1983
Li and Lin, 1996
Mishraetal., 1978
Myhr and Caspary,
1988
Nesnow et al., 1984
Raveh and Huberman,
1983
Raveh et al., 1982
Slagaetal., 1978
Tong et al., 1983
Tongetal., 1981b
Cell type
V79 Chinese hamster cells
MCL-3 human
lymphoblastoid cells
V79 Chinese hamster cells
HS1 HeLa cells (human
epithelial cells)
Fischer rat embryo cells
infected with Rauscher
leukemia virus
Mouse lymphoma cells
(L5178Y)
V79 Chinese hamster cells
V79 Chinese hamster cells
V79 Chinese hamster cells
V79 Chinese hamster cells
Rat liver epithelial cells
(ARL-18)
Rat liver epithelial cells
(ARL-18)
Metabolic activation
Rat MC S9
Transfected with cypla2 and
cyp2a6 cDNA
Cocultivation with primary rodent
cells from liver, lung, kidney, and
bladder
None
Rat Ar S9
Rat Ar and noninduced S9
Rat Ar S9
Hamster embryo fibroblasts
Hamster embryo fibroblasts
Hamster embryo cells

None
Mutagenesis assay
6-Thioguanine resistance (HPRT)
Trifluorothymidine resistance (TK)
Ouabain resistance (HPRT)
6-Thioguanine resistance (HPRT)
Ouabain resistance (HPRT)
Trifluorothymidine resistance (TK)
6-Thioguanine resistance (HPRT)
6-Thioguanine resistance (HPRT);
phorbol myristate acetate used to
enhance recovery
Ouabain and 6-thioguanine
resistance (HPRT)
Ouabain resistance (HPRT)
6-Thioguanine resistance (HPRT)
6-Thioguanine resistance (HPRT)
Positive
result
BaA,
DBacA,
DBahA
CPcdP,
ACEA,
CPhiACEA

BaA

AC, BaA,
BeP
B1AC,
BeAC,
BjAC
CPcdP
CPcdP
BaA (weak)


Non-
positive
result

APA,
CPhiAPA,
BghiF
AC

AC, PH,
Pyr, BeP


BaA


BaA, BeP,
Pyr
BeP, Pyr,
BaA
Meets selection
criteria?
Yes
Yes
Yes
Yes
Yes
No
Yes
Yes
Yes
Yes
No
Yes
Comments
DBacA and
DBahA data
not useable;
treatment
different than
BaP.




Results
reported as
ranges.


Mutagenicity
correlated with
skin tumor
initiation.

Repeats data
from
Record 21730
Tong et al.,
1981b

                                          77
DRAFT - DO NOT CITE OR QUOTE

-------
        Table 4-10. Study summaries:  in vitro mammalian mutagenicity assays with benzo[a]pyrene and at least one
        other PAH
Record
number
16190
21900
Reference
Vacaetal., 1992
Wangenheim and
Bolcsfoldi, 1988
Cell type
UV-sensitive Chinese hamster
ovary (CHO) cells
Mouse lymphoma cells
(L5178Y)
Metabolic activation
Rat Ar S9
Rat Ar S9
Mutagenesis assay
6-Thioguanine resistance (HPRT)
Trifluorothymidine resistance (TK)
Positive
result
FA
Pyr, FE
Non-
positive
result


Meets selection
criteria?
Yes
Yes
Comments


HPRT = hypoxanthine-guanine phosphoribosyl transferase mutagenicity assay (resistance to 6-thioguanine, 8-azaguanine, or ouabain); TK = thymidine kinase mutagenicity assay (resistance to
trifluorothymidine)
                                                              78
DRAFT - DO NOT CITE OR QUOTE

-------
Table 4-11. Study summaries: in vitro morphological/malignant cell transformation with benzo[a]pyrene and
at least one other PAH
Record
number
13390
17610
17730
24750
17970
17990
23630
18020
18080
23640
18260
14130
23890
14640
14700
14850
19870
24710
24700
7980
7990
8000
20120
Reference
Atchison et al., 1985
Casto, 1979
Chen and
Heidelberger, 1969
Davis, 1999
DiPaolo et al., 1969
DiPaolo et al., 1972
DiPaolo et al., 1973
Dunkeletal., 1981
Emuraetal., 1980
Evans and DiPaolo,
1975
Gehlyetal., 1982
Greb et al., 1980
Kakunaga, 1973
Krolewskietal., 1986
Laaksonen et al., 1983
Lubetetal., 1983
Mishraetal., 1978
Mohapatra et al., 1987
Nesnow et al., 1990
Nesnow et al., 1997
Nesnow et al., 1994
Nesnow et al., 1993a
Nesnow et al., 1991
Cell type
BALB/3T3 mouse embryo fibroblasts
Syrian golden hamster embryo cells
Adult C3H mouse ventral prostate cells
C3H10T1/2 cells
Syrian golden hamster embryo cells
BALB/3T3
Syrian golden hamster embryo cells
Balb/3T3, Syrian golden hamster embryo,
and Rauscher murine leukemia virus-
infected F344 rat embryo cells
Syrian golden hamster fetal lung cells
Strain 2 guinea pig fetal cells
C3H10T1/2CL8 mouse embryo fibroblasts
BHK 21/CL 13
BALB/3T3 subclone A3 1-714
C3H10T1/2CL8 mouse embryo fibroblasts
Newborn NMRI nu/nu nude mouse skin
fibroblasts
C3H10T1/2CL8 mouse embryo fibroblasts
Rauscher leukemia virus-infected Fischer
rat embryo
C3H10T1/2CL8 mouse embryo fibroblasts
Human neonatal foreskin fibroblasts
C3H10T1/2CL8 mouse embryo fibroblasts
C3H10T1/2CL8 mouse embryo fibroblasts
C3H10T1/2CL8 mouse embryo fibroblasts
C3H10T1/2CL8 mouse embryo fibroblasts
Metabolic activation system
None
None
Cocultivated irradiated C3H
mouse embryonic fibroblasts
None
Cocultivated irradiated
Sprague-Dawley rat fetal
cells
None
In vivo (transplacental)
exposure
None
None
None
None
Rat Ar S9
None
None
None
None
None
None
None
None
None
None
None
Positive result

DBahA
DBahA
DBalP, DBaeP,
BcC, BgC, BcPH
DBahA, BaA,
BeP, DBacA


BaA
BbF, BaA, IP


CH, BaA, BbF,
DBahA, BeP

CPcdP
BaA
BeP

Be AC, BjAC,
B1AC
B1AC
DBalP
DBahA
DBkmnoAPH

Nonpositive
result
FA, Pyr
Pyr
DBacA, Pyr

Pyr, PH
AC, Pyr
AC, PH, Pyr
BeP, PH, AC
BkF, BeP
AC, Pyr, PH
BeP
PH, AC
PH, Pyr

AC
AC, DBahA, PH
AC, PH, Pyr,
BeP
BkAC



DBjmnoAPH,
N123mnoAPH
ACEA
Meets selection
criteria?
Yes
Yes
No
No
Yes
Yes
No
Yes
Yes
No
Yes
Yes
No
Yes
Yes
Yes
No
Yes
Yes
Yes
Yes
Yes
Yes
Comments


Control data not provided.
Control data not provided.


No quantitative information.
Qualitative data only for R-MuLV-
RE cells. BaA positive in SHEM,
equivocal in Balb/3T3.

No quantitative information.


Not clear if BaP administered
simultaneously.



No quantitative information.






                                          79
DRAFT - DO NOT CITE OR QUOTE

-------
Table 4-11. Study summaries: in vitro morphological/malignant cell transformation with benzo[a]pyrene and
at least one other PAH
Record
number
23720
8490
Reference
Pienta et al., 1977
Sheuetal., 1994
Cell type
Syrian golden hamster embryo
BALB/3T3 A31-1-1
Metabolic activation system
Cocultivated X-irradiated
cells of same type
None
Positive result
BaA, DBahA

Nonpositive
result
CH, BeP, Pyr,
AC, DBacA, PH
Pyr, BaA, CH
Meets selection
criteria?
Yes
Yes
Comments


                                          80
DRAFT - DO NOT CITE OR QUOTE

-------
       Table 4-12. Study summaries:  in vitro DNA adducts with benzo[a]pyrene and at least one other PAH
Record
number
16890
6300
9510
6570
13780
22800
10660
10670
13200
7870
7990
20120
21200
24810
Reference
Allen and Coombs,
1980
Binkova et al., 2000
Bryla and Weyand,
1992
Cherngetal., 2001
Cooper et al., 1982
Grover and Sims, 1968
Johnsen et al., 1998
Johnsen et al., 1997
Li et al., 2002
Melendez-Colon et al.,
2000
Nesnow et al., 1994
Nesnowetal., 1991
Segerback and
Vodicka, 1993
Baird et al., 2002
Cell type or DNA source
Mouse embryo cells from TO mice
Human diploid lung fibroblast
cells
Calf thymus DNA
Human hepatoma HepG2 cells
Fibroblasts and epithelial cells
from Wistar rat mammary tissue
Salmon testes DNA
Human lymphocytes and human
promyelocytic HL-60 cells
Rat lung Clara cells, Type 2 cells,
and macrophages
MCF-7 cells or rat lung DNA
Human mammary carcinoma
MCF-7 cells and leukemia HL-60
cells
C3H10T1/2CL8 fibroblasts
C3H10T1/2 cells
Calf thymus DNA
MCF-7 cells
Incubation
time
24 hr
Various up to
24 hr
Ihr
24 hr
24 hr
Not specified
24 hr
2hr
7-24 hr
4 or 24 hr
24 hr
24 hr
3hr
24 hr
Activation system
None
None
None
None
None
Rat liver microsomes
None
PCB pretreatment of
whole animals
Human mammary
microsomes with rat
lung DNA
None
None
None
Rat Ar S9
Morpholinos
inhibition (antisense
oligomer that blocks
protein synthesis of
CYPIA1)
Method of analysis
[3H] prelabeling
[32P] postlabeling
[32P] postlabeling
[32P] postlabeling
[3H] prelabeling
[3H] prelabeling
[32P] postlabeling
[32P] postlabeling
[32P] postlabeling
[32P] postlabeling
[32P] postlabeling
[32P] postlabeling
[32P] postlabeling,
3H-binding
[32P] postlabeling
PAHs
evaluated"
BaA
DBalP
BaA, DBacA,
PH
BghiP
BaA
DBahA, DBacA,
BaA, Pyr, PH
BjAC, B1AC
BjAC, B1AC
DBalP, BcPH,
DBahA
DBalP
DBahA
ACEA
CH, BaA, BbF,
DBahA, FA,
BghiP, Pyr
DBalP
Meets selection
criteria?
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
No
Yes
No
No
Yes
No
Comments


PH did not form measurable
DNA adducts. Adduct
formation enhanced when
reacted under white light.
BghiP did not form measurable
DNA adducts.
BaA formed little or no
measurable DNA adducts.



No quantitative results.
No adducts formed in HL-60
cells that lack significant P450
activity.
No quantitative results.
Measures repair of adducts only,
not synthesis.

Confounded by CYP1A1
inhibition by morpholinos.
"Except where noted, positive findings were reported for all PAHs evaluated.
                                                      81
DRAFT - DO NOT CITE OR QUOTE

-------
Table 4-13. Study summaries:  in vitro DNA damage, repair, or synthesis with benzo[a]pyrene and at least one
other PAH
Record
number
16840
17610
24030
18030
23790
10670
10660
19270
19680
19690
19730
19740
23800
Reference
Agrelo and Amos,
1981
Casto, 1979
De Flora etal., 1984
Dunkel et al., 1984
Ichinotsubo etal.,
1977
Johnsen et al., 1997
Johnsenetal., 1998
Lake etal., 1978
Mamber et al., 1983
Mane et al., 1990
Martin and
McDermid, 1981
Martin etal., 1978
McCarroll et al.,
1981
Cell type
Human fibroblasts
Syrian golden hamster
embryo
Escherichia coli WP2,
WP67, and CM871
E. coli WP-2 uvrA
E. coli Rec BC
Rat lung Clara cells, Type 2
cells, and macrophages
Human lymphocytes and
human promyelocytic HL-
60 cells
Human foreskin epithelial
cells
E. coli WP2 and WP100
Human and rat mammary
epithelial cells
HeLa S3 cells
HeLa S3 cells
E. coli WP2, WP2 uvrA,
WP67, CM611, WP100,
W3110polA+, and
p3478pola-
Metabolic activation
Rat Ar S9
Intrinsic
Rat Ar S9
Rat, mouse, hamster
ArS9
S9 (origin unknown)
PCB pretreatment of
whole animals
Rat or human liver
microsomes
None
Rat Ar S9
None
PB-induced rat liver
postmitochondrial
supernatant
3-MC induced rat
liver
postmitochondrial
supernatant
Rat Ar S9
1 ml point
Unscheduled DNA
synthesis
Unscheduled DNA
synthesis
DNA damage
DNA damage
DNA damage
DNA damage
DNA damage
Unscheduled DNA
synthesis
DNA damage
Inhibition of DNA
synthesis
Unscheduled DNA
synthesis
Unscheduled DNA
synthesis
DNA damage
Assay
[3H] Thymidine
uptake
[3H] Thymidine
uptake
Differential
killing repair-
deficient strains
Differential
killing repair-
deficient strains

Alkaline elution
Alkaline elution
[3H] Thymidine
uptake
Growth
inhibition of
repair deficient
strains
[3H] Thymidine
uptake
[3H] Thymidine
uptake
[3H] Thymidine
uptake
Differential
killing repair-
deficient strains
Positive result
Pyr

AC, BaA
BaA, BeP, PH, Pyr
DBaiP, DBahA

BjAC, B1AC
DBahA

BaA (in human MEC
only)
Pyr (authors:
"dubious" result)
BeP, BaA, DBacA,
DBahA

Nonpositive
result

DBahA, Pyr,
PH
Pery, BeP
AC

BjAC, B1AC

AC, BeP, PH,
Pyr
AC, FE, Pyr
BeP
AC
Pyr, AC
AC, PH
Meets selection
criteria?
Yes
Yes
No
No
Yes
No
Yes
No
Yes
No
No
Yes
Yes
Comments


Semiquantitative
data.
Dose-response
data not
provided.

No untreated
control.

Doses reported
as ranges.

Positive response
for BaA not
observed
consistently.
No quantitative
information.


                                          82
DRAFT - DO NOT CITE OR QUOTE

-------
Table 4-13. Study summaries:  in vitro DNA damage, repair, or synthesis with benzo[a]pyrene and at least one
other PAH
Record
number
19830
19850
20050
20560
20810
23900
20880
20940
21380
21720
21730
21790
16190
22260
Reference
Mersch-
Sundermann et al.,
1992
Miloetal., 1978
Nagabhushan et al.,
1990
Probst etal., 1981
Robinson and
Mitchell, 1981
Rosenkranz and
Leifer, 1980
Rosenkranz and
Poirier, 1979
Rossman et al.,
1991
Simmon, 1979b
Tongetal., 1983
Tongetal., 1981b
Tweats, 1981
Vacaetal., 1992
Williams et al.,
1982
Cell type
E. coli PQ37
Human skin fibroblast NF
and Detroit 550 cells
Hamster buccal pouch
epithelial cells and tissue
fragments
Rat hepatocyte primary
culture
Human fibroblasts WI-38
cells
E. coli pol Al-
E. coli pol Al-
E. coli WP2s(X)
S. cerevisiae D3
Rat hepatocyte primary
culture
Rat hepatocyte primary
culture
E. coli WP2, WP67(uvrA
polA), CM871 (uvrAlexA
recA)
CHO cells
Rat hepatocyte primary
culture
Metabolic activation
Rat Ar S9
None
Not specified
None
Rat Ar S9
Rat liver S9
Uninduced rat S9
Rat liver S9
Rat Ar S9
None
None
Rat Ar S9
Rat Ar S9
None
1 ml point
Induction of SOS
system
DNA damage
Inhibition of DNA
synthesis
Unscheduled DNA
synthesis
Unscheduled DNA
synthesis
DNA damage
DNA damage
DNA damage
induced
recombination
Unscheduled DNA
synthesis
Unscheduled DNA
synthesis
DNA damage
DNA damage
Unscheduled DNA
synthesis
Assay
SOS chromotest
Alkaline elution
[3H] Thymidine
uptake
[3H] Thymidine
uptake
[3H] Thymidine
uptake
Differential
killing repair-
deficient strains
Differential
killing repair-
deficient strains
A prophage
induction
Colony
pigmentation on
adenine medium
[3H] Thymidine
uptake
[3H] Thymidine
uptake
Differential
killing repair-
deficient strains
Alkaline elution
[3H] Thymidine
uptake
Positive result
AA, BaA, BbF, BghiF,
BjF, BbFE, BghiP,
BeP, CH, DBacA,
DBahA, DBalP,
DBahP, DBaiP, FA,
IP, PH, Tphen


BbA, DBacA
Pyr (with activation)


AC, DBacA, DBahA,
PH

BaA
BaA

FA

Nonpositive
result
AC, BaFE,
CO, FE, Pery,
Pyr
AC, Pyr, PH,
BeP
BaA
AC, DBahA,
PH, Pyr,
DBaiP, FE,
BeP

AC, BaA, BeP,
CH, PH
AC, BaA, BeP,
CH, PH
BeP, FA, Pyr
AC, BaA, BeP,
CH, PH
BeP, AC, CH,
Pyr
BeP, AC, CH,
Pyr
Pyr, AC

Pyr, BeP
Meets selection
criteria?
Yes
Yes
No
No
Yes
Yes
Yes
Yes
Yes
No
Yes
No
No
No
Comments


Abstract only.
BaA inhibited
synthesis 4%.
Artifact of
counting method
resulted in
control responses
reported as
negative values.





Repeats data
from21730Tong
etal., 1981b.

No quantitative
information.
No untreated or
vehicle control.
No quantitative
information.
                                          83
DRAFT - DO NOT CITE OR QUOTE

-------
Table 4-14. Study summaries: in vitro clastogenicity or sister chromatid exchange with benzo[a]pyrene and at
least one other PAH
Record
number
16740
17890
17930
18120
23640
18260
14620
14640
19690
19770
20020
20340
Reference
Abe and Sasaki, 1977
Dean, 1981
DeSalvia et al., 1988
Evans and Mitchell,
1981
Evans, and DiPaolo,
1975
Gehlyetal., 1982
Kochhar, 1982
Krolewski et al., 1986
Mane et al., 1990
Matsuokaetal., 1979
Murison, 1988
Perry and Thomson,
1981
Cell type
Pseudodiploid Chinese
hamster D-6
Near-diploid epithelial-
type rat liver RLi
Male Chinese hamster
liver epithelial cells
CHO
Diploid strain 2 guinea
pig fetal cells
CH3/10T1/2 clone 8
mouse fibroblasts
Chinese hamster V79
CH3/10T1/2 clone 8
mouse embryo cells
Chinese hamster V79
cells
Male Chinese hamster
lung
P3 clonal isolate from
human epithelial
teratocarcinoma
CHO cells
Metabolic activation
None
None
None
Rat Ar S9
None
None
None
None
With and without rat mam-
mary epithelial cell coculture
Rat Ar S9
BJ-0 1 5 human breast
epithelial cell coculture
Rat Ar S9
Clastogenic
endpoint(s)
Aberrations and sister
chromatid exchanges
Various aberrations
Sister chromatid
exchanges
Sister chromatid
exchanges
Aneuploidy
Sister chromatid
exchanges
Aberrations including
gaps, rings, breaks,
fragments, exchanges
Sister chromatid
exchanges
Sister chromatid
exchanges
Aberrations and sister
chromatid exchanges
Sister chromatid
exchanges
Sister chromatid
exchanges
Positive
results



Pyr (with
activation)


BaA
CPcdP
BaA

CPcdP
Pyr
Non-
positive
results
AC, Pyr
AC, Pyr
Pyr, FA

AC
BeP


BeP
PH
BeP
AC
Meets selection
criteria?
Yes
No
Yes
No
No
Yes
Yes
Yes
Yes
No
No
No
Comments

Semiquantitative results.

No untreated or vehicle control.
No quantitative data. Pyr, PH also
evaluated using different protocol without
BaP reference.

Dose-dependent increase in the percentage
cells with aberrations.
CPcdP appears to increase sister chromatid
exchanges in dose-dependent fashion (two
doses).

Not clear if BaP administered simultane-
ously. No untreated control.
Not clear if BaP administered
simultaneously; no concurrent control.
No untreated control.
                                           84
DRAFT - DO NOT CITE OR QUOTE

-------
Table 4-14. Study summaries: in vitro clastogenicity or sister chromatid exchange with benzo[a]pyrene and at
least one other PAH
Record
number
20500
21710
21720
8780
8850
21980
Reference
Popescu et al., 1977
Tongetal., 1981a
Tongetal., 1983
Vienneau et al., 1995
Warshawsky et al.,
1995
Weinstein et al., 1977
Cell type
Chinese hamster
V79-4 cells
Adult rat liver
epithelial (ARL 18)
cells
Adult rat liver
epithelial (ARL 18)
cells
UDP-Glucuronosyl-
transferases-deficient
rat (RHA-J/J) skin
fibroblasts
Human lymphocytes
Human diploid
fibroblasts (WI-38)
Metabolic activation
With or without irradiated
Syrian golden hamster
secondary embryo feeder cells
None
None
None
None
With or without rat Ar s9
Clastogenic
endpoint(s)
Aberrations and sister
chromatid exchanges
Sister chromatid
exchanges
Sister chromatid
exchanges
Micronuclei
Micronuclei and sister
chromatid exchanges
Chromosomal
damage, mitotic
index, abnormal
metaphases
Positive
results
Pery, Pyr
BaA
BaA



Non-
positive
results
PH
BeP, Pyr,
AC
BeP, Pyr,
AC
BeP
BaA
Pyr
Meets selection
criteria?
No
Yes
No
Yes
Yes
Yes
Comments
BaP increased sister chromatid exchanges
but Pyr and Pery increased aberrations.
Pery increased aberrations w/o activation.
60% of Pyr treated cells (activated)
polyploid. Increased aberrations in
polyploid cells.

Repeats data from Record 21710 Tong et
al., 1981a.



                                           85
DRAFT - DO NOT CITE OR QUOTE

-------
 1          If the above criteria were met, studies were selected for use in the analysis regardless of
 2   whether positive or nonpositive results were reported.  Studies with positive findings were used
 3   for calculation of RPFs. Studies with nonpositive findings were used in a weight of evidence
 4   evaluation for selecting PAHs for inclusion in the RPF approach (discussed later in Section 6.1).
 5   To be considered adequate for use in the analysis, nonpositive bioassays were selected only if
 6   two additional conditions were met: (1) at least 20 animals were used per dose group, and
 7   (2) animals were observed for at least 6 months. More strict criteria were applied to nonpositive
 8   studies due to the difficulty in demonstrating the absence of an effect.  For example, if a positive
 9   tumor response (i.e., statistically significant increase in incidence) was observed after 3 months
10   of treatment with a given PAH,  the positive finding is clear; however,  if no response (or a
11   nonsignificant response) was observed after 3 months, the absence of response might reflect a
12   lack of carcinogenic action, but might also have resulted from inadequate follow-up time. The
13   use of these additional criteria for nonpositive studies served to ensure that PAHs would not be
14   treated as noncarcinogenic based on inadequate nonpositive bioassays.
15          Study design details, findings, limitations, and a determination of whether the study met
16   selection criteria are presented in Tables 4-1 through 4-14 for each study reviewed in each
17   category.  Except where noted, positive and nonpositive findings reported in the table are based
18   on the author's determination. When statistical analysis of tumor bioassay data was not included
19   in the pertinent publication, statistical analysis was conducted to determine whether the response
20   differed from control.  In the sections that follow, overviews of the data available in each
21   category are presented. The overviews address the nature of the studies available, concise
22   information on  general study methods, general findings for the tested compounds, and key
23   strengths and limitations of the available data for relative potency development.
24
25   4.3.1. In Vivo  Cancer Bioassays in Animals
26          The PAH database contained a large number of cancer bioassay studies in which one or
27   more PAHs was evaluated along with benzo[a]pyrene. The vast majority of the tumor bioassay
28   studies were  mouse skin painting studies (n = 43). In addition, there were 12 intraperitoneal
29   studies, 9 subcutaneous exposure studies, 3 oral studies, and 9 studies  using miscellaneous
30   exposure routes.
31
32   4.3.1.1. Dermal Exposure
33          A summary of the 43 dermal bioassays is  provided in  Table 4-1. These studies were all
34   conducted in mice. Fifteen studies tested the complete carcinogenicity of PAHs, while
35   23  studies tested PAHs as initiators in initiation-promotion protocols.  In some cases, both
36   complete and initiation-promotion studies were reported in the same reference. For these
37   references, two entries are included in the table.
                                             86           DRAFT - DO NOT CITE OR QUOTE

-------
 1          Complete carcinogenicity studies were conducted in mice using either dropper or
 2   paintbrush application. Swiss mice were typically preferred for these studies. PAHs, usually in
 3   acetone, were applied to the shaved interscapular skin 2 or 3 times/week.  The duration of
 4   exposure varied from 10 weeks up to about 70 weeks; most studies continued exposure for at
 5   least 30 weeks. Skin tumor counts were recorded on a weekly basis, and animals were sacrificed
 6   when tumors reached a minimum size (e.g., 2 cm) or when the animals were moribund.  These
 7   studies generally focused exclusively on skin papillomas and  carcinomas. Skin tumor data were
 8   reported as incidence (i.e., number of animals with tumors) and/or tumor count (mean number of
 9   tumors per animal) (indicated in Table 4-1).
10          Several PAHs consistently (in two or more studies) proved to be complete carcinogens in
11   mouse skin painting assays, including benzo[b]fluoranthene, benzo[j]fluoranthene,
12   cyclopenta[c,d]pyrene, dibenzo[a,e]pyrene, dibenzo[a,h]pyrene, dibenzo[a,i]pyrene, and
13   dibenzo[a,l]pyrene.  Chrysene gave positive results in two complete carcinogenicity studies
14   (LaVoie et al., 1979; Wynder and Hoffmann, 1959) and equivocal results in a third (Hecht et al.,
15   1974). Anthanthrene, dibenzo[a,e]fluoranthene, and dibenz[a,h]anthracene each gave positive
16   turnorigenicity results in a single assay (Cavalieri et al., 1977; Hoffmann and Wynder, 1966; and
17   Wynder and Hoffmann, 1959; respectively). Nonpositive or equivocal results were reported for
18   benzo[k]fluoranthene, benzo[g,h,i]fluoranthene, dibenzo[e,l]pyrene, indeno[l,2,3-c,d]pyrene,
19   benzo[g,h,i]perylene, naphtho[2,3-e]pyrene, anthracene, pyrene, fluoranthene, 2,3-acepyrene,
20   benz[a]anthracene, coronene, and benzo[e]pyrene (see Table 4-1).
21          According to LaCassagne et al. (1968), in studies conducted prior to 1966, the compound
22   reported as dibenzo[a,l]pyrene was actually dibenzo[a,e]fluoranthene.  In the text and tables of
23   this report, data from Hoffmann and Wynder (1966) are reported as dibenzo[a,e]fluoranthene in
24   Table 4-1.
25          The initiation studies in Table 4-1 were performed under a generally consistent protocol,
26   as follows. During the early part of the second telogen phase of the hair cycle (at about 7-
27   8 weeks of age), PAHs in acetone were applied to the shaved  interscapular skin of mice.  In
28   general, female Swiss, CD-I, or SENCAR mice were used. Some studies used dropper
29   administration, but the majority employed a painting method using a camel's hair brush.  About
30   half of the initiation studies used a single initiation dose, while the other half administered the
31   initiating compound in 10 subdoses given every other day. One to 2 weeks after the final
32   initiating dose, promotion was begun with twice or thrice weekly applications of a promoting
33   agent, usually TPA or croton oil. The dose of the promoting agent varied by study.  Promotion
34   usually continued for about 20 weeks (with a range across studies from 11 to 26 weeks).  The
35   incidence of skin papillomas was recorded on a weekly basis until the promotion period was
36   ended. Papillomas were removed at random for histological verification.  Some studies reported
37   the number of tumors per animal; some reported only the incidence.
                                            87          DRAFT - DO NOT CITE OR QUOTE

-------
 1          The initiation studies in Table 4-1 consistently showed positive tumorigenicity across two
 2   or more studies for the following compounds: benzo[j]fluoranthene, benzo[b]fluoranthene,
 3   chrysene, dibenzo[a,e]pyrene, dibenzo[a,h]pyrene, dibenzo[a,i]pyrene, dibenzo[a,l]pyrene, and
 4   cyclopenta[d,e,f]chrysene. In at least one study, benzo[k]fluoranthene, benz[l]aceanthrylene,
 5   benz[e]aceanthrylene, naphtho[2,3-e]pyrene, dibenz[a,h]anthracene, dibenz[a,c]anthracene, and
 6   benz[b,c]aceanthrylene showed positive initiating activity. Nonpositive results were reported for
 7   pyrene, perylene, benzo[g,h,i]fluoranthene, fluoranthene,  anthanthrene, dibenzo[e,l]pyrene,
 8   benzo[g,h,i]perylene, indeno[l,2,3-c,d]pyrene, benzo[e]pyrene, anthracene, 2,3-acepyrene, and
 9   phenanthrene. Cyclopenta[c,d]pyrene gave nonpositive results in one study (Wood et al., 1980)
10   and positive results in two studies (Raveh et al.,  1982; Cavalieri et al., 1981b) (see Table 4-1).
11          The vast majority of the initiation and complete carcinogenicity studies were conducted
12   in female mice; thus, data on gender differences in skin tumor susceptibility are not available.
13          A few studies using dermal application (Warshawsky et al., 1993; Slaga et al.,  1979; Van
14   Duuren and Goldschmidt, 1976; Horton and Christian, 1974; Van Duuren et al., 1973) were
15   designed to evaluate the cocarcinogenicity of two or more PAHs, or of a single PAH with
16   dodecane as a vehicle. These were primarily complete carcinogenicity studies, wherein PAHs
17   were administered together over a chronic time period, although Slaga et al. (1979) used an
18   initiation-promotion design.  Study design was similar to  other complete carcinogenicity
19   experiments.  In these studies, the carcinogenicity of single PAHs was evaluated for comparison
20   with the results obtained when the PAHs were administered with a cocarcinogen.  Data on single
21   PAHs (without a cocarcinogen) were generally limited to single dose levels. In the
22   cocarcinogenesis studies, only dibenz[a,c]anthracene,  benzo[e]pyrene, and pyrene gave positive
23   results when administered without a cocarcinogen; results for pyrene were judged to be
24   equivocal in the absence of statistical confirmation. The PAHs chosen for cocarcinogenesis
25   studies were often those traditionally understood to be nontumorigenic or weakly  tumorigenic
26   when administered alone (e.g., perylene, pyrene, benzo[e]pyrene, benzo[g,h,i]perylene,
27   phenanthrene, fluoranthene).
28          Several  issues relating to the potential use of the dermal bioassay data for relative
29   potency development were identified during study review. Several studies did not include a
30   concurrent untreated or vehicle-treated control group (Masuda and Kagawa, 1972; Bingham and
31   Falk, 1969; Wynder and Hoffmann,  1959a, b).  In a number of reports, it appears that bioassays
32   were done in batches and reported in a single publication.  In these cases, it appears that
33   benzo[a]pyrene treatment may not have been undertaken concurrently with all of the compounds
34   in the report.  For some of these studies (Horton and Christian, 1974; Bingham and Falk, 1969),
35   there are differences in the choice of vehicle or promoter, or other issues that argue against using
36   the benzo[a]pyrene data for direct comparison.  In several other studies, however (Rice et al.,
37   1988; Slaga et al., 1980; Van Duuren and Goldschmidt, 1976; Wynder and Hoffmann, 1959), the
38   protocols (including vehicle and promoting agent) appear to have been the same.

                                             88           DRAFT - DO NOT CITE OR QUOTE

-------
 1          Among the dermal tumor bioassay studies in Table 4-1, 24 studies met the selection
 2   criteria for use in this analysis.
 o
 J
 4   4.3.1.2. Intraperitoneal Exposure
 5          Twelve cancer bioassay s in the literature used intraperitoneal injection.  Six of these
 6   studies were carried out in newborn mice, while the other six used adult A/J mice. The studies
 7   were focused on lung and liver tumorigenicity after PAH exposure; one study also examined
 8   forestomach lesions.  Study summaries for all of these references are reported in Table 4-2.
 9   Tumor data were reported as incidence (i.e., number of animals with tumors) and/or tumor count
10   (mean number of tumors per animal) (indicated in Table 4-2).
11          Newborn mouse studies.  Six cancer bioassays in newborn mice were identified (LaVoie
12   et al., 1994, 1987; Busby et al., 1989,  1984; Weyand and LaVoie, 1988; Wislocki et al., 1986).
13   In general, PAHs were administered intraperitoneally to newborn mice (usually of the Swiss or
14   CD-I strains). The dosing schedule called for l/7th, 2/7ths, and 4/7ths of the total dose to be
15   administered  on the 1st, 8th, and 15th days of life. Typically, the mice were sacrificed at  either
16   6 months or 1 year, and lung and/or liver tumors were identified and classified.
17          The studies in newborn mice showed a distinct gender difference in liver tumorigenicity.
18   Male mice appear to be substantially more susceptible to liver tumor induction than females. In
19   contrast, both male and female mice developed lung tumors after exposure.  Three studies
20   (LaVoie et al., 1994; Busby et al., 1989, 1984) reported that fluoranthene induced lung tumors in
21   both male and female mice, while one study reported that fluoranthene induced liver tumors in
22   male mice only (LaVoie et al., 1994).  LaVoie et al. (1987) reported that benzo[b]fluoranthene
23   and benzo[j]fluoranthene induced lung adenomas in both male and female mice, but induced
24   liver tumors only in males.  Wislocki et al. (1986) reported that treatment with benz[a]anthracene
25   resulted in a significant increase in liver tumors in male mice. In this study, benz[a]anthracene
26   treatment resulted in an increased incidence of lung tumors in both males and females, although
27   the tumor  incidence was significantly increased only for females. The same authors (Wislocki et
28   al., 1986) reported a significant increase in liver tumors in male mice treated with chrysene, but
29   no increase in lung tumorigenicity.  The lack of lung tumorigenicity in mice treated with
30   chrysene was also reported by Busby et al. (1989).
31          Nonpositive tumorigenicity results in newborn mouse assays were reported for pyrene,
32   chrysene, benzo[k]fluoranthene, and indeno[l,2,3-c,d]pyrene (Busby et al., 1989; LaVoie et al.,
33   1987).
34          Most of the data from the newborn mouse assays met the  criteria for relative potency
35   development, although Weyand and LaVoie (1988) is an abstract and does not provide dose-
36   response information.  LaVoie et al. (1994) noted that liver tumorigenicity in newborn mice
37   exposed to weak tumorigenic agents may not be fully realized for 12 months; thus, the failure to
                                            89           DRAFT - DO NOT CITE OR QUOTE

-------
 1   observe liver tumors in studies of shorter duration (Busby et al., 1989, 1984) may result from the
 2   longer latency and should be taken into consideration in using these data.
 3          Lung adenoma A/J mouse studies.  Six studies (Nesnow et al., 1998a, b, 1996, 1995; Ross
 4   et al., 1995; Mass et al., 1993) were carried out in 6- to 8-week-old A/J mice by the same
 5   laboratory using a standard protocol (Table 4-2). Mice were given a single intraperitoneal
 6   injection of PAH in tricaprylin and followed for 8 months. Upon sacrifice, the lungs were
 7   removed and adenomas were counted. Tumor multiplicity was reported, while tumor incidence
 8   was not.  Several of these studies include estimates of relative potency based on statistical
 9   analysis of the tumor multiplicity data. These studies report positive tumor findings (reported as
10   an increase in the number of tumors per animal) for all of the PAHs tested (benz[j]aceanthrylene,
11   benzo[b]fluoranthene, dibenz[a,h]anthracene, cyclopenta[c,d]pyrene, and dibenzo[a,l]pyrene).
12   One additional study by a different group (Weyand et al.,  2004) used the same study design to
13   assess effects of benzo[c]fluorene. In this study, both lung adenomas and forestomach lesions
14   were evaluated after 8 months. Both benzo[c]fluorene and benzo[a]pyrene were associated with
15   increased incidences of lung adenomas but not with increased forestomach lesions.
16          Among the intraperitoneal tumor bioassay studies  in Table 4-2, nine studies met the
17   selection criteria for use in this analysis.
18
19   4.3.1.3. Subcutaneous Injection Exposure
20          Nine studies employing a subcutaneous exposure design were identified.  All of the
21   subcutaneous exposure studies are more than 25 years old; the most recent is Pfeiffer (1977).
22   Study descriptions are presented in Table 4-3.
23          Two studies utilized newborn mice (Roe and Waters,  1967; Grant and  Roe,  1963). In
24   these  studies, phenanthrene was administered subcutaneously to newborn albino mice on the first
25   day of life.  Ten mice of each group were sacrificed after 52 weeks, and the remaining animals
26   were sacrificed at 62 weeks.  Grant and Roe (1963) evaluated lung tumorigenicity and observed
27   no increase with phenanthrene, while Roe and Waters (1967) reported liver tumors in the same
28   group of mice. Roe and Waters (1967) reported an elevated incidence of liver tumors in male
29   mice exposed subcutaneously to phenanthrene; however, it is not clear whether the difference
30   was significant. Roe and Waters (1967) is a brief communication with limited details of the
31   study design and results.
32          In most of the remaining studies, single subcutaneous doses of one or more PAH and
33   benzo[a]pyrene were administered to mice, followed 1-2.5 years later by an evaluation of
34   injection site and other tumors. Tumors at the injection site were most commonly reported;
35   however, in some studies, investigators also examined other organs for tumors (Homburger et
36   al., 1972; Roe and Waters, 1967; Grant and Roe, 1963; Rask-Nielsen, 1950; Pfeiffer and Allen,
37   1948).
                                            90          DRAFT - DO NOT CITE OR QUOTE

-------
 1          Most of the subcutaneous bioassays suffer from critical shortcomings in design or
 2   reporting. One study used "aged" mice for controls, allowing these animals to live 16 weeks
 3   longer than the treated group (Homburger et al., 1972). Three studies gave apparently positive
 4   results for dibenz[a,h]anthracene (i.e., substantial tumor induction) (Pfeiffer, 1977; Steiner, 1955;
 5   Bryan and Shimkin, 1943). However, neither Bryan and Shimkin (1943) nor Steiner (1955)
 6   included untreated control groups.  Pfeiffer (1977) included an untreated control group in which
 7   there was 90% mortality prior to sacrifice of the treated animals; data on tumor incidence in
 8   controls were not reported.  Several other studies (Pfeiffer and Allen, 1948;  Barry et al., 1935)
 9   also did not include a concurrent untreated or vehicle-treated  control  group.  These studies were
10   not used for dose-response assessment due to the lack of appropriate  controls.
11          Fundamental flaws were observed in two older studies. Pfeiffer and Allen (1948)
12   examined the effects of PAHs in Rhesus monkeys.  Individual animals were exposed
13   sequentially to several PAHs via multiple exposure routes; thus, the effect of any individual PAH
14   or benzo[a]pyrene cannot be discerned. Barry et al. (1935) treated mice with PAHs from varying
15   sources and of varying purity. Given the age of the study and the attendant issues with
16   nomenclature, purity, and analysis of the  treatment compounds, data from this study are excluded
17   from use in relative potency development.
18          Among the subcutaneous tumor bioassay studies in Table 4-3, only a single study met
19   selection criteria for use in this analysis.
20
21   4.3.1.4. Oral Exposure
22          The literature search identified three oral bioassays that included benzo[a]pyrene and at
23   least one other PAH.  Critical aspects of the study design for these  studies are reported in
24   Table 4-4.
25          Biancifiori and Caschera (1962) compared the induction of mammary tumors in virgin
26   and pseudopregnant mice (female mice mated with vasectomized males) after gavage exposure
27   to dibenz[a,h]anthracene or benzo[a]pyrene. Tumor incidence was increased in pseudopregnant
28   mice given 1 mg/week of either  compound  for 15 weeks, but  not in virgin mice given the same
29   dose. The relevance of the positive findings in pseudopregnant mice is uncertain given that an
30   increased incidence of tumors was not observed in virgin mice treated at the same dose. One
31   possible explanation for the disparate findings is that circulating hormones in pseudopregnant
32   mice differed from those in virgin mice and interacted with the PAH to enhance tumor
33   formation.  Huggins and Yang (1962) also evaluated mammary tumor incidence after a single
34   oral PAH exposure. Sprague-Dawley rats were given gavage doses of benzo[a]pyrene,
35   benz[a]anthracene, or phenanthrene. This study did not include an untreated or vehicle-treated
36   control group. No tumors were  observed in the rats treated with either benz[a]anthracene or
37   phenanthrene, while mammary tumors were observed in eight of the nine benzo[a]pyrene-treated
38   animals.

                                             91           DRAFT - DO NOT CITE OR QUOTE

-------
 1          Weyand et al. (2004) conducted an oral bioassay in which female A/J mice were fed diets
 2   containing benzo[c]fluorene or benzo[a]pyrene throughout the study.  At sacrifice after 260 days,
 3   lung adenomas were counted and forestomach lesions were characterized.  Exposure to
 4   benzo[c]fluorene and benzo[a]pyrene resulted in significantly increased incidences of lung
 5   adenomas, but only benzo[a]pyrene exposure resulted in forestomach neoplasms. This was the
 6   only oral study that met the selection criteria for use in this analysis.
 7
 8   4.3.1.5. Other Routes
 9          Nine bioassays were available that did not fit into other exposure route categories (i.e.,
10   dermal, intraperitoneal, subcutaneous, or oral) (see Table 4-5). Among these were studies using
11   intramammillary,  intramuscular, and intravenous injection as well as lung implantation, tracheal
12   implantation, and transplacental exposure after subcutaneous injection. Seven studies were in
13   rats, with one each in mice and hamsters.
14          Deutsch-Wenzel et al. (1983) and Wenzel-Hartung et al. (1990) implanted
15   PAH-containing pellets (consisting of beeswax and trioctanoin) into the lungs of inbred female
16   Osborne-Mendel rats.  Lung tumor incidence was reported for a total of 10 PAHs and
17   benzo[a]pyrene.  The authors reported relative potency estimates based on the lung tumor data.
18   Lung tumors were induced by benzo[b]fluoranthene, benzo[j]fluoranthene, benzo[k]fluor-
19   anthene, benzo[g,h,i]perylene, indeno[l,2,3-c,d]pyrene, anthanthrene, chrysene, and
20   dibenz[a,h]anthracene.  Nonpositive findings were reported for benzo[e]pyrene and
21   phenanthrene.
22          Cavalieri et al. (1991) treated Sprague-Dawley rats with single intramammillary
23   injections of dibenzo[a,l]pyrene into the left mammary glands and followed them for up to
24   24 weeks. Tumors of the mammary gland,  mesenchymal tissue, or skin were recorded.
25   Dibenzo[a,l]pyrene produced tumors in all animals at both doses.
26          In six studies, tumors were not induced after exposure to any target PAH.
27   Intramammillary injection of dibenz[a,h]anthracene and benz[a]anthracene did not induce
28   mammary tumors in rats (Cavalieri et al., 1988b). Pregnant mice receiving subcutaneous
29   injection of pyrene did not develop tumors, nor did their offspring (Nikonova, 1977).  Rats
30   treated either intravenously or intramuscularly with benz[a]anthracene did not develop either
31   mammary or injection site tumors (Pataki and Huggins, 1969).  Similarly, benz[a]anthracene was
32   not tumorigenic after intramuscular injection  in rats (Sugiyama, 1973) or buccal pouch painting
33   in hamsters (Solt et al., 1987).  Finally, benzo[e]pyrene was not tumorigenic when it was
34   implanted into tracheas transplanted subcutaneously into isogenic rats (Topping et al.,  1981).
35          Among the tumor bioassays that used alternative exposure routes in Table 4-5,  four
36   studies met the selection criteria for use in this analysis.
37
                                             92          DRAFT - DO NOT CITE OR QUOTE

-------
 1   4.3.2. In Vivo Studies of Cancer-Related Endpoints
 2          The database of cancer-related endpoints measured after in vivo exposure to PAHs is
 3   much smaller than the in vitro database. Endpoints examined after in vivo exposure include
 4   mutagenicity, DNA adducts, and clastogenicity or sister chromatid exchange. As with the in
 5   vitro database, only studies of selected PAHs that included benzo[a]pyrene as a reference
 6   compound were reviewed. Each study that was reviewed for consideration in relative potency
 7   development is presented in tabular format in subsequent sections. The tables summarize study-
 8   specific information and indicate whether a particular study is considered useful for dose-
 9   response assessment. The text provides an overall description of the available studies, including
10   a general description of the methodology used for each study type, the results, and the
11   weaknesses or problems associated with specific studies or study types.
12
13   4.3.2.1. DNA Adducts
14          Nineteen studies evaluating DNA adduct formation for PAHs and benzo[a]pyrene were
15   identified in the database (Table 4-6). Nine studies presented quantitative data for DNA adduct
16   formation and are discussed below. Among studies with data potentially useful for RPF
17   derivation, the route of exposure was  intramammillary injection in one study (Arif et al., 1997),
18   intraperitoneal injection in seven studies (Weyand et al., 2004; Kligerman et al., 2002; Nesnow
19   et al., 1998a, 1996,  1995; Ross et al.,  1995; Mass et al., 1993), dermal in three studies (Hughes
20   and Phillips, 1990; Cavalieri et al., 1981b; Phillips et al., 1979), and oral in two studies (Weyand
21   et al., 2004; Kligerman et al., 2002).  Adducts were identified by [32P]-postlabeling in all of the
22   studies except for two by Phillips et al. (1979) and Cavalieri et al. (1981b), which utilized
23   [3H]- or [14C]-radiolabeled PAHs. Three papers described experiments with a single time
24   point(s) at 24 or 48 hours or  14 days (Weyand et al., 2004; Arif et al., 1997; Hughes and Phillips,
25   1990), whereas the rest had multiple time points. The duration of exposure was as short as
26   4 hours (Cavalieri et al., 1981b), although 24 hours was usually the first time point(s) in time-
27   course studies. The longest duration for a time-course study was 84 days (Hughes and Phillips,
28   1990), but most were <3 weeks. The  tissues evaluated included mammary epithelium (Arif et
29   al., 1997), skin (Hughes and  Phillips,  1990; Cavalieri et al., 1981b; Phillips et al., 1979), liver
30   and peripheral blood lymphocytes (Kligerman et al., 2002; Nesnow et al., 1993b), lung (Weyand
31   et al., 2004; Nesnow et al., 1998a, 1993b;  Arif et al.,  1997; Ross et al.,  1995; Mass et al., 1993;
32   Hughes and Phillips, 1990), and forestomach (Weyand et al., 2004).
33          Dermal exposure studies typically  involved application of the chemical in solution to the
34   shaved dorsal skin of mice (Hughes and Phillips, 1990; Cavalieri et al., 1981b; Phillips et al.,
35   1979). After the scheduled sacrifice,  the treated skin was excised and frozen; a scalpel was used
36   to scrape away the dermis from the epidermis that was subsequently powdered in liquid nitrogen.
37   In one study, the lung was also excised and frozen in liquid nitrogen  (Hughes and Phillips,
38   1990). DNA was isolated from the frozen epidermis or lung.  Liquid scintillation counting was

                                             93           DRAFT - DO NOT CITE OR QUOTE

-------
 1   used to quantify DNA adducts to PAH labeled with [3H] or [14C] (Cavalieri et al., 1981b; Phillips
 2   etal., 1979).  For [32P]-postlabeling, DNA was treated to selectively dephosphorylated
 3   nonadducted nucleotides; after postlabeling, adducts were resolved by sequential anion-exchange
 4   thin layer chromatography on polyethyleneimine-cellulose plates in several directions using three
 5   solvents (Hughes and Phillips, 1990).  Adduct spots on chromatograms were located by
 6   autoradiography, after which the spots were excised and radioactivity levels were determined by
 7   Cerenkov counting.
 8          Most studies reported the mean number of adducts formed within a tissue per unit of
 9   DNA, with time-course data displayed graphically.  Peak values were sometimes called out
10   specifically in the text or tables.  As the shapes of dose-response curves differ among different
11   PAHs, the peak value is an imprecise measure for comparing the relative adduct-forming
12   potency of the different compounds. The TIDAL has also been used for reporting results for a
13   time-course study (Ross et al., 1995).  The TIDAL value is the area under the curve (AUC) for
14   adduct persistence  (based on the rate of adduct formation and repair) for the duration of the
15   study.  The TIDAL value expresses the total DNA adduct burden experienced by the tissue from
16   the time of treatment to the end of the study. The TIDAL versus administered dose curve
17   provides a convenient way to compare adduct-forming potency for different PAHs in time-
18   course experiments. An important limitation of the TIDAL approach is the inherent assumption
19   that the ratios of specific adducts are relatively constant across dose and  time course. Ross et al.
20   (1995) demonstrated that this assumption was valid for several different  PAHs; however, it was
21   also noted that two adducts of benzo[a]pyrene in rat liver did not conform to this general pattern.
22          Ross et al. (1995) presented data for lung adenoma incidence (measured at 8 months) in
23   several ways: as a function of administered dose, as a function of adduct levels per dose
24   measured 24 hours after dosing (results for 3 days postdosing were mentioned but not shown),  as
25   a function of TIDAL values measured over 21 days (during which period, adduct  levels were
26   specifically quantified), and as a function of TIDAL values extrapolated to 8 months.  The
27   relative tumor induction potencies of the studied PAHs were similar for each assay for a single
28   PAH when described as functions of administered dose, the adduct levels per dose at 3 days, the
29   TIDAL values over 21 days, or the TIDAL values extrapolated to 8 months. The  relative
30   potencies for tumor incidence as a function of adduct levels at 24 hours were not similar to those
31   associated with  the other measures of exposure. Ross et al. (1995) suggested that
32   pharmacokinetic differences in adduct formation among the PAHs were  responsible for the
33   discrepancy, but suggested that peak levels could be used to compare the potencies of different
34   PAHs if adduct formation for those PAHs followed similar kinetics.
35          DNA adduct experiments were carried out in replicate and were usually analyzed
36   statistically. It should be noted that, based on the work of Ross et al. (1995), relative potencies
37   determined from studies that administered a single dose level and measured adducts at a single
38   time point will be less reliable unless the shapes of the adduct formation curves are similar.

                                             94          DRAFT - DO NOT CITE OR QUOTE

-------
 1   However, the single dose and single measurement studies were also used for dose-response
 2   assessment.
 3          Among the in vivo DNA adduct studies shown in Table 4-6, nine studies met the
 4   selection criteria for use in this analysis.
 5
 6   4.3.2.2. Clastogenicity or Sister Chromatid Exchange Frequency
 7          The database included 13 studies in which clastogenic effects or frequency of sister
 8   chromatid exchanges of benzo[a]pyrene and at least one other PAH were tested in whole animal
 9   systems. Table 4-7 lists the studies along with important study design details.  The clastogenic
10   endpoints measured in these studies were micronuclei, chromosome gaps and breaks, and
11   nonspecific aberrations; sister chromatid exchanges were also measured. These studies were all
12   conducted in rodents, including mice, rats, and hamsters.
13          Eight of the studies evaluated micronuclei, sister chromatid exchanges, or chromosome
14   gaps or breaks in bone marrow from treated mice or hamsters (Allen et al., 1999; Katz et al.,
15   1981; Paika et al.,  1981; Salamone et al., 1981; Tsuchimoto and Matter, 1981;  Roszinsky-Kocher
16   et al., 1979; Bayer, 1978; Sugiyama, 1973).  In these studies, one or two doses of PAH were
17   injected intraperitoneally into the animals, and sacrifice occurred at various time points thereafter
18   (typically 24 hours after). Bone marrow smears were examined microscopically and scored for
19   micronuclei, sister chromatid exchanges, gaps, or breaks.
20          He and Baker (1991) applied multiple dose levels of chrysene or phenanthrene  to the skin
21   of hairless mice and harvested keratinocytes upon sacrifice 24 hours later. The keratinocytes
22   were incubated for 2 days and treated with cytochalasin B to identify binucleated cells. After
23   4 days in vitro, cells were mounted on slides and examined microscopically for micronuclei.
24   Results were reported as the percent of binucleated cells with one or more micronuclei among
25   the total number of binucleated cells scored. Chrysene treatment resulted in a dose-related
26   increase in micronuclei, while pyrene did not.
27          Kligerman  et al.  (2002, 1986) measured sister chromatid exchanges and/or micronuclei in
28   the blood of mice or rats given a single dose of PAH either orally or intraperitoneally.  The study
29   by Oshiro et al. (1992) involved two or four oral  doses of pyrene  or anthracene in mice. Blood
30   obtained from the tail 24 hours after the last treatment was examined microscopically and
31   micronuclei were scored in polychromatic erythrocytes. In an unusual study design, Sirianni and
32   Huang (1978) measured sister chromatid exchanges in V79 cells placed in a diffusion chamber
33   implanted in the peritoneal cavity of mice.
34          Thirteen individual PAHs were evaluated in these studies.  Only chrysene gave positive
35   results for more than one endpoint (for sister chromatid exchange and micronucleus frequency;
36   He and Baker, 1991; Roszinsky-Kocher et al.,  1979). Five other PAHs (phenanthrene,
37   dibenz[a,h]anthracene, benz[a]anthracene, benzo[b]fluoranthene, and benzo[e]pyrene) increased
38   the frequency of sister chromatid exchange in hamster bone marrow after intraperitoneal

                                             95          DRAFT - DO NOT  CITE OR QUOTE

-------
 1   administration (Roszinsky-Kocher et al., 1979). Bayer (1978) also reported an increase in sister
 2   chromatid exchange frequency in hamster bone marrow after phenanthrene administration (high
 3   dose only). Anthracene and pyrene consistently gave nonpositive results in several studies
 4   (Oshiro et al., 1992; He and Baker,  1991; Katz et al., 1981; Paika et al., 1981; Salamone et al.,
 5   1981; Tsuchimoto and Matter,  1981; Roszinsky-Kocher et al., 1979; Sirianni and Huang, 1978).
 6   Dibenzo[a,i]pyrene and benzo[g,h,i]perylene each gave nonpositive results in an assay for bone
 7   marrow micronuclei (Katz et al., 1981).
 8          Among studies with positive results, only He and Baker (1991), Kligerman et al. (1986),
 9   and Bayer (1978) administered PAHs at multiple dose levels.  Bayer (1978) observed a positive
10   response only with the highest dose of phenanthrene. Of the single dose studies, only
11   Roszinsky-Kocher et al. (1979) reported responses clearly differing from controls.
12          Among the in vivo clastogenicity or sister chromatid exchange studies shown in
13   Table 4-7, 10 studies met the selection criteria for use in this analysis.
14
15   4.3.2.3. In Vivo Mutagenicity
16          The PAH database contains several  studies that evaluate specific mutagenic endpoints
17   following in vivo exposure to PAHs (see Table 4-8). These studies include mutagenicity
18   experiments in Drosophila melanogaster, an intraperitoneal host-mediated assay using
19   Salmonella strains or yeast, and DNA sequence analysis of specific codons in the Ki-ras
20   oncogene in mouse lung tumors.
21          Most Drosophila studies administered PAH compounds to either the suspension media or
22   to the diet for 48-72 hours prior to cross-mating and analysis of mutations (Frolich and Wurgler,
23   1990; Valencia and Houtchens, 1981; Fahmy and Fahmy, 1980). One study used abdominal
24   injection as an exposure pathway (Zijlstra and Vogel, 1984). The mutagenic endpoints evaluated
25   included somatic mutations (i.e., eye color mosaicism, wing spots) (Frolich and Wurgler, 1990;
26   Fahmy and Fahmy, 1980) or sex-linked recessive lethal mutations (Zijlstra and Vogel, 1984;
27   Valencia and Houtchens,  1981). Only two  PAHs were evaluated in the Drosophila studies in
28   addition to benzo[a]pyrene (benz[a]anthracene and pyrene), and the results were either
29   nonpositive or inconsistent in all studies (Frolich and Wurgler, 1990; Zijlstra and Vogel, 1984;
30   Valencia and Houtchens,  1981; Fahmy and Fahmy,  1980). A significant effect was seen for
31   benz [a] anthracene only with cross-breeding of strains selected for enhanced metabolic activity
32   (Frolich and Wurgler, 1990). No effect was observed using the standard strains.
33          An intraperitoneal host-mediated assay was described by Simmon et al. (1979).  Five
34   PAHs (anthracene, benz [a] anthracene, benzo[e]pyrene, chrysene, and phenanthrene) were
35   administered to Swiss Webster mice by gavage or intramuscular injection (single dose only).
36   Microorganisms (S. typhimurium and Saccharomyces cerevisiae) were injected intraperitoneally
37   into exposed mice and were recovered 4 hours later  for mutation analysis.  Nonpositive results
                                             96          DRAFT - DO NOT CITE OR QUOTE

-------
 1   were observed and the host-mediated assay system was considered insensitive for detecting
 2   carcinogenic PAHs.
 3          A series of studies have investigated the mutation sequence in codons 12 and 61 of the
 4   Ki-ras oncogene from PAH-induced lung adenomas in A/J mice (Nesnow et al., 1998a, 1996,
 5   1995; Mass et al., 1993). As discussed in Section 2.4 (Similarities in Mode of Carcinogenic
 6   Action for PAHs), the purpose of these studies was to correlate the tumorigenic potency of
 7   specific PAHs with the formation of DNA adducts and the mutation of specific codons in the
 8   Ki-ras oncogene.  Six non-alkylated PAHs were utilized in these studies (benzo[a]pyrene,
 9   benz[j]aceanthrylene, benzo[b]fluoranthene, dibenz[a,h]anthracene, cyclopenta[c,d]pyrene,  and
10   dibenzo[a,l]pyrene).  Mutation analysis of the Ki-ras oncogene at codons 12 and 61 was carried
11   out in PAH-induced lung adenomas using PCR amplification and dideoxy nucleotide sequencing
12   methods. The primary mutation type for benzo[a]pyrene, benzo[b]fluoranthene, and
13   dibenzo[a,l]pyrene was the GGT^TGT mutation. This guanine mutation was correlated with
14   the formation of diol epoxide guanine adducts. The GGT—>CGT mutation was the primary
15   mutation type for benz[j]aceanthrylene and cyclopenta[c,d]pyrene. The CGT mutation was
16   associated with the formation of cyclopenta-guanine adducts and increased tumorigenic potency
17   (i.e., >90 adenomas per mouse) in A/J mice. Dibenz[a,h]anthracene was the only PAH evaluated
18   that did not induce mutations in Ki-ras codons 12 or 61. This compound produced diol epoxide
19   guanine adducts and lung adenomas in A/J mice, suggesting a possible interaction at a different
20   genetic target. The Ki-ras mutation analysis data were presented as percent of tumors with a
21   specific mutation at either codon 12 or 61. No dose-response data were provided.
22          Among the in vivo mutagenicity studies shown in Table 4-8, only one study met the
23   selection criteria for use in this analysis.
24
25   4.3.3. In Vitro Studies of Cancer-Related Endpoints
26          Many in vitro studies of cancer-related endpoints are present in the PAH database. As
27   previously  discussed, only those studies that included at least one selected PAH and
28   benzo[a]pyrene as a reference compound were reviewed. Each study that was reviewed for the
29   purpose of RPF development is included in Tables 4-9 through 4-14. The tables summarize
30   study-specific information and indicate whether a particular study is considered useful for dose-
31   response assessment. The text provides an overall description of the available studies, including
32   a general description of the methodology used for each study type, the results, and the
33   weaknesses or problems  associated with specific studies or study types.
34
35   4.3.3.1. Bacterial Mutagenicity
36          The bacterial mutagenicity of many PAHs has been extensively studied (39 studies with
37   benzo[a]pyrene; see Table 4-9).  All of the studies used the Ames assay in S. typhimurium.  A
38   total of 38 PAHs have been evaluated for their ability to induce mutations in bacterial systems.

                                             97          DRAFT - DO NOT CITE OR QUOTE

-------
 1          The Ames Salmonella assay is a bacterial reverse mutation assay, which measures the
 2   frequency at which histidine-independent bacteria arise from histidine-requiring bacterial strains
 3   in the presence of a chemical mutagen.  The results are generally expressed as either the number
 4   of revertant colonies per plate or the number of revertants/nmol of the test compound (calculated
 5   from the linear portion of the dose-response curve).  Several strains of S. typhimurium have been
 6   used to evaluate specific PAH mutation types; for example, TA98, TA1537, and TA1538 detect
 7   various frameshift mutations, TA1535 responds to base-pair substitution, and TA100 responds to
 8   a broad spectrum of mutations.  Metabolism to reactive intermediates is required for PAH
 9   mutagenicity in Salmonella and many metabolic activation systems have been employed. Rat
10   liver postmitochondrial supernatant (known as S9) from Aroclor-induced rats is most often used,
11   although other rodent species and enzyme inducers are sometimes employed.  Isolated rat
12   hepatocytes or purified mixed-function oxidase enzymes were occasionally utilized for metabolic
13   activation of PAHs.
14          Of the PAHs tested for bacterial mutagenicity, most were considered positive in at least
15   one study  under optimal study conditions. Compounds that produced nonpositive results in
16   multiple studies include anthracene, fluorene, phenanthrene, and pyrene. The primary weakness
17   of the bacterial mutagenicity database for PAHs is the limited amount of multiple-dose data for
18   many PAHs. Many studies report findings at a single dose level for several PAHs.
19          Among the in vitro bacterial mutagenicity  studies shown in Table 4-9, 29 studies met the
20   selection criteria for use in this analysis.
21
22   4.3.3.2.  Mammalian Mutagenicity
23          Studies that evaluate the mutagenicity of target PAHs in mammalian cells are described
24   in Table 4-10 (29  studies). The most common cell types used in these studies were the
25   V79 Chinese hamster cells and the L5178Y mouse lymphoma cells. Other cell types include
26   human epidermal  keratinocytes, TK6 human lymphoblasts, human epithelial cells (HS1 HeLa),
27   human foreskin fibroblasts (D-550), mouse fibroblasts, rat embryo cells, rat liver epithelial cells
28   (ARL-18), and Chinese hamster ovary (CHO) cells.  A total of 14 PAHs have been evaluated for
29   their ability to induce mutations in mammalian cell systems.
30          Each of the mammalian  cell assays detects forward mutations that confer resistance to a
31   toxic chemical.  Mutations in the hypoxanthine-guanine phosphoribosyl transferase gene (HPRT)
32   result in resistance to purine analogs such as 6-thioguanine, 8-azaguanine, and ouabain. HPRT
33   mutations induced by PAHs were most often measured in V79 Chinese hamster cells, but have
34   also been detected in human, rat, and mouse cell lines. Forward mutation at the thymidine
35   kinase (TK) locus is measured as colony growth in the presence of thymidine analogs (e.g.,
36   trifluorothymidine or 5-bromo-2'-deoxyuridine). PAH-induced TK mutations were measured in
37   mouse lymphoma cells (L5178Y) and human lymphoblasts. Forward mutation assays are
38   considered to respond to a variety of mutation types (including frameshift, base-pair substitution,

                                            98          DRAFT - DO NOT CITE OR QUOTE

-------
 1   deletions, and rearrangements or complex mutations). Exogenous metabolic activation is
 2   required for PAH mutagenicity in most mammalian cell assays. This was accomplished using a
 3   rat liver S9 mix or cocultivation with other rodent cells able to metabolize PAHs to reactive
 4   intermediates (i.e., hamster embryo cells, fibroblasts, or hepatocytes; rat hepatocytes). The
 5   results of forward mutation assays in mammalian cell lines are generally expressed as mutant
 6   frequency/ 10X survivors.
 7          Of the 26 PAHs tested for mammalian cell mutagenicity, all were considered positive in
 8   at least one study under optimal study conditions. Compounds that produced nonpositive results
 9   in some studies include anthracene, benzo[e]pyrene, phenanthrene, and pyrene. Benzo[a]-
10   anthracene produced positive findings in seven studies and nonpositive findings in four studies.
11   The mammalian mutagenicity studies generally provide more multidose data than the bacterial
12   mutagenicity studies.
13          Among the in vitro mammalian mutagenicity studies shown in Table 4-10,  27 studies met
14   the selection criteria for use in this analysis.
15
16   4.3.3.3. Morphological/Malignant Cell Transformation
17          Twenty-five studies examined the capacity of benzo[a]pyrene and other PAHs to
18   transform cells in culture (Table 4-11).  All of these studies were conducted using mammalian
19   cells, most commonly mouse or hamster embryo cells.  A few studies added feeder cells or rat
20   liver homogenate to enhance metabolic activation in the test system; however, the majority relied
21   on the intrinsic metabolic capacity of the cells.  The general test protocol involved  seeding the
22   cultured cells in Petri dishes followed by exposure to a solution of the test compound, usually for
23   a period of 24 hours. The cells were then cultured for about  6 weeks before being fixed and
24   stained. Transformed colonies (foci) were scored based on characteristics such as cell piling,
25   criss-crossing, basophilic staining, and/or invasion of surrounding (nontransformed) cell
26   monolayer.  In studies conducted by some laboratories, foci were classified  as Type II or
27   Type III; the latter category included those with invasion of the surrounding monolayer, highly
28   criss-crossed arrays, and deep staining.  Data were generally reported as the number of foci
29   (colony of transformed cells) per dish or per surviving cells and/or the percent of dishes with
30   foci.
31          In a few cases (e.g., Greb et al., 1980), transformation was assessed by growth of treated
32   cells in soft agar.  Transformed cell colonies growing in semi-solid agar are capable of
33   anchorage-independent growth.
34          Three studies (Evans and DiPaolo, 1975; Kakunaga,  1973; DiPaolo et al., 1972)
35   confirmed the identification of malignant cells by injecting the transformed  cells into rodents and
36   following tumor induction in the animals. In all three cases,  cells identified as transformed gave
37   rise to tumors, while the cells without these characteristics did not.
                                             99           DRAFT - DO NOT CITE OR QUOTE

-------
 1          Cell transformation assays were identified that included 22 individual PAHs other than
 2   benzo[a]pyrene. Dibenz[a,h]anthracene consistently gave rise to transformed cells in all but one
 3   of the seven studies in which it was tested.  Cyclopenta[c,d]pyrene, indeno[l,2,3-c,d]pyrene,
 4   benzo[j]aceanthralene, benz[e]aceanthrylene, and dibenz[k,mno]acephenanthrylene were each
 5   tested in a single study and gave positive results. Benz[a]anthracene, pyrene, phenanthrene,
 6   benzo[e]pyrene, and anthracene each gave nonpositive results in a number of studies, while
 7   fluoranthene, benzo[k]fluoranthene, dibenz[j,mno]acephenanthrylene, naphth[ 1,2,3-mno]ace-
 8   phenanthrylene, and aceanthrylene were each tested in a single study and gave nonpositive
 9   results.  Only a single dose of the target PAH was applied in 8 of the 26 studies of in vitro
10   morphological/malignant cell transformation.
11          Among the in vitro morphological/malignant transformation studies shown in Table 4-11,
12   19 studies met the selection criteria for use in this analysis.
13
14   4.3.3.4. DNAAdducts
15          Several studies (14) were identified in which DNA adducts were measured after either
16   whole cells or extracted DNA were incubated with benzo[a]pyrene and at least one other PAH.
17   Table 4-12 shows general study details for these studies. Most of the studies involved
18   measurement of DNA adducts in whole mammalian cells, while some measured adducts formed
19   when PAHs were incubated with extracted DNA. Whole cells were usually incubated with
20   PAHs for about 24 hours, while extracted DNA was exposed to PAH solutions for a shorter time
21   period (1-3 hours).  Some of the studies added metabolic activation (usually rat liver
22   microsomes) to the incubation solution. Melendez-Colon et al. (2000) evaluated DNA adduct
23   formation after dibenzo[a,l]pyrene exposure in two cell types: one having significant CYP450
24   activity (MCF-7 cells) and one lacking significant CYP450 activity (HL-60). The authors
25   reported that adducts were formed in the  cells having CYP450 activity, but no adducts were
26   formed in the cells lacking such activity.
27          Identification and quantification of adducts was generally done using a [32P]-postlabeling
28   assay as follows.  After exposure, DNA was isolated and digested to mononucleotides.
29   Mononucleotides were radiolabeled with [32P]-ATP, separated with thin layer chromatography,
30   and visualized by autoradiography.  Relative adduct labeling was measured using a scintillation
31   counter.  A few early studies used [3H]-labeled PAHs to identify and quantify adducts. In some
32   cases, adducts were identified by high-performance liquid chromatography and gas
33   chromatography-mass spectrometry.
34          The 14 studies reviewed examined 15 PAHs other than benzo[a]pyrene. Apart from
35   phenanthrene, which did not result in measurable DNA adducts when incubated with calf thymus
36   DNA under various conditions (Bryla and Weyand, 1992), each of the PAHs produced
37   measurable DNA adducts in at least one study.
                                            100          DRAFT - DO NOT CITE OR QUOTE

-------
 1          Major limitations associated with some of the in vitro DNA adduct data for relative
 2   potency development include the lack of data at multiple PAH exposure levels, the use of
 3   extracted DNA rather than whole cell assays, and the inconsistent use of extrinsic metabolic
 4   activation sources. Only three studies with positive adduct findings reported adduct
 5   measurements at multiple doses (concentrations) of PAH (Binkova et al., 2000; Melendez-Colon,
 6   2000; Bryla and Weyand, 1992). Three studies used extracted DNA rather than whole cells to
 7   measure DNA binding (Segerback and Vodicka, 1993; Bryla and Weyand, 1992; Grover and
 8   Sims, 1968). Finally, the available studies on DNA adduct formation use  cell types with varying
 9   degrees of PAH metabolic capacity, with and without added metabolic activation sources. Both
10   the types and the quantities of DNA adducts formed are likely to depend on the level of
11   metabolic activation for most PAHs.
12          Among the in vitro DNA adduct studies shown in Table 4-12, 10 studies met the
13   selection criteria for use in this analysis.
14
15   4.3.3.5. DNA Damage/Repair
16          Twenty-four reports in the database evaluated the effects of one or more PAHs on DNA
17   damage, repair, or synthesis. Table 4-13 summarizes the study design information and results of
18   these studies.  Studies included measures of unscheduled DNA synthesis and DNA damage.
19   Unscheduled DNA synthesis was generally measured by increased radiolabeled (3H) thymidine
20   uptake in treated cells versus untreated cells.  DNA damage was measured either using the
21   alkaline elution assay for DNA strand breakage in mammalian cells, or using the differential
22   killing of DNA repair-deficient bacterial strains.  Metabolic activation of PAHs was most often
23   accomplished using a rat liver S9 mix.
24          Twenty-eight different PAHs have been tested for effects on DNA in one or more assays.
25   In general, pyrene, anthracene, phenanthrene, perylene, fluorene, and benzo[e]pyrene gave
26   nonpositive  results in multiple studies. Chrysene gave nonpositive results in four assays and
27   positive results in one assay (Mersch-Sundermann  et al., 1992).  More positive than nonpositive
28   results were reported for benz[a]anthracene, dibenz[a,h]anthracene, and dibenz[a,c]anthracene.
29   Other PAHs were tested only once, or gave roughly an equal frequency of positive and
30   nonpositive  responses in these assays.
31          Although a large number of PAHs have been tested for DNA damage/repair, the  database
32   includes both bacterial and mammalian cells and several different genotoxic endpoints. In
33   addition, the use of external metabolic activation, or cell types with intrinsic metabolic capacity,
34   was inconsistent across these studies. These limitations make it difficult to compare studies
35   using the same target PAHs.
36          Among the in vitro DNA damage/repair studies shown in Table 4-13, 15 studies met the
37   selection criteria for use in this analysis.
38

                                            101          DRAFT - DO NOT  CITE OR QUOTE

-------
 1   4.3.3.6. Clastogenicity or Sister Chromatid Exchange Frequency
 1          The database contains 18 studies in which clastogenicity or sister chromatid exchange
 3   frequency was measured in cultured cells after exposure to benzo[a]pyrene and at least one other
 4   PAH (Table 4-14). A wide variety of cell types was used in these assays, including hamster
 5   liver, lung, CHO, and V79 cells; rat liver epithelial cells; human teratocarcinoma epithelial cells;
 6   rat and human mammary epithelial cells; mouse, rat, and human fibroblasts; human
 7   lymphocytes; and guinea pig fetal cells. A number of the studies used a metabolic activation
 8   system, typically either rat liver S9 or coculture with a cell type able to metabolize PAHs. While
 9   laboratory methods varied widely, the general approach involved treating the cultured cells with
10   a solution of the test compound, either with or without metabolic activation. Usually,
11   bromodeoxyuridine was added to the growth medium to provide a means of staining metaphase
12   chromosomes, and colcemid was used to arrest mitotic cells.  Chromosomes were examined
13   microscopically and aberrations or exchanges were scored visually.  In most cases, the endpoint
14   examined was frequency of sister chromatid exchanges.  Other endpoints included frequency of
15   micronuclei  and scoring of chromosomal aberrations such as breaks, gaps, deletions, etc.
16          Only eight PAHs (anthracene, benz[a]anthracene, benzo[e]pyrene, cyclopenta-
17   [c,d]pyrene,  fluoranthene, perylene, phenanthrene, and pyrene) have been tested for clastogenic
18   effects in vitro. In many cases, the available studies were aimed at evaluating the validity of a
19   given test system to predict carcinogenicity.  In these studies, a range of compounds of known or
20   believed carcinogenicity were used. Often, benzo[a]pyrene was included as a known carcinogen,
21   and other PAHs were chosen because they were known or believed to be noncarcinogenic or
22   weakly carcinogenic.
23          Among the tested compounds, four gave positive results in at least one study.  With few
24   exceptions, PAHs administered without metabolic activation gave nonpositive responses in these
25   assays.  Cyclopenta[c,d]pyrene was reported to  increase the frequency of sister chromatid
26   exchanges in two assays, one with and one without metabolic activation (Murison, 1988;
27   Krolewski et al., 1986). Benz[a]anthracene gave positive results in three  studies of sister
28   chromatid exchange induction (Mane et al., 1990; Tong et al., 1983, 1981a) and nonpositive
29   results in a fourth (Warshawsky et al., 1995).  Kochhar (1982) reported a dose-dependent
30   increase in chromosomal aberrations in V79 cells treated with benz[a]anthracene in the absence
31   of metabolic activation. Perylene increased aberrations in one system (Popescu et al., 1977), but
32   did not increase sister chromatid exchanges in another (Sirianni and Huang, 1978).  Likewise,
33   pyrene gave positive results in a number of studies that included metabolic activation (Evans and
34   Mitchell, 1981; Perry and Thomson, 1981; Popescu et al., 1977) and nonpositive results in
35   several that did not include activation (DeSalvia et al., 1988; Tong et al.,  1983, 1981a; Dean,
36   1981; Abe and Sasaki, 1977).
37          The clastogenicity and sister chromatid exchange data for PAHs are variable with respect
38   to cell type and use of extrinsic metabolic activation.  Some cells have intrinsic metabolic

                                            102          DRAFT - DO NOT CITE OR QUOTE

-------
 1   activity, while others require activation from an external source. The degree to which metabolic
 2   activation is required for PAHs to exert a clastogenic effect in cell cultures is not well
 3   established. Another limitation of these data stems from the fact that a small number of PAHs,
 4   many traditionally believed to be noncarcinogenic or weakly carcinogenic, have been tested for
 5   clastogenic effects in vitro.
 6          Among the in vitro clastogenicity/sister chromatid exchange studies shown in Table 4-14,
 7   10 studies met the selection criteria for use in this analysis.
 8
 9   4.4.  SUMMARY OF INFORMATION AVAILABLE TO DEVELOP RPFs FOR
10   INDIVIDUAL PAHs
11          The PAH database contains several different types of data that may be used to estimate
12   relative potencies of individual PAHs. The data were summarized in Section 4.3 and include in
13   vivo tumor bioassays using various routes of exposure and data for cancer-related endpoints
14   from both in vivo and in vitro studies. As discussed above, the concurrent testing of
15   benzo[a]pyrene as a reference compound was considered essential to allow for RPF calculation.
16   The introduction to Section 4.3 lists criteria for selecting studies or data sets for use in the
17   analysis. Studies that met these criteria were used in the development of the RPF approach.
18   Chapter 5 discusses methods used for dose-response assessment and RPF calculation from each
19   study or dataset, and Chapter 6 discusses the selection of PAHs to be included in the RPF
20   approach using a weight of evidence evaluation of the available data.  Chapter 7 describes the
21   derivation of final RPFs for each PAH included in the analysis.
22
23
                                            103          DRAFT - DO NOT CITE OR QUOTE

-------
 1      5. METHODS FOR DOSE-RESPONSE ASSESSMENT AND RTF CALCULATION
 2
 3
 4          A discussion of the available data on PAH carcinogenicity and cancer-related endpoints
 5   and criteria for selection of studies was presented in Chapter 4. This section describes the
 6   selection of dose-response data and methods for dose-response assessment and RPF calculation
 7   from the selected datasets. The dose-response data extracted from each study with positive
 8   results and the results of the statistical analyses are shown in Appendix C. Appendix C also
 9   contains information regarding the source of the dose-response data (i.e., the figure or table
10   number from the study and the particular data points that were used in the dose-response
11   assessment) and additional comments on the use of the data for dose-response assessment and
12   RPF calculation.  The results of the RPF  calculations are shown in tables in Appendix E.  These
13   tables provide summary information for each study, including the PAHs that were tested, the
14   data used to estimate the slopes (point estimate4 or BMD model result), the calculated RPF
15   value, and any specific comments related to the data analysis.
16
17   5.1.  CHOICE OF DOSE-RESPONSE DATA
18          For each of the endpoints evaluated in Chapter 4 (dermal, intraperitoneal,  subcutaneous,
19   oral, and other route bioassays; in vivo DNA adducts; in vivo clastogenicity or sister chromatid
20   exchange frequency; in vitro bacterial and mammalian  mutagenicity; in vitro morphological/
21   malignant transformation; in vitro clastogenicity or sister chromatid exchange frequency; and
22   other in vitro endpoints [DNA adducts, unscheduled DNA synthesis, DNA damage, etc.]), there
23   was  at least one study that met selection criteria. For those studies with positive findings, dose-
24   response data were extracted for dose-response assessment and calculation of RPFs.
25
26   5.1.1. Dose-Response Data for Tumor Bioassays
27          Data on both benign and malignant tumors were included in the dose-response
28   assessment.  In cases where the combined incidence of benign and malignant tumors was
29   reported, these data were selected; however, in some cases, only benign or only malignant tumor
30   incidence was reported.  These data were also considered appropriate for derivation of RPFs.
31   There is evidence for progression from benign to malignant tumors (e.g., dermal papillomas
32   progressing to carcinomas) in studies of benzo[a]pyrene (for example, see Albert  et al., 1991),
33   and other PAHs  are assumed to be lexicologically similar to benzo[a]pyrene.  Thus, even when a
34   study reported only the incidence  of benign tumors, these data were used in the dose-response
35   assessment.
     4For the purpose of this report, the term "point estimate RPF" is used to describe an RPF calculated from a single
     point on the dose-response curve for both the PAH of interest and benzo[a]pyrene. This term distinguishes the RPF
     from one calculating using a BMD modeling result from multidose data.

                                            104           DRAFT - DO NOT CITE OR QUOTE

-------
 1          While tumor multiplicity data from tumor bioassays are not generally used to estimate
 2   cancer potency., these data were included in the dose-response assessment in order to determine
 3   whether they could serve as a reliable measure of relative cancer potency.  Several bioassays
 4   reported data on both tumor incidence and tumor number, providing information that could later
 5   be used to compare relative potencies estimated from these two endpoints.
 6          As discussed in Section 4.3, statistics were used for tumor bioassay data to determine
 7   whether the tumor incidence or multiplicity observed at a particular dose represented a
 8   statistically significant increase over controls.  If statistical analyses were not described in the
 9   original report, incidence data were analyzed using Fisher's exact test and the Cochran-Armitage
10   trend test. Positive findings were indicated by a significant (p < 0.05) difference for at least one
11   dose group by comparison to control (in Fisher's exact or an equivalent test) or a significant
12   dose-response trend (Cochran-Armitage or equivalent) for multidose studies. For tumor bioassay
13   data reported as tumor count, a t-test was conducted (when variance data were available) to
14   determine whether the count was significantly different from control (p < 0.05).  The results of
15   the statistical analyses are shown with the dose-response data in Appendix  C.
16          The tumor bioassays that reported both incidence and tumor count were unique in
17   offering two different datasets for the same study. For each dose of each PAH in the tumor
18   bioassays, the decision to calculate an RPF, and in some instances,  the selection of the point of
19   departure, was based on whether the tumor incidence or count was  statistically significantly
20   increased over the control; if there was a significant increase, an RPF was calculated.  There was
21   a single instance where the tumor count was statistically significantly increased, but the
22   incidence of tumors was not.  In female mice exposed at the high dose  of fluoranthene in the
23   study by Busby et al. (1984), the lung tumor count was significantly increased (albeit borderline,
24   p = 0.0343) while the incidence was not, and neither was statistically significantly increased at
25   the lower dose. As there were no higher doses in this study, it is possible that the two measures
26   might have produced consistent findings at higher doses.  For the purpose of this analysis, the
27   multiplicity data from this study were treated as an independent measure of carcinogenic
28   potency, and an RPF was calculated for the statistically increased tumor count irrespective of the
29   analysis of incidence. It should be noted that average tumor count can be skewed by an unusual
30   response in a single animal, and no information was available to determine whether such
31   response represented an anomaly unrelated to exposure or an unusual susceptibility to the
32   exposure. Thus, reliance on statistical analysis of mean tumor count alone  as a measure  of
33   carcinogenic response may be subject to additional uncertainty.
34
35   5.1.2. Dose-Response Data for Cancer-Related Endpoint Studies
36          For cancer-related endpoint data, each study authors' conclusions regarding a positive or
37   nonpositive response for each PAH were accepted, and RPFs were  calculated when positive
38   results were reported. Data that were reported in graphical format in published studies of cancer-

                                             105           DRAFT - DO NOT CITE OR QUOTE

-------
 1   related endpoints were digitized (Grab It!™ Graph Digitizer, Datatrend Software) to identify the
 2   dose-response data points.  In a few cases, the only cancer-related endpoint data in a given
 3   publication were reported as relative potency (relative to benzo[a]pyrene). For these
 4   publications, which included only in vitro cancer-related endpoint data (primarily mutagenicity),
 5   the relative potency estimates calculated by the authors were used without modification (except
 6   for dose adjustment where  appropriate; see Section 5.5).
 7
 8   5.2. OVERALL FORM OF RTF ESTIMATE
 9          The overall goal of the dose-response analysis was to calculate ratios representing the
10   relative potency of a given PAH compared with benzo[a]pyrene (i.e., RPFs).  For all datasets, the
11   RPF was defined as the ratio (PAH;:BaP) of the slopes of the dose-response curves in the low-
12   dose region, following Equation 5-1 below:
13
14                              RPF = slope PAH; + slope BaP                           (5-1)
15
16          Data available for calculation of RPFs consisted of both quantal and continuous
17   endpoints.  Quantal endpoints included tumor incidence or incidence of cancer-related endpoints
18   (including frequency of mutations).  Continuous endpoint  datasets included tumor counts
19   (number of tumors per animal) or cancer-related endpoints of a continuous-variable nature (e.g.,
20   number of sister chromatid exchanges, number of morphologically transformed colonies). Dose-
21   response assessment methods were specific to each type of endpoint (quantal or continuous) and
22   differed depending on whether there were multiple dose groups or a single dose group in the
23   dataset. Methods for multidose and single dose quantal and continuous data are described below.
24
25   5.3. RPF CALCULATION FOR MULTIDOSE DATASETS
26          Dose-response modeling using U.S. EPA's Benchmark Dose Software (Version 2.1.1 or
27   1.3.2) was conducted on multiple-dose data sets to estimate potency for both the target PAHs and
28   benzo[a]pyrene. Modeled estimates consider information  about the shape of the dose-response
29   curve and are thus preferred over using a single dose group as the point of departure.
30          Dose-response modeling.  For multidose quantal data, the multistage model was used and
31   the degree of the polynomial was assumed to equal the number of dose groups minus 2.  The
32   multistage model was selected because it is the preferred model for cancer risk assessment of
33   animal bioassay data, and it provided a consistent model form for all of the datasets. For tumor
34   bioassay data, the multistage-cancer model was selected, while other quantal data were modeled
35   using the multistage model (both have the same model form and yield the same result). For
36   multidose continuous data, the linear model was selected for all datasets, as it is the simplest
37   model form for continuous data. For both quantal and continuous datasets, the goodness-of-fit
38   criteria were used to evaluate model fit.  If the model did not provide adequate fit to the data,

                                            106          DRAFT - DO NOT CITE OR QUOTE

-------
 1   high-dose groups were sequentially eliminated in an effort to achieve adequate fit, except when
 2   truncating the data would result in the loss of datapoints at response levels in the range of the
 3   benzo[a]pyrene response. The focus of the modeling effort is on the low dose and response
 4   region, so doses and responses much higher than the benchmark response (BMR) are not as
 5   informative and can be eliminated to improve model fit. If dose-group elimination did not
 6   improve the model fit, a point-estimate ratio approach was used (see Section 5.4). The BMD
 7   modeling outputs for all datasets that were successfully modeled are shown in Appendix D.
 8          Selection of BMR: Multidose data for both PAH and benzo[a]pyrene.  For tumor
 9   incidence data, the BMR used in estimating the point of departure was a 10% increase in tumor
10   incidence over controls (extra risk form). For cancer-related endpoints such as frequency of
11   mutations, endpoint-specific points of departure were selected based on the background/control
12   frequency of the endpoint and the detection limit of the assay. For example, a 1% frequency was
13   selected for a control mutation frequency of 1/10,000 and a detection limit of two- to threefold
14   above background.
15          For multidose continuous data, the BMR used in estimating the point of departure was a
16   change of 1 standard deviation (1 SD) from the control mean. In the event that multiple-dose
17   continuous data were reported in the absence of SD values, a point estimate  ratio approach was
18   employed to calculate the slope (see Section 5.4).
19          Selection of BMR: Multidose data for PAH, single dose benzo [a]pyrene.  Some studies
20   included only one dose  of benzo[a]pyrene as a positive control,  while providing multiple-dose
21   data for a selected PAH. In these cases, dose-response modeling was performed for the selected
22   PAH and the BMR used for modeling was the observed response for benzo[a]pyrene adjusted for
23   background response. For tumor incidence data, for example, if the benzo [a] pyrene dose was
24   associated with a 60% extra risk for tumors, the BMR chosen for modeling the data for the PAH
25   was 60% extra risk.  RPFs were then calculated using a ratio of the slope factors calculated with
26   equivalent points of departure (e.g., BMD6o). The goal of this approach was to compare PAH
27   potencies at similar response locations on the dose-response curve. There is uncertainty
28   associated with relative potency estimates calculated at the high end of the dose-response curves
29   and using the resultant RPF for low-exposure scenarios, because the relative potency relationship
30   between any two PAHs may be different at the low end, compared with the high end, of the
31   dose-response curves. The uncertainties and limitations associated with the  use of high-dose
32   data to estimate relative potency are further discussed in Chapter 7. Data sets for which tumor
33   incidence was >90% in the lowest dose group were not used to calculate potency estimates and
34   RPFs, because the response is near plateau and such data provide insufficient information on the
35   slope of the dose-response relationship.
36          For continuous data, when a point estimate was used to estimate the slope for
37   benzo[a]pyrene and  modeling was used to estimate the slope for a given PAH, the BMR used for
38   BMD modeling was a point value set at the response (e.g., mean number of tumors per animal

                                            107          DRAFT - DO NOT CITE OR QUOTE

-------
 1   for tumor multiplicity data) observed in the benzo[a]pyrene group, adjusted for response in the
 2   control group. This approach is consistent with the BMR used for quantal data when only a
 3   single benzo[a]pyrene dose group was available.  Provided that a linear model is fit to continuous
 4   data, the choice of a higher BMR would not appreciably change the RPF.
 5          Selection of point of departure.  The point of departure selected for slope estimation was
 6   the BMD estimate rather than the lower confidence limit on the BMD.  The BMD, as the central
 7   or "best" estimate of the dose associated with the selected BMR, was considered a more stable
 8   basis for comparison between the potency of the selected PAH and benzo[a]pyrene, and thus for
 9   calculation of relative potency, than the lower confidence limit.
10          Extrapolation from point of departure.  The slopes of the dose-response curves in the
11   low-dose regions were calculated by linear  extrapolation to the origin from the model-predicted
12   points of departure.  Equation 5-2 below shows the calculation of slope from multidose quantal
13   data.
14
15                                     Slope = [0.1/BMDi<,]                             (5-2)
16
17          Equation 5-3 below shows the calculation of slope from multidose continuous data.
18
19                                 Slope = [lSDchange]/[BMD1SD]                      (5-3)
20
21   5.4. RPF CALCULATION FOR SINGLE DOSE DATASETS
22          A number of studies reported data for only single doses of benzo[a]pyrene and other
23   PAHs; for these studies, a point estimate approach was used to calculate the RPF.  A point
24   estimate approach was also used to calculate RPFs for multidose datasets when model fit was not
25   achieved, when variance data were not available for continuous data, or when  problems with
26   model implementation were encountered.
27          Selection of point of departure.  When only one dose of each compound was used, there
28   was only one choice for the point of departure. However, when multidose data were available,
29   but a point estimate approach was used, the point of departure was chosen as follows.  For tumor
30   bioassay data, the lowest dose associated with a statistically significant increase in tumor
31   incidence or multiplicity over control values was selected as the point of departure. Variance
32   was not reported for tumor multiplicity data in any of the dermal studies and for some of the
33   intraperitoneal studies, so the corresponding incidence data were used to determine the dose at
34   which a significant difference from control  was observed.
35          The benzo[a]pyrene dose chosen in  most instances was the lowest dose associated with a
36   significant increase in tumor count or incidence.  For tumor multiplicity data, the PAH dose
37   chosen for the point estimate RPF calculation was the lowest dose associated with  a tumor count
38   similar to that observed at the selected benzo[a]pyrene dose (similar to selecting a BMR similar

                                            108          DRAFT - DO NOT CITE OR QUOTE

-------
 1   to the benzo[a]pyrene incidence).  In the case of two dermal initiation studies conducted by
 2   Cavalieri et al. (1991), however, the tumor count at the lowest dose of dibenzo[a,l]pyrene was
 3   much higher than the tumor count at the lowest benzo[a]pyrene dose associated with statistical
 4   significance.  In order to compare the doses associated with similar tumor counts (i.e., at a
 5   similar place on the dose-response curve), a higher benzo[a]pyrene dose was chosen for the RPF
 6   calculation. A comparison of the RPFs calculated using this approach with RPFs calculated
 7   using the lowest dose  associated with a statistically significant increase over controls for both
 8   dibenzo[a,l]pyrene and benzo[a]pyrene showed only small differences in the RPF values
 9   (9 versus 10 in the 16-week study and 39 versus 42 in the 27-week study). A similar approach
10   was used to calculate the RPF for Bj AC using the intraperitoneal multiplicity data from Mass et
11   al. (1993).
12          For cancer-related endpoint data, statistical analysis was not always available for each
13   dose group. For these data, the lowest dose that produced a near maximal change in the assay of
14   concern was selected as the point of departure.  That is, the highest dose in the linear portion of
15   the dose-response curve (identified by visual display of the data) was selected in these cases.
16          Extrapolation from point of departure. As with multiple dose slope estimations, point
17   estimate slope calculations also used the extra risk  form.  Thus, for single dose quantal data, the
18   slope was calculated by linear extrapolation to the origin after an extra risk adjustment of the
19   observed response (Equation 5-4):
20
21         Slope = [(response at dose - control response) ^ (1 - control response)] + dose      (5-4)
22
23          For single dose continuous data, the slope was calculated by linear extrapolation to the
24   origin after adjustment of the observed response in the PAH-treated animals for the control
25   response (Equation 5-5).
26
27         Slope = [(value of variable at dose) - (value of variable)COntroi] ^ dose               (5-5)
28
29   5.5. DOSE CONVERSION FOR RPF CALCULATION
30          Some of the studies used to calculate RPFs  reported doses or test  concentrations on a
31   molar basis (e.g.,  (imol per mouse,  |imol/L), rather than a mass basis (mg or jig). The molar
32   ratio differs from the mass ratio for any PAH with a molecular weight that differs from that of
33   benzo[a]pyrene; thus,  for these compounds, an RPF expressed on a mass basis will differ from
34   that expressed on a molar basis. Table 5-1 shows a hypothetical example for fluoranthene, a
35   PAH with a molecular weight that differs from benzo[a]pyrene by 20%.  As the table shows, the
36   RPF differs depending on which dose units are used.
37
                                             109          DRAFT - DO NOT CITE OR QUOTE

-------
            Table 5-1.  Comparison between molar and mass-based RPF


FA
BaP

Response
0.1
0.1

Dose in mol
5
1
Molecular
weight
(g/mol)
202.26
252.32

Dose in g
1,011
252

Molar RPF
0.20
1

Mass RPF
0.25
1
 1
 2          In order to ensure that comparisons across endpoints used consistent units, the doses used
 3   to calculate RPFs were converted to mass-based units using the molecular weight of the relevant
 4   PAH prior to estimating the RPF. While the RPF ratio is nominally unitless, it should be
 5   interpreted as the ratio of the dose of PAH to the dose of benzo[a]pyrene. Since RPFs will be
 6   used in conjunction with a PAH dose and benzo[a]pyrene cancer potency in mass units (oral
 7   slope factors and inhalation unit risks reported in units of [mg/kg-day]"1 and [jig/m3]"1,
 8   respectively); it is important to use mass-based RPFs. Alternatively, if a molar RPF ratio were to
 9   be used, it would be applied with PAH doses and benzo[a]pyrene cancer potency values
10   estimated on a  molar basis; this would require  a significant shift in the way PAH risks are
11   calculated compared to other carcinogens.  Therefore, the mass-based RPF was selected to be
12   consistent with dose metrics used to calculate cancer risk.
13
14   5.6.  SPECIAL CONSIDERATIONS FOR RPF CALCULATION USING TUMOR
15   BIOASSAY DATA
16          Several dermal bioassays reported significant mortality prior to the appearance of the first
17   skin tumor. For these data  sets, an assumption was made that the number of animals  at risk for
18   tumor development was equal to the total number of animals alive at the time of the appearance
19   of the first tumor.  Benign and malignant tumor types within the same target organ were
20   combined for calculation of the RPF.  The total incidence of animals with either a benign or
21   malignant lesion was directly reported in each  study (i.e., the number of animals with adenoma
22   or carcinoma).
23          Tumor incidence data reported for different target organs within the  same group of
24   animals were analyzed separately unless the joint incidence (incidence of either tumor type in
25   each dose group) was reported in the publication. Liver and lung tumors were reported in
26   newborn mice exposed to PAHs by intraperitoneal injection (LaVoie et al., 1994, 1987; Busby et
27   al.,  1989, 1984; Weyand and LaVoie, 1988; Wislocki et al., 1986). In most studies, tumor
28   incidence was reported separately for the different target organs and could not be combined as
29   the joint incidence was unknown. A gender difference was observed in the newborn  mouse
30   studies, with liver tumors observed in male mice only, and lung tumors reported for both male
31   and female mice.  The tumor incidence data were, therefore, evaluated separately for male and
                                           110
DRAFT - DO NOT CITE OR QUOTE

-------
 1   female mice. RPF values were calculated separately for male and female mice and for lung
 2   tumor incidence and liver tumor incidence in these studies.
 o
 3
 4   5.7.  SPECIAL CONSIDERATIONS FOR RPF CALCULATION USING CANCER-
 5   RELATED ENDPOINT DATA
 6          The in vitro studies of cancer-related endpoints included measurements of bacterial
 7   mutagenicity, mammalian mutagenicity, morphological/malignant cell transformation, DNA
 8   adduct formation, DNA damage or repair, and clastogenicity or sister chromatid exchange
 9   frequency.  Many of the studies describing in vitro cancer-related endpoints provide dose-
10   response data under varying study conditions. For example, bacterial mutagenesis studies used
11   multiple strains, different metabolic activation processes, and/or varying assay systems.  In order
12   to limit the number of datasets used for dose-response analysis of in vitro mutagenicity studies,
13   and to provide a consistent basis for comparing RPFs for different PAHs, data associated with
14   the conditions that maximized the benzo[a]pyrene response within a particular study were used
15   for the dose-response assessment of PAHs.  It should be noted that in several studies, test
16   conditions that were optimal for benzo[a]pyrene were not necessarily optimal for the selected
17   PAH (see Appendix C for specific studies). The uncertainties and limitations associated with
18   this approach are discussed further in Chapter 8.
19          For time-course studies of DNA adducts, results were reported as either AUC or peak
20   formation of adducts. AUC was considered preferable for dose-response assessment, because
21   this measure considers both adduct formation and  repair. Adducts measured in more than one
22   organ were summed to derive a total measure of adduct formation (standardized per unit amount
23   of DNA).
24          The data for bacterial and mammalian cell  mutagenicity and malignant cell
25   transformation were sometimes expressed as a mutation or transformation frequency (i.e.,
26   mutants/total cell count or transformed cells/total cells). For multiple-dose studies, these quantal
27   variables were evaluated using the multistage model as described above. Problems were
28   sometimes encountered when using the multistage model for incidence data of this type.  In some
29   cases, modifying the initial parameters in the multistage algorithm facilitated convergence.  In a
30   select few cases, the quantal linear model was used when the multistage model would not
31   converge. If neither the multistage nor quantal linear models provided adequate fit, a point
32   estimate approach was used.  If possible, the point estimates for both benzo[a]pyrene and the
33   target PAH were chosen at a comparable response level (e.g., the doses of benzo[a]pyrene and
34   the target PAH that both gave two mutants in 105 cells). However, in many cases, a comparable
35   response rate was not available. In these instances, the RPF was derived from slopes calculated
36   by linear extrapolation from the peak response.
37          As noted earlier, for studies that included only one dose of benzo[a]pyrene and multiple
38   dose data for a selected PAH, the BMR selected for dose-response modeling for the selected

                                            111          DRAFT - DO NOT CITE OR QUOTE

-------
 1    PAH was the benzo[a]pyrene response with the background or control response subtracted. In
 2    some instances, when the benzo[a]pyrene response level greatly exceeded the response at the
 3    highest dose of the selected PAH, the software would fail to calculate the BMD at the
 4    benzo[a]pyrene response level. In these instances, a point estimate approach using the peak
 5    response for the selected PAH was used.
 6          The individual study RPFs calculated for each PAH were used in a weight of evidence
 7    evaluation to select PAHs for inclusion in the RPF approach (see Chapter 6) and in the derivation
 8    of a final RPF for each compound (Chapter 7).
 9
10
                                           112          DRAFT - DO NOT CITE OR QUOTE

-------
       6. SELECTION OF PAHs FOR INCLUSION IN RELATIVE POTENCY APPROACH
 2
 3
 4
 5
 6
 7
 8
 9
10
11
       The selection of PAHs to be included in the RPF approach began with an evaluation of
whether the available data were adequate to assess the carcinogen!city of each compound.  At
least one RPF value was calculated for each of 51 PAHs. For 16 of these compounds, only a
single RPF value derived from an in vitro cancer-related endpoint (primarily mutagenicity
assays) was available.  These PAHs are shown in Table 6-1.  Due to the limited data available for
these 16 compounds, no further evaluation of these PAHs was conducted, and they were not
selected for inclusion in the RPF approach.
            Table 6-1. PAHs with only one RPF from a single in vitro cancer-related
            endpoint study and excluded from RPF approach
PAH
Aceanthrylene
Acenaphthene
Acenaphthylene
Acephenanthrylene
Benzo[a]perylene
Benz [b] anthracene
Benzo[b]perylene
Benzo [c]phenanthrene
Cyclopent[h,i]aceanthrylene
Cyclopent[h,i]acephenanthrylene
Dibenzo [a,f]fluoranthene
Dibenz[aj ] anthracene
Dibenzo [b,e]fluoranthene
Dibenzo [e,l]pyrene
Dibenz[k,mno]acephenanthrylene
Naphtho[2,3-a]pyrene
CASRN
202-03-09
83-32-9
208-96-8
201-06-9
191-85-5
92-24-9
197-70-6
195-19-7
131581-33-4
114959-37-4
203-11-2
224-41-9
2997-45-7
192-51-8
153043-81-3
196-42-9
Abbreviation
ACEA
AN
ANL
APA
BaPery
BbA
BbPery
BcPH
CPhiACEA
CPhiAPA
DBafF
DBajA
DBbeF
DBelP
DBkmnoAPH
N23aP
12
13
14
15
16
17
18
19
20
21
22
       The remaining 35 PAHs had RPF values calculated from at least one in vivo dataset or at
least two in vitro cancer-related endpoint datasets. For these compounds, a weight of evidence
approach was used to determine whether the available data (including the calculated RPFs as
well as nonpositive studies that met selection criteria) were adequate to include each compound
in the RPF approach. Using the calculated RPFs in the weight of evidence evaluation allowed
consideration of the magnitude of calculated RPFs in assessing carcinogenicity.  When data were
not considered adequate, the PAH was excluded from the RPF approach. When data were
considered adequate for a given PAH, it was selected for inclusion.
       A PAH with adequate evidence to suggest no carcinogenicity was selected for inclusion
in the RPF approach and assigned an RPF of zero. While there is little quantitative difference
                                           113
                                                   DRAFT - DO NOT CITE OR QUOTE

-------
 1   between selecting a final RPF of zero for a given PAH and excluding that PAH from the RPF
 2   approach, this is an important distinction for uncertainty analysis.  There is substantial
 3   uncertainty in the risk associated with a PAH that is excluded from the RPF approach due to
 4   inadequate data; this compound could be of low or high potency. However, for a PAH with an
 5   RPF of zero, there is evidence to suggest that this compound is not carcinogenic, and the
 6   uncertainty associated with the cancer risk is markedly reduced. For anthracene, phenanthrene,
 7   and pyrene, it has been determined that the available data support a practical RPF of zero. The
 8   weight of evidence analysis is outlined in Section 6.1 and the results are described in narratives
 9   for each of the 35 individual PAHs (Section 6.2). Chapter 7 describes how the RPFs from
10   multiple datasets were used to derive final RPFs for those PAHs selected for inclusion in the
11   approach, and reports the final RPF information for each PAH.
12
13   6.1.  METHOD FOR SELECTING PAHs FOR INCLUSION IN RELATIVE POTENCY
14   APPROACH
15          For each of the 35 PAHs, a weight of evidence  evaluation was conducted to assess the
16   evidence that each PAH could induce a carcinogenic response. For the purposes of this analysis,
17   PAHs were assumed to be carcinogenic by inferring toxicological similarity to the indicator
18   compound, benzo[a]pyrene. The weight of evidence approach was developed to determine
19   whether the available information  for each PAH was adequate for inclusion of the PAH in the
20   RPF approach. Figure 6-1  shows the decision tree that was used to evaluate the data for each
21   PAH and to determine whether it should be included in the RPF approach.  The weight of
22   evidence evaluation concluded with one of two possible outcomes:
23
24          (1) The data reviewed are adequate to evaluate  carcinogen!city and the PAH should be
25             included in the RPF analysis,  or
26
27          (2) The data reviewed are inadequate to assess carcinogenicity and the PAH should be
28             excluded from the RPF analysis.
29
                                           114          DRAFT - DO NOT CITE OR QUOTE

-------
                                             Has PAH been tested in
                                              tumor bioassay with
                                                  BaPa?
                                             Yes
                                                               No
                                  Did PAH give positive
                                   result in any tumor
                                   bioassays with BaP?
                                Yes
                                                   No
                                                                Do other tumor bioassaysb
                                                                  and/or cancer-related
                                                                  endpoint data0 provide
                                                                  adequate data to assess
                                                                   carcinogenicity?
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
aBioassays with benzo[a]pyrene that met study quality criteria (includes studies with
nonpositive results).
bOther bioassays include those that did not test benzo[a]pyrene and/or those that were not
suitable for RPF derivation (e.g., incidence at lowest dose exceeded 90%).
cCancer-related endpoint data examined in this process included studies of DNA adducts,
clastogenicity or sister chromatid exchange, mutagenicity, morphological transformation,
DNA damage, unscheduled DNA synthesis, etc. that included the selected PAH and
benzo[a]pyrene.

Figure 6-1. Weight of evidence analysis of for selection of PAHs to be
included in the RPF approach.
                                               115
                                                DRAFT - DO NOT CITE OR QUOTE

-------
 1          In vivo tumor bioassays that included benzo[a]pyrene were given the greatest weight in
 2   assessing the carcinogenicity of a given PAH; data from other bioassays and cancer-related
 3   endpoint studies were used to supplement the weight of evidence when the bioassay data that
 4   included benzo[a]pyrene were conflicting or nonpositive. Structural alerts for PAH
 5   carcinogenicity or mutagenicity (specifically, at least four aromatic rings, or the presence of a
 6   classic bay or fjord region formed entirely by aromatic rings) were noted in the evaluation for
 7   each PAH, but were not used explicitly in the weight of evidence evaluation.
 8          When there were bioassays including benzo[a]pyrene with positive findings, and none
 9   with nonpositive findings for a given PAH, that compound was selected for  inclusion in the RPF
10   approach, and no further evaluation of cancer-related endpoint data was conducted.  However,
11   the cancer-related endpoint findings for these compounds were noted in the individual PAH
12   narratives (Section 6.2). Among  the PAHs included in this analysis, there were none with
13   positive bioassay data and robust nonpositive cancer-related  endpoint data.  Were this instance to
14   arise, it would require special consideration, as it might imply a different mode of carcinogenic
15   action than the PAHs addressed herein.
16          Bioassays that met selection criteria (see Section 4.3) were included  in the weight of
17   evidence analysis, regardless of whether positive or nonpositive results were found.  However,
18   the weight of evidence evaluation assumed that a given compound may be active in  one system
19   (e.g., newborn mouse) and inactive or weakly active in another (e.g., dermal initiation).  Thus,
20   when conflicting results were observed in different test systems, different species, or different
21   genders, the PAH was assumed to be carcinogenic based on the positive findings and was
22   included in the RPF approach.
23          In order to evaluate the results of bioassays with positive and nonpositive results in the
24   same test system, an "RPF detection limit" was conceptualized as a means of approximating the
25   minimum RPF that could be determined with respect to the design of the study. The "RPF
26   detection limit" was defined as the RPF determined by the lowest response that would have been
27   statistically significant for the subject PAH and the actual benzo[a]pyrene response. The lowest
28   statistically significant response was calculated using the incidence of tumors in the control
29   group, number of animals  in the group treated with the subject PAH, and Fisher's exact test5
30   (employing a one-sided^-value < 0.05).  Appendix F provides an example calculation of an
31   "RPF detection limit." The utility of this concept is in weighing positive and nonpositive
32   bioassay results.  If all of the nonpositive studies for a subject PAH had "RPF detection limits" in
33   excess of or in the range of what  is observed in the positive studies, then it is plausible that the
34   nonpositive studies may not have been sufficiently sensitive  to estimate the RPF appropriate to
35   the subject PAH. In this event, the PAH was considered carcinogenic and was included in the
36   RPF approach.
     5This calculation was implemented using trial and error within the Fisher's exact test in the online statistical
     calculator, GraphPad6.

                                             116           DRAFT - DO NOT CITE OR QUOTE

-------
 1          If there were no bioassays with benzo[a]pyrene for a given compound, all of the selected
 2   bioassays gave nonpositive results, or inconsistent results could not be explained by test system
 3   or "RPF detection limit", then the results of other bioassays (those without benzo[a]pyrene, or
 4   those rejected from dose-response assessment exclusively because of concerns associated with
 5   benzo[a]pyrene) and cancer-related endpoint data were evaluated. The weight of evidence
 6   analysis then considered all of the following information: bioassays with benzo[a]pyrene, other
 7   bioassays, and cancer-related endpoint data. If these data were determined to be inadequate to
 8   assess the carcinogenicity for a given PAH, then that compound was excluded from the RPF
 9   approach. If the data were considered adequate to assess the carcinogenicity, the compound was
10   retained and a final RPF was derived. Section 6.2 below describes the weight of evidence
11   evaluation for each of the 35 PAHs. Section 7.1 describes how final RPFs were derived for the
12   27 PAHs selected for inclusion in the RPF approach.
13
14   6.2.  WEIGHT OF EVIDENCE EVALUATION FOR 35 INDIVIDUAL PAHs
15          For each PAH, the structure is shown along with a brief reference to any structural alerts
16   for carcinogenicity (specifically, more than three aromatic rings and/or bay or fjord region in
17   alternant PAH).  Next, a brief narrative describing the weight of evidence  evaluation is given,
18   with a graphical representation of the data that were available for RPF  calculation (Figures 6-2 to
19   6-35). The graph for each compound provides a visual representation of the database of studies
20   that included both the subject PAH and benzo[a]pyrene.  The solid bars show the values of the
21   RPFs calculated from all studies with positive findings.  The x-axis label shows the reference for
22   the pertinent study.  The RPFs are color-coded to distinguish among in vivo tumor bioassays
23   based on incidence data, in vivo tumor bioassays based on multiplicity data, in vivo cancer-
24   related endpoint studies, and in vitro cancer-related endpoint studies. Within these categories,
25   the RPFs are ordered (left to right in the graph) from highest to lowest, with positive results
26   shown before nonpositive results.
27          For each nonpositive bioassay, an empty, dotted bar shows what is termed the "RPF
28   detection limit" (see Section 6.1 for description). Missing bars designate cancer-related studies
29   that resulted in nonpositive findings. An RPF detection limit for nonpositive cancer-related
30   studies was not included, because comparisons between nonpositive and positive studies were
31   complicated by the wide variety of study conditions (e.g., test species and strains, metabolic
32   activation sources, assay systems).
33          Each narrative concludes with a statement as to whether the subject PAH was selected for
34   inclusion in the PAH RPF approach. The weight of evidence evaluation for the 35 PAHs with at
35   least one in vivo RPF or at least two in vitro cancer-related endpoint RPFs resulted in the
36   selection of 27 PAHs for inclusion in the RPF approach (see Table 6-2) and the exclusion of
37   8 PAHs from the approach.
38

                                            117           DRAFT - DO NOT CITE OR QUOTE

-------
Table 6-2. Results of weight of evidence evaluation for 27 PAHs selected for
inclusion in the RPF approach
Adequate data: selected for inclusion in RPF approach
PAH
Benzo[a]pyrene
Anthanthrene
Anthracene
B enz [a] anthracene
Benz[b,c]aceanthrylene,
11H-
Benzo [bjfluoranthene
Benzo[c]fluorene
Benz [e] aceanthry lene
Benzo [g,h,i]pery lene
Benz |j ] aceanthry lene
BenzoO]fluoranthene
Benzo [kjfluoranthene
Benz [1] aceanthry lene
Chrysene
CASRN
50-32-8
191-26-4
120-12-7
56-55-3
202-94-8
205-99-2
205-12-9
199-54-2
191-24-2
202-33-5
205-82-3
207-08-9
211-91-6
218-01-9
Abbreviation
BaP
AA
AC
BaA
BbcAC
BbF
BcFE
BeAC
BghiP
BjAC
BjF
BkF
B1AC
CH
PAH
Cyclopenta[c,d]pyrene
Cyclopenta[d,e,f]chrysene,
4H-
Dibenz [a,c] anthracene
Dibenzo[a,e]fluoranthene
Dibenzo[a,e]pyrene
Dibenz [a,h] anthracene
Dibenzo[a,h]pyrene
Dibenzo[a,i]pyrene
Dibenzo[a,l]pyrene
Fluoranthene
Indeno [ 1 ,2,3 -c,d]pyrene
Naphtho [2, 3 -e]py rene
Phenanthrene
Pyrene
CASRN
27208-37-3
202-98-2
215-58-7
5385-75-1
192-65-4
53-70-3
189-64-0
189-55-9
191-30-0
206-44-0
193-39-5
193-09-9
85-01-8
129-00-0
Abbreviation
CPcdP
CPdefC
DBacA
DBaeF
DBaeP
DBahA
DBahP
DBaiP
DBalP
FA
IP
N23eP
PH
Pyr
Inadequate data
PAH
Acepyrene, 2,3-
Benzo[b]fluorene, 11H-
Benzo[e]pyrene
Benzo [g,h,i]fluoranthene
CASRN
25732-74-5
243-17-4
192-97-2
203-12-3
Abbreviation
ACEP
BbFE
BeP
BghiF
PAH
Coronene
Fluorene
Perylene
Triphenylene
CASRN
191-07-1
86-73-7
198-55-0
217-59-4
Abbreviation
CO
FE
Pery
Tphen
                             118
DRAFT - DO NOT CITE OR QUOTE

-------
                                          2,3-Acepyrene (ACEP)
 1
 3
 4          2,3-Acepyrene (CASRN 25732-74-5) is a nonalternant PAH comprised of four aromatic
 5   rings and one five-membered ring.  2,3-Acepyrene does not contain a classic bay or fjord region
 6   in its structure.
 7          Five datasets for 2,3-acepyrene met selection criteria and included benzo[a]pyrene
 8   (shown in Figure 6-2).  Dermal initiation and complete carcinogenicity bioassays in mice
 9   resulted in nonpositive findings (both published by Cavalieri et al., 1981b).  RPF detection limits
10   for these studies were 0.09 and 0.02, respectively.  The limited cancer-related data are mixed,
11   with one positive dataset for in vivo DNA adduct formation, one positive bacterial mutagenicity
12   dataset (both published by Cavalieri et al., 1981a),  and one nonpositive mammalian mutagenicity
13   dataset (Barfknecht et al., 1982).  There are no bioassays of 2,3-acepyrene without
14   benzo[a]pyrene. Overall, the database for 2,3-acepyrene is both limited and inconsistent. The
15   database for 2,3-acepyrene does not provide adequate  information with which to assess
16   carcinogenicity; this PAH was not selected for inclusion in the RPF approach.
17
                                            119           DRAFT - DO NOT CITE OR QUOTE

-------
      100
       10  -
        1  -
o

-------
                                           Anthanthrene (AA)
 2
 3
 4          Anthanthrene (CASRN 191-26-4) is an alternant PAH comprised of six fused aromatic
 5   rings. Anthanthrene does not have a bay or fjord region in its structure.
 6          There are seven datasets for anthanthrene that met selection criteria and included
 7   benzo[a]pyrene (Figure 6-3). The database includes three in vivo tumor bioassays, three
 8   bacterial mutagenicity datasets, and one in vitro DNA damage dataset. Statistically increased
 9   tumor incidences were reported in both a rat lung implantation bioassay (Deutsch-Wenzel et al.,
10   1983) and a dermal complete carcinogenicity bioassay in mice (Cavalieri et al., 1977). No
11   increase over control tumor incidence was reported in a dermal initiation study (Hoffmann and
12   Wynder, 1966), but the RPF detection limit for this study was 0.3. All of the cancer-related
13   endpoint studies gave positive results. Because conflicting bioassay data can be explained by
14   differences in study design (initiation versus complete dermal carcinogenicity), anthanthrene was
15   considered carcinogenic and selected for inclusion in the RPF approach.
16
                                            121           DRAFT - DO NOT CITE OR QUOTE

-------
      100
       10 --
         1  -•
 s
I
0.1  --
     0.01  --
    0.001
                                                  RPF detection limit
                                                   fornonpositive
                                                      bioassay
                                                                                                              Positive bioassay (incidence)
                                                                                                              Positive bioassay (multiplicity)
                                                                                                              In vivo cancer-related endpoint
                                                                                                              In vitro cancer-related endpoint
             Cavalieriet al., 1977   Deutsch-Wenzeletal.,  Hoffman and Wynder,   Andrews etal., 1978      Hermann,1981     Mersch-Sundermann et   Kadenetal., 1979
                                      1983               1966                                                   al., 1992

                                                                        Reference
            Figure 6-3.  Anthanthrene (AA) RTFs.
                                                              122
                                                                                             DRAFT - DO NOT CITE OR QUOTE

-------
                                             Anthracene (AC)
 2
 3
 4          Anthracene (CASRN 120-12-7) is an alternant PAH comprised of three fused aromatic
 5   rings.  Anthracene does not have a bay or fjord region in its structure, and contains less than four
 6   aromatic rings.
 7          Thirty-seven datasets for anthracene met selection criteria and included benzo[a]pyrene,
 8   including 1 dermal initiation tumor bioassay, 3 in vivo clastogenicity or sister chromatid
 9   exchange datasets, 10 bacterial mutagenicity datasets, 4 mammalian mutagenicity datasets,
10   6 morphological/malignant cell transformation datasets, and 13 in vitro DNA adduct, DNA
11   damage, or clastogenicity datasets (Figure 6-4). The single dermal initiation bioassay gave a
12   nonpositive result, with an RPF detection limit of 0.2 (LaVoie et al., 1985). Only two datasets
13   gave positive results:  an in vitro bacterial mutagenicity assay and an in vitro study of DNA
14   damage. The remaining 35 datasets reported nonpositive findings.  To confirm the nonpositive
15   findings in the one tumor bioassay that included benzo[a]pyrene, other bioassays and cancer-
16   related endpoint data for anthracene were considered in the weight  of evidence evaluation.  In
17   bioassays without benzo[a]pyrene, anthracene did not induce a statistically significant increase in
18   tumor incidence in two dermal initiation studies (LaVoie et al., 1983; Salaman and Roe, 1956)
19   and a lung implantation bioassay (Stanton, 1972).  Scribner (1973)  reported a weak tumorigenic
20   response in a dermal initiation study in mice (4/28 mice developed  papillomas by week 35 after
21   dermal treatment with 10 (imol  anthracene in benzene followed by  twice weekly treatment with
22   TPA, as compared with 0/30 control mice, p = 0.048).
23          In vitro assays of mutagenicity (both bacterial and mammalian) are nearly all nonpositive
24   for anthracene (13/14  studies).  Studies  of morphological/malignant cell transformation were all
25   nonpositive. Finally,  in numerous in vitro studies of DNA damage or clastogenicity, anthracene
26   has given nonpositive results (12/13). Sakai et al. (1985) reported a mutagenic response in
27   bacteria treated with anthracene, and Rossman et al. (1991) observed evidence of unscheduled
28   DNA synthesis in Escherichia coli treated with anthracene. Overall, the weight of evidence
29   suggests that anthracene is not carcinogenic. In addition, anthracene lacks all three known
30   structural alerts (at least four rings, bay  or fjord region) for PAH carcinogen!city and/or
31   mutagenicity.  Because the weight of evidence  evaluation suggests  that the data are adequate to
32   assess the carcinogenicity of anthracene, this compound was selected for inclusion in the RPF
33   approach and assigned an RPF of zero.
                                             123          DRAFT - DO NOT CITE OR QUOTE

-------
    100
     10 -
      1 -
o
OJ
3
£   0.1 -
    0.01 -
   0.001
                                                                            Positive bioassay (incidence)
                                                                            Positive bioassay (multiplicity)
                                                                            In vivo cancer-related endpoint
                                                                            In vitro cancer-related endpoint
RPF detection limits
 fornonpositive
   bioassays
       f <**  *"
       w>:
                  .*" ^ .**
                  e!" .*•  *
,-  «• ^r  f .e  «• * ^ «•  «• **>*'<*' S******** **S******* S* **********J?**^^S +*
-------
                                      Benzfajanthracene (BaA)
 2
 3
 4          Benz[a]anthracene (CASRN 56-55-3) is an alternant PAH comprised of four fused
 5   aromatic rings. Benz[a]anthracene contains a bay region but no fjord region in its structure.
 6          There are 65 datasets for benz[a]anthracene that met selection criteria and included
 7   benzo[a]pyrene (Figure 6-5).  Included in the database are tumor bioassays (5), in vivo DNA
 8   adduct studies (4), in vivo clastogenicity studies (4), an in vivo mutagenicity study (1), bacterial
 9   mutagenicity (15), mammalian mutagenicity (14), morphological/malignant cell transformation
10   assays (6), and in vitro studies of DNA damage, adducts, or clastogenicity (16). There are five
11   tumor bioassay datasets of benz[a]anthracene that included benzo[a]pyrene; four gave positive
12   results and one gave a nonpositive result. The positive findings were in different genders tested in
13   a newborn mouse study using intraperitoneal injection (Wislocki et al., 1986); the datasets
14   included both tumor incidence and multiplicity data for both sexes. Positive results were also
15   reported in a dermal initiation study (Slaga et al., 1978). The one nonpositive bioassay (Cavalieri
16   et al., 1977) was a dermal complete carcinogenicity study with an RPF detection limit of 0.2.
17   Benz[a]anthracene was shown to form DNA adducts when administered in vivo in both rats and
18   mice via injection  and gavage (Kligerman et al., 2002). Mutagenicity and morphological/
19   malignant cell transformation assays of benz[a]anthracene were predominantly positive,  as were
20   studies of other cancer-related endpoints.
21          Given that  the differing bioassay results can be attributed to different test systems and
22   study design, benz[a]anthracene was considered carcinogenic and was selected for inclusion in
23   the RPF approach.
24
                                             125          DRAFT - DO NOT CITE OR QUOTE

-------
       100
        10  --
         1  •-
 s
I
0.1  -
      0.01  •
              RPF detection limit
                fornonpositive
                  bioassay
                                                                                                          Positive bioassay (incidence)
                                                                                                          Positive bioassay (multiplicity)
                                                                                                          In vivo cancer-related endpoint
                                                                                                          In vitro cancer-related endpoint
                                                                                                           JL
     0.001
                   ^?<^^^
                   B^ K'fr I
                   S" & e
      4V5^ 'c$^«^
         * Missing bar indicates nonpositive cancer-related endpoint study
            Figure 6-5. Benz[a]anthracene (BaA) RTFs*.
                                                                      Reference
                                                            126
                                                                                        DRAFT - DO NOT CITE OR QUOTE

-------
                                   HH-Benz[b,c]aceanthrylene (BbcAC)
 2
 3
 4          1 lH-Benz[b,c]aceanthrylene (CASRN 202-94-8) is a nonalternant PAH comprised of
 5   four aromatic rings and one five-membered ring. 1 lH-Benz[b,c]aceanthrylene does not contain
 6   a classic bay or fjord region in its structure.
 7          There was only one dataset for benz[b,c]aceanthrylene that met selection criteria and
 8   included benzo[a]pyrene (Figure 6-6).  This multidose dermal initiation study resulted in an RPF
 9   estimate of 0.05 (Rice et al., 1988).  Benz[b,c]aceanthrylene has not been tested in any bioassay
10   without benzo[a]pyrene. There are no cancer-related endpoint data for benz[b,c]aceanthrylene.
11   As the only available bioassay of this PAH was positive, benz[b,c]aceanthrylene was considered
12   carcinogenic and was selected for inclusion in the RPF approach.
13
                                            127          DRAFT - DO NOT CITE OR QUOTE

-------
     100
      10  --
       1  --
o

-------
                                       Benzo[b]fluoranthene (BbF)
 2
 3
 4          Benzo[b]fluoranthene (CASRN 205-99-2) is a nonalternant PAH comprised of four
 5   aromatic rings and one five-membered ring. Benzo[b]fluoranthene contains one classic bay
 6   region but no fjord region in its structure.
 7          There were 22 datasets of benzo[b]fluoranthene that met selection criteria and included
 8   benzo[a]pyrene (Figure 6-7). Included in the database are in vivo tumor bioassay datasets (8), in
 9   vivo DNA adduct datasets (7), in vivo clastogenicity datasets (3), mutagenicity and
10   morphological/malignant cell transformation datasets (3), and an in vitro DNA damage dataset
11   (1).  Statistically significant increases in tumor incidence and/or multiplicity were reported in
12   male mice tested in two newborn mouse bioassays using intraperitoneal injection (Nesnow et al.,
13   1998b; LaVoie et al., 1987), in dermal initiation (LaVoie et al., 1982) and dermal complete
14   carcinogenicity (Habs et al.,  1980) bioassays, and in a rat lung implantation bioassay (Deutsch-
15   Wenzel et al., 1983). The one nonpositive result was in female mice tested in the newborn
16   mouse bioassay; the RPF detection limit was 0.8 (LaVoie et al.,  1987). A number of studies
17   showed that benzo[b]fluoranthene forms DNA adducts  when administered in vivo to rats or mice
18   via injection or gavage (Kligerman et al., 2002; Nesnow et al., 1998b, 1993b).  One mutagenicity
19   assay and two morphological/malignant cell transformation assays of benzo[b]fluoranthene were
20   positive, as were studies of other cancer-related endpoints; there were no nonpositive studies of
21   cancer-related endpoints. Given that the differing bioassay results can be attributed to different
22   genders, benz[a]anthracene was considered carcinogenic and was selected for inclusion in the
23   RPF approach.
                                            129           DRAFT - DO NOT CITE OR QUOTE

-------
     100
      10  -•
       1  -•
o
u
s
     0.1  -•
    o.oi  --
   0.001  -I
RPF detection limit
  fornonpositive
    bioassay
Positive bioassay (incidence)
Positive bioassay (multiplicity)
In vivo cancer-related endpoint
In vitro cancer-related endpoint
                     1
       4°
      S?
   Figure 6-7. Benzo[b]fluoranthene (BbF) RTFs.
                                                         **•   y  **•
                                                                Reference
                                                          ?'   
-------
                                      HH-Benzo[b]fluorene (BbFE)
 4          1 lH-Benzo[b]fluorene (CASRN 243-17-4) is a nonalternant PAH comprised of three
 5   aromatic rings and one five-membered ring.  1 lH-Benzo[b]fluorene does not contain a classic
 6   bay or fjord region in its structure.
 7          There were three datasets for 1 lH-benzo[b]fluorene that met selection criteria and
 8   included benzo[a]pyrene (Figure 6-8): two mutagenicity datasets and an in vitro DNA damage
 9   dataset. There are no bioassays of 1 lH-benzo[b]fluorene that included benzo[a]pyrene, so
10   bioassays without benzo[a]pyrene and cancer-related endpoint data were considered.  LaVoie et
11   al. (1981) conducted a study of skin tumor initiation in mice treated with 1 mg 1 lH-benzo[b]-
12   fluorene followed by 20 weeks of treatment with TPA. The incidence of tumor-bearing animals
13   (4/20) was not significantly increased over controls (1/20) (LaVoie et al., 1981).  The limited
14   cancer-related endpoint data were mixed, with one positive  mutagenicity study (Kaden et al.,
15   1979), one nonpositive mutagenicity study (Hermann, 1981), and one positive in vitro study of
16   DNA damage (Mersch-Sundermann et al.,  1992). Overall, the database for 1 lH-benzo[b]-
17   fluorene is both limited and inconsistent. Because the database for 1 lH-benzo[b]fluorene does
18   not provide adequate information with which to assess carcinogenicity, this PAH was not
19   selected for inclusion in the RPF approach.
20
                                            131          DRAFT - DO NOT CITE OR QUOTE

-------
     100
       10 -
        1  -
o
3J
s
      0.1  -
    0.01  -
   0.001
                                                                                                         Positive bioassay (incidence)

                                                                                                         Positive bioassay (multiplicity)

                                                                                                         In vivo cancer-related endpoint

                                                                                                         In vitro cancer-related endpoint
                          Kadenetal., 1979



        * Missingbar indicatesnonpositive cancer-related endpoint study




        Figure 6-8.  llH-Benzo[b]fluorene (BbFE) RTFs*.
Mersch-Sundermann etal., 1992


 Reference
Hermann, 1981
                                                          132
                                DRAFT - DO NOT CITE OR QUOTE

-------
                                        Benzo[c]fluorene (BcFE).
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
       Benzo[c]fluorene (CASRN 205-12-9) is a nonalternant PAH comprised of three aromatic
rings and one five-membered ring. Benzo[c]fluorene does not contain a classic bay or fjord
region in its structure.
       There were six datasets for benzo[c]fluorene that met selection criteria and included
benzo[a]pyrene (Figure 6-9); all gave positive results.  The database includes oral and
intraperitoneal in vivo tumor bioassays (each reporting both incidence and multiplicity) and in
vivo DNA adduct data. Significantly increased lung tumor incidence and tumor multiplicity
were reported after both oral and intraperitoneal exposure (Weyand et al., 2004).  As the
available bioassays that included benzo[a]pyrene were positive, benzo[c]fluorene was considered
carcinogenic and was selected for inclusion in the RPF approach.
                                            133
                                                    DRAFT - DO NOT CITE OR QUOTE

-------
     100
       10 -•
        1  -•
o
u
j3
13
      0.1  -•
     0.01  -•
   0.001
                                                                                    Positive bioassay (incidence)
                                                                                    Positive bioassay (multiplicity)
                                                                                    In vivo cancer-re la ted endpoint
                                                                                    In vitro cancer-related endpoint
            Weyandetal.,2004   Weyandetal., 2004   Weyandetal., 2004   Weyandetal., 2004   Weyandetal., 2004   Weyandetal., 2004
                                                              Reference
    Figure 6-9. Benzo[c]fluorene (BcFE) RTFs.
                                                     134
DRAFT - DO NOT CITE OR QUOTE

-------
                                      Benzfejaceanthrylene (BeAC).
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
       Benz[e]aceanthrylene (CASRN 199-54-2) is a nonalternant PAH comprised of four
aromatic rings and one five-membered ring.  Benzfejaceanthrylene contains a classic bay region
but no fjord region in its structure.
       There were six datasets for benz[e]aceanthrylene that met selection criteria and included
benzo[a]pyrene (Figure 6-10); all gave positive results.  The database includes an in vivo tumor
bioassay in two sexes (each reporting both incidence and multiplicity), a mammalian
mutagenicity study, and a morphological/malignant cell transformation study. Significantly
increased tumor incidence and tumor multiplicity were reported for both male and female mice
in a dermal initiation bioassay in mice (Nesnow et al., 1984). As the available bioassay that
included benzo[a]pyrene was positive, benz[e]aceanthrylene was considered carcinogenic and
was selected for inclusion in the RPF approach.
                                            135
                                                    DRAFT - DO NOT CITE OR QUOTE

-------
     100
      10  -•
       1  -•
o
u
s
     0.1  -•
    0.01  -•
   0.001
                                                                                            Positive bioassay (incidence)
                                                                                            Positive bioassay (multiplicity)
                                                                                            In vivo cancer-related endpoint
                                                                                            In vitro cancer-related endpoint
            Nesnowet al., 1984   Nesnowet al., 1984   Nesnowet al., 1984   Nesnowet al., 1984   Nesnowet al., 1984b Mohapatraetal., 1987
                                                              Reference
     Figure 6-10. Benz[e]aceanthrylene (BeAC) RTFs.
                                                      136
DRAFT - DO NOT CITE OR QUOTE

-------
                                          Benzofejpyrene (BeP)
 4          Benzo[e]pyrene (192-97-2) is an alternant PAH comprised of five fused aromatic rings.
 5   Benzo[e]pyrene contains two bay regions but no fjord region in its structure.
 6          Thirty-seven datasets for benzo[e]pyrene met selection criteria and included
 7   benzo[a]pyrene: 2 tumor bioassays, 1 in vivo clastogenicity dataset, 12 bacterial mutagenicity
 8   datasets, 4 mammalian mutagenicity datasets, 7 morphological/malignant cell transformation
 9   datasets, and 11 in vitro DNA damage or clastogenicity datasets (Figure 6-11). No increase in
10   tumor incidence was observed when benzo[e]pyrene was tested alone as part of a dermal
11   cocarcinogenicity bioassay (Van Duuren and Goldschmidt, 1976). When tested in a lung
12   implantation bioassay in rats, benzo[e]pyrene exposure did not result in a significant increase in
13   tumor incidence (Deutsch-Wenzel et al.,  1983).  The RPF detection limits of these studies were
14   approximately 0.01 and 0.1.  To confirm the nonpositive findings in the available tumor
15   bioassays that included benzo[a]pyrene, other bioassays and cancer-related endpoint data were
16   considered.  In bioassays without benzo[a]pyrene, benzo[e]pyrene gave nonpositive results in a
17   dermal initiation bioassay (1 mg/mouse; Van Duuren et al., 1968) and a newborn mouse bioassay
18   (0.7 (imol; Chang et al.,  1981). A significant increase in tumor incidence was reported in a
19   single-concentration dermal initiation study in mice; 11/13 surviving mice (20 were treated) had
20   papillomas by week 35 after dermal treatment with 10 |imol benzo[e]pyrene in benzene
21   (p < 0.0001), followed by twice weekly treatment with TPA; no control mice had papillomas
22   (Scribner, 1973).
23          In vitro assays of mutagenicity (both bacterial and mammalian) and morphological/
24   malignant cell transformation give inconsistent results for benzo[e]pyrene; 11/23 studies were
25   positive and the rest were nonpositive.  Positive studies include a mix of bacterial mutagenicity
26   and morphological/malignant cell transformation assays; four mammalian mutagenicity assays
27   were nonpositive.  One study of in vivo clastogenicity and two studies of in vitro DNA damage
28   were positive, while nine studies of in vitro DNA damage or clastogenicity were nonpositive.
29          While the database for benzo[e]pyrene is quite large, the results are inconsistent; as a
30   result, no conclusion can be drawn as to carcinogenicity. This PAH was not selected for
31   inclusion in the RPF approach.
                                            137           DRAFT - DO NOT CITE OR QUOTE

-------

   0.
1UU


10 -


1 -


0.1 -





0.01 -



nm -


| | Positive bioassay (incidence)
| | Positive bioassay (multiplicity)
I In vivo cancer-related endpoint

RPF detection
limits for
nonpositive
bioassays
1\
r~\
i \
|
j
i
I
•





r
I p
j
j


j
i
j_!_. 	 LH





















H-
-

















| | In vitro cancer-related endpoint
































































-





















-
























































^_^



fl n
1-1 1-1



     <^ -
-------
                                     Benzo[g,h,i]fluoranthene (BghiF)
 2
 o
 5
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
       Benzo[g,h,i]fluoranthene (CASRN 203-12-3) is a nonalternant PAH comprised of four
aromatic rings and one five-membered ring. Benzo[g,h,i]fluoranthene does not contain a classic
bay or fjord region in its structure.
       There were six datasets for benzo[g,h,i]fluoranthene that met selection criteria and
included benzo[a]pyrene (Figure 6-12). A dermal initiation bioassay in mice (Van Duuren et al.,
1966) did not result in a statistically significant increase in tumor incidence; the RPF detection
limit was 0.06.  There were no other bioassays that met selection criteria.  There were three
positive bacterial mutagenicity studies (Chang et al., 2002; Lafleur et al., 1993; Carver et al.,
1986), one  positive study of in vitro DNA damage (Mersch-Sundermann et al., 1992), and a
mammalian mutagenicity study with nonpositive results (Lafleur et al., 1993). The RPF values
for the positive cancer-related endpoint datasets ranged from 0.6 to 1.  Overall, the database for
benzo[g,h,i]fluroanthene is both limited and inconsistent. Because the database for
benzo[g,h,i]fluoranthene does not provide adequate information with which to assess
carcinogenicity, this PAH was not selected for inclusion in the RPF approach.
                                             139
                                                     DRAFT - DO NOT CITE OR QUOTE

-------
      100
        10  -
         1  -
 o
 OJ
 3
I
       0.1  -
     0.01  -
    0.001
RPF detection limits
  fornonpositive
    bioassays
                                                                                                             Positive bio assay (incidence)
                                                                                                             Positive bioassay (multiplicity)
                                                                                                             In vivo cancer-related endpoint
                                                                                                             In vitro cancer-related endpoint
              Van Duurenetal., 1966      LaFleuretal., 1993        Carver etal., 1986        Chang etal., 2002
                                                                     Reference
    * Missing bar indicates nonpositive cancer-related endpoint study


             Figure 6-12. Benzo[g,h,i]fluoranthene (BghiF) RTFs*.
                                                                                     Mersch-Sundermannetal.,    LaFleuretal., 1993
                                                                                             1992
                                                                 140
                                                                                      DRAFT - DO NOT CITE OR QUOTE

-------
                                       Benzo[g,h,i]perylene (BghiP)
 2
 3
 4          Benzo[g,h,i]perylene (CASRN 191-24-2) is an alternant PAH comprised of six fused
 5   aromatic rings. Benzo[g,h,i]perylene contains a bay region but no fjord region in its structure.
 6          There were 10 datasets for benzo[g,h,i]perylene that met selection criteria and included
 7   benzo[a]pyrene (Figure 6-13).  The database includes three in vivo tumor bioassays, four
 8   bacterial mutagenicity datasets, an in vitro DNA damage dataset, and two in vitro DNA adduct
 9   datasets. Of the three bioassays, positive findings were only reported in one: a rat lung
10   implantation bioassay (Deutsch-Wenzel et al., 1983) that resulted in an RPF estimate of 0.009.
11   In a dermal initiation bioassay (Hoffmann and Wynder,  1966) and a dermal cocarcinogenicity
12   bioassay (Van Duuren and Goldschmidt, 1976), there was no statistically  significant increase in
13   tumor incidence, but these studies had relatively insensitive RPF detection limits (around 0.1)
14   compared with the positive study. There were four positive mutagenicity studies; all were
15   conducted in bacterial systems.  Studies of in vitro DNA adducts and DNA damage were
16   positive. Because the inconsistent bioassay  results can be attributed to different test systems
17   (different species and route), benzo[g,h,i]perylene was considered carcinogenic and was selected
18   for inclusion in the RPF approach.
                                             141           DRAFT - DO NOT CITE OR QUOTE

-------
      100
       10  --
        1  --
o
0>
s
      0.1  --
     0.01  --
   0.001
                                                                                                            Positive bioassay (incidence)
                                                                                                            Positive bioassay (multiplicity)
                                                                                                            In vivo cancer-related endpoint
                                                                                                            In vitro cancer-related endpoint
                  RPF detection limits for
                   nonpositive bioassays
Deutsch-Wenzel  Hoffmannand VanDuurenand Andrews etal., Sakaietal., 1985  Gibson etal.,   Kadenetal.,
  etal.,1983    Wynder, 1966   Goldschmidt,      1978                       1978         1979
                               1976
                                                                                                           Mersch-     Segerbackand   Cherngetal.
                                                                                                         Sundermannet  Vodicka, 1993     2001
                                                                                                           al.,1992
             * Missing bar indicates nonpositive cancer-related endpoint study


            Figure 6-13. Benzo[g,h,i]perylene (BghiP) RTFs*.
                                                                        Reference
                                                                 142
                                                                                           DRAFT - DO NOT CITE OR QUOTE

-------
                                       Benz[j]aceanthrylene (BjAC)
 2
 3
 4          Benz[j]aceanthrylene (CASRN 202-33-5) is a nonalternant PAH comprised of four
 5   aromatic rings and one five-membered ring.  Benz[j]aceanthrylene contains a classic bay region
 6   but no fjord region in its structure.
 7          There were 12 datasets for benz[j]aceanthrylene that met selection criteria and included
 8   benzo[a]pyrene (Figure 6-14); all of the studies gave positive results. The database includes one
 9   in vivo tumor bioassay dataset, one in vivo DNA adduct dataset, four mutagenicity or
10   morphological/malignant cell transformation datasets, and six in vitro DNA damage or DNA
11   adduct datasets. In a bioassay of benz[j]aceanthrylene that used intraperitoneal injection in an
12   A/J mouse system (Mass et al., 1993), all mice treated with benz[j]aceanthrylene developed
13   tumors (incidence of 100% at doses of 20-100 mg/kg; incidence for benzo[a]pyrene was 63-
14   100% across the same dose range), precluding the derivation of an RPF using incidence data.
15   However, tumor multiplicity (average number of tumors per animal) data were available for
16   dose-response modeling and resulted in an RPF estimate of 60.  Benz[j]aceanthrylene treatment
17   resulted in a pronounced increase in the average number of tumors per animal  (59.45 tumors per
18   animal at 20 mg/kg), much higher than benzo[a]pyrene treatment (5.05 tumors per animal at
19   100 mg/kg), indicating that this compound is very potent in this test system. In a dermal
20   initiation bioassay that did not include benzo[a]pyrene, benz[j]aceanthrylene induced papillomas
21   in 90% of mice treated with an initiating dose of 40 jig (compared with 5% incidence in
22   controls). As the available bioassay that included benzo[a]pyrene was positive and suggested
23   that this compound is very potent, benz[j]aceanthrylene was considered carcinogenic and was
24   selected for inclusion in the RPF approach.
25
                                            143          DRAFT - DO NOT CITE OR QUOTE

-------

iuu -
10 -
-
0.1 -
0.01 -





























^ Positive bioassay (incidence)
I | Positive bioassay (multiplicity)
I | In vivo cancer-related endpoint
| | In vitro cancer-related endpoint













 Mass etal.,   Mass etal., Johnsenetal., Johnsenetal., Johnsenetal., Sangaiah et Nesnowetal., Mohapatraet Johnsenetal., Johnsenetal., Johnsenetal., Johnsenetal.
   1993       1993       1997        1997       1997     al., 1983       1984      al., 1987      1998        1998       1998       1998


                                                            Reference
Figure 6-14. Benz[j]aceanthrylene (BjAC) RTFs.
                                                   144
DRAFT - DO NOT CITE OR QUOTE

-------
                                       Benzofjjfluoranthene (BjF)
 2
 3
 4          Benzo[j]fluoranthene (CASRN 205-82-3) is a nonalternant PAH comprised of four
 5   aromatic rings and one five-membered ring.  Benzo[j]fluoranthene does not contain a classic bay
 6   or fjord region in its structure.
 7          There were eight datasets for benzo[j]fluoranthene that met selection criteria and
 8   included benzo[a]pyrene (Figure 6-15):  seven in vivo tumor bioassay datasets and one in vitro
 9   study of DNA damage. Of the seven bioassay datasets, significant increases in tumor incidence
10   or count were observed in all but one. Significant increases in tumor incidence were reported in
11   both male and female mice tested in a newborn mouse bioassay using intraperitoneal injection of
12   single doses (LaVoie et al., 1987), a mouse dermal initiation study (LaVoie et al., 1982), and a
13   rat lung implantation bioassay (Deutsch-Wenzel et al.,  1983).  Significant increases in tumor
14   multiplicity were reported in two mouse dermal initiation studies (Weyand et al., 1992; LaVoie
15   et al., 1982).  The one  nonpositive bioassay was a mouse dermal complete carcinogenicity
16   bioassay with an RPF detection limit of 0.1 (Habs et al., 1980). The in vitro study of DNA
17   damage gave positive results (Mersch-Sundermann et al., 1992). Because the inconsistent
18   bioassay results can be attributed to different test systems or study design, benzo[j]fluroanthene
19   was considered carcinogenic and was selected for inclusion in the RPF approach.
20
                                            145          DRAFT - DO NOT CITE OR QUOTE

-------
      100
       10  --
        1  --
PH
rt
!*.
 o
 OJ
"08
      0.1  --
     0.01  --
    0.001
                                                                                   B
                                                        Positive bio assay (incidence)
                                                        Positive bioassay (multiplicity)
                                                        In vivo cancer-related endpoint
                                                        In vitro cancer-related endpoint
n
RPF detection
  limit for
 nonpositive
  bioassay
  JL
                                                                                      -H
            LaVoieet al., 1987 LaVoieet al., 1987 LaVoieet al., 1982 LaVoieet al., 1982 Deutsch-Wenzel et Weyandet al., 1992  Habsetal., 1980      Mersch-
                                                                             al., 1983                                   Sundermannetal..
                                                                                                                            1992
                                                                   Reference
     Figure 6-15. Benzo[j]fluoranthene (BjF) RTFs.
                                                       146
                                                            DRAFT - DO NOT CITE OR QUOTE

-------
                                       Benzo[k]fluoranthene (BkF)
 1
 3
 4          Benzo[k]fluoranthene (CASRN 207-08-9) is a nonalternant PAH comprised of four
 5   aromatic rings and one five-membered ring.  Benzo[j]fluoranthene does not contain a classic bay
 6   or fjord region in its structure.
 7          There were five datasets for benzo[k]fluoranthene that met selection criteria and included
 8   benzo[a]pyrene (Figure 6-16).  The database includes four in vivo tumor bioassay datasets and
 9   one morphological/malignant cell transformation dataset.  Statistically significant increases in
10   tumor incidence and tumor count were reported in a mouse dermal initiation study (LaVoie et al.,
11   1982) and increased tumor incidence was reported in a rat lung implantation bioassay (Deutsch-
12   Wenzel et al., 1983). No significant increase in tumor incidence was observed in a dermal
13   complete carcinogenicity study with an RPF detection limit of 0.1 (Habs et al., 1980). The
14   morphological/malignant cell transformation study (Emura et al., 1980) was nonpositive.
15   Because the inconsistent bioassay results can be attributed to different test systems or study
16   design (dermal initiation versus dermal complete carcinogenicity), benzo[k]fluroanthene was
17   considered carcinogenic and was selected for inclusion in the RPF approach.
18
                                            147          DRAFT - DO NOT CITE OR QUOTE

-------
      100
       10  --
        1  --
 o
 3J


I
      0.1  --
     0.01  --
    0.001
                                                                                                   Positive bioassay  (incidence)
                                                                                                   Positive bioassay  (multiplicity)
                                                                                                   In vivo cancer-related endpoint
                                                                                                   In vitro cancer-related endpoint
                                                                                      RPF detection limits for
                                                                                       nonpositive bioassays
              Deutsch-Wenzel et al..
                     1983
                             LaVoieetal., 1982


* Missingbar indicatesnonpositive cancer-related endpoint study
LaVoieetal., 1982
   Reference
HabsetaL 1980
EmuraetaL 1980
          Figure 6-16. Benzo[k]fluoranthene (BkF) RTFs*.
                                                           148
                                                                                      DRAFT - DO NOT CITE OR QUOTE

-------
                                       Benzfljaceanthrylene (BIAC)
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
       Benzfljaceanthrylene (CASRN 211-91-6) is a nonalternant PAH comprised of four
aromatic rings and one five-membered ring. Benz[l]aceanthrylene does not contain a classic bay
or fjord region in its structure.
       There were 16 datasets for benz[l]aceanthrylene that met selection criteria and included
benzo[a]pyrene (Figure 6-17); all of the studies gave positive results.  The database includes four
in vivo tumor bioassay datasets, five mutagenicity or morphological/malignant cell
transformation datasets, one in vivo clastogenicity dataset, and six in vitro DNA adduct or DNA
damage datasets.  Significant increases in tumor count and multiplicity were reported in both
male and female mice in a dermal initiation bioassay (Nesnow et al., 1984). All of the cancer-
related endpoint studies were positive as well. Relative potency estimates for most of the
available datasets were >1.0, suggesting equivalent or greater potency than benzo[a]pyrene.  As
the available bioassays that included benzo[a]pyrene were positive, benz[l]aceanthrylene was
considered carcinogenic and was selected for inclusion in the RPF approach.
                                             149
                                                     DRAFT - DO NOT CITE OR QUOTE

-------
  100
   10  -•
    1  -•
  0.1  -•
 0.01  -•
0.001


                                                                                            Positive bioassay (incidence)
                                                                                            Positive bioassay (multiplicity)
                                                                                            In vivo cancer-related endpoint
                                                                                            In vitro cancer-related endpoint
                                             ^T     ^    ^    ^    ^     ^     ^    ^    ^    ^

                                                  ^     ^     ^     **    ^     &     f     >n     ^

                                                            Reference


      Figure 6-17. Benz[l]aceanthrylene (B1AC) RTFs.
                                                     150
DRAFT - DO NOT CITE OR QUOTE

-------
                                              Chrysene (CH)
 1
 3
 4          Chrysene (CASRN 218-01-9) is an alternant PAH comprised of four fused aromatic
 5   rings.  Chrysene contains two bay regions but no fjord region in its structure.
 6          There were 40 datasets for chrysene that met selection criteria and included
 7   benzo[a]pyrene (Figure 6-18). Included in the database are 13 in vivo tumor bioassay datasets,
 8   4 in vivo DNA adduct datasets, 3 in vivo clastogenicity datasets, 11 mutagenicity datasets,
 9   3 morphological/malignant cell transformation datasets, and 6 in vitro studies of DNA damage,
10   adducts, or clastogenicity. Among the bioassays that included benzo[a]pyrene, 11 reported
11   significant increases in tumor incidence or tumor multiplicity, and 3 did not. Significant
12   increases in tumor incidence and/or multiplicity were reported in three dermal  initiation studies
13   in mice (Rice et al., 1988; Slaga et al., 1980; Hecht et al., 1974), a newborn mouse study in
14   males (Wislocki et al., 1986), and a rat lung implantation bioassay (Wenzel-Hartung et al.,
15   1990). Female mice tested in the newborn mouse assay published by Wislocki et al. (1986) did
16   not have a significant increase in tumor incidence, resulting in one of the three nonpositive
17   studies. The other two nonpositive findings were in males and females tested in another
18   newborn mouse bioassay (Busby et al.,  1989). The bioassays with nonpositive findings had RPF
19   detection limits between 0.06 and 0.2. Conflicting results in male mice were reported in the two
20   newborn mouse bioassays (Busby et al., 1989; Wislocki et al., 1986). The major difference
21   between the two studies is the duration of follow-up; Busby et al. (1989) sacrificed the mice at
22   26 weeks, while Wislocki et al. (1986) followed the mice for a full year.  LaVoie et al. (1994)
23   observed that liver tumor induction in the newborn mouse bioassay is not fully realized until the
24   mice have reached 1 year of age, and the positive findings by Wislocki et al. (1986) indeed
25   reflect liver tumors in the male mice. Chrysene was shown to form DNA adducts when
26   administered in vivo in both rats and mice via injection and gavage (Kligerman et al., 2002).
27   Bacterial and mammalian mutagenicity and morphological/malignant cell transformation assays
28   of chrysene were all positive, as were studies of clastogenicity tested in vivo. In contrast, results
29   from in vitro studies of DNA adducts, DNA damage, and clastogenicity were not consistent.
30          Because the inconsistent bioassay results can be attributed to different study designs
31   (gender, follow-up time), chrysene was considered carcinogenic and was selected for inclusion in
32   the RPF approach.
33
                                            151           DRAFT - DO NOT CITE OR QUOTE

-------
  100
   10 -•
     1  --
 0.01  --
0.001
                                                                                                       Positive bioassay (incidence)
                                                                                                       Positive bioassay (multiplicity)
                                                                                                       In vivo cancer-related endpoint
                                                                                                       In vitro cancer-related endpoint
                          RPF detection limits for
                          nonpositive bioassays
- -  I^M 11 I M ^ i I M F 11  I M M^I! -i •  '\' -\> M I I M M I  I M M I I M M I M I  M I  M I I M M I I M  M I  I M M I I M M—i—i—i—i—i—i—i—h

>^^X^

^^^^^^^^^^^^^^^%^*'
                   A&                 ^              (^       NN                 ^^^   6,^^   AJ^         ^   GT^
                                     ^                              ^° ^°
            * Mis sing bar indicates nonpositive cancer-related endpoint study


           Figure 6-18. Chrysene (CH) RTFs*.
                                                              Reference
                                                              152
                                                                                            DRAFT - DO NOT CITE OR QUOTE

-------
                                              Coronene (CO)
 2
 3
 4          Coronene (CASRN 191-07-1) is an alternant PAH comprised of seven fused aromatic
 5   rings.  Coronene contains no bay or fjord regions in its structure.
 6          There were six datasets for coronene that met selection criteria and included
 7   benzo[a]pyrene (Figure 6-19).  A dermal complete carcinogenicity bioassay in mice did not
 8   result in a statistically significant increase in tumor incidence (Habs et al., 1980); the RPF
 9   detection limit was 0.06.  To confirm the nonpositive findings in the one tumor bioassay that
10   included benzo[a]pyrene, other bioassays and cancer-related endpoint data were considered.
11   There was one bioassay of coronene that did not include benzo[a]pyrene. Van Duuren et al.
12   (1968) conducted a dermal initiation bioassay of coronene using groups of 20 mice (0.5 mg
13   coronene in 0.5 mL benzene, followed by croton resin treatment until death).  Although the
14   authors characterized coronene as a weak tumor initiator, the incidence of tumors was not
15   significantly increased over concurrent controls. The limited cancer-related endpoint data were
16   mixed, with three positive bacterial mutagenicity studies (with RPFs ranging from 0.01 to 0.5),
17   one nonpositive bacterial mutagenicity study, and a nonpositive in vitro DNA damage study.
18          Overall, the database for coronene is both limited and inconsistent.  Because the database
19   for coronene does not provide adequate information with which to assess carcinogenicity, this
20   PAH was not selected for inclusion in the RPF approach.
                                            153           DRAFT - DO NOT CITE OR QUOTE

-------
o
V
s
     100
       10 -
        1 -
      0.1 -
    0.01 -
   0.001
                                                                                                         Positive bioassay (incidence)
                                                                                                         Positive bioassay (multiplicity)
                                                                                                         In vivo cancer-related endpoint
                                                                                                         In vitro cancer-related endpoint
   RPF detection limits for
    nonpositive bioassays
           J
                     -+-
                                          -+-
                                                               -+-
                                                                                    -+-
                                                                                                          •+•
              Habsetal., 1980       Hermann,1981        Sakaietal., 1985      Florin etal., 1980     Kadenetal., 1979   Mersch-Sundermannet
                                                                                                                            al., 1992
* Missingbarindicatesnonpositive cancer-related endpoint study
                                                                 Reference
          Figure 6-19. Coronene (CO) RTFs*.
                                                            154
                                                                                       DRAFT - DO NOT CITE OR QUOTE

-------
                                       Cyclopenta[c,d]pyrene (CPcdP)

                                                          \
 2
 3
 4          Cyclopenta[c,d]pyrene (CASRN 27208-37-3) is a nonalternant PAH comprised of four
 5   aromatic rings and one five-membered ring.  Cyclopenta[c,d]pyrene does not contain a classic
 6   bay or fjord region in its structure.
 7          There were 25 datasets for cyclopenta[c,d]pyrene that met selection criteria and included
 8   benzo[a]pyrene (Figure 6-20). The database includes 11 in vivo tumor bioassay datasets, 2 in
 9   vivo DNA adduct datasets, 11 studies of mutagenicity or morphological/malignant cell
10   transformation, and a single study of in vitro clastogenicity. Nine of the 11 tumor bioassay
11   datasets and all of the cancer-related  endpoint studies gave positive results.  Statistically
12   significant increases in tumor incidence and/or multiplicity were reported in two dermal
13   complete carcinogenicity bioassay (Cavalieri et al., 1983, 1981b), two dermal initiation
14   bioassays (Raveh et al., 1982; Cavalieri  et al., 1981b), and an intraperitoneal study using adult
15   A/J mice (Nesnow et al., 1998b).  Bioassays in which no significant increase in tumorigenicity
16   was observed included a dermal initiation (Wood et al., 1980) and complete carcinogenicity
17   study (Habs et al., 1980); these studies had RPF detection limits of 0.1 and 0.03, respectively.
18   After obtaining nonpositive results for low initiating doses of cyclopenta[c,d]pyrene, Wood et al.
19   (1980) repeated their experiment with higher doses and observed statistically significant
20   increases in tumor incidence. In the latter experiment, benzo[a]pyrene was not included, so an
21   RPF could not be calculated from these data. The study design of the nonpositive complete
22   carcinogenicity bioassay was quite similar to that of the two positive studies of this type, with the
23   exception of the mouse strain used; Habs et al.  (1980) used NMRI mice, while Cavalieri et al.
24   (1983, 198Ib) used Swiss mice. Although the differing results in dermal complete
25   carcinogenicity studies may be explained by slight differences in strain susceptibility, these two
26   strains are of common origin, which argues against this explanation.
27          The available cancer-related endpoint data indicate that cyclopenta[c,d]pyrene is
28   mutagenic and capable of morphological/malignant cell transformation in vitro; a single study of
29   in vitro clastogenicity was also positive.  Overall, the data supporting  a finding of
30   carcinogenicity for cyclopenta[c,d]pyrene are very consistent, and this compound was selected
31   for inclusion in the RPF approach.
                                             155          DRAFT - DO NOT CITE OR QUOTE

-------
         100
          10 -•
           i --n
PH
«
!*.
 o

          0.1 -•
        0.01 -•
                                                 RPF detection limits
                                                   fornonpositive
                                                     bioassays
       0.001
                                                                                                   Positive bioassay (incidence)
                                                                                                   Positive bioassay (multiplicity)
                                                                                                   In vivo cancer-related endpoint
                                                                                                   In vitro cancer-related endpoint
I '  '  I '  ' I  '  ' I '  '  I '  ' I  '—' I  ' '  I '  ' I  '  ' I  ' '  I
                                      c?
                                                                                             0° _**
                                                                   Reference
           Figure 6-20.  Cyclopenta[c,d]pyrene (CPcdP) RTFs.
                                                            156
                DRAFT - DO NOT CITE OR QUOTE

-------
                                  4H-Cyclopenta[d, e,fjchrysene (CPdefC)
 1
 3
 4          4H-Cyclopenta[d,e,f]chrysene (CASRN 202-98-2) is a nonalternant PAH comprised of
 5   four aromatic rings and one five-membered ring. 4H-Cyclopenta[d,e,f]chrysene contains a
 6   classic bay region but no fjord region in its structure.
 7          There were two datasets for 4H-cyclopenta[d,e,f]chrysene that met selection criteria and
 8   included benzo[a]pyrene (Figure 6-21); both were multidose dermal initiation datasets (Rice et
 9   al., 1988, 1985). Rice et al. (1988) reported a statistically significant increase in tumor incidence
10   in a multidose dermal initiation study.  In the second study, the incidence of tumors after
11   treatment with cyclopenta[d,e,f]chrysene exceeded 90%, precluding RPF derivation from
12   incidence data, but tumor multiplicity data were available for RPF calculation (Rice et al., 1985).
13   Cyclopenta[d,e,f]chrysene has not been tested in a bioassay without benzo[a]pyrene; however,
14   sterically hindered diol epoxides of this compound have given positive results in a newborn
15   mouse assay (Amin et al., 1995).  Because the bioassay of cyclopenta[d,e,f]chrysene was
16   positive, this PAH was considered carcinogenic and was selected for inclusion in the RPF
17   approach.
                                            157          DRAFT - DO NOT CITE OR QUOTE

-------
    100
     10 --
      1 --
J3
13
    o.i --
   o.oi -•
  0.001
                               RiceetaL 1988
                                                                                            Positive bioassay (incidence)
                                                                                            Positive bioassay (multiplicity)
                                                                                            In vivo cancer-relatedendpoint
                                                                                            In vitro cancer-related endpoint
RiceetaL, 1985
                                                             Reference
           Figure 6-21. Cyclopenta[d,e,f]chrysene (CPdefC) RTFs.
                                                          158
    DRAFT - DO NOT CITE OR QUOTE

-------
                                      Dibenz[a,c]anthracene (DBacA)
 2
 3
 4          Dibenz[a,c]anthracene (CASRN 215-58-7) is an alternant PAH comprised of five fused
 5   aromatic rings. Dibenz[a,c]anthracene contains three bay regions but no fiord region in its
 6   structure.
 7          There were 15 datasets for dibenz[a,c]anthracene that met selection criteria and included
 8   benzo[a]pyrene (Figure 6-22).  The database includes a single in vivo study of DNA adducts,
 9   nine mutagenicity or morphological/malignant cell transformation studies, and five studies of in
10   vitro DNA damage or adducts. One morphological/malignant cell transformation assay gave
11   nonpositive results, while the remaining studies were positive.  In the absence of positive
12   bioassays with benzo[a]pyrene, other bioassays and cancer-related data were considered to
13   evaluate the carcinogenicity of dibenz[a,c]anthracene.
14          Conflicting results were reported in three dermal initiation bioassays of
15   dibenz[a,c]anthracene in which benzo[a]pyrene was not included. Van Duuren et al. (1970)
16   observed a tumor incidence of 95% (19/20, compared to 1/20 controls) when mice were treated
17   with an initiating dose of 1 mg dibenz [a, c] anthracene in benzene followed by thrice weekly
18   treatment with phorbol myristate acetate.  In contrast, there was no significant increase in tumor
19   formation when the same initiating dose was followed by thrice weekly application of croton
20   resin (Van Duuren et al.,  1968); however, the latency to first tumor was substantially reduced
21   (65 versus 150 days in controls). Latency was also substantially reduced in the study by Van
22   Duuren et al. (1970), in which the first tumor appeared after 74 days, compared with 338 days in
23   controls.
24          Cancer-related endpoint data for dibenz[a,c]anthracene are predominantly positive
25   (8/9 mutagenicity or morphological/malignant cell transformation studies and 5/5 studies of in
26   vitro DNA adducts or DNA damage). Although the conflicting  bioassay data are not easily
27   explained, the high incidence of tumors (19/20) in the study by Van Duuren et al. (1970) and the
28   reduced latency to tumor formation in both studies, coupled with predominantly positive cancer-
29   related endpoint data, suggest that dibenz [a, c] anthracene is carcinogenic.  Contributing to this
30   conclusion is the observation that dibenz[a,c]anthracene is an alternant PAH with known
31   structural alerts for carcinogenicity (more than three rings, and three bay regions).  Thus,
32   dibenz [a, c]anthracene was selected for inclusion in the RPF approach.
                                             159          DRAFT - DO NOT CITE OR QUOTE

-------
       10 --
tf
!*.
 o

_s


>
0.1  --
     0.01  --
    0.001
                                                                                                       Positive bioassay (incidence)


                                                                                                       Positive bioassay (multiplicity)

                                                                                                       In vivo cancer-relatedendpoint


                                                                                                       In vitro cancer-related endpoint
      /  /  /
                                /

                                                                                     /
                                                                                            jf    Jr     ^
            * Missing bar in dicatesnonpositive cancer-relatedendpoint study
                                                                 Reference
            Figure 6-22. Dibenz[a,c]anthracene (DBacA) RTFs*.
                                                             160
                                                                                          DRAFT - DO NOT CITE OR QUOTE

-------
                                    Dibenzo[a,e]fluoranthene (DBaeF)
 2
 3
 4          Dibenzo[a,e]fluoranthene (CASRN 5385-75-1) is a nonalternant PAH comprised of five
 5   aromatic rings and one five-membered ring. Dibenzo[a,e]fluoranthene contains a classic bay
 6   region but no fiord region in its structure.
 7          There were three datasets for dibenzo[a,e]fluoranthene that met selection criteria and
 8   included benzo[a]pyrene (Figure 6-23); all gave positive results.  The database includes two in
 9   vivo tumor bioassays and one mammalian mutagenicity study.  Statistically significant increases
10   in tumor incidence were reported in dermal initiation and complete carcinogenicity bioassays in
11   mice (both reported by Hoffmann and Wynder, 1966). As the available bioassays for
12   dibenzo[a,e]fluoranthene were positive, this compound was considered carcinogenic and was
13   selected for inclusion in the RPF approach.
14
                                            161          DRAFT - DO NOT CITE OR QUOTE

-------
     100
       10 -•
        1  -•
 s
>    o.i  -•
     0.01  -•
   0.001
                   Hoffmann and Wynder 1966
Hoffmann and Wynder 1966

       Reference
                                                                                                  Positive bioassay (incidence)
                                                                                                  Positive bioassay (multiplicity)
                                                                                                  In vivo cancer-related endpoint
                                                                                                  In vitro cancer-related endpoint
Durantetal., 1999
           Figure 6-23. Dibenzo[a,e]fluoranthene (DBaeF) RTFs.
                                                          162
                                  DRAFT - DO NOT CITE OR QUOTE

-------
                                       Dibenzo[a,e]pyrene (DBaeP)
 2
 3
 4          Dibenzo[a,e]pyrene (CASRN 192-65-4) is an alternant PAH comprised of six fused
 5   aromatic rings.  Dibenzo[a,e]pyrene contains three bay regions but no fjord region in its
 6   structure.
 7          There were three datasets for dibenzo[a,e]pyrene that met selection criteria and included
 8   benzo[a]pyrene (Figure 6-24).  The database includes two in vivo tumor bioassay datasets and
 9   one in vitro bacterial mutagenicity dataset, all of which gave positive results.  Statistically
10   significant increases in tumor incidence were reported in dermal initiation and complete
11   carcinogeni city bioassays in mice (Hoffmann and Wynder, 1966). The complete carcinogenicity
12   bioassay was confounded by significant toxicity-related mortality unrelated to tumors (Hoffmann
13   and Wynder, 1966). The one bacterial mutagenicity study reported positive results.  Because the
14   available bioassays with benzo[a]pyrene were both positive,  dibenzo[a,e]pyrene was considered
15   carcinogenic and was selected for inclusion in the RPF approach.
16
                                            163           DRAFT - DO NOT CITE OR QUOTE

-------
      100
       10 --
        1  --
o
a;
s

3
      0.1  --
     o.oi  --
   0.001
                   Hoffmann and Wynder 1966
Hoffmann and Wynder 1966



      Reference
     Figure 6-24. Dibenzo[a,e]pyrene (DBaeP) RTFs.
                                                                                                 Positive bioassay (incidence)


                                                                                                 Positive bioassay (multiplicity)

                                                                                                 In vivo cancer-related endpoint

                                                                                                 In vitro cancer-re la ted endpoint
Teranishiet al., 1975
                                                    164
                             DRAFT - DO NOT CITE OR QUOTE

-------
                                      Dibenz[a,h]anthracene (DBahA)
 2
 3
 4          Dibenz[a,h]anthracene (CASRN 53-70-3) is an alternant PAH comprised of five fused
 5   aromatic rings. Dibenz [a, h] anthracene contains two bay regions but no fiord region in its
 6   structure.
 7          There were 31 datasets for dibenz[a,h]anthracene that met selection criteria and included
 8   benzo[a]pyrene (Figure 6-25). Included in the database are in vivo tumor bioassay datasets (5),
 9   in vivo DNA adduct datasets (2), an in vivo clastogenicity dataset, mutagenicity datasets (10),
10   morphological/malignant cell transformation datasets (6), and in vitro DNA damage, adducts, or
11   clastogenicity datasets (7).  There were three tumor bioassays for dibenz[a,h]anthracene that
12   included benzo[a]pyrene, and all resulted in statistically significant increases in tumor incidence
13   and/or multiplicity. The bioassays were in three different test systems: a rat lung implantation
14   study (Wenzel-Hartung et al., 1990), a mouse dermal initiation  study reporting both incidence
15   and multiplicity (Slaga et al., 1980), and an intraperitoneal study in A/J mice (Nesnow et al.,
16   1998b).  Dibenz[a,h]anthracene was shown to form DNA adducts when administered in vivo to
17   mice via intraperitoneal injection (Nesnow et al., 1998b) and dermal application (Phillips et al.,
18   1979). Mutagenicity and morphological/malignant cell transformation assays of
19   dibenz[a,h]anthracene were predominantly positive (13/16), as were studies of other cancer-
20   related endpoints.  Because the available bioassays with benzo[a]pyrene were positive,
21   dibenz [a,h]anthracene was  considered carcinogenic and was selected for inclusion in the RPF
22   approach.
23
                                             165          DRAFT - DO NOT CITE OR QUOTE

-------
      10  --
PH
rt
 o
 s
        1  --
      0.1  --
    0.01  --
n
                                                                   Positive bioassay  (incidence)
                                                                   Positive bioassay  (multiplicity)
                                                                   In vivo cancer-related endpoint
                                                                   In vitro cancer-related endpoint
   0 001   I™!1 M™!™!1  M1  M1 M1 M11!1  M1  M1 M1 M1 M1 M11!1  M1  M1 l|l l|l l|l  l|ll|l l|l l|l M   I	1    I    I
                                 "  *"*"  *"  *"  *"  *"  *"      *
                                                   *" *"" **"
         * Mis sing bar indicates nonpositive cancer-related endpoint study

   Figure 6-25. Dibenz[a,h]anthracene (DBahA) RTFs*.
                                                                 Rcf crcn CG
                                                    166
                                                        DRAFT - DO NOT CITE OR QUOTE

-------
                                       Dibenzo[a,h]pyrene (DBahP)
 2
 3
 4          Dibenzo[a,h]pyrene (CASRN 189-64-0) is an alternant PAH comprised of six fused
 5   aromatic rings.  Dibenzo[a,h]pyrene contains two bay regions but no fjord region in its structure.
 6          There were five datasets for dibenzo[a,h]pyrene that met selection criteria and included
 7   benzo[a]pyrene (Figure 6-26); all gave positive results.  The database includes one in vivo
 8   bioassay dataset, one in vivo DNA adduct dataset, two in vitro mammalian mutagenicity
 9   datasets, and one in vitro DNA damage dataset.  A statistically significant increase in tumor
10   incidence was reported in a dermal initiation bioassay in mice (Hoffmann and Wynder, 1966).
11   In addition, two dermal studies of complete carcinogenicity that included benzo[a]pyrene gave
12   positive results, but no RPF could be calculated because the incidence of tumors in the mice
13   exposed to dibenzo[a,h]pyrene was >90% at the lowest dose tested (Cavalieri et al., 1977;
14   Hoffmann and Wynder, 1966) and tumor multiplicity was not reported. As all of the available
15   bioassays that included benzo[a]pyrene showed exposure-related tumorigenic responses,
16   dibenzo[a,h]pyrene was considered carcinogenic and was  selected for inclusion in the RPF
17   approach.
18
                                            167          DRAFT - DO NOT CITE OR QUOTE

-------
       10  --
fan
PH
*
!*.
 O
 OJ
_s
"08
>
.1  --
     0.01  --
    0.001
                                                                                                        Positive bioassay  (incidence)
                                                                                                        Positive bioassay  (multiplicity)
                                                                                                        In vivo cancer-relatedendpoint
                                                                                                        In vitro cancer-related endpoint
             Hoffmann and Wynder 1966    Hughes and P hillips 1990
                                                           Durantet al., 1999
                                                                    Reference
Hassetal., 1982
Mersch-Sundermann et al..
         1992
            Figure 6-26. Dibenzo[a,h]pyrene (DBahP) RPFs.
                                                             168
                                                                                         DRAFT - DO NOT CITE OR QUOTE

-------
                                        Dibenzo[a,i]pyrene (DBaiP)
 2
 3
 4          Dibenzo[a,i]pyrene (CASRN 189-55-9) is an alternant PAH comprised of six fused
 5   aromatic rings. Dibenzo[a,i]pyrene contains two bay regions but no fjord region in its structure.
 6          There were 12 datasets for dibenzo[a,i]pyrene that met selection criteria and included
 7   benzo[a]pyrene (Figure 6-27); all gave positive results. The database includes two in vivo
 8   bioassay datasets, one in vivo DNA adduct dataset, seven in vitro mutagenicity datasets, and two
 9   in vitro DNA damage datasets. Statistically significant increases in tumor incidence were
10   reported in dermal initiation and complete carcinogenicity bioassays in mice, both published by
11   Hoffmann and Wynder (1966). The cancer-related endpoint studies were all positive. As the
12   available bioassays that included benzo[a]pyrene were both positive, dibenzo[a,i]pyrene was
13   considered carcinogenic and was selected for inclusion in the RPF approach.
14
                                             169          DRAFT - DO NOT CITE OR QUOTE

-------
        10  ••
ta
PH
 s
£
         1  --
       0.1  --
     0.01  -•
    0.001
                n
                                                                                                                  Positive bioassay  (incidence)
                                                                                                                  Positive bioassay  (multiplicity)
                                                                                                                  In vivo cancer-related endpoint
                                                                                                                  In vitro cancer-related endpoint
                           1
                                      n
                                                         -+-
                                                                    -+-
                                                                               -+-
                                                                                          •+•
                                                                                                      •+•
                                                                                                                 •+•
                                                                                                                             •+•
                                                                                                                                        -H
            Hoffmann an dHoffmann and Hughes and  Durantetal.,  Hassetal.,   Teranishiet   Hermann,    Mersch-   Phillipsonand Baker etal.,  IchinotsuboetMcCann etal.
             Wynder 1966 Wynder 1966  Phillips 1990    1999       1982      al., 1975      1981    Sundermannetloannides 1989    1980      al., 1977       1975
                                                                                              al., 1992)
             * Missingbar indicatesnonpositive genotoxicity study
            Figure 6-27.  Dibenzo[a,i]pyrene (DbaiP) RTFs*.
                                                                           Reference
                                                                 170
                                                                                                        DRAFT - DO NOT CITE OR QUOTE

-------
                                       Dibenzo[a,l]pyrene (DBalP).
 2
 3
 4          Dibenzo[a,l]pyrene (CASRN 191-30-0) is an alternant PAH comprised of six fused
 5   aromatic rings. Dibenzo[a,l]pyrene contains both a bay region and a fjord region in its structure.
 6          There were 16 datasets for dibenzo[a,l]pyrene that met  selection criteria and included
 7   benzo[a]pyrene (Figure 6-28); all of the studies gave positive results.  The database includes four
 8   in vivo tumor bioassay datasets, three in vivo DNA adduct datasets, one bacterial mutagenicity
 9   dataset, one morphological/malignant cell transformation dataset, four in vivo clastogenicity
10   datasets, and three in vitro DNA adduct or DNA damage datasets.
11          In three bioassays of dibenzo[a,l]pyrene included benzo[a]pyrene, RPFs could not be
12   calculated using incidence data, because the incidence of tumors associated with the lowest dose
13   of dibenzo[a,l]pyrene exceeded 90% (two dermal initiation experiments in mice and an
14   intramammilary  injection study in rats, both reported by Cavalieri et al., 1991); however, tumor
15   multiplicity data were reported for the dermal initiation experiments and were used to calculate
16   RPFs of 10 and 40.  Nesnow et al. (1998b) provided tumor multiplicity and incidence data6 in
17   A/J mice exposed intraperitoneally; both endpoints indicated an RPF of-30.  Because the
18   available studies indicated that dibenzo[a,l]pyrene may be much more potent benzo[a]pyrene,
19   other studies were also examined to confirm the potency of this compound.
20          Dibenzo[a,l]pyrene treatment resulted in significant increases in tumor incidence in seven
21   bioassays that did not include benzo[a]pyrene, including two dermal initiation studies (Gill et al.,
22   1994; Cavalieri et al., 1989), a dermal complete carcinogenicity study  (Nakatsuru et al., 2004),
23   an intramammilary injection study in rats (Cavalieri et al., 1989), a newborn mouse bioassay
24   (Platt et al., 2004), an intraperitoneal bioassay using A/J mice (Prahalad et al., 1997), and a
25   gavage bioassay comparing the responses of cyplBl wild-type and null mice (Buters et al.,
26   2002). In several of these studies, there was significant toxicity associated with dibenzo[a,l]-
27   pyrene treatment. Tumor incidences were very high in most of the studies, including the gavage
28   study (Buters et al., 2002), which reported an overall tumor incidence of 100% in cyplBl wild-
29   type mice treated with a single dose of dibenzo[a,l]pyrene. A recent study  examining in utero
30   and/or lactational exposure to dibenzo[a,l]pyrene showed that mouse pups exposed during late
31   gestation develop T-cell lymphomas between 3 and 6 months of age, as well multiple lung and
32   liver tumors (Castro et al., 2008).  All of the cancer-related data for dibenzo[a,l]pyrene were
33   positive and resulted in high RPF estimates, including in vivo and in vitro studies of DNA
     6Data were obtained courtesy of S. Nesnow.

                                             171          DRAFT - DO NOT CITE OR QUOTE

-------
1   adducts, in vivo clastogenicity studies, morphological/malignant cell transformation studies,
2   bacterial mutagenicity studies, and in vitro DNA damage or DNA adduct studies.
3          The weight of evidence supporting a finding of carcinogenicity for dibenzo[a,l]pyrene is
4   strong and suggests that this compound is very potent; thus, it was selected for inclusion in the
5   RPF approach.
6
                                            172          DRAFT - DO NOT CITE OR QUOTE

-------
 o

J3

>
       100
        10  ••
          1  ••
0.1  ••
       0.01  ••
      0.001
                                                  n
                                                                   I Positive bioassay (incidence)
                                                                   I Positive bioassay (multiplicity)
                                                                   ] In vivo cancer-related endpoint
                                                                   | In vitro cancer-related endpoint



      c/                    c/       v     v

                                                   ^

v



                                                                  Reference
           Figure 6-28. Dibenzo[a,l]pyrene (DBalP) RTFs.
                                                          173
                                                                                     DRAFT - DO NOT CITE OR QUOTE

-------
                                            Fluoranthene (FA)
 2
 3
 4          Fluoranthene (CASRN 206-44-0) is a nonalternant PAH comprised of three aromatic
 5   rings and one five-membered ring.  Fluoranthene does not contain a classic bay or fjord region in
 6   its structure.
 7          There were 21 datasets for fluoranthene that met selection criteria and included
 8   benzo[a]pyrene (Figure 6-29). Included in the database are in vivo tumor bioassay datasets (11),
 9   bacterial and mammalian mutagenicity datasets (5), a morphological/malignant cell
10   transformation assay, and in vitro studies of DNA damage, DNA adducts, or clastogenicity (4).
11   Of the bioassay datasets that included benzo[a]pyrene, nine gave positive results and two gave
12   nonpositive results. Statistically significant increases in tumor incidence and tumor multiplicity
13   were reported in newborn mouse bioassays (in male and female mice [LaVoie et al.,  1994] and in
14   female mice [Busby et al., 1989]).  The tumor incidence was not significantly increased by
15   fluoranthene in a mouse dermal initiation study with an RPF detection limit of 0.01 (Hoffman et
16   al., 1972) and when fluoranthene was tested alone in a dermal cocarcinogenicity bioassay with
17   an RPF detection limit of 0.1 (Van  Duuren and Goldschmidt, 1976). In another newborn mouse
18   bioassay (Busby et al., 1984) that reported both incidence and multiplicity, the lowest dose of
19   benzo[a]pyrene resulted in a tumor incidence of >90%, precluding RPF calculation from the
20   incidence data;  however, multiplicity data were available.  Statistical analysis of the data  for
21   fluoranthene demonstrated positive findings for both incidence and multiplicity in male mice, but
22   the results for the two endpoints were inconsistent in females.  In female mice exposed at the
23   high dose of fluoranthene in a newborn mouse bioassay reported by Busby et al. (1984), the lung
24   tumor  count was significantly increased (albeit borderline, p =  0.0343) while the incidence was
25   not (p  > 0.05), and neither was statistically significantly increased at the lower dose.  For  the
26   purpose of this analysis, the multiplicity data were treated  as an independent measure of
27   carcinogenic potency, and an RPF was calculated for the statistically increased tumor count in
28   female mice.
29          The mutagenicity studies of fluoranthene were all positive, but in vitro studies of DNA
30   damage, DNA adducts, and clastogenicity gave inconsistent results. Because the inconsistent
31   bioassay results can be attributed to different test systems (different exposure route and/or
32   gender) or study design, fluoranthene was considered carcinogenic and was selected for
33   inclusion in the RPF approach.
34
                                             174          DRAFT - DO NOT CITE OR QUOTE

-------
      100
       10  ••
ta
PH
        1  ••
     0.01  --
    0.001
                                                         RPF detection limits for
                                                          nonpositive bioassays
                                                                                                          Positive bioassay (incidence)
                                                                                                          Positive bioassay (multiplicity)
                                                                                                          In vivo cancer-related endpoint
                                                                                                          In vitro cancer-related endpoint



                                                                                                      &

           Mis sing bar indicates nonpositive cancer- related en dpoint study
Rpfpl'PIlCP
           Figure 6-29. Fluoranthene (FA) RTFs*.
                                                              175
                             DRAFT - DO NOT CITE OR QUOTE

-------
                                             Fluorene (FE)
 2
 o
 J
 4          Fluorene (CASRN 86-73-7) is a nonalternant PAH comprised of two aromatic rings and
 5   one five-membered ring. Fluorene does not contain a classic bay or fjord region in its structure.
 6          There were nine datasets for fluorene that met selection criteria and included
 7   benzo[a]pyrene (Figure 6-30). There were no tumor bioassays of fluorene that included
 8   benzo[a]pyrene, so other bioassays and cancer-related endpoint data were considered. LaVoie et
 9   al. (1980) conducted a study of skin tumor initiation in mice treated with 1 mg fluorene followed
10   by 20 weeks of treatment with TPA; the study did not include benzo[a]pyrene.  The incidence of
11   tumor-bearing animals (5%) was not significantly increased over controls (0%) (LaVoie et al.,
12   1980). The limited cancer-related endpoint data were mixed, with three positive and four
13   nonpositive mutagenicity datasets,  and two nonpositive in vitro DNA damage datasets.  Overall,
14   the database for fluorene is both limited and inconsistent. Because the database for fluorene does
15   not provide adequate information with which to assess carcinogenicity, this PAH was not
16   selected for inclusion in the RPF approach.
17
18
                                            176          DRAFT - DO NOT CITE OR QUOTE

-------
 o
 OJ
J3
13
      100
       10 -
         1  -
0.1  -
     0.01 -
    0.001
                                                                                                        Positive bioassay (incidence)
                                                                                                        Positive bioassay (multiplicity)
                                                                                                        In vivo cancer-related endpoint
                                                                                                        In vitro cancer-related endpoint
             Sakaietal., 1985 Gibson etal., 1978 Wangenheim and Hermann, 1981 Kaden etal., 1979  McCann etal.,    Pahlman and    Mamber etal.,      Mersch-
                                         Bolcsfoldi 1988                                  1975      Pelkonen, 1987      1983     SundermannetaL
                                                                                                                                 1992
           : Missing bar indicates nonpositive cancer-related endpoint study
                                                             Reference
  Figure 6-30.  Fluorene (FE) RTFs*.
                                                       177
                                                                                      DRAFT - DO NOT CITE OR QUOTE

-------
                                       Indeno[l,2,3-c,d]pyrene (IP)
 1
 3
 4          Indeno[l,2,3-c,d]pyrene (CASRN 193-39-5) is a nonalternant PAH comprised of five
 5   aromatic rings and one five-membered ring. Indeno[l,2,3-c,d]pyrene does not contain a classic
 6   bay or fiord region in its structure.
 7          There were five datasets for indeno[l,2,3-c,d]pyrene that met selection criteria and
 8   included benzo[a]pyrene (Figure 6-31). There are three tumor bioassays, one in vitro study of
 9   morphological/malignant cell transformation (Emura et al., 1980), and one in vitro study of DNA
10   damage (Mersch-Sundermann et al., 1992). Of the three tumor bioassays, only one, a rat lung
11   implantation study (Deutsch-Wenzel et al.,  1983), reported a statistically significant increase in
12   tumor incidence or multiplicity; the RPF was 0.07. Nonpositive findings were reported in mouse
13   dermal initiation (Hoffmann and Wyner, 1966) and complete carcinogenicity (Habs et al., 1980)
14   studies with RPF detection limits in the range of 0.1-0.3.  Because the inconsistent bioassay
15   results can be attributed to different test systems (different species and route), and the
16   nonpositive studies may not have been sufficiently sensitive to detect an effect, indeno-
17   [l,2,3-c,d]pyrene was considered carcinogenic and was selected for inclusion in the RPF
18   approach.
19
                                            178          DRAFT - DO NOT CITE OR QUOTE

-------
      100
       10  --
        1  --
PH
«
!*.
 o
      0.1  --
     0.01  --
    0.001
                                                   RPF detection limits for
                                                    nonpositive bioassays
                                                                                                            Positive bioassay (incidence)
                                                                                                            Positive bioassay (multiplicity)
                                                                                                            In vivo cancer-related endpoint
                                                                                                            In vitro cancer-related endpoint
              Deutsch-Wenzel et al., 1983    Hoffmannand Wynder 1966         Habs etal., 1980       Mersch-Sundermann et al., 1992      Emura et al., 1980

                                                                       Reference
          Figure 6-31. Indeno[l,2,3-c,d]pyrene (IP) RTFs.
                                                             179
DRAFT - DO NOT CITE OR QUOTE

-------
                                      Naphtho[2,3-e]pyrene (N23eP)
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
       Naphtho[2,3-e]pyrene (CASRN 193-09-9) is an alternant PAH comprised of six fused
aromatic rings. Naphtho[2,3-e]contains two bay regions but no fjord region in its structure.
       There were two datasets for naphtho[2,3-e]pyrene that met selection criteria and included
benzo[a]pyrene (Figure 6-32): a tumor bioassay dataset and an in vitro mammalian mutagenicity
dataset (both were positive). The tumor bioassay was a single dose dermal initiation bioassay
(Hoffmann and Wynder, 1966).  As the available bioassay reported a statistically significant
increase in tumor incidence, naphtho[2,3-e]pyrene was considered carcinogenic, and was
selected for inclusion in the RPF approach.
                                            180
                                                    DRAFT - DO NOT CITE OR QUOTE

-------
     100  T
      10  -•
       1  -•
o
u
s
     0.1  --
    0.01  --
   0.001
                             Hoffmann and Wynder 1966
                                                                Reference
        Figure 6-32. Naphtho[2,3-e]pyrene (N23eP) RTFs.
         Positive bioassay (incidence)
         Positive bioassay (multiplicity)
         In vivo cancer-related endpoint
         In vitro cancer-related endpoint
Durantet al., 1999
                                                         181
DRAFT - DO NOT CITE OR QUOTE

-------
                                              Perylene (Pery)
 4          Perylene (CASRN 198-55-0) is an alternant PAH comprised of five fused aromatic rings.
 5   Perylene contains two bay regions but no fjord region in its structure.
 6          There were 11 datasets for perylene that met selection criteria and included
 7   benzo[a]pyrene (Figure 6-33). The database includes an in vivo tumor bioassay dataset, an in
 8   vivo clastogenicity dataset, eight bacterial mutagenicity datasets, and an in vitro DNA damage
 9   dataset. The single tumor bioassay, a dermal initiation study, gave nonpositive results for
10   perylene (El-Bayoumy et al.,  1982); the RPF detection limit was 0.01. To confirm the
11   nonpositive bioassay findings, other bioassays and cancer-related endpoint data were considered.
12   In a study that did not include benzo[a]pyrene, Van Duuren et al. (1970) did not observe an
13   increase in tumor incidence over controls when mice were treated by dermal application with an
14   initiating dose of 0.8 mg perylene in benzene followed by thrice weekly treatment with phorbol
15   myristate acetate for 58 weeks.  However, seven of the eight bacterial mutagenicity studies gave
16   positive results, while perylene tested nonpositive in one bacterial mutagenicity study, the
17   clastogenicity study, and the DNA damage study.  Overall, the database for perylene is both
18   limited and inconsistent.  Because the database for perylene does not provide adequate
19   information with which to assess carcinogenicity, this PAH was not selected for inclusion in the
20   RPF approach.
21
                                             182          DRAFT - DO NOT CITE OR QUOTE

-------
     10 -
       1  -
fan
PH
£  0.1
   0.01  -
  0.001
                                                                                                                       Positive bioassay  (incidence)
                                                                                                                       Positive bioassay  (multiplicity)
                                                                                                                       In vivo cancer-related endpoint
                                                                                                                       In vitro cancer-related endpoint
            RPF detection limits for
            nonpositive bioassays
•+•
             •+•
                          -+-
                                       -+-
                                                    -+-
                                                                 -+-
                                                                              •+•
                                                                                           •+•
                                                                                                         •+•
                                                                                                                      -H
           El-Bayoumyet  Sirianniand   Kadenetal.,   Carveretal.,  Sakaietal., 1985 Hermann, 1981  LaVoieetal.,   DeFloraetal.,   Florinetal.,   Pahlman and     Mersch-
             al., 1982     Huang, 1978      1979         1986                                   1979         1984         1980      Pelkonen, 1987  Sundermannet
                                                                                                                                                al., 1992
        * Missingbar indicatesnonpositive cancer-related endpoint study


             Figure 6-33. Perylene (Pery) RTFs*.
                                                                           Reference
                                                                    183
                                                                                     DRAFT - DO NOT CITE OR QUOTE

-------
                                           Phenanthrene (PH)
 2
 3
 4          Phenanthrene (CASRN 85-01-8) is an alternant PAH comprised of three fused aromatic
 5   rings. Phenanthrene contains a bay region in its structure, but has less than four aromatic rings.
 6          There were 34 datasets for phenanthrene that met selection criteria and included
 7   benzo[a]pyrene, including 3 in vivo tumor bioassay datasets, 2 in vivo clastogenicity datasets,
 8   11 mutagenicity datasets, 6 morphological/malignant cell transformation datasets, and 12 in vitro
 9   studies of DNA adducts, DNA damage, or clastogenicity (Figure 6-34). Only 7 studies reported
10   positive results; the remaining 27 studies reported nonpositive findings, including all 3 bioassays.
11   Nonpositive findings were reported in the three bioassays that included benzo[a]pyrene,
12   including a lung implantation study in rats (Wenzel-Hartung et al., 1990), a dermal initiation
13   study in mice (LaVoie et al., 1981), and a subcutaneous study in mice (Grant and Roe, 1963).
14   To confirm the nonpositive findings, other bioassays and cancer-related endpoint data were
15   considered.  In bioassays without benzo[a]pyrene, phenanthrene did not induce significant
16   increases in tumors in a newborn mouse assay using a total dose of 1.4  (imol (Buening et al.,
17   1979) or in two dermal initiation assays (Wood et al., 1979; Salaman and Roe, 1956) using doses
18   of 10 (imol and 540 mg, respectively. However, 12/30 mice developed papillomas by week 35
19   after dermal treatment with  10 |imol phenanthrene (in benzene) followed by twice weekly
20   treatment with TPA; no control mice had papillomas (Scribner, 1973).  The response was
21   statistically significantly increased over controls (p < 0.01).
22          In vitro assays of mutagenicity and morphological/malignant cell transformation were
23   predominantly nonpositive for phenanthrene. One of the two positive studies (Sakai et al., 1988)
24   reported a poor dose-response relationship for phenanthrene. Two studies found evidence of
25   clastogenicity after in vivo administration of phenanthrene (Roszinsky-Kocher et al., 1979;
26   Bayer, 1978).  However, in the study by Bayer (1978), only the high dose gave a significant
27   response, and there was not a significant dose-response trend. When phenathrene was tested in
28   in vitro studies of DNA adducts, DNA damage, and clastogenicity, the  results were
29   predominantly nonpositive (9/12 studies). Overall, the database for phenanthrene is substantial,
30   and the weight of evidence suggests that this PAH is not carcinogenic.  Based on the large
31   number of nonpositive bioassays and the abundant evidence that phenanthrene lacks genotoxic
32   action, this compound was selected for inclusion in the RPF approach and assigned an RPF of
33   zero.
34
                                            184          DRAFT - DO NOT CITE OR QUOTE

-------
   100
    10 -
      1  -
o
3j
s
   0.1  -
  0.01  -
 0.001
                                                                                                          Positive bioassay (incidence)
                                                                                                          Positive bioassay (multiplicity)
                                                                                                          In vivo cancer-related endpoint
                                                                                                          In vitro cancer-related endpoint
          RPF detection limits for
           nonpositive bioassays
              r~\
     1 Missingbar indicates nonpositive cancer-related endpoint study
        Figure 6-34. Phenanthrene (PH) RPFs*.
                                                         Reference
                                                           185
DRAFT - DO NOT CITE OR QUOTE

-------
 1                                             Pyrene (Pyr)
 2
 4
 5          Pyrene (CASRN 129-00-0) is an alternant PAH comprised of four fused aromatic rings.
 6   Pyrene does not contain a bay or fjord region in its structure.
 7          There were 49 datasets for pyrene that met study quality criteria and included
 8   benzo[a]pyrene (Figure 6-35). Included in the database are in vivo tumor bioassay datasets (7),
 9   in vivo clastogenicity datasets (5), bacterial and mammalian mutagenicity datasets (14),
10   morphological/malignant cell transformation datasets (7), and in vitro DNA damage, DNA
11   adducts, or clastogenicity datasets (16).  There were seven bioassay s of pyrene that included
12   benzo[a]pyrene; all gave nonpositive results. Nonpositive results were reported in two newborn
13   mouse bioassays in which both males and females were tested (Busby et al.,  1989; Wislocki et
14   al., 1986), two studies of dermal initiation (El-Bayoumy et al., 1982; Wood et al., 1980), and a
15   dermal cocarcinogenesis bioassay (Van Duuren and Goldschmidt, 1976). RPF detection limits in
16   these studies ranged from about 0.01 to 0.1 (see Figure 6-35). In an intraperitoneal bioassay
17   using A/J mice that included benzo[a]pyrene, the authors reported that pyrene treatment did not
18   induce lung adenomas (Ross et al., 1995); data were not reported, so an RPF detection limit
19   could not be estimated. In bioassays without benzo[a]pyrene, pyrene did not induce a significant
20   increase in tumors in a dermal initiation bioassay (Salaman and Roe, 1956).  Scribner (1973)
21   reported a weak tumorigenic response in a dermal initiation study in mice (5/29 mice developed
22   papillomas 35 weeks after dermal treatment with 10 |imol pyrene in benzene followed by twice
23   weekly treatment with TPA as compared with 0/30 control mice, p = 0.02).
24          In vitro assays of bacterial and mammalian mutagenicity and morphological/malignant
25   cell transformation were predominantly nonpositive for pyrene.  In five studies of clastogenicity
26   in animals exposed in vivo to pyrene, no evidence of clastogenic effects was reported. Further,
27   in vitro studies of DNA adducts, DNA damage, and clastogenicity using pyrene also largely
28   reported nonpositive results. Overall, the database for pyrene is substantial, and the  weight of
29   evidence suggests that this PAH is not carcinogenic. Based on the large number of nonpositive
30   bioassays and the abundant evidence that pyrene lacks genotoxic action, this compound was
31   selected for inclusion in the RPF approach and assigned an RPF of zero.
32
                                            186          DRAFT - DO NOT CITE OR QUOTE

-------
iuu -




10 -

1 -
«*•
O
0>
s
"cS 0.1 -
>



0.01 -







o nni
U.UUl ^
y
*> c
X *°
^






RPF detection limits for
nonpositive bioassays

n

r
rl
•~l
1
i n
i ! n
i
: n
i i
! i
i
i :
i
i i • i

9* s# ** ^ «* ^ / ^ ^S S*V ^ ^ 4
^*y^^^^A*vVV»**^

/ ^ a?^ ^^ "o^
(^' A? (


























*
4-s

^
^
?

























s^
y"1
tf.
$
r


























$>
y

>


























^
*"/
«T
V
«/


























^
^
^
0









-
















§>
^

/


























s^
V
i
.4
^


























0*
^

<


























-9
V
JS*
r












,-,













8^
^

f


^ Positive bioassay (incidence)
S Positive bioassay (multiplicity)
In vivo cancer-related endpoint
| | In vitro cancer-related endpoint







n n












* * ^ «P ^ ^ # 4P 
-------
                                          Triphenylene (TPhen)
 4          Triphenylene (CASRN 217-59-4) is an alternant PAH comprised of four fused aromatic
 5   rings. Triphenylene contains several bay regions but no fjord region in its structure.
 6          There were six datasets for triphenylene that met selection criteria and included
 7   benzo[a]pyrene (Figure 6-36); all but one of the studies gave positive results.  The database
 8   includes five mutagenicity studies (four positive and one nonpositive) and a study of in vitro
 9   DNA damage. There were no bioassays of triphenylene that met selection criteria, and no
10   bioassays without benzo[a]pyrene. Although all of the available cancer-related endpoint studies
11   for triphenylene gave positive results, the database is very limited, consisting of only a few in
12   vitro mutagenicity and DNA damage studies.  The RPFs for cancer-related endpoints ranged
13   from 0.02 to 0.4. Because the database for triphenylene does not provide adequate information
14   with which to assess carcinogenicity, this PAH was not selected for inclusion in the RPF
15   approach.
16
                                            188          DRAFT - DO NOT CITE OR QUOTE

-------
      100
       10  •
PH
«
it-
 o

      0.1
     0.01
    0.001
                                                                                                         Positive bioassay (incidence)
                                                                                                         Positive bioassay (multiplicity)
                                                                                                         In vivo cancer-related endpoint
                                                                                                         In vitro cancer-related endpoint
             Mersch-Sundermann et     Hermann, 1981
                   al., 1992
Kadenet al., 1979     Barfknecht et al., 1982   Pahlman and Pelkonen     Gibsonetal., 1978
                                               1987
            * Missingbar indicatesnonpositive cancer-related endpoint study
         Figure 6-36. Triphenylene (Tphen) RTFs*.
                                                                    Reference
                                                            189
                                        DRAFT - DO NOT CITE OR QUOTE

-------
 1                  7.  DERIVATION OF FINAL RTFs FOR SELECTED PAHs
 2
 3
 4          The weight of evidence evaluation (Chapter 6) indicates that the available data are
 5   adequate to suggest that 24 of the 27 PAHs are carcinogenic, 3 PAHs (anthracene, phenanthrene,
 6   and pyrene) exhibited no carcinogenicity, and data are inadequate to  evaluate the carcinogenicity
 7   of eight PAHs.  The 8 PAHs with inadequate data are excluded from the RPF analysis.
 8          For the three PAHs for which there were sufficient data to conclude that they were not
 9   carcinogenic (i.e., robust nonpositive tumor bioassay data and cancer-related endpoint data), a
10   final RPF of zero was recommended. While there is little quantitative difference between
11   selecting a final RPF of zero for a given PAH and excluding that PAH from the RPF approach,
12   this is an important distinction for uncertainty analysis. There is substantial uncertainty in the
13   risk associated with PAHs that are excluded from the RPF analysis due to inadequate data, as
14   these compounds could be of low or high potency. However, for PAHs with an RPF of zero,
15   there is evidence to suggest that these compounds are not carcinogenic, and the uncertainty
16   associated with the cancer risk for these compounds is markedly reduced.
17          For each of the remaining 24 compounds, a final nonzero RPF was derived. A number of
18   options were considered for deriving a final RPF from among the numerous values calculated for
19   each individual PAH. These options included: prioritizing bioassay  RPFs from different
20   exposure routes based on  environmentally relevant routes; prioritizing bioassay RPFs based on
21   target organs considered relevant to human susceptibility to PAH carcinogenesis; prioritizing
22   RPFs based on quality of the underlying study; prioritizing cancer-related endpoints by their
23   correlation with bioassay potency (i.e., ability to predict bioassay potency); and combining (i.e.,
24   averaging) RPFs across all bioassays, across all cancer-related endpoints, or across all endpoints.
25   Appendix G details analyses that were undertaken to assess various options for ranking or
26   prioritizing RPFs.  It was  concluded that the available data did not provide a basis  for prioritizing
27   RPFs except for a preference for bioassay  data over cancer-related endpoints. As a consequence,
28   final RPFs were derived from bioassay data for any PAH that had at least one RPF based on a
29   bioassay.  For carcinogenic PAHs without bioassay data, final RPFs were calculated from all
30   cancer-related endpoint datasets with positive results (see next section).
31
32   7.1.  METHODS FOR DERIVING FINAL RPFs
33          For each carcinogenic PAH with bioassay data, the average RPF was calculated from
34   bioassay datasets with positive results (nonpositive bioassay results were not included in the
35   calculation). For those PAHs that did not have any RPF based on a bioassay, but for which the
36   weight of evidence evaluation indicated a carcinogenic response (e.g., dibenz[a,c]anthracene),
37   the average RPF was calculated from all cancer-related endpoint datasets with  positive results
38   (again, nonpositive results were not included in the calculation).  The range of  RPF values was
                                            190          DRAFT - DO NOT CITE OR QUOTE

-------
 1   also reported. Presenting the average and the range provides an average and maximum estimate
 2   for each PAH that has data from multiple studies.
 3          Several options were considered for the estimation of a final RPF, including arithmetic
 4   mean, geometric mean, weighted average, maximum, or order of magnitude estimates. The
 5   arithmetic mean and range were chosen as a simple approach to describing the calculated RPF
 6   values available for each PAH.  Other estimates were not considered due to the limited number
 7   of individual RPF values calculated for most PAHs and the variability in the RPF estimates.
 8   There were usually not enough  data (3 or fewer RPFs for 17/23 PAHs with  nonzero RPFs) to
 9   assess the shape of the RPF distribution for any given PAH; thus, a geometric mean was not
10   considered. Further, the range of RPF values from tumor bioassays was greater than an order of
11   magnitude for several compounds (6/23 PAHs). The variability in RPF estimates is likely due to
12   differences in study design parameters (e.g., route, species/strain, exposure  duration, exposure
13   during sensitive time periods, initiation versus complete carcinogenesis protocol, tumor
14   incidence versus tumor multiplicity reporting) and dose-response methods (modeled versus point
15   estimates). Calculation of a weighted average was considered, but without a rationale for
16   assigning weights among study types or among tumor data outcomes, using a weighting
17   approach might increase uncertainty.
18          Several previous approaches for generating RPF values for PAHs have used order-of-
19   magnitude estimates (Collins et al., 1998; Malcolm and Dobson, 1994; U.S. EPA,  1993; Nisbet
20   and LaGoy, 1992, see Chapter 3).  The  presentation of the arithmetic mean  (and range) of RPFs
21   for each PAH reflects the available data better than an order-of-magnitude approach.
22          The range was reported  as a measure of variability instead of a confidence interval on the
23   average RPF. The input data for each average RPF (bioassay RPFs of different route, species,
24   sex, and target organ, or cancer-related  endpoint data across a wide variety of assays and test
25   conditions) reflect such heterogeneity in study design that confidence limits would not provide
26   the statistical precision that they typically convey.  All tumor bioassay RPFs (across all exposure
27   routes, species, and sexes, and including both tumor incidence and tumor multiplicity RPFs)
28   were combined to estimate the mean and range for each PAH, except as follows.  Only nonzero
29   RPFs were included in the calculation of the final RPF and range for each PAH
30          While tumor multiplicity data from tumor bioassays are not generally used to estimate
31   cancer potency, these data were included in the dose-response assessment in order to determine
32   whether they could serve as a reliable measure of relative cancer potency.  Several bioassays
33   reported  data on both tumor incidence and tumor number, providing information that was used to
34   compare  relative potencies estimated from these two endpoints.  The comparison between RPFs
35   calculated from incidence and tumor multiplicity data from the same experiment showed these
36   values to be highly correlated (r2 = 0.76; see further discussion in Chapter 8), indicating that
37   multiplicity RPFs are reasonably predictive of incidence RPFs. When both incidence and
38   multiplicity RPFs were calculated for the same group of animals, the results for each endpoint

                                           191           DRAFT - DO NOT CITE OR QUOTE

-------
 1   could not be considered independent, so the higher of the two values was included in the average
 2   and the lower value was excluded.  As discussed further in Chapter 8, in 70% of the cases where
 3   data for both incidence and multiplicity were used to calculate RPFs, the RPF associated with
 4   incidence was the higher of the two (or the two values were equal) and was therefore included in
 5   the average, omitting the corresponding multiplicity RPF.
 6          When separate RPFs were calculated for different target organs in the same group of
 7   animals, the higher value of the two RPFs was included in the average and range, and the lower
 8   value was dropped from the combined data. Different RPFs were calculated for liver and lung
 9   tumors in male mice (females did not develop liver tumors) in newborn mouse studies. This
10   occurrence  applied only to benz[a]anthracene, chrysene, and fluoranthene tested in studies
11   reported by LaVoie et al. (1994) and Wislocki et al. (1986).
12          When separate RPFs were calculated for male and female animals in the same study
13   (generally, these were also newborn mouse studies), both sex-specific RPFs were included in the
14   aggregation, as these were two separate groups of animals.  In the one dermal study that included
15   both sexes (Nesnow et al., 1984), the male and female RPFs differed by only -50% for both
16   benz[c]aceanthrylene and benz[l]aceanthrylene. In the newborn mouse studies that resulted in
17   nonzero RPFs for both males and females (LaVoie et al., 1994, 1987; Wislocki et al., 1986), the
18   male RPF was typically three- to fivefold higher than the female RPF. Final RPFs that included
19   both male and female values from the same study were calculated for three PAHs:
20   benzo[j]fluoranthene, benz[a]anthracene, and fluoranthene.
21          Table 7-1 shows the average RPFs based on tumor bioassay data with their associated
22   range, and an overview of the tumor bioassay database (total number of studies, exposure routes
23   tested, species tested, and sexes tested) for each PAH. Table 7-2 shows the average RPF for
24   dibenz[a,c]anthracene, the only RPF based on cancer-related endpoint data, with its associated
25   range, and an overview of the database for this compound.
                                            192         DRAFT - DO NOT CITE OR QUOTE

-------
Table 7-1. Final RTFs based on tumor bioassay data
PAH
Anthanthrene
Anthracene
B enz [a] anthracene
Benz[b,c]aceanthrylene, 11H-
Benzo [bjfluoranthene
Benzo[c]fluorene
Benz [e] aceanthry lene
Benzo [g,h,i]pery lene
Benz |] ] aceanthry lene
Benzo tj]fluoranthene
Benzo [kjfluoranthene
Benz [1] aceanthry lene
Chrysene
Cyclopenta[c,d]pyrene
Cyclopenta[d,e,f]chrysene, 4H-
Dibenzo [a,e]fluoranthene
Dibenzo [a,e]py rene
Dibenz[a,h]anthracene
Dibenzo [a,h]pyrene
Dibenzo [a,i]py rene
Dibenzo [a,l]py rene
Fluoranthene
Indeno [ 1,2,3 -c,d]pyrene
Naphtho[2,3-e]pyrene
Phenanthrene
Pyrene
Average RPF
0.4
0
0.2
0.05
0.8
20
0.8
0.009
60
0.3
0.03
5
0.1
0.4
0.3
0.9
0.4
10
0.9
0.6
30
0.08
0.07
0.3
0
0
Range of RPFs
0.2-0.5
0
0.02-0.4
0.05
0.1-2
1-50
0.6-0.9
0.009
60
0.01-1
0.03-0.03
4-7
0.04-0.2
0.07-1
0.2-0.5
0.7-1
0.3-0.4
1-40
0.9
0.5-0.7
10-40
0.009-0.2
0.07
0.3
0
0
Number of datasets
2
1 (nonpositive)
3
1
5
2
2
1
1
5
2
2
7
5
2
2
2
3
1
2
3
5
1
1
3 (nonpositive)
7 (nonpositive)
Exposure routes tested
Dermal, lung implantation
Dermal
Dermal, intraperitoneal
Dermal
Dermal, intraperitoneal, lung
implantation
Oral, intraperitoneal
Dermal
Lung implantation
Intraperitoneal
Dermal, intraperitoneal, lung
implantation
Dermal, lung implantation
Dermal
Dermal, intraperitoneal, lung
implantation
Dermal, intraperitoneal
Dermal
Dermal
Dermal
Dermal, intraperitoneal, lung
implantation
Dermal
Dermal
Dermal, intraperitoneal
Intraperitoneal
Lung implantation
Dermal
Dermal, intraperitoneal, lung
implantation
Dermal, intraperitoneal
Species tested
Mouse, rat
Mouse
Mouse
Mouse
Mouse, rat
Mouse
Mouse
Rat
Mouse
Mouse, rat
Mouse, rat
Mouse
Mouse, rat
Mouse
Mouse
Mouse
Mouse
Mouse, rat
Mouse
Mouse
Mouse
Mouse
Rat
Mouse
Mouse, rat
Mouse
Sexes tested
Female
Female
Female, male
Female
Female, male
Female
Female, male
Female
Male
Female, male
Female
Female, male
Female, male
Female, male
Female
Female
Female
Female, male
Female
Female
Female, male
Female, male
Female
Female
Female, male
Female, male
                                          193
DRAFT - DO NOT CITE OR QUOTE

-------
            Table 7-2. Final RPFs based on cancer-related endpoint data (no tumor
            bioassay data available)
PAH
Dibenz[a,c]anthracene


Average RPF
4


Range of RPFs
0.04-50


Types of studies
Total =14 studies
One in vivo DNA adduct
Six in vitro bacterial
mutagenicity
One in vitro mammalian
mutagenicity
One in vitro morphological/
malignant transformation
Three in vitro DNA damage
Two in vitro DNA adducts
Multiple dose studies
Total = 6 studies
Four in vitro bacterial
mutagenicity
One in vitro DNA damage
One in vitro DNA adduct

 1
 2    7.2. CONFIDENCE RATINGS FOR FINAL RPFs
 3           Once a final RPF was derived for a given PAH, the resulting value was assigned a
 4    relative confidence rating of high, medium, low, or very low.  The relative confidence rating
 5    characterized the nature of the database upon which the final RPF was based.  Confidence
 6    rankings were based on the robustness of the database. For final RPFs based on tumor bioassay
 7    data, confidence ratings considered both the available tumor bioassays and the availability of
 8    supporting data for cancer-related endpoints. The most important factors that were considered
 9    included the availability of in vivo data and whether multiple exposure routes were represented.
10    Other database characteristics that were considered included the availability of more than one  in
11    vivo study, and whether effects were evident in more than one sex or species.  The database
12    characteristics of exposure route, species, and gender are somewhat related (i.e., not independent
13    variables). For example, intraperitoneal injection studies were generally performed in both male
14    and female mice while lung implantation studies were conducted in rats only.  An increase in the
15    number of exposure routes tested also results in generation of data for multiple species and
16    genders.  The factors that were considered in the relative confidence rating for each RPF are
17    illustrated in Table 7-3.
                                           194
DRAFT - DO NOT CITE OR QUOTE

-------
       Table 7-3. Relative confidence ratings for RTFs
PAH
Benzo [b]fluoranthene
Benzo tj]fluoranthene
Chrysene
Dibenz[a,h]anthracene
Phenanthrene
Anthanthrene
Anthracene
B enz [a] anthracene
Benzo [c]fluorene
Benzo [kjfluoranthene
Cyclopenta[c,d]pyrene
Dibenzo [a,l]pyrene
Pyrene
Benz[b,c]aceanthrylene, 11H-
Benz [e] aceanthry lene
Benzo [g,h,i]pery lene
Benz \j ] aceanthry lene
Benz [1] aceanthry lene
Cyclopenta[d,e,f]chrysene, 4H-
Dibenzo [a,e]fluoranthene
Dibenzo [a,e]py rene
Dibenzo [a,h]pyrene
Dibenzo [a,i]py rene
Fluoranthene
Indeno [ 1,2,3 -c,d]pyrene
Naphtho[2,3-e]pyrene
Dibenz[a,c]anthracene
Relative
confidence
High
High
High
High
High
Medium
Medium
Medium
Medium
Medium
Medium
Medium
Medium
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Low
Very low
Tumor bioassay data
In vivo data
•/
•/
•/
•/
•/
•/
•/
•/
•/
•/
•/
•/
•/
•/
•/
•/
•/
•/
•/
•/
•/
•/
•/
•/
•/
•/

>1 Exposure
route
•/
•/
•/
•/
•/
•/
S*
•/
•/
•/
•/
•/
•/














>2 Exposure
routes
•/
S
S
s
s






















>1 Species
•/
•/
•/
•/
•/
•/
S*


•/

















>1 Gender
•/
•/
•/
•/
•/


•/


•/
•/
•/

•/


•/





•/



Supporting data
for cancer-related
endpoints
•/
S
S
s
s
s
•/
•/
•/

•/
•/
s

s
s
s
s

•/
•/
•/
•/
•/
s
s
s
aBioassays of anthracene without benzo[a]pyrene included dermal studies in mice and a lung implantation study in rats.
                                                           195
DRAFT - DO NOT CITE OR QUOTE

-------
 1          Very low relative confidence was used to describe final RPFs based on cancer-related
 2   endpoint data only (e.g., dibenz[a,c]anthracene).
 3          For RPFs of zero, the confidence rating considered both the available tumor bioassays
 4   (with and without benzo[a]pyrene) and the size and consistency of the cancer-related endpoint
 5   database. An RPF of zero was only applied if the data implied high or medium relative
 6   confidence.  For anthracene, phenanthrene, and pyrene, the available data support a practical
 7   RPF of zero.
 8
 9   7.3.  APPLICATION OF RPFs FOR ASSESSING CANCER RISKS FROM EXPOSURE
10   TO PAH MIXTURES
11          In the proposed RPF approach, the cancer risk associated with exposure to a particular
12   mixture of PAHs is assumed to equal the sum of the risks associated with exposure to individual
13   carcinogenic components. Because quantitative cancer risk values are available only for
14   benzo[a]pyrene, exposure units (either concentrations or doses, in  units of mass) for other PAHs
15   found in the mixture are expressed in terms of benzo[a]pyrene equivalents. These are summed
16   with benzo[a]pyrene to obtain an estimate of the total benzo[a]pyrene equivalents (in
17   concentration or dose) presented by the mixture. Benzo[a]pyrene  equivalents for PAH
18   components in a particular mixture are calculated by multiplying the concentration (or dose) of a
19   particular PAH component in the mixture by its RPF. The total benzo[a]pyrene equivalents for a
20   particular mixture of PAHs is calculated as follows:
21
22                                        E = ZRPFjCj + X
23
24      where:
25      E    =    the benzo[a]pyrene equivalent exposure presented by the mixture
26      RPFj =    relative potency factor of the jth PAH detected in the mixture
27      Cj    =    dose or concentration of the jth PAH detected in the mixture
28      X    =    dose or concentration of benzo[a]pyrene in the mixture.
29
30          The cancer risk for the PAH mixture is determined by multiplying the benzo[a]pyrene
31   equivalent dose or concentration by the benzo[a]pyrene cancer toxicity value (e.g., oral slope
32   factor). The proposed RPF approach considers each of the bioassay types used for RPF
33   derivation to be equivalent for the purpose of determining relative potency to benzo[a]pyrene.
34   The uncertainty associated with using a single RPF to derive benzo[a]pyrene equivalents for
35   multiple exposure routes is discussed in Section 8.6.
36
37   7.4.  SUSCEPTIBILITY FROM EARLY LIFE EXPOSURE TO CARCINOGENS
38          According to the Supplemental Guidance for Assessing Susceptibility from Early Life
39   Exposure to Carcinogens (U.S. EPA, 2005b), benzo[a]pyrene is carcinogenic by  a mutagenic
                                            196          DRAFT - DO NOT CITE OR QUOTE

-------
 1
 2
 o
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
mode of action. For example, an acute dosing study using benzo[a]pyrene suggests that early-
lifestage exposure would lead to an increased incidence of tumors compared with adult
exposures of a similar dose and duration (EPA, 2005b). Mice that were treated with
benzo[a]pyrene (75 or 150 ug/g body weight intraperitoneal) within 24 hours of birth or at
15 days of age developed hepatomas at a higher incidence than similarly treated animals at
42 days of age (Vesselinovitch et al.,  1975, as cited in EPA 2005b).
       The Supplemental Guidance establishes age-dependent adjustment factors (ADAFs) for
three specific age groups. The ADAFs and their age groupings are 10 for <2 years, 3 for 2-<16
years, and 1 for >16 years (U.S. EPA, 2005b). The 10- and 3-fold adjustments in slope factor are
to be combined with age-specific exposure estimates when estimating cancer risks from early life
(<16 years age) exposure to PAHs.
       Because a mutagenic mode of action for benzo[a]pyrene carcinogenicity is sufficiently
supported in laboratory animals and relevant to humans, and in the absence of chemical-specific
data to evaluate differences in susceptibility, increased early-life susceptibility is assumed and
the ADAFs should be applied, as appropriate.  A common mutagenic mode of action for
carcinogenic PAHs is hypothesized based on information  available for the indicator chemical,
benzo[a]pyrene (U.S. EPA, 2005b).  In the absence of chemical-specific data to evaluate
differences in susceptibility, increased early-life susceptibility to the 24 PAHs (for which RPFs
were derived) in this analysis is assumed and the ADAFs  should be applied, along with exposure
information, as appropriate (see Table 7-4 for example).
       Some of the studies used to derive RPFs for the PAHs were conducted in newborn mice.
The RPFs calculated from the newborn mouse studies reflect only the potency of the tested PAH
relative to that ofbenzofajpyrene, and do not take into account the potency of the PAH
administered in newborn or young animals relative to the potency of the same PAH administered
to adult animals.  The ADAF should be  applid to account  for the latter difference.
            Table 7-4.  Sample calculation of estimated cancer risk for
            benz [a] anthracene with the application of ADAFs
Age group
0-<2
2-<16
>16
ADAF
10
3
1
Benzo[a]pyrene oral slope
factor (per mg/kg-d)
7.3
7.3
7.3
Adjusted
benzo[a]pyrene
cancer risk estimate
73
24
7.3
RPF
0.2
0.2
0.2
Benz[a]anthracene estimated
cancer risk (per mg/kg-d)
15
4.8
1.5
27
28
                                            197
                                                   DRAFT - DO NOT CITE OR QUOTE

-------
 1         8. UNCERTAINTIES AND LIMITATIONS ASSOCIATED WITH THE RTF
 2                                        APPROACH
 o
 J
 4
 5          A description of uncertainties and limitations is an important component of the RPF
 6   approach for PAH mixtures risk assessment. Many of the general uncertainties related to
 7   chemical-specific risk assessment are also applicable to the proposed RPF approach for PAHs.
 8   These include issues related to selection of an appropriate animal model, low-dose and
 9   interspecies extrapolation, and variability within the human population. Use of a component-
10   based approach to mixtures risk assessment leads to additional uncertainties, e.g., the lack of
11   experimental data on potential interactions among individual components within the mixture
12   (i.e., among PAHs and with other chemicals).
13          The feasibility of conducting a robust component-based approach for PAH mixtures
14   (RPF approach) was evaluated by a PAH mixtures peer consultation workshop (U.S. EPA,
15   2002). Included in the discussion was a general evaluation of U.S. EPA's Provisional Guidance
16   (U.S. EPA, 1993). Workshop participants highlighted the following limitations of the 1993
17   guidance:
18
19          (1) The approach only considered a small subset of PAHs (i.e., unsubstituted PAHs only,
20             no heterocyclic compounds or nitro- or alkyl- substituted PAHs);
21
22          (2) There are no human toxicity data for any individual PAH;
23
24          (3) The assumption of additivity may not be valid, and there may be interactions among
25             PAHs or between PAHs and other components of a mixture (e.g., metals);
26
27          (4) PAHs may generally have a common mode of action (i.e., mutagenicity), but multiple
28             modes of action for carcinogenesis are possible; and
29
30          (5) The EOPP approach was limited to the oral exposure route  (i.e., a recommendation
31             was made not to apply the factors to dermal and inhalation exposures).
32
33          The current analysis represents a  significant improvement upon the previous component-
34   based approach for PAH mixtures risk assessment. One of the most important improvements is a
35   comprehensive review of the  scientific literature dating from the 1950s through 2009 on the
36   carcinogenicity and genotoxicity of PAHs. The search identified over  900 individual
37   publications for a target list of 74 PAHs that had been identified in environmental media or for
38   which toxicological data were available.  Review of these publications resulted in the
39   identification of more than 600 papers that included carcinogenicity or cancer-related endpoint
40   data on at least one PAH and  benzo[a]pyrene tested at the same time. Dose-response data were
41   extracted, and individual RPFs were calculated from over 300  data sets representing
                                           198          DRAFT - DO NOT CITE OR QUOTE

-------
 1   51 individual PAHs.  For 35 PAHs, a weight of evidence evaluation was conducted to select
 2   compounds for inclusion in the RPF approach; data were inadequate to conduct such an
 3   evaluation for the remaining 16 compounds. A final RPF was derived for each PAH based on
 4   tumor bioassay data (if available) or cancer-related endpoint data if no tumor bioassay RPFs
 5   were available. Final RPFs were derived for 27 PAHs (see Table 7-2), significantly increasing
 6   the number of PAHs that can be addressed through this approach. Each RPF was assigned a
 7   relative confidence rating reflecting the size and diversity of the tumor bioassay or cancer-related
 8   endpoint database that was used to derive the final RPF for that PAH.
 9          Despite these improvements, many of the uncertainties highlighted during the 2002 peer
10   consultation workshop (U.S. EPA, 2002) also apply to the current analysis. The following
11   sections describe some specific uncertainties and limitations associated with the development
12   and use of RPFs for PAHs.  The uncertainties that are specific to the approach presented herein
13   are discussed below in Sections 8.1 and 8.2. Sections 8.3-8.6 discuss the general uncertainties
14   associated with a component-based approach to PAH mixtures risk assessment. These include
15   the number of PAHs included in the approach, human relevance of animal data, assumptions
16   regarding mode of action and dose additivity, and cross-route extrapolation.
17
18   8.1. DOSE-RESPONSE ASSESSMENT FOR INDIVIDUAL PAHs
19          Several uncertainties and limitations are specifically associated with the selection of data
20   and dose-response assessment methodology used in this analysis to derive RPFs for PAHs.
21   Uncertainties are associated with the following decisions:
22
23          •  Inclusion of data from studies reporting the occurrence of benign tumors in derivation
24             of RPFs;
25
26          •  Use of a single dose-response model for quantal or continuous data;
27
28          •  Use of varying BMR levels;
29
30          •  Use of tumor incidence data at the upper end of the dose-response curve (e.g., >75%
31             incidence) to calculate some RPFs;
32
33          •  Use of tumor multiplicity data to calculate some RPFs;
34
35          •  Use of single-dose point estimates7 to calculate some RPFs;
36
37          •  Reliance on data from cancer-related endpoint studies in the absence of bioassays;
38             and
39
     7In this report, the term "point estimate RPF" is used to describe an RPF calculated from a single point on the dose-
     response curve for both the PAH of interest and benzo[a]pyrene.  This term distinguishes the RPF from one
     calculating using a BMD modeling result from multidose data.

                                            199          DRAFT - DO NOT CITE OR QUOTE

-------
 1          •  Use of cancer-related data from assay conditions that maximize the benzo[a]pyrene
 2             response, even though these conditions were not necessarily optimal for other PAHs.
 3
 4          The decision was made to employ a single dose-response model for either quantal or
 5   continuous data due to the large number of data sets that needed be analyzed from the PAH
 6   database.  The multistage model for incidence data  and the linear model for continuous data were
 7   considered to be broadly applicable to different types of data as simple curve-fitting models. In
 8   some cases, the goodness-of-fit criteria indicated that the selected model did not fit the data. In
 9   these cases, high-dose groups were sequentially eliminated until an adequate fit was achieved,
10   but other model structures (e.g., gamma, probit, logistic, etc.) were not considered.
11          Tumor bioassay data were modeled at a BMR of 10% in order to target the low end of the
12   dose-response curve as the point of departure for slope estimation. When this was not feasible,
13   usually because only a single dose was used for benzo[a]pyrene, an attempt was made to match
14   individual target PAH response levels to the benzo[a]pyrene response chosen for the point
15   estimate. This assumes that the shape of the dose-response curve is similar for the target PAH
16   and benzo[a]pyrene (also a necessary assumption of dose additivity) and that the slope is
17   constant across the dose-response curve.  These assumptions may not hold, especially in studies
18   of tumor incidence where the point estimate benzo[a]pyrene response was very high or near
19   maximal.  In many cases, the dose of benzo[a]pyrene selected as the positive control produced
20   near maximal tumor incidence in exposed animals (i.e., >75%).  There is uncertainty associated
21   with comparing potency estimates at the high end of the dose-response curves and using the
22   resultant RPF to estimate risks associated with low  environmental exposures.  The relative
23   potency relationship between any two PAHs may be different at the low end, compared  with the
24   high end, of the dose-response curves.
25          It is not clear whether relative potency values  estimated at the high end of the dose-
26   response curve are reasonably predictive of relative potency at low environmental exposure
27   levels. For this reason, additional uncertainty is involved in using RPFs that are not based on a
28   BMR of 10% (especially those RPFs that are based on responses exceeding 75%) to estimate
29   risks associated with low exposures.
30          If model fit was not achieved, then a point-estimate ratio approach was used.  Point
31   estimate ratios were also used for several other reasons:
32
33          (1) Only a single dose group was tested;
34
35          (2) When the standard deviation or number  of replicates were not reported for continuous
36             data sets; or
37
38          (3) High-dose groups from multiple dose data sets were not usable due to a saturated
39             tumor response (>90% incidence in the lowest exposure group).
40

                                            200          DRAFT - DO NOT CITE OR QUOTE

-------
 1          The point estimate approach is most reliable when the chosen point is in the linear
 2   portion of the dose-response curve.  In many cases, however, especially for single-dose data, it
 3   was not possible to determine whether the chosen point was in a linear or nonlinear portion of
 4   the dose-response curve. The dose-response relationship observed in many studies of cancer-
 5   related endpoints was nonlinear at high doses. Whenever possible, the point estimate was chosen
 6   from the linear portion of the dose-response curve (i.e., before the response plateau that occurs at
 7   high doses). Of 50 individual RPFs calculated from tumor incidence data,  21 were calculated
 8   using a point of departure incidence <25%, 19 were calculated using a point of departure
 9   incidence between 25 and 75%, and the remaining 10 were calculated using a point of departure
10   incidence between 75 and 90%. Thus, only 20% of the individual RPFs for tumor incidence data
11   were calculated from a point high (>75 and <90% incidence) on the dose-response curve.
12          For a few PAHs tested in older dermal bioassays, the authors reported mortality prior to
13   the appearance of the first tumor. For these data sets, an assumption was made that the number
14   of animals at risk for tumor development was equal to the total number of animals alive at the
15   time of the  appearance of the first tumor. This approach ensures that the incidence is not
16   underestimated by including animals that did not survive long  enough to develop tumors. As this
17   assumption applied to a small number of RPFs (specifically, individual RPFs for chrysene,
18   dibenzo[a,e]pyrene, dibenzo[a,e]fluoranthene, and dibenzo[a,h]pyrene calculated from data
19   reported by Hecht et al. [1974]  and Hoffmann and Wynder [1966]), it had little impact on the
20   overall analysis.
21          RPFs were also calculated for many cancer-related endpoints. Many of the studies
22   describing in vitro cancer-related endpoints provided dose-response data under varying study
23   conditions.  For example, bacterial mutagenesis  studies utilized multiple strains,  different
24   metabolic activation processes, and varying assay systems.  In order to minimize the amount of
25   data used for dose-response analysis of in vitro mutagenicity studies, and to provide a consistent
26   basis for comparing RPFs for different PAHs, the data from conditions that maximize the
27   benzo[a]pyrene response within a particular study were used for the dose-response assessment.
28   In several studies, the conditions that were optimal for benzo[a]pyrene were not necessarily
29   optimal for the target PAH. For example, the concentration of S9 mix that produced the highest
30   mutation rate for benzo[a]pyrene did not produce a maximal response for perylene or
31   cyclopenta[c,d]pyrene (Carver et al., 1986; Eisenstadt and Gold, 1978). In vitro  data were only
32   used in the  derivation of a single final RPF (for dibenz[a,c]anthracene; see  Table 7-2); thus, the
33   uncertainties associated with the use of cancer-related endpoint data are important for
34   dibenz[a,c]anthracene, but have minimal impact on the proposed RPFs for  the other 26 PAHs.
35
36   8.2. SELECTION OF PAHs FOR INCLUSION IN RPF APPROACH
37          One of the uncertainties highlighted by the peer consultation workshop (U.S. EPA, 2002)
38   stemmed from the fact that U.S. EPA's 1993 provisional EOPP approach only considered a small

                                            201          DRAFT - DO NOT CITE OR QUOTE

-------
 1   subset of PAHs (i.e., unsubstituted PAHs only, no heterocyclic compounds or nitro- or alkyl-
 2   substituted PAHs), and EOPPs were available for only seven PAHs. Although the present report
 3   considered a larger number of PAHs than previous analyses (the toxicological literature was
 4   searched for data on 74 individual PAHs identified in environmental media or for which there
 5   were toxicological data), the focus of this analysis remains limited to unsubstituted PAHs with
 6   three or more fused aromatic rings containing only carbon and hydrogen atoms.  Thus, the RPF
 7   analysis presented here does not account for the possible carcinogenicity of substituted or
 8   heterocyclic PAHs that may be present in complex mixtures. This may result in an
 9   underestimation of PAH mixture cancer risk.
10          Of the 74 unsubstituted PAHs with three or more aromatic rings, there were studies
11   including benzo[a]pyrene that were suitable for RPF calculation for 51 compounds. The
12   methodology for selecting PAHs for inclusion in the RPF approach from among these 51 PAHs
13   is described in Chapter 6. At the outset, 16 PAHs were excluded because only one or two in
14   vitro cancer-related endpoint RPFs were available. The remaining 35 PAHs were evaluated
15   using a weight of evidence approach.  The primary uncertainties associated with the selection
16   process relate to:
17
18          (1) The use of a weight of evidence approach that focused on tumor bioassays including
19             benzo[a]pyrene as opposed to a comprehensive cancer assessment to select PAHs for
20             inclusion in the approach; and
21
22          (2) The exclusion of PAHs with limited or inconclusive data.
23
24          The weight of evidence approach was used due to the large number of compounds that
25   were under consideration.  The approach was structured as a decision tree that focused primarily
26   on cancer bioassays that included benzo[a]pyrene, and only  considered other data (e.g., bioassays
27   that did not include benzo[a]pyrene, or cancer-related data) when  cancer bioassays with
28   benzo[a]pyrene were unavailable, nonpositive, or inconsistent (see Figure 6-1).  The data
29   collection for this analysis was centered on studies that included benzo[a]pyrene, as these studies
30   would be most useful for RPF calculation. Consequently, information from bioassays that
31   included benzo[a]pyrene were readily available for use in the weight of evidence determinations.
32   Bioassays that did not include benzo[a]pyrene and cancer-related  endpoint data were considered
33   only when there were conflicting or nonpositive results in the studies that did include
34   benzo[a]pyrene.  There is uncertainty in drawing conclusions as to carcinogenicity based on a
35   narrow subset of the available database. Other elements of a more comprehensive weight of
36   evidence determination that were not considered include:  cancer-related endpoint data from
37   studies that did not include benzo[a]pyrene; information on tumorigenicity of metabolites;
38   information on formation of reactive metabolites; other mechanistic data (e.g., AhR reactivity,
39   inhibition of gap junction intercellular communication, etc.); and QSAR assessment.

                                            202          DRAFT - DO NOT CITE OR QUOTE

-------
 1          A number of PAHs (24 of 51 PAHs that had at least one RPF value) were excluded from
 2   the relative potency approach because the available data were inadequate to draw a conclusion as
 3   to carcinogenicity (see Tables 6-1 and 6-2).  All of these PAHs had at least one RPF, indicating
 4   that the compounds were active in at least one cancer-related endpoint assay. Excluding these
 5   PAHs from the approach increases the uncertainty in assessing risks from a mixture that includes
 6   them, particularly if the excluded PAHs constitute a large fraction of the mixture.
 7          In summary, RPFs were proposed for only 27 of the 74 PAHs initially considered,
 8   because the remaining 47 compounds did not have adequate data. Thus, even among the subset
 9   of PAHs upon which this analysis was focused, RPFs were only recommended for only about
10   one-third of the compounds. Because only a fraction of any given PAH mixture can be
11   evaluated using the RPF approach, it is important to note as part of the uncertainty evaluation of
12   a risk assessment using these RPFs that there is some proportion of the total mixture (i.e., mass
13   fraction) that is comprised of compounds that are not considered in the component-based
14   approach.
15
16   8.3.  DERIVATION OF A FINAL RPF FOR EACH PAH
17          The methodology for deriving a final RPF value and assigning a relative confidence
18   rating is described in Sections 7.1 and 7.2.  The primary uncertainties associated with RPF
19   derivati on rel ate to:
20
21          (1) Combining RPFs across multiple exposure routes, species, sexes, tumor types, and
22             studies;
23
24          (2) Inclusion of RPFs based on tumor multiplicity data in the combined data;
25
26          (3) Inclusion of RPFs from female newborn mice when male RPF values were
27             demonstrably higher;
28
29          (4) Use of an arithmetic mean to  derive final RPFs; and
30
31          (5) Use of cancer-related endpoint data to derive final RPFs for compounds without
32             tumor bioassay RPFs.
33
34          A variety of options were considered for prioritizing and/or combining RPFs.
35   Appendix G describes analyses that were undertaken to assess options for prioritizing RPFs. As
36   the appendix indicates, the current state of knowledge does not suggest a clear biological basis
37   for prioritizing RPFs.  As a result, RPFs were combined across exposure routes, species, sexes,
38   tumor types, dose-response methods, and studies.
39          In addition to tumor incidence data, tumor multiplicity data were used to calculate RPFs.
40   The relationship between tumor incidence RPFs and tumor multiplicity RPFs is not known;
41   however, this analysis resulted in the calculation of both incidence and multiplicity RPFs for

                                           203          DRAFT - DO NOT CITE OR QUOTE

-------
 1    24 individual datasets.  These data were plotted, and a linear regression analysis was performed
 2    to assess the correlation between these two relative potency estimates. Figure 8-1 shows the
 3    results.
     10
                                              1:1 Correspondence Line
                                          Incidence RPF
                                                                                      100
 5          Figure 8-1. Correlation between incidence and multiplicity RTFs.
 6
 7          As shown in Figure 8-1, there is a high degree of correspondence between incidence and
 8   multiplicity RPFs calculated from results in the same animals, with one exception (see circled
 9   data point).  The regression analysis indicated an r2 of 0.76 for the correlation when the outlier
10   was excluded, or only 0.28 when it was included.  The outlier datapoint reflects the incidence
11   and multiplicity RPFs for benzo[c]fluorene calculated for the one oral study (Weyand et al.,
12   2004). All of the other datapoints reflect incidence and multiplicity RPFs for dermal or
13   intraperitoneal exposure studies; thus, one possible explanation for the outlier is that the
14   relationship between incidence and multiplicity after oral exposure differs from the relationship
15   after exposure via other routes.  However, there was good correspondence between incidence
16   and multiplicity in dermal and intraperitoneal studies, despite the marked differences in
17   absorption, distribution, and metabolism of PAHs administered by these two exposure routes.
                                             204
DRAFT - DO NOT CITE OR QUOTE

-------
 1   Compound-specific differences in the association between incidence and multiplicity RPFs also
 2   seem unlikely; the dataset shown in Figure 8-1 also includes a comparison between incidence
 3   and multiplicity RPFs for benzo[c]fluorene in an intraperitoneal exposure study, and there is
 4   good correspondence between the two (RPF = 1 for incidence and RPF =  0.6 for multiplicity).
 5   The most plausible explanation for the outlier is that the basis for the multiplicity RPF in the  oral
 6   study of benzo[c]fluorene (RPF = 50) was estimated using a point high on the dose-response
 7   curve (incidence was 100%), at which a large mean number of tumors per animal (46 ± 2.8) was
 8   recorded, while the incidence RPF (RPF = 5) for the same study was estimated using BMD
 9   modeling at a response point lower on the curve (BMR of 0.7). All of the other comparisons
10   between incidence and multiplicity RPFs from the same set of animals were based  on
11   multiplicity responses <10 tumors per animal.  Although there is little information with which to
12   explore this hypothesis, it is possible that RPFs for multiplicity that are calculated using
13   unusually high tumor number are not reliable measures of relative incidence potency.  This could
14   result from changes in the slope of the tumor number versus dose curve at high tumor number, or
15   from methodology limitations that hamper accurate measurement of high tumor numbers.
16          Notwithstanding the one outlier, as the remaining incidence and multiplicity RPFs from
17   the same study were highly correlated, only one of the two metrics (the higher of the incidence or
18   multiplicity RPF from the same study) was included in the average and range. Figure 8-1 shows
19   that multiplicity RPFs exhibit a slight tendency to underestimate the RPF from incidence data
20   (more points are to the right of the 1:1 correspondence line); thus, the higher value was usually
21   calculated from incidence data.  Specifically, 15/24 incidence RPFs were higher than the
22   corresponding multiplicity RPF from the same study, and 2/24 of the incidence and multiplicity
23   RPFs were identical. Thus, only 7/24 multiplicity RPFs were higher than their corresponding
24   incidence RPFs.
25          As discussed in Section 7.1, in newborn mouse studies that resulted in nonzero RPFs  for
26   both males and females (LaVoie et al., 1994, 1987; Wislocki  et al., 1986), the male RPF was
27   typically three- to fivefold higher than the female RPF, but both were included in the final RPF
28   calculation. Final RPFs that included both male and female values from the same study were
29   calculated for three PAHs: benzo[a]anthracene, benzo[j]fluoranthene, and fluoranthene. An
30   alternative approach would be to select the RPF associated with the most sensitive  sex (i.e.,
31   males) and to omit the female RPF from the final calculation.  The net effect of including female
32   RPFs for these three compounds is to reduce the average RPF and, in some cases, to reduce the
33   lower limit of the range of RPFs.  For benzo[a]anthracene and benzo[j]fluoranthene, the final
34   RPF is unchanged whether or not the female RPF is included. For fluoranthene, inclusion of the
35   female RPFs yields a final RPF of 0.08, while excluding the female RPFs would result in a final
36   RPF of 0.1.
37          Final RPFs were calculated as the arithmetic mean and range of RPFs from tumor
38   bioassay data when such data were available. Presenting the  average and  the range provides  both

                                           205           DRAFT - DO NOT CITE OR QUOTE

-------
 1
 2
 o
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
an average and a maximum estimate for each PAH that has data from multiple studies. Other
options for deriving a central tendency RPF include geometric mean, median, weighted average,
and order of magnitude estimates. The arithmetic mean represents a simple approach to
describing the calculated RPF values available for each PAH.  There were usually not enough
data (<3 RPFs for 18/24 PAHs with nonzero RPFs) to assess the shape of the RPF distribution
for any given PAH, so a geometric mean was not considered. Calculation of a weighted average
was considered, but without a clear biological rationale for assigning weights among study types
or tumor data outcomes, using a weighting approach might increase uncertainty. Finally, the use
of simple means and ranges of estimated RPFs rather than order of magnitude estimates, as has
been previously done for estimating RPFs for PAHs, was considered to better reflect the
available data and provide a clearer characterization of uncertainty.
       Cancer-related endpoint data were relied upon for the derivation of an RPF for only one
PAH (dibenz[a,c]anthracene).  For this compound, there were no tumor bioassay data suitable for
the determination of an RPF. However, cancer-related endpoint data provided qualitative
support for the finding of carcinogenicity for this compound (see individual narrative for this
compound in Section 6.2). Although the mutagenic mode of action for benzo[a]pyrene (U.S.
EPA, 2005b) suggests that, in general, these endpoints may be relevant to PAH carcinogenicity,
the predictive value of a positive response in these tests has not been conclusively demonstrated.
Thus, there is considerable uncertainty in an RPF based on cancer-related endpoint data.
Appendix G includes analysis of the correlation between average RPFs calculated from cancer-
related endpoint data and tumor bioassay data. As shown in Table 8-1, and further discussed in
Appendix G, cancer-related endpoint RPFs are reasonably predictive of tumor bioassay RPFs;
however, the relationship between these RPFs and the relative potency of a given PAH in
humans exposed via environmentally relevant routes is unknown.

       Table 8-1.  Results of simple linear  regression of log-transformed average
       tumor  bioassay RPF versus log average genotoxicity RPF
Genotoxicity endpoint
All in vivo DNA adducts
All in vivo nonbioassays
All nonbioassay endpoints (in vitro and in vivo)
All in vitro nonbioassays
All in vivo micronuclei and sister chromatid exchanges
All in vitro mutagenicity
r2
0.64
0.55
0.40
0.39
0.39
0.032
Slope
1.22
1.16
1.10
0.91
0.81
0.33
/7-Value
O.01
0.01
0.01
0.01
>0.05 (nonsignificant)
>0.05 (nonsignificant)
n
10
11
20
19
6
17
26
27
28
29
30
       For three PAHs (anthracene, phenanthrene, and pyrene), a final RPF of zero was
recommended. As noted earlier in Chapter 6, there is little quantitative difference between
selecting a final RPF of zero for a given PAH and excluding that PAH from the RPF approach.
However, excluding PAHs from the RPF approach implies substantial uncertainty (these
                                           206
                                                   DRAFT - DO NOT CITE OR QUOTE

-------
 1
 2
 o
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
compounds could be of low or high potency), while assigning an RPF of zero suggests lower
uncertainty because there is evidence to suggest that these compounds are not carcinogenic.
Nevertheless, there remains uncertainty in the RPFs for these three compounds, as all of them
included one or more studies suggesting activity in cancer-related endpoint assays.  In addition, it
is possible that available bioassay studies for these compounds may not provide sufficient
sensitivity to allow for a potency comparison with benzo[a]pyrene; thus, the RPF of zero should
not be considered a characterization of the inherent carcinogen!city of anthracene, phenanthrene,
or pyrene.
       In the present analysis, RPFs for individual PAHs were based on data of varying quality
and reproducibility, so there is additional uncertainty in risks estimated for mixtures containing
differing concentrations of individual PAHs.  Confidence ratings were assigned to each RPF to
qualitatively characterize the uncertainty in each individual RPF.  Table  8-2 shows the
distribution of PAHs with RPFs of each confidence rating.  As the table indicates, there are
5 PAHs with RPFs of high confidence, 8 PAHs with RPFs of medium confidence, 13 PAHs with
RPFs of low confidence, and 1 PAH with an RPF of very low confidence. The confidence
ratings assigned to the RPFs may be used to qualitatively assess the uncertainty in a mixtures risk
assessment that utilizes the RPFs. For example, if a high proportion of the total cancer risk
predicted for a given mixture is attributable to benzo[a]pyrene and other PAHs with RPFs of
high or medium confidence, then the confidence in the overall cancer risk assessment will be
relatively high. If, in contrast, benzo[a]pyrene contributes a relatively small fraction of the
overall risk, and/or the mixture consists primarily of PAHs with RPFs of low confidence, then
the confidence in the overall cancer risk assessment will be correspondingly lower.  Thus, it will
be important to consider the relative contribution of benzo[a]pyrene to the total risk, as well as
the relative confidence ratings of the RPF values for component PAHs, in the uncertainty
evaluation for cancer risk assessments that employ these RPFs.

       Table 8-2. PAHs with RPFs of varying relative confidence
       High confidence RPF
                      Medium confidence RPF
    Low confidence RPF
Very low confidence RPF
      Benzo [b]fluoranthene
      Benzo tj]fluoranthene
      Chrysene
      Dibenz[a,h]anthracene
      Phenanthrene
                     Anthanthrene
                     Anthracene
                     B enz [a] anthracene
                     Benzo [c]fluorene
                     Benzo fkjfluoranthene
                     Cyclopenta[c,d]pyrene
                     Dibenzo [a,l]py rene
                     Pyrene
Benz[b,c]aceanthrylene, 11H-
Benz [e] aceanthry lene
Benzo[g,h,i]perylene
Benz |] ] aceanthry lene
Benz [1] aceanthry lene
Cyclopenta[d,e,f]chrysene, 4H-
Dibenzo [a,e]fluoranthene
Dibenzo [a,e]pyrene
Dibenzo [a,h]pyrene
Dibenzo [a,i]pyrene
Fluoranthene
Indeno[l,2,3-c,d]pyrene
Naphtho [2,3 -e]pyrene
Dibenz [a,c] anthracene
27
                                             207
                                                      DRAFT - DO NOT CITE OR QUOTE

-------
 1   8.4.  USE OF ANIMAL DATA TO PREDICT HUMAN CANCER RISK FOR PAHs
 2          Section 4.2 briefly summarizes the epidemiology and human biomarker data related to
 3   exposure to PAH mixtures and carcinogenicity. Exposure to certain PAH mixtures is clearly
 4   associated with cancer in humans.  Epidemiology studies evaluating emissions from coke
 5   production, coal  gasification, aluminum production, iron and steel founding, coal tars, coal tar
 6   pitches, and soot have demonstrated associations between exposure and increased risk of lung
 7   cancer in humans (see review of Bostrom et al., 2002).  Skin and scrotal cancers have been
 8   associated with exposure to coal tar, coal tar pitches, nonrefmed mineral oils, shale oils, and soot
 9   (Larsen and Larsen, 1998; WHO, 1998; ATSDR, 1995). While human epidemiology data may
10   be sufficient for the purpose of quantifying the cancer risks associated with exposure to a few
11   PAH mixtures, there are no data for many mixtures; hence the need for other approaches
12   including surrogate-mixture and component-based approaches. As noted by the peer
13   consultation workshop (U.S. EPA, 2002), there are no human data on cancer response to
14   individual PAHs that could be used as the basis for, or as a supplement to,  a component-based
15   approach. As a result, the RPF approach relies on animal bioassay data to  predict human cancer
16   risk associated with individual PAHs.
17          The use of animal bioassays in predicting relative carcinogenic potency in humans
18   represents a source of uncertainty in this approach.  As there are no human data on cancer
19   response to individual PAHs, including benzo[a]pyrene, there can be no quantitative evaluation
20   of uncertainty in extrapolating from RPFs based on animal bioassay data to relative potency in
21   humans. Possible species differences in toxicokinetics,  toxicodynamics, and mode of action
22   contribute to the  uncertainty. Cancer-related endpoint data are available using human cells (e.g.,
23   epidermal keratinocytes, lymphoblasts, human epithelial cells) for the evaluation of
24   mutagenicity, DNA adducts, unscheduled DNA synthesis, DNA damage, and clastogenicity or
25   sister chromatid  exchange frequency (see Section 4.3).  Findings in human cells were generally
26   consistent with those in other mammalian cells; however, whether this finding of consistency
27   extends to effects in vivo, and specifically to formation  of tumors, is not known.
28          In addition, animal  bioassays use various routes  of administration (e.g., intraperitoneal
29   and subcutaneous injection), which may not be directly  relevant to expected routes of exposure
30   for humans. It is difficult to determine whether the relative potency based  on animal bioassays
31   using injection routes of exposure is predictive of relative potency that would be observed in
32   humans exposed through environmentally relevant  exposure routes (see further discussion of
33   exposure-route uncertainties in Section 8.6). An additional source of uncertainty in the use of
34   animal bioassay  data stems from differences in the  doses used in animal bioassays as compared
35   with low doses received by humans exposed in the  environment. Mechanistic data, primarily
36   obtained using benzo[a]pyrene, provide support for the  human relevance of PAH tumorigenicity
37   in animals. There is evidence linking three pathways activating benzo[a]pyrene to DNA-reactive
38   agents [(+)-anti-BPDE, radical cations, benzo[a]pyrene-7,8-dione, and reactive oxygen species]

                                            208          DRAFT - DO NOT CITE OR QUOTE

-------
 1   with key mutational events in genes (p53 tumor suppressor gene and H-ras or K-ras oncogenes)
 2   that can lead to tumor initiation.  Results in support of mutagenic modes of action via the diol
 3   epoxide and radical cation pathways include in vivo results in animals. All of these activation
 4   pathways occur in human tissues, and associations have been made between spectra of mutations
 5   in the p53 tumor suppressor gene or ras oncogenes induced by benzo[a]pyrene metabolites with
 6   spectra of mutations in these genes in tumor tissue from benzo[a]pyrene-exposed animals or
 7   tumor tissue in humans.
 8          Support for the association between the diol epoxide pathway and tumor initiation
 9   includes observation that: (+)-anti-BPDE activated the H-ras-1 proto-oncogene to transform
10   NIH/3T3 cells via G^T point mutations in the 12th codon (Marshall et al., 1984); (+)-anti-
11   BPDE reacts with the p53 tumor suppressor gene at several hotspots mutated in lung cancer
12   patients (Denissenko et al., 1996; Puisieux et al., 1991); the spectra of p53 and K-ras mutations
13   in lung tumors of nonsmoking patients, chronically exposed to smoky coal emissions, was
14   consistent with (+)-anti-BPDE mutations in these genes (DeMarini et al., 2001); elevated BPDE-
15   DNA adducts have been observed in coke oven workers and chimney sweepers (Pavanello et al.,
16   1999); and the spectra of mutation in the K-ras, H-ras, and p53 genes in forestomach tumors of
17   mice fed benzo[a]pyrene in the diet for 2 years were consistent with (+)-anti-BPDE DNA
18   reactions (Culp et al., 2000).
19          Support for the radical cation pathway includes observations that depurinated adducts,
20   (expected products from reactions of benzo[a]pyrene radical cations with DNA) accounted for
21   74% of identified DNA adducts in mouse skin exposed to benzo[a]pyrene (Rogan et al., 1993)
22   and 9/13 examined tumors from mice exposed to dermal applications of benzo[a]pyrene had
23   H-ras oncogene mutations attributed to depurinated DNA adducts from benzo[a]pyrene radical
24   cations (Chakravarti et al., 1995).
25          Support for the aldo-keto reductase pathway includes in vitro demonstration that several
26   types of DNA damage can occur from o-quinones and reactive oxygen species (Park et al., 2006;
27   Balu et al., 2004; McCoull et al., 1999; Flowers-Geary et al., 1997, 1996), benzo[a]pyrene-
28   7,8-dione can induce mutations in the p53 tumor suppressor gene using an in vitro yeast reporter
29   gene assay (Park et al., 2008; Shen et al., 2006; Yu et al., 2002), and dominant p53 mutations
30   induced by benzo[a]pyrene,7,8-dione in this system corresponded with p53 mutation hotspots
31   observed in human lung cancer tissue (Park, 2008).
32          All three activation pathways are expected to  occur in human tissues (Jiang et al., 2007),
33   and associations have been made between spectra of mutations in the p53 tumor suppressor gene
34   or ras oncogenes induced by benzo[a]pyrene metabolites with spectra of mutations in these genes
35   in tumor tissue from benzo[a]pyrene-exposed animals or humans. In particular, DeMarini et al.
36   (2001) demonstrated mutations in the p53 tumor suppressor gene and the K-ras oncogene in the
37   lung tumors of nonsmokers, whose tumors were  associated with exposure to smoky coal.
                                           209          DRAFT - DO NOT CITE OR QUOTE

-------
 1          The available information supporting these actions for benzo[a]pyrene is consistent with
 2   what is known about the mode of action for other PAHs demonstrated to induce cancer in
 3   animals, including cyclopenta[cd]pyrene, dibenz[a,h]anthracene, and dibenzo[a,l]pyrene
 4   (Cogliano et al., 2008; Straif et al., 2005).  All PAHs that have been studied require metabolic
 5   activation to produce carcinogenic responses in animals, and there is evidence for activation to
 6   DNA reactive intermediates via several pathways (Straif et al., 2005; Xue and Warshawsky,
 7   2005; WHO,  1998; Cavalieri and Rogan,  1995). For example, incubation of rat liver
 8   microsomes with dibenzo[a,l]pyrene, a PAH that is more tumorigenically potent than
 9   benzo[a]pyrene in mouse skin and rat mammary tissue, formed depurinated DNA adducts from
10   the radical cation pathway, as well as DNA adducts from the diol epoxide pathway (Cavalieri
11   and Rogan, 1995).
12          In summary, the relevance of animal bioassay data to the prediction of human
13   carcinogenic potency remains a significant area of uncertainty in the use of this and other
14   approaches to PAH cancer risk assessment. However, mechanistic data on benzo[a]pyrene and
15   other PAHs provide evidence that the molecular events leading to PAH-induced tumor formation
16   in animals are relevant to humans.
17
18   8.5. ASSUMPTIONS OF A COMMON MODE OF ACTION AND DOSE  ADDITIVITY
19          A discussion of the potential modes of action for PAH carcinogenicity is presented in
20   Section 2.4. Individual carcinogenic PAHs are linked by a common effect (i.e., tumorigenicity),
21   which may occur through multiple mechanisms. Reactive metabolites produced during
22   metabolic transformations of PAHs include diol epoxides, reactive oxygen species, radical
23   cations, and o-quinones. The formation of these metabolites is not mutually exclusive, and the
24   carcinogenic process for PAHs is likely to be related to some combination of molecular events
25   resulting from formation of several reactive species.  Reactive metabolites of PAHs interact with
26   DNA to form adducts and produce DNA damage resulting in mutations in cancer-related genes
27   such as tumor suppressor genes or oncogenes. These events appear to reflect the initiation
28   potency of an individual PAH (e.g., strong mutagens are generally potent initiators) (Sjogren et
29   al., 1996). Certain PAHs exhibit promotional effects that may be related to cytotoxicity and the
30   formation of reactive oxygen species, AHR affinity, and the upregulation of genes related to
31   biotransformation (i.e., induction of CYP1A1), growth, and differentiation (Bostrom et al.,
32   2002). The inhibition of gap junctional intracellular communication is also related to tumor
33   promotion by PAHs (Bostrom et al., 2002). The ability of certain PAHs to act as tumor
34   promoters as well as initiators may increase their carcinogenic potency in animal bioassays
35   conducted at high doses. Initiation potency may be more relevant to low-level environmental
36   exposure in humans (Bostrom et al., 2002; Sjogren et al., 1996); however, the proposed RPF
37   approach is not unduly affected by this as it relies largely on high-dose animal bioassay data for
                                            210          DRAFT - DO NOT CITE OR QUOTE

-------
 1   selecting RPF values.  This represents an uncertainty in the use of the RPF approach in
 2   estimating human cancer risks from PAHs.
 3          Conceptually, the uncertainty related to relative potency for initiation versus promotion
 4   could be reduced by using separate RPF schemes for each part of the carcinogenic process.  This
 5   would require selection of indicator compounds that best represent the initiation and promotion
 6   processes, and use of mechanistic data to determine relative potency for each process (i.e.,
 7   mutagenicity for initiation, AhR binding, or enzyme induction for promotion).  There are several
 8   problems with this approach, including the lack of data to support the selection of indicator
 9   compounds and the complete carcinogenic nature of many PAHs (i.e., they act as both initiators
10   and promoters). The initiation and promotion potency of an individual PAH is determined by its
11   chemical structure. Some PAHs are strong mutagens, but have low affinity for the AhR (e.g.,
12   fjord-region PAHs) (Bostrum et al., 2002; Sjogren et al.,  1996). Other PAHs are complete
13   carcinogens, with initiating properties (i.e., mutagenesis) and AhR affinity leading to tumor
14   promotion (e.g., benzo[a]pyrene, dibenz[a,h]anthracene)  (Bostrum et al., 2002; Sjogren et al.,
15   1996). Benzo[a]pyrene is considered a good indicator compound for similar PAHs with
16   complete carcinogenic activity. However, the relative potency of other PAHs, especially those
17   that act primarily via either initiation or promotion, may be over- or underestimated.
18          There is evidence that an assumption of similar toxicological action is reasonable for
19   PAHs; however, the carcinogenic process for individual PAHs is likely to be related to some
20   unique combination of multiple molecular events resulting from formation of several reactive
21   species.  The absence of a clearly-defined common mode of action increases the level of
22   uncertainty associated with the use of an RPF approach.  It is not possible to determine whether
23   cancer risks would be under- or overestimated by using a PAH RPF approach that assumes a
24   common mode of action. The assumption that interactions among PAH mixture components do
25   not occur at low levels of exposure cannot be conclusively demonstrated using experimental
26   approaches. The experimental data relating to dose additivity for PAH carcinogenicity are
27   discussed in Section 2.8. It appears that interactions may occur at higher doses of PAH mixtures
28   given in combination. This remains a significant uncertainty  in the proposed RPF approach.
29
30   8.6.  EXTRAPOLATION OF RPFs ACROSS EXPOSURE ROUTES
31          The peer consultation workshop (U.S.  EPA, 2002) also identified uncertainty in
32   extrapolation of RPFs across exposure routes. As with the 1993 Provisional Guidance., RPFs
33   proposed in this analysis are also based on in vivo bioassay data collected using various  routes of
34   administration (e.g., dermal, intraperitoneal, subcutaneous, intramammillary, intramuscular, or
35   intravenous injection, as well as lung implantation, tracheal implantation, and transplacental
36   exposure after subcutaneous injection). The RPF approach considers each bioassay type
37   equivalent for the purpose of determining relative potency to  benzo[a]pyrene.
                                            211          DRAFT - DO NOT CITE OR QUOTE

-------
1          Table 8-3 compares the average RPFs (calculated from raw numbers and rounded to one
2   significant digit) based on tumor bioassay data for each PAH across exposure routes. Dermal
3   studies are shown collectively as well as separated by study type (complete or initiation).
4   Likewise, intraperitoneal studies are shown grouped as well as separated by target organ (lung
5   and liver).
                                          212          DRAFT - DO NOT CITE OR QUOTE

-------
Table 8-3. Comparisons among average tumor bioassay RPF values by exposure route and target organ
PAH
AA
AC
BaA
BbcAC
(1,12-MBA)
BbF
BcFE
BeAC
BghiP
BjAC
BjF
BkF
B1AC
CH
CPcdP
CPdefC
DBacA
DBaeF
DBaeP
DBahA
DBahP
DBaiP
DBalP
FA
IP
N23eP
Dermal
n
1
-
1
1
2
-
2
-
-
2
1
2
5
4
2
-
2
2
1
1
2
2
-
-
1
Average
0.5
-
0.02
0.05
0.4
-
0.8
-
-
0.03
0.03
5
0.1
0.3
0.3
-
0.9
0.4
1
0.9
0.6
30
-
-
0.3
Dermal
complete
n
1
-
-
-
1
-
-
-
-
-
-
-
-
2
-
-
1
1
-
-
1
-
-
-
-
Average
0.5
-
-
-
0.3
-
-
-
-
-
-
-
-
0.4
-
-
1
0.3
-
-
0.7
-
-
-
-
Dermal initiation
n
-
-
1
1
1
-
2
-
-
2
1
2
5
2
2
-
1
1
1
1
1
2
-
-
1
Average
-
-
0.02
0.05
0.4
-
0.8
-
-
0.03
0.03
5
0.1
0.2
0.3
-
0.7
0.4
1
0.9
0.5
30
-
-
0.3
Intraperitoneal
n
-
-
2
-
2b
1
-
-
1
2b
-
-
1
1
-
-
-
-
1
-
-
1
5
-
-
Average
-
-
0.2a
-
lc
ld
-
-
60d
OT
-
-
0.2a
ld
-
-
-
-
40d
-
-
30d
0.08a
-
-
Intraperitoneal,
target organ =
lung
n
-
-
1
-
1
1
-
-
1
1
-
-
-
1
-
-
-
-
1
-
-
1
4
-
-
Average
-
-
0.08
-
1
1
-
-
60
0.4
-
-
-
1
-
-
-
-
40
-
-
30
0.05
-
-
Intraperitoneal,
target organ =
liver
n
-
-
2
-
-
-
-
-
-
1
-
-
1
-
-
-
-
-
-
-
-
-
1
-
-
Average
-
-
0.4
-
-
-
-
-
-
1
-
-
0.2
-
-
-
-
-
-
-
-
-
0.2
-
-
Lung
implantation
n
1
-
-
-
1
-
-
1
-
1
1
-
1
-
-
-
-
-
1
-
-
-
-
1
-
Average
0.2
-
-
-
0.1
-
-
0.009
-
0.03
0.03
-
0.04
-
-
-
-
-
2
-
-
-
-
0.07
-
Oral
n
-
-
-
-
-
1
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
Average
-
-
-
-
-
50
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
                                          213
DRAFT - DO NOT CITE OR QUOTE

-------
       Table 8-3. Comparisons among average tumor bioassay RPF values by exposure route and target organ
PAH
PH
Pyr
Dermal
n
-
-
Average
-
-
Dermal
complete
n
-
-
Average
-
-
Dermal initiation
n
-
-
Average
-
-
Intraperitoneal
n
-
-
Average
-
-
Intraperitoneal,
target organ =
lung
n
-
-
Average
-
-
Intraperitoneal,
target organ =
liver
n
-
-
Average
-
-
Lung
implantation
n
-
-
Average
-
-
Oral
n
-
-
Average
-
-
"Newborn mouse model.
bNumber of intraperitoneal RPFs includes those calculated for combined lung and liver incidence; these are not included in numbers of RPFs with lung or liver tumors.
Includes both newborn mouse and adult A/J mouse models.
dAdult A/J mouse model.
                                                         214
DRAFT - DO NOT CITE OR QUOTE

-------
 1
 2          The table shows a marked difference between the oral and intraperitoneal RPFs for
 3   benzo[c]fluorene (BcFE) (RPF = 50 for oral multiplicity and RPF = 1 for intraperitoneal
 4   incidence). However, as discussed earlier, this difference may result more from the use of a high
 5   tumor number to calculate the oral multiplicity RPF for this compound than route differences; if
 6   the oral incidence RPF is used for comparison, the two routes are more similar (RPF = 1 for
 7   intraperitoneal incidence versus RPF = 5 for oral incidence). Based on the latter comparison,
 8   which represents the only data with which to compare oral RPFs with those calculated from
 9   other routes, there appears to be fairly good correspondence between intraperitoneal and oral
10   RPFs; however, this is based on only one PAH.
11          Based on the comparisons in the table, RPFs based on initiation and complete dermal
12   carcinogenicity studies are similar (within a factor of 2).  However, there are few PAHs with
13   both types of dermal studies.
14          With respect to other route comparisons, the table generally shows that RPFs calculated
15   from lung implantation and dermal studies are of the same order of magnitude, while RPFs
16   calculated from intraperitoneal studies are higher for most compounds. The intraperitoneal RPF
17   for dibenzo[a,l]pyrene is similar to its dermal RPF. At first glance, one might attribute the
18   higher intraperitoneal RPFs calculated from newborn mouse assays (footnoted "a" in the table)
19   to greater sensitivity of the newborn mouse, compared with an adolescent or adult mouse, to the
20   carcinogenic action of PAHs.  However, since the RPFs reflect potency of the PAH relative to
21   benzo[a]pyrene, and not potency of the newborn mouse relative to other systems, the higher RPF
22   cannot reflect a greater sensitivity of the animal model, since both the PAH of interest and
23   benzo[a]pyrene have been tested in the same model.  There is little information to evaluate
24   whether RPFs from newborn mouse studies tend to be higher or lower than the adult A/J mouse
25   model when both are exposed via intraperitoneal injection.  Only one compound,
26   benzo[b]fluoranthene (BbF), had RPFs calculated from both newborn mouse  and adult A/J
27   mouse models, and the values were similar; the newborn mouse RPF was 2, while the A/J mouse
28   RPF was 1. In summary, it is not clear whether the intraperitoneal RPFs are higher than dermal
29   or lung implantation RPFs due to route-specific differences or animal model differences (for
30   example, differential metabolism in various animal systems).
31          Cross-route extrapolation of relative potency estimates is a necessary, though uncertain,
32   aspect of the RPF approach.  It is difficult to determine which of the available study types (e.g.,
33   dermal, intraperitoneal, intratracheal) is most predictive of potential risks from oral and
34   inhalation exposure in humans. In order to prioritize bioassays by exposure route, robust data
35   are needed on relative potencies for oral and inhalation exposures for comparison with relative
36   potencies based on other exposure routes.
37          The inhalation RPF scheme used by the California EPA (2004) employed a hierarchy of
38   bioassay data based on exposure route (inhalation studies were preferred, followed by

                                            215          DRAFT - DO NOT CITE OR QUOTE

-------
 1   intratracheal or intrapulmonary instillation, oral administration, skin-painting, and subcutaneous
 2   or intraperitoneal injection). Apart from the obvious preference for exposure routes that targeted
 3   the respiratory tract (inhalation, intratracheal, intrapulmonary), the basis for prioritizing the other
 4   exposure routes is not evident. Pufulete et al. (2004), who were also focused on PAHs as air
 5   contaminants, suggested that the clearance of PAHs after intratracheal instillation may be similar
 6   to clearance after inhalation exposure. The authors acknowledged that the high concentrations of
 7   PAHs used in intratracheal and intrapulmonary instillation studies may lead to major differences
 8   in pharmacokinetics, compared with inhalation exposure (Pufulete et al., 2004).  Nevertheless,
 9   the authors suggested that intratracheal instillation of low doses of PAHs might be an appropriate
10   surrogate exposure model for assessing relative potency of inhalation exposure.  It is important
11   to note that no intratracheal instillation studies were identified in the search for studies from
12   which to calculate RPFs; thus, the information provided by Pufulete et al. (2004) is not directly
13   useful for suggesting route-specific RPFs. Pufulete et al. (2004) did not provide any specific
14   information on  the relevance of intrapulmonary administration (a route used in several of the
15   bioassays used  to calculate RPFs) to inhalation exposure.
16          As noted by U.S. EPA (2004), cross-route extrapolation would be contraindicated if there
17   were convincing toxicokinetic evidence that absorption of PAHs does  not occur by one or more
18   exposure routes.  Available data on the absorption of PAHs indicate that, in general, PAHs are
19   readily absorbed via ingestion, inhalation, and dermal exposure routes; however, the rate of
20   uptake varies with route and other factors (e.g., matrix, intake of fats and oils) (ATSDR, 1995).
21   Evidence for absorption of PAHs through these routes includes measurement of PAH-DNA
22   adducts at sites distal from the route of entry, measurement of urinary  metabolites, and
23   radiotracer studies in animals (ATSDR, 1995). U.S. EPA (2004) indicated that demonstration of
24   any degree of uptake for each of the routes of interest is sufficient to allow the qualitative
25   judgment to apply the route-to-route extrapolation; thus, cross-route extrapolation is supported
26   by current data on the bioavailability of PAHs across several exposure routes.
27          U.S. EPA (2004, 1994) also noted that point-of-entry toxicity may be considered contrary
28   evidence for cross-route extrapolation. With respect to PAHs, available information on this issue
29   is mixed.  The one inhalation bioassay of benzo[a]pyrene (Thyssen et  al., 1981) identified the
30   upper respiratory tract as the site of tumor formation, suggesting a point-of-entry effect;
31   however, the authors did not specify the organs that were examined histologically in the study.
32   Dermal bioassays of benzo[a]pyrene have generally evaluated only skin tumors, precluding their
33   use in determining whether distal tumors are induced. A number of early oral cancer bioassays
34   of benzo[a]pyrene suggested that tumor formation was limited to the forestomach (Rigdon and
35   Neal, 1969, 1966; Neal and Rigdon, 1967).  In oral carcinogenicity bioassays of MGP residue
36   (Weyand et al., 1995) and coal tar preparations (Culp et al., 1998; Gaylor et al., 1998) that
37   included separate groups exposed to benzo[a]pyrene, there were significant differences in target
38   organ distribution of tumors between benzo[a]pyrene and the complex mixtures.

                                            216           DRAFT - DO NOT CITE OR QUOTE

-------
 1   Benzo[a]pyrene-induced tumors were observed primarily at the point of contact (i.e., the
 2   forestomach), while MGP residue and coal tar produced tumors in the lung, liver, forestomach,
 3   skin, and other organs.  Other PAHs (e.g., benzo[c]fluorene) were proposed as the primary
 4   compounds responsible for tumors at distal sites such as the lung (Koganti et al., 2000; Gulp et
 5   al., 1998). However, a gavage study in rats (Kroese et al., 2001) and a dietary study in A/J mice
 6   (Weyland et al., 2004) each demonstrated that oral exposure to benzo[a]pyrene could induce
 7   tumors at distal sites, including the lung, liver, and auditory canal.  Tissue-specific differences in
 8   metabolic activation and DNA binding of PAHs may  contribute to the observed differences in
 9   target organ sensitivity (Weyand and Wu, 1995; Gulp and Beland, 1994).
10          In summary, available information provides some support for cross-route extrapolation.
11   Absorption of PAHs across  oral, inhalation, and dermal routes is evident and, while many of the
12   cancer bioassays of benzo[a]pyrene suggested tumor formation limited to the point-of-entry, at
13   least one recent study (Kroese et al., 2001) suggests that tumors may also be induced at distal
14   sites.  Furthermore, there is  evidence that other PAHs (e.g., benzo[c]fluorene) may induce
15   tumors at distal sites after oral exposure (Weyand et al., 2004; Koganti et al., 2000;  Gulp et al.,
16   1998). However, cross-route extrapolation of RPFs is a significant source of uncertainty in this
17   approach.
18          Another approach to the issue of route-to-route extrapolation would be to prefer RPFs
19   derived from particular target tissues deemed relevant to  the exposure route of interest. For
20   example, RPFs based on lung tumor data might be preferred for use in inhalation risk
21   assessment.  To examine whether lung tumor RPFs were consistent across routes, RPFs
22   calculated from lung tumor  potency in intraperitoneal studies (both newborn mouse and adult
23   A/J mouse models) were compared with RPFs from lung implantation studies in Table 8-3.
24   RPFs for both intraperitoneal-lung and lung implantation studies were available for only four
25   compounds (benzo[b]fluoranthene, benzo[j]fluoranthene, chrysene, and dibenz[a,h]anthracene);
26   for each of these, the intraperitoneal lung tumor RPF exceeded the lung implantation RPF. No
27   information assessing the concordance between lung tumor potency after intraperitoneal
28   administration and inhalation cancer potency was identified in the literature.  The use of the final
29   RPFs derived in this analysis across all routes of exposure is recommended given the information
30   outlined above and in the absence of data to indicate otherwise.
                                            217          DRAFT - DO NOT CITE OR QUOTE

-------
                                         9. REFERENCES
Abe, S; Sasaki, M. (1977) Chromosome aberrations and sister chromatic! exchanges in Chinese hamster cells
exposed to various chemicals. J Natl Cancer Inst 58:1635-1641.

Agrelo, C; Amos, H. (1981) DNA repair in human fibroblasts.  Prog Mutat Res 1:528-532.

Albert, RE; Lewtas, J; Nesnow,  S; et al. (1983) Comparative potency method for cancer risk assessment: application
to diesel paniculate emissions. Risk Anal 3:101-117.

Albert, RE; Miller, ML; Cody, TE; et al. (1991) Benzo[a]pyrene-induced skin damage and tumor promotion in the
mouse. Carcinogenesis 12:1273-1280.

Allen, JA; Coombs, MM. (1980) Covalent binding of poly cyclic aromatic hydrocarbons to mitochondrial and
nuclear DNA. Nature 287:244-245.

Allen, CCR; Boyd, DR; Hempenstall, F; et al. (1999) Contrasting effects of a nonionic surfactant on the
biotransformation of poly cyclic aromatic hydrocarbons to cis-dihydrodiols by soil bacteria. Appl Environ Microbiol
65:1335-1339.

Allen-Hoffmann, BL; Rheinwald, JG. (1984) Poly cyclic aromatic hydrocarbon mutagenesis of human epidermal
keratinocytes in culture. Proc Natl Acad Sci USA 81:7802-7806.

Amacher, DE; Paillet, SC. (1982) Hamster hepatocyte-mediated activation of procarcinogens to mutagens in the
L5178Y/TK mutation assay.  Mutat Res 106:305-316.

Amacher, DE; Paillet, SC. (1983) The activation of procarcinogens to mutagens by cultured rat hepatocytes in the
L5178Y/TK mutation assay.  Mutat Res 113:77-88.

Amacher, DE; Turner, GN. (1980) Promutagen activation by rodent-liver postmitochondrial fractions in the
L5178Y/TK cell mutation assay. Mutat Res 74:485-501.

Amacher, DE; Paillet, SC; Turner, GN; et al. (1980) Point mutations at the thymidine kinase locus in L5178Y
mouse lymphoma cells. II. Test validation and interpretation. Mutat Res 72:447-474.

Amin, S; Desai, D; Dai, W; et al. (1995) Tumorigenicity in newborn mice of fjord region and other sterically
hindered diol epoxides of benzo[g]chrysene, dibenzo[a,l]pyrene (dibenzo[def,p]chrysene),
4H-cyclopenta[def]chrysene and fluoranthene. Carcinogenesis 16:2813-2817.

Andrews, AW; Thibault, LH; Lijinsky, W. (1978) The relationship between carcinogenicity and mutagenicity of
some polynuclear hydrocarbons. Mutat Res 51:311-318.

Archibong, AE; Inyang, F; Ramesh, A; et al. (2002) Alteration of pregnancy related hormones and fetal survival in
F-344 rats exposed by inhalation to benzo[a]pyrene.  Reprod Toxicol 16:801-808.

Arif, JM; Smith, WA; Gupta, RC. (1997) Tissue distribution of DNA adducts in rats treated by intramammillary
injection with dibenzo[a,l]pyrene, 7,12-dimethylbenz[a]anthracene and benzo[a]pyrene.  Mutat Res  378:31-39.

Atchison, M; Atchison, ML; VanDuuren, BL. (1985) Cocarcinogenesis in vitro using Balb/3T3 cells and aromatic
hydrocarbon cocarcinogens. Cell Biol Toxicol 1:323-331.

ATSDR (Agency for Toxic Substances and Disease Registry). (1995) Toxicological profile for polycyclic aromatic
Hydrocarbons (PAHs).  Public Health Service, U.S. Department of Health and Human Services.

Ayrton, AD; McFarlane, M; Walker, R; et al. (1990) Induction of the P-450 I family of proteins by polycyclic
aromatic  hydrocarbons: possible relationship to their carcinogenicity.  Toxicology 60:173-186.
                                             218            DRAFT - DO NOT CITE OR QUOTE

-------
Baird, WM; Salmon, CP; Diamond, L. (1984) Benzo[e]pyrene-induced alterations in the metabolic activation of
benzo[a]pyrene and 7,12-dimethylbenz[a]anthracene by hamster embryo cells. Cancer Res 44:1445-1452.

Baird, WM; Hooven, LA; Mahadevan, B; et al. (2002) Responses of human cells to PAH-induced DNA damage.
Polycycl Aromat Compd 22:771-780.

Baker, RS; Bonin, AM; Stupans, I; et al. (1980) Comparison of rat and guinea pig as sources of the S9 fraction in
the Salmonella/mammalian microsome mutagenicity test. Mutat Res 71:43-52.

Balu, N; Padgett, WT, Lambert, GR; et al. (2004) Identification and characterization of novel stable deoxyguanosine
and deoxyadenosine adducts of benzo[a]pyrene-7,8-quinone from reactions of physiological pH.  Chem Res Toxicol
17(6):827-838.

Balu, N; Padgett, WT; Nelson, GB. (2006) Benzo[a]pyrene-7,8-quinone-3'-mononucleotide adduct standards for 32P
postlabeling analysis: detection of benzo[a]pyrene-7,8-quinone-calf thymus DNA adducts. Anal Biochem
15(2):213-223.

Barfknecht, TR; Hites, RA; Cavaliers, EL; et al. (1982) Human cell mutagenicity of polycyclic aromatic
hydrocarbon components of diesel emissions. Dev Toxicol Environ Sci 10:277-294.

Barrai, I; Barale, R; Scapoli, C; et al. (1992) The analysis of the joint effect of substances on reversion systems and
the assessment of antimutagenicity. Mutat Res 267:173-182.

Barry, G; Cook, JW; Haslewood, GAD; et al. (1935) The production of cancer by pure hydrocarbons. Part III.  Proc
Royal Soc London 117:318-351.

Bartsch, H; Malaveille, C; Camus, AM; et al. (1980) Validation and comparative studies on 180 chemicals with S.
typhimurium strains and V79 Chinese hamster cells in the presence of various metabolizing systems. Mutat Res
76:1-50.

Bayer, U. (1978) In vivo induction of sister chromatid exchanges by three polyaromatic hydrocarbons.
Carcinogenesis 3:423-428.

Beland,  F; Gulp, S. (1998) Chronic bioassay of two composite samples from selected manufactured gas plant waste
sites.  Jefferson, AK: Division of Biochemical Toxicology, National Center for Toxicological Research. Technical
Report 6722.02. Unpublished report.

Biancifiori, C; Caschera, F. (1962) The relation between pseudopregnancy and the chemical induction by four
carcinogens of mammary and ovarian tumours in BALB/c mice. Br J Cancer 16:722-730.

Bingham, E; Falk, HL. (1969) The modifying effects of carcinogens on the threshold response. Arch Environ
Health 19:779-783.

Binkova, B; Giguere, Y; Rossner,  P, Jr.; et al. (2000) The effect of dibenzo[a,l]pyrene andbenzo[a]pyrene on
human diploid lung fibroblasts: the induction of DNA adducts, expression of p53 and p21(WAFl) proteins and cell
cycle distribution. Mutat Res 471:57-70.

Blackburn, GL; Roy, TA; Bleicher, WT, Jr.; et al. (1996) Comparison of biological and chemical predictors of
dermal carcinogenicity of petroleum oils. Polycycl Aromat Compd 11:201-210.

Blaha, L; Kapplova, P; Vondracek, J; et al. (2002) Inhibition of gap-junctional intercellular communication by
environmentally occurring polycyclic aromatic hydrocarbons. Toxicol Sci 65:43-51.

Bols, NC; Schirmer, K; Joyce, EM; et al. (1999) Ability of polycyclic aromatic hydrocarbons to induce
7-ethoxyresorufin-o-deethylase activity in a trout liver cell line. Ecotoxicol Environ Saf 44:118-128.

Bolton,  JL; Trush, MA; Penning, TM; et al. (2000) Role of quinones in toxicology. Chem Res Toxicol 13(3):2-17.
                                             219            DRAFT - DO NOT CITE OR QUOTE

-------
Bos, RP; Theuws, JLG; Jongeneelen, FJ; et al. (1988) Mutagenicity of bi-, tri- and tetra-cyclic aromatic
hydrocarbons in the 'taped-plate assay' and in the conventional Salmonella mutagenicity assay. Mutat Res 204:203-
206.

Bostrom, E; Engen, S; Eide, I. (1998) Mutagenicity testing of organic extracts of diesel exhaust particles after
spiking with polycyclic aromatic hydrocarbons (PAH). Arch Toxicol 72:645-649.

Bostrom, CC; Gerde, P; Hanberg, A; et al. (2002) Cancer risk assessment, indicators, and guidelines for polycyclic
aromatic hydrocarbons in the ambient air.  Environ Health Perspect 110(Suppl 3):451-488.

Bosveld, AT; de Bie, PA; van den Brink, NW; et al. (2002) In vitro EROD induction equivalency factors for the
10 PAHs generally monitored in risk assessment studies in The Netherlands. Chemosphere 49:75-83.

Brookes, P; Lawley, PD. (1964) Evidence of the binding of polynuclear aromatic hydrocarbons to the nucleic acids
of mouse skin: relation between carcinogenic power of hydrocarbons and their binding to deoxyribonucleic acid.
Nature 202:781-784.

Bruce, ED; Austenrieth, RL; Burghardt, RC; et al.  (2008) Using quantitative structure-activity relationships  (QSAR)
to predict toxic endpoints for polycyclic aromatic hydrocarbons (PAH). J Toxicol Environ Health Part A 71:1073-
1084.

Brune, K; Kalin, H; Schmidt, R; et al. (1978) Inflammatory, tumor initiating and promoting activities of polycyclic
aromatic hydrocarbons and diterpene esters in mouse skin as compared with their prostaglandin releasing potency in
vitro. Cancer Lett 4:333-342.

Brune, H; Deutsch-Wenzel, RP; Habs, M; et al. (1981) Investigation of the tumorigenic response to benzo[a]pyrene
in aqueous caffeine solution applied orally to Sprague-Dawley rats. J Cancer Res Clin Oncol 102:153-157.

Bryan, WR; Shimkin, MB. (1943) Quantitative analysis  of dose-response data obtained with three carcinogenic
hydrocarbons in strain C3H male mice.  J Natl Cancer Inst 3:503-531.

Bryla, P; Weyand, EH. (1992) Detection of PAH:DNA adducts from auto-oxidation using 52P-postlabeling.  Cancer
Lett 65:35^1.

Buening, MK; Levin, W; Wood, A; et al. (1979) Tumorigenicity of the dihydrodiols of dibenz[a,h]anthracene on
mouse skin and in newborn mice. Cancer Res 39:1310-1314.

Burdick, AD; Davis, JW; Liu, KJ; et al.  (2003) Benzo[a]pyrene quinones increase cell proliferation, generate
reactive oxygen species, and transactivate the epidermal growth factor receptor in breast epithelial cells. Cancer Res
63:7825-7833.

Busby, WFJ; Goldman, ME; Newberne, PM; et al. (1984) Tumorigenicity of fluoranthene in a newborn mouse lung
adenomabioassay.  Carcinogenesis 5:1311-1316.

Busby, WFJ; Stevens, EK; Martin, CN; et al. (1989) Comparative lung tumorigenicity of parent and mononitro-
polynuclear aromatic hydrocarbons in the BLU:Ha newborn mouse assay. Toxicol Appl Pharmacol 99:555-563.

Buterin, T; Hess, MT; Luneva, N; et al. (2000) Unrepaired fjord region polycyclic aromatic hydrocarbon-DNA
adducts  in ras codon 61 mutational  hot spots. Cancer Res 60:1849-1856.

Buters, JT; Mahadevan, B; Quintanilla-Martinez, L; et al. (2002) Cytochrome P450 1B1 determines susceptibility to
dibenzo[a,l]pyrene-induced tumor formation. ChemRes Toxicol 15:1127-1135.

California EPA (California Environmental Protection Agency). (2002) Air toxics hot spots program risk assessment
guidelines. Part II. Technical support document for describing available cancer potency factors.  Office of
Environmental Health Hazard Assessment, Air Toxicology and Epidemiology Section, Oakland, CA.
                                             220            DRAFT - DO NOT CITE OR QUOTE

-------
California EPA (California Environmental Protection Agency). (2004) No Significant Risk Levels (NSRLs) for the
Proposition 65 carcinogens benzo[b]fluoranthene,benzo[j]fluoranthene, chrysene, dibenzo[a,h]pyrene,
dibenzo[a,i]pyrene, and 5-methylchrysene by the oral route. Office of Environmental Health Hazard Assessment,
Reproductive and Cancer Hazard Assessment Section, Oakland, CA.

Carver, JH; Machado, ML; MacGregor, JA. (1985) Petroleum distillates suppress in vitro metabolic activation:
higher (S-9) required in the Salmonella/microsome mutagenicity assay. Environ Mutagen 7:369-379.

Carver, JH; Machado, ML; MacGregor, JA. (1986) Application of modified Salmonella/microsome prescreento
petroleum-derived complex mixtures and polynuclear aromatic hydrocarbons (PAH). Mutat Res 174:247-253.

Casale, GP; Higginbotham, S; Johansson, SL; et al. (1997) Inflammatory response of mouse skin exposed to the
very potent carcinogen dibenzo[a,l]pyrene:  a model for tumor promotion.  Fundam Appl Toxicol 36(l):71-78.

Casale, GP; Cheng, Z; Liu, J; et al. (2000) Profiles of cytokine mRNAs in the skin and lymph nodes of SENCAR
mice treated epicutaneously with dibenzo[a,l]pyrene or dimethylbenz[a]anthracene reveal a direct correlation
between carcinogen-induced contact hypersensitivity and epidermal hyperplasia. Mol Carcinog 27(2): 125-140.

Casto, BC. (1979) Polycyclic hydrocarbons and Syrian hamster embryo cells: cell transformation, enhancement of
viral transformation and analysis of DNA-damage. In: Jones, PW; Leber, P, eds. Polynuclear aromatic
hydrocarbons. Ann Arbor, MI: Ann Arbor Science Publishers, pp. 51-66.

Castro, DJ; Lohr, CV; Fischer, KA; et al. (2008) Lymphoma and lung cancer in offspring born to pregnant mice
dosed with dibenzo[a,l]pyrene: the importance of in utero vs.  lactational exposure. Toxicol Appl Pharmacol
233:454-458.

Cavalieri, EL; Rogan, EG. (1992) The approach to understanding aromatic hydrocarbon carcinogenesis. The central
role of radical cations in metabolic activation.  Pharmacol Ther 55:183-199.

Cavalieri, EL; Rogan, EG. (1995) Central role of radical cations in metabolic activation of polycyclic aromatic
hydrocarbons. Xenobiotica 25:677-688.

Cavalieri, EL; Mailander, P; Pelfrene, A. (1977) Carcinogenic activity of anthanthrene on mouse skin. Z
Krebsforsch Klin Onkol Cancer Res Clin Oncol 89:113-118.

Cavalieri, E; Rogan, E; Thilly, WG. (1981a) Carcinogenicity, mutagenicity and binding studies of the environmental
contaminant cyclopenteno(c,d]pyrene and some of its derivatives. In: Cook, M; Dennis, AJ, eds. Chemical analysis
and biological fate: polynuclear aromatic hydrocarbons. Columbus, OH: Battelle Press, pp. 487-499.

Cavalieri, E; Rogan, E; Toth, B; et al. (1981b) Carcinogenicity of the environmental pollutants cyclopenteno-
[cd]pyrene and cyclopentano[cd]pyrene in mouse skin.  Carcinogenesis 2:277-281.

Cavalieri, E; Munhall, A; Rogan, E; et al. (1983) Syncarcinogenic effect of the environmental pollutants
cyclopenteno[cd]pyrene and benzo[a]pyrene in mouse skin. Carcinogenesis 4:393-397.

Cavalieri, EL; Rogan, EG; Cremonesi, P; et al. (1988a) Radical cations as precursors in the metabolic formation of
quinones frombenzo[a]pyrene and 6-fluorobenzo[a]pyrene. Fluoro substitution as a probe for one-electron oxidation
in aromatic substrates.  Biochem Pharmacol 37(11):2173-2182.

Cavalieri, E; Rogan, E;  Sinha, D. (1988b) Carcinogenicity of aromatic hydrocarbons directly applied to rat
mammary gland. Cancer Res Clin Oncol 114:3-9.

Cavalieri, EL; Rogan, EG; Higginbotham, S; et al. (1989) Tumor-initiating activity in mouse skin and
Carcinogenicity in rat mammary gland of dibenzo[a]pyrenes: the very potent environmental carcinogen
dibenzo[a,l]pyrene. J Cancer Res Clin Oncol 115:67-72.

Cavalieri, EL; Higginbotham, S; RamaKrishna, NV; et al. (1991) Comparative dose-response tumorigenicity studies
of dibenzo[a,l]pyrene versus 7,12-dimethylbenz[a]anthracene, benzo[a]pyrene and two dibenzo[a,l]pyrene
dihydrodiols in mouse skin and rat mammary gland. Carcinogenesis  12:1939-1944.
                                              221            DRAFT - DO NOT CITE OR QUOTE

-------
Cavalieri, EL; Rogan, EG; Ramakrishna, NVS; et al. (1993) Mechanisms of benzo[a]pyrene and
7,12-diemthylbenz[a]antrhacene activation: qualitative aspects of the stable and depurination DNA adducts obtained
from radical cations and diol epoxides.  In: Polycyclic aromatic hydrocarbons: synthesis, properties, analytical
measurements, occurrence and biological effects. Bordeaux, France: Gordon and Breach Science Publishers, pp.
725-732.

Cavalieri, EL; Rogan, EG; Li, KM; et al. (2005) Identification and quantification of the depurinating DNA adducts
formed in mouse skin treated with dibenzo[a,l]pyrene (DB[a,l]P) or its metabolites and in rat mammary gland
treated with DB[a,l]P.  Chem Res Toxicol 18(6):976-983.

CCME (Canadian Council of the Ministers of the Environment). (2003) Canadian soil quality guidelines for
potentially carcinogenic and higher molecular weight poly cyclic aromatic hydrocarbons (environmental and human
health aspects).  Scientific supporting document. UMA Group, Ltd., Victoria, British Columbia.

Chakravarti, D; Felling, JC; Cavalieri, EL; et al. (1995) Relating aromatic hydrocarbon-induced DNA adducts and
c-H-ras mutations in mouse skin papillomas: the role of apurinic sites. Proc Natl Acad Sci USA 92(22): 10422-
10426.

Chakravarti, D; Mailander, PC; Cavalieri, EL; et al. (2000) Evidence that error-prone DNA repair converts
dibenzo[a,l]pyrene-induced depurinating lesions into mutations: formation, clonal proliferation and regression of
initiated cells carrying H-ras oncogene mutations in early preneoplasia. Mutat Res 456(1-2): 17-32.

Chakravarti, D; Venugopal, D; Mailander, PC; et al. (2008) The role of polycyclic aromatic hydrocarbon-DNA
adducts in inducing mutations in mouse skin. Mutat Res 649(1-2):161-178.

Chang, RL; Levin, W; Wood, AW; et al. (1981) Tumorigenicity of the diastereomeric bay-region benzo(e)pyrene
9,10-diol-ll,12-epoxides in newborn mice. Cancer Res 41:915-918.

Chang, HF; Huffer, DM; Chiarelli, MP; et al. (2002) Characterization of DNA adducts derived from syn-
benzo[ghi]fluoranthene-3,4-dihydrodiol-5,5a-epoxide and comparative DNA binding studies with structurally-
related anti-diolepoxides of benzo[ghi]fluoranthene and benzo[c]phenanthrene. Chem Res Toxicol 15:198-208.

Chen, TT; Heidelberger, C. (1969) Quantitative studies on the morphological/malignant cell transformation of
mouse prostate cells by carcinogenic hydrocarbons in vitro. Int J Cancer 4:166-178.

Chen, S; Nguyen, N; Tamura, K; et al. (2003) The role of the Ah receptor and p38 inbenzo[a]pyrene-
7,8-dihydrodiol and benzo[a]pyrene-7,8-dihydrodiol-9,10-epoxide-induced apoptosis. J Biol Chem 278:19526-
19533.

Cherng, SH; Lin, P; Yang, JL;  et al. (2001) Benzo[g,h,i]perylene synergistically transactivates benzo[a]pyrene-
induced CYP1 Al gene expression by aryl hydrocarbon receptor pathway.  Toxicol Appl Pharmacol 170:63-68.

Chu, ML; Chen, CW. (1984) Evaluation and estimation of potential carcinogenic risks of polynuclear aromatic
hydrocarbons.  In: Polynuclear aromatic hydrocarbons in the workplace: proceedings of a symposium; Chemical
Congress of Pacific Basin Societies.

Clement Associates. (1988) Comparative potency approach for estimating the cancer risk associated with exposure
to mixtures of polycyclic aromatic hydrocarbons.  Fairfax, VA: ICF Clement Associates.

Clement Associates. (1990) Development of relative potency estimates for PAHs and hydrocarbon combustion
product fractions compared to benzo[a]pyrene and their use in carcinogenic risk assessments. Fairfax, VA: ICF
Clement Associates.

Cogliano, VJ; Baan, RA; Straif, K; et al. (2008) Use of mechanistic data in IARC evaluations.  Environ Mol
Mutagen 49(2): 100-109.

Collins JF; Brown, JP; Alexeeff, GV; et al. (1998) Potency equivalency factors for some polycyclic aromatic
hydrocarbons and polycyclic aromatic hydrocarbon derivatives.  Regul Toxicol Pharmacol 28:45-54.
                                              222            DRAFT - DO NOT CITE OR QUOTE

-------
Conney, AH; Chang, RL; Cui, XX; et al. (2001) Dose-dependent differences in the profile of mutations induced by
carcinogenic (R,S,S,R) bay- and fjord-region diol epoxides of polycyclic aromatic hydrocarbons.  Adv Exp Med
Biol 500:697-707.

Cooper, CS; Pal, K; Hewer, A; et al. (1982) The metabolism and activation of polycyclic aromatic hydrocarbons in
epithelial cell aggregates and fibroblasts prepared from rat mammary tissue.  Carcinogenesis 3:203-210.

Gulp,  SJ; Beland, FA. (1994) Comparison of DNA adduct formation in mice fed coal tar orbenzo[a]pyrene.
Carcinogenesis 15:247-252.

Gulp,  SJ; Gay lor, DW; Sheldon, WG; et al. (1996a) DNA adduct measurements in relation to tumor incidence
during the chronic feeding of coaltarorbenzo[a]pyrene to mice. Polycycl Aromat Compd 11:161-168.

Gulp,  SJ; Gay lor, DW; Sheldon, WG; et al. (1996b) Relationship between DNA adduct levels and tumor incidence
in mice fed coal tar or benzo[a]pyrene for two years. Proc Am Assoc Cancer Res 37:101.

Gulp,  SJ; Gay lor, DW; Sheldon, WG; et al. (1998) A comparison of the tumors induced by coal tar and
benzo[a]pyrene in a 2-year bioassay. Carcinogenesis 19:117-124.

Gulp,  SJ; Warbritton, AR; Smith, BA; et al. (2000) DNA adduct measurements, cell proliferation and tumor
mutation induction in relation to tumor formation in B6C3F1 mice fed coal tar or benzo[a]pyrene. Carcinogenesis
21(7):1433-1440.

Dasgupta, PS; Lahiri, T. (1992) Alteration of brain catecholamines during growth of benzo[a]pyrene induced murine
fibrosarcoma. Neoplasma 39:163-165.

Davis, C; Desai, D; Amin, S; et al. (1999) Comparison of the morphological transforming activities of fjord-region
PAHs with dibenzo[a,e]pyrene andbenzo[a]pyrene. Polycycl Aromat Compd 16:141-149.

Dean, BJ. (1981) Activity of 27 coded compounds in the RL1 chromosome assay. Prog Mutat Res 1:570-579.

De Flora, S; Zanacchi, P; Camoirano, A; et al. (1984) Genotoxic activity and potency of 135 compounds in the
Ames reversion test and in a bacterial DNA-repair test. Mutat Res 133:161-198.

DeMarini, DM; Landi, S; Tian, D; et al.  (2004) Lung tumor KRAS and TP53 mutations in nonsmokers reflect
exposure to PAH-rich coal combustion emissions. Cancer Res 61:6679-6681.

Denissenko, MF; Pao, A; Tang, M; et al. (1996) Preferential formation of benzo[a]pyrene adducts at lung cancer
mutational hotspots in P53. Science 274(5286):430^32.

DeSalvia, R; Meschini, R; Fiore, M; et al. (1988) Induction of sister-chromatid exchanges by procarcinogens in
metabolically competent Chinese hamster epithelial liver cells. Mutat Res 207:69-75.

Deutsch-Wenzel, RP; Brune, H; Grimmer, G; et al.  (1983) Experimental studies in rat lungs on the carcinogenicity
and dose- response relationships of eight frequently occurring environmental polycyclic aromatic hydrocarbons. J
Natl Cancer Inst 71:539-544.

Devanesan, PD; Cremonesi, P; Nunnally, JE; et al. (1990) Metabolism and mutagenicity of dibenzo[a,e]pyrene and
the very potent environmental carcinogen dibenzo[a,l]pyrene. Chem Res Toxicol 3:580-586.

DiGiovanni, J; Rymer, J; Slaga, TJ; et al. (1982) Anticarcinogenic  and cocarcinogenic effects of benzo[e]pyrene and
dibenz[u,c]anthracene on skin tumor initiation by polycyclic hydrocarbons. Carcinogenesis 3:371-375.

DiPaolo, JA; Donovan, JP; Nelson, RL.  (1969) Quantitative studies of in vitro transformation by chemical
carcinogens. J Natl Cancer Inst 42:867-874.

DiPaolo, JA; Takano, K; Popescu, NC. (1972) Quantitation of chemically induced neoplastic transformation of
BALB/3T3 cloned cell lines. Cancer Res 35:2686-2695.
                                             223            DRAFT - DO NOT CITE OR QUOTE

-------
DiPaolo, JA; Nelson, RL; Donovan, PJ; et al. (1973) Host-mediated in vivo-in vitro assay for chemical
carcinogenesis.  ArchPathol 95:380-385.

Dunkel, VC; Pienta, RJ; Sivak, A; et al. (1981) Comparative neoplastic transformation responses of Balb 3T3 cells,
Syrian hamster embryo cells, and Rauscher mm-me leukemia virus-infected Fischer 344 rat embryo cells to
chemical carcinogens. J Natl Cancer Inst 67:1303-1315.

Dunkel, VC; Zeiger, E; Brusick, D; et al. (1984) Reproducibility of microbial mutagenicity assays: tests with
Salmonella typhimurium and Escherichia coli using a standardized protocol.  Environ Mutagen 6:1-251.

Durant, JL; Lafleur, AL; Busby, WF, Jr.; et al. (1999) Mutagenicity of C24H14 PAH in human cells expressing
CYP1A1.  Mutat Res 446:1-14.

Eisenstadt, E; Gold, A. (1978) Cyclopenta[c,d]pyrene: a highly mutagenic polycyclic aromatic hydrocarbon. Proc
Natl Acad Sci USA 75:1667-1669.

El-Bayoumy, K; Hecht, SS; Hoffmann, D. (1982) Comparative tumor initiating activity on mouse skin of
6-nitrobenzo[a]pyrene, 6-nitrochrysene, 3-nitroperylene, 1-nitropyrene and their parent hydrocarbons. Cancer Lett
16:333-337.

Emura, M; Richter-Reichhelm, HB; Schneider, P; et al. (1980) Sensitivity of Syrian golden hamster fetal lung cells
to benzo[a]pyrene and other polycyclic hydrocarbons in vitro. Toxicology 17:149-155.

Evans, CH; DiPaolo, JA. (1975) Neoplastic transformation of guinea pig fetal cells in culture induced by chemical
carcinogens. Cancer Res 35:1035-1044.

Evans, EL; Mitchell, AD. (1981) Effects of 20 coded chemicals  on sister chromatid exchange frequencies in
cultured Chinese hamster cells. Prog Mutat Res 1:538-550.

Fahmy, M; Fahmy, OG. (1980) Altered control of gene activity in the somaby carcinogens.  Mutat Res 72:165-172.

Falk, HL; Kotin, P; Thompson, S.  (1964) Inhibition of carcinogenesis. The effect of hydrocarbons and related
compounds.  Arch Environ Health 13:169-179.

Flesher, JW; Harvey, RG; Sydnor, KL. (1976) Oncogenicity of K-region epoxides of benzo[a]pyrene and
7,12-dimethylbenz[a]anthracene. Int J Cancer 18:351-353.

Florin, I; Rutberg, L; Curvall, M; et al. (1980) Screening of tobacco smoke constituents for mutagenicity using the
Ames'test. Toxicology 18:219-232.

Flowers, L; Ohnishi, T; Penning, TM. (1997) DNA strand scission by polycyclic aromatic hydrocarbon o-quinones:
role of reactive oxygen species, Cu(II)/Cu(I) redox cycling and o-semiquinone anion radicals.  Biochemistry
36:8640-8648.

Flowers-Geary, L; Harvey, RG; Penning, TM. (1993) Cytotoxicity of polycyclic aromatic hydrocarbon o-quinones
in rat and human hepatoma cells. Chem Res Toxicol 6(3):252-260.

Flowers-Geary, L; Bleczinki, W; Harvey, RG; et al. (1996) Cytotoxicity and mutagenicity of polycyclic aromatic
hydrocarbon ortho-quinones produced by dihydrodiol dehydrogenase. Chem Biol Interact 99(l-3):55-72.

Frolich, A; Wurgler, FE. (1990) Drosophila wing-spot test: improved detectability of genotoxicity of polycyclic
aromatic hydrocarbons. Mutat Res 234:71-80.

Gaylor, DW; Moolgavkar, S; Krewski, D;  et al. (1998) Recent bioassay results on coal tars and benzo[a]pyrene:
implications for risk assessment. Regul Toxicol Pharmacol 28:178-179.
                                             224            DRAFT - DO NOT CITE OR QUOTE

-------
Geacintov, NE; Cosman, M; Hingerty, BE; et al. (1997) NMR solution structures of stereoisomeric covalent
polycyclic aromatic carcinogen - DNA adducts: principles, patterns, and diversity. Chem Res Toxicol 10(2): 111-
146.

Gehly, EB; Landolph, JR; Heidelberger, C; et al. (1982) Induction of cytotoxicity, mutation, cytogenetic changes
and neoplastic transformation by benzo[a]pyrene and derivatives in C3H/10T 1/2 clone 8 mouse fibroblasts. Cancer
Res 42:1866-1875.

Gibson, TL; Smart, VB; Smith, LL. (1978) Non-enzymic activation of polycyclic aromatic hydrocarbons as
mutagens. Mutat Res 49:153-161.

Gill, HS; Kole, PL; Wiley, JC; et al. (1994) Synthesis and tumor-initiating activity in mouse skin of
dibenzo[a,l]pyrene syn- and anti-fjord-region diolepoxides. Carcinogenesis 15:2455-2460.

Gold, A; Eisenstadt, E. (1980) Metabolic activation of cyclopenta[cd]pyrene to 3,4-epoxycyclopenta[cd]pyrene by
rat liver microsomes.  Cancer Res 40:3940-3944.

Goldschmidt, BM; Katz, C; VanDuuren, BL. (1973) The cocarcinogenic activity of noncarcinogenic aromatic
hydrocarbons. Proc Am Assoc Cancer Res 14:84.

Goldstein, LS; Safe, S; Weyand, EH. (1994) Carcinogenicity of coal tars: a multidisciplinary approach. Polycycl
Aromat Compd 7:161-174.

Grant, G; Roe, FJC. (1963) The effect of phenanthrene on tumour induction by 3,4-benzopyrene administered to
newly born mice.  Br J Cancer 17:261-265.

Greb,W; Strobel, R; Rohrborn, G. (1980) Transformation of BHK 21/CL 13 cells by various polycyclic aromatic
hydrocarbons using the method of Styles.  Toxicol Lett 7:143-148.

Grimmer, G; Brune, H; Deutsch-Wenzel, R; et al. (1984) Contribution of polycyclic aromatic hydrocarbons to the
carcinogenic impact of gasoline engine exhaust condensate evaluated by implantation into the lungs of rats. J Natl
Cancer Inst 72:733-739.

Grimmer, G; Brune, H; Deutsch-Wenzel, R; et al. (1987a) Contribution of polycyclic aromatic hydrocarbons and
nitro-derivatives to the carcinogenic impact of diesel engine exhaust condensate evaluated by implantation into the
lungs of rats.  Cancer Lett 37:173-180.

Grimmer, G; Brune, H; Deutsch-Wenzel, R; et al. (1987b) Contribution of polycyclic aromatic hydrocarbons and
polar polycyclic aromatic compounds to the carcinogenic impact of flue gas condensate from coal-fired residential
furnaces evaluated by implantation into the rat lung. J Natl Cancer Inst 78:935-942.

Grimmer, G; Brune, H; Dettbarn, G; et al. (1988) Contribution of polycyclic aromatic compounds to the
carcinogenicity of sidestream smoke of cigarettes evaluated by implantation into the lungs of rats. Cancer Lett
43:173-177.

Grover, PL; Sims, P. (1968) Enzyme-catalysed reactions of polycyclic hydrocarbons with deoxyribonucleic acid and
protein in vitro. BiochemJ 110:159-160.

Guthrie, J; Robertson, IG; Zeiger, E; et al. (1982) Selective activation of some dihydrodiols of several polycyclic
aromatic hydrocarbons to mutagenic products by prostaglandin synthetase. Cancer Res 42:1620-1623.

Habs, M; Schmahl, D; Misfeld, J.  (1980) Local  carcinogenicity of some environmentally relevant polycyclic
aromatic hydrocarbons after lifelong topical application to mouse skin. Arch Geschwulstforsch 50:266-274.

Habs, M; Jahn, SA; Schmahl, D. (1984) Carcinogenic activity  of condensate from coloquint seeds (Citrullus
colocynthis) after chronic epicutaneous administration to mice. J Cancer Res Clin Oncol 108:154-156.

Harvey, RG. (1996) Mechanisms of carcinogenesis of polycyclic aromatic hydrocarbons.  Polycycl Aromat Compd
9:1-23.
                                              225            DRAFT - DO NOT CITE OR QUOTE

-------
Hass, BS; Brooks, EE; Schumann, KE; et al. (1981) Synergistic, additive, and antagonistic mutagenic responses to
binary mixtures of benzo[a]pyrene and benzo[e]pyrene as detected by strains TA98 and TA100 in the
Salmonella/microsome assay. Environ Mutagen 3:159-166.

Hass, BS; McKeown, CK; Sardella, DJ; et al. (1982) Cell-mediated mutagenicity in Chinese hamster V79 cells of
dibenzopyrenes and their bay-region fluorene-substituted derivatives. Cancer Res 42:1646-1649.

He, SL; Baker, R. (1991) Micronuclei in mouse skin cells following in vivo exposure to benzo[a]pyrene,
7,12-dimethylbenz[a]anthracene, chrysene, pyrene andurethane.  Environ Mol Mutagen 17:163-168.

Hecht,  SS; Bondinell, WE; Hoffman, D. (1974) Chrysene and methylchrysenes: presence on tobacco smoke and
carcinogenicity. JNatl Cancer Inst 53:1121-1133.

Hermann, M. (1981) Synergistic effects of individual polycyclic aromatic hydrocarbons on the mutagenicity of their
mixtures. Mutat Res 90:399^109.

Higginbotham, S; RamaKrishna, NV; Johansson, SL; et al. (1993) Tumor-initiating activity and carcinogenicity of
dibenzo[a,l]pyrene versus 7,12-dimethylbenz[a]anthracene andbenzo[a]pyrene at low doses in mouse skin.
Carcinogenesis 14:875-878.

Hoffmann, D; Wynder, EL. (1966) [Contribution on the carcinogenic effect of dibenzopyrenes].  Z Krebsforsch
68:137-149.

Hoffmann, D; Rathkamp, G; Nesnow, S; et al. (1972) Fluoranthenes: quantitative determination in cigarette smoke,
formation by pyrolysis and tumor initiating activity. J Natl Cancer Inst 49:1165-1175.

Homburger, F; Hsueh, SS; Kerr, CS et al. (1972) Inherited susceptibility of inbred strains of Syrian hamsters to
induction of subcutaneous sarcomas and mammary  and gastrointestinal carcinomas by subcutaneous and gastric
administration of polynuclear hydrocarbons. Cancer Res 32:360-366.

Horton, AW; Christian, GM. (1974) Cocarcinogenic versus incomplete carcinogenic activity among aromatic
hydrocarbons: contrast between chrysene andbenzo[b]-triphenylene. JNatl Cancer Inst 53:1017-1020.

Huberman,  E. (1975) Mammalian cell transformation and cell-mediated mutagenesis by carcinogenic polycyclic
hydrocarbons.  Mutat Res 29:285-291.

Huberman,  E; Sachs, L. (1974) Cell-mediated mutagenesis of mammalian cells with chemical carcinogens.  Int J
Cancer 13:326-333.

Huberman,  E; Sachs, L. (1976) Mutability of different genetic loci in mammalian cells by metabolically activated
carcinogenic polycyclic hydrocarbons. Proc Natl Acad Sci USA 73:188-192.

Huggins, C; Yang, NC. (1962) Induction and extinction of mammary cancer. A striking effect of hydrocarbons
permits analysis of mechanisms of causes and cure of breast cancer.  Science 137:257-262.

Hughes, NC; Phillips, DH. (1990) Covalent binding of dibenzpyrenes and benzo[a]pyrene to DNA: evidence for
Synergistic and inhibitory interactions when applied in combination to mouse skin.  Carcinogenesis 11:1611-1620.

Hughes, NC; Phillips, DH. (1991) Dependence on dose of initial and persistent levels of benzo[a]pyrene and
dibenzo[a,e]pyrene DNA adducts in mouse tissues.  Proc Am Assoc Cancer Res 32:98.

Hughes, NC; Phillips, DH. (1993) 32P-postlabeling analysis of the covalent binding of benzo[ghi]perylene to DNA
in vivo and in vitro. Carcinogenesis  14:127-133.

I ARC (International Agency for Research on Cancer). (1973) Certain polycyclic aromatic hydrocarbons and
heterocyclic compounds.  In: IARC monographs on the evaluation of carcinogenic risk of chemicals to humans. Vol.
3. Lyon, France.
                                              226            DRAFT - DO NOT CITE OR QUOTE

-------
IARC (International Agency for Research on Cancer). (1983) Polynuclear aromatic compounds. Part 1. Chemical,
environmental and experimental data. In: IARC monographs on the evaluation of carcinogenic risk of chemicals to
humans. Vol. 32. Lyon, France.

IARC (International Agency for Research on Cancer). (1984a) Polynuclear aromatic compounds. Part 2. Carbon
black, mineral oils (lubricant base oils and derived products) and some nitroarenes. In: IARC monographs on the
evaluation of carcinogenic risk of chemicals to humans. Lyon, France, pp. 87-168.

IARC (International Agency for Research on Cancer). (1984b) Polynuclear aromatic compounds. Part 3. Industrial
exposures in aluminum production, coal gasification, coke production, and iron and steel founding. In: IARC
monographs on the evaluation of carcinogenic risk of chemicals to humans. Lyon, France, pp. 37-111.

IARC (International Agency for Research on Cancer). (1985) Polynuclear aromatic compounds. Part 4. Bitumens,
coal-tars and derived products, shale-oils and soots.  In: IARC monographs on the evaluation of carcinogenic risk of
chemicals to humans. Lyon, France, pp. 65-159.

IARC (International Agency for Research on Cancer). (1986) Some halogenated hydrocarbons and pesticide
exposures.  In: IARC monographs on the evaluation of carcinogenic risk of chemicals to humans. Lyon, France.

IARC (International Agency for Research on Cancer). (1987) Overall evaluation of carcinogenicity: an updating of
IARC Monographs volumes 1 to 42.  In: IARC monographs on the evaluation of carcinogenic risks to humans.
Suppl. 7.  Lyon, France.

IARC (International Agency for Research on Cancer). (1989) Occupational exposures  in petroleum refining; crude
oil and major petroleum fuels. In: IARC monographs on the evaluation of carcinogenic risk of chemicals to
humans. Vol. 45. Lyon, France, pp. 239-270.

Ichinotsubo, D; Mower, HF; Setliff, J; et al. (1977) The use of rec-bacteria for testing of carcinogenic substances.
Mutat Res 46:53-56.

Jeffy, BD; Chen, EJ; Gudas, JM; et al. (2000) Disruption of cell cycle kinetics by benzo[a]pyrene: inverse
expression patterns of BRCA-1 and p53 in MCF-7 cells arrested in S and G2. Neoplasia 2:460-470.

Jeffy, BD; Chirnomas, RB; Chen, EJ; et al. (2002) Activation of the aromatic hydrocarbon receptor pathway is not
sufficient for transcriptional repression of BRCA-1: requirements for metabolism of benzo [a]pyrene to
7r,8t-dihydroxy-9t,10-epoxy-7,8,9,10-tetrahydrobenzo[a]pyrene.  Cancer Res 62:113-121.

Jerina, DM; Lehr, RE. (1977) The bay-region theory: a quantum mechanical approach to aromatic hydrocarbon-
induced carcinogenicity.  In: Microsomes and drug oxidations. Oxford: Pergamon Press, pp. 709-720.

Jerina, DM; Yagi, H; Lehr, RE; et al. (1978) The bay-region theory of carcinogenesis by polycyclic aromatic
hydrocarbons. In:  Polycyclic hydrocarbons and cancer: environment, chemistry and metabolism. Boca Raton,
Florida: CRC Press, pp. 173-188.

Jerina, DM; Chadha, A; Cheh, AM; et al. (1991) Covalent bonding of bay-region diol epoxides to nucleic acids.
Adv Exp Med Biol 283:533-553.

Jiang, H;  Shen, YM; Quinn, AM; et al.  (2005) Competing roles of cytochrome P450 1A1/1B1 and aldo-keto
reductase 1A1 in the metabolic activation of (±)-7,8-dihydroxy-7,8-dihydro-benzo[a]pyrene inhuman
bronchoalveolar cell extracts.  Chem Res Toxicol 18:365-374.

Jiang, H;  Gelhaus, SL; Mangal, D; et al. (2007) Metabolism of benzo [a]pyrene in human bronchoalveolar H358
cells using liquid chromatography-mass spectrometry. Chem Res Toxicol 20:1331-1341.

Johnsen, NM; Schwarze,  PE; Nyholm,  SH; et al. (1997) Genotoxic effects of cyclopenta-fused polycyclic aromatic
hydrocarbons in different types of isolated rat lung cells.  Carcinogenesis 18:193-199.

Johnsen, NM; Brunborg,  G; Haug, K; et al. (1998) Metabolism and activation of cyclopenta polycyclic aromatic
hydrocarbons in isolated human lymphocytes, HL-60 cells and exposed rats.  Chem Biol Interact 114:77-95.
                                             227            DRAFT - DO NOT CITE OR QUOTE

-------
Jotz, MM; Mitchell, AD. (1981) Effects of 20 coded chemicals on the forward mutation frequency at the thymidine
kinase locus in L5178Y mouse lymphoma cells. Prog Mutat Res 1:580-593.

Kaden, DA; Hites, RA; Thilly, WG. (1979) Mutagenicity of soot and associated poly cyclic aromatic hydrocarbons
to Salmonella typhimurium.  Cancer Res 39:4152-4159.

Kakunaga, T.  (1973) A quantitative system for assay of morphological/malignant cell transformation by chemical
carcinogens using a clone derived from BALB-3T3. Int J Cancer 12(2):463-473.

Katz, M; Heddle, JA; Salamone, MF. (1981) Mutagenic activity of polycyclic aromatic hydrocarbons and other
environmental pollutants.  Columbus, OH: Battelle Press, pp. 519-528.

Kligerman, AD; Moore, MM; Erexson, GL; et al. (1986) Genotoxicity studies of benz[i]aceanthrylene.  Cancer Lett
31:123-131.

Kligerman, AD; Nelson, GB; Ross, JA; et al. (2002) Effect of the route of administration on the induction of
cytogenetic damage and DNA adducts  in peripheral blood lymphocytes of rats and mice by polycyclic aromatic
hydrocarbons.  Polycycl Aromat Compd 22:814-851.

Kochhar, TS. (1982) Effects of polycyclic hydrocarbons on the induction of chromosomal aberrations in absence of
an exogenous metabolic activation system in cultured hamster cells. Experientia 38:845-846.

Koganti, A; Singh, R; Rozett, K; et al. (2000) 7H-benzo[c]fluorene: a major DNA adduct-forming component of
coal tar. Carcinogenesis 21:1601-1609.

Kondraganti, SR; Fernandez-Salguerro, P; Gonzalez, FJ; et al. (2003) Polycyclic aromatic hydrocarbon-inducible
DNA adducts: evidence by 32P-postlabeling and use of knockout mice for AH receptor-independent mechanisms of
metabolic activation in vivo. Int J Cancer 103:5-11.

Krahn, DF; Heidelberger, C. (1977) Liver homogenate-mediated mutagenesis in Chinese hamster V79 cells by
polycyclic aromatic hydrocarbons and aflatoxins. Mutat Res 46:27-44.

Krewski, D; Thorslund,  T; Withey, J. (1989) Carcinogenic risk assessment of complex mixtures.  Toxicol Ind Health
5:851-867.

Kroese, ED; Muller, JJA; Mohn, GR; et al. (2001) Tumorigenic effects in Wistar rats orally administered
benzo[a]pyrene for two years (gavage studies). Implications for human cancer risks associated with oral exposure to
polycyclic aromatic hydrocarbons. National Institute of Public Health and the Environment, Bilthoven, Netherlands.

Krolewski, B; Nagasawa, H; Little, JB. (1986) Effect of aliphatic amides on oncogenic transformation, sister
chromatid exchanges, and mutations induced by cyclopenta[cd]-pyrene and benzo[a]pyrene.  Carcinogenesis
7:1647-1650.

Laaksonen, AM; Mantyjarvi, RA; Hanninen, OO. (1983) Fibroblast cultures of nude mouse skin as targets for
transformation by chemical carcinogens. Med Biol 61:59-64.

Lacassagne, A; Buu-Hoi, NP; Zajdela,  F; et al. (1968) The true dibenzo[a,l]pyrene, a new, potent carcinogen.
Naturwissenschaften 55:43.

Lafleur, AL; Longwell, JP; Marr, JA; et al. (1993) Bacterial and human cell mutagenicity study of some C18H10
cyclopenta-fused polycyclic aromatic hydrocarbons associated with fossil fuels combustion. Environ Health
Perspect 101:146-153.

Lake, RS; Kropko, ML; Pezzutti, MR;  et al. (1978)  Chemical induction of unscheduled DNA synthesis in human
skin epithelial cell cultures.  Water Res 38:2091-2098.

Langenbach, R; Hix, C;  Oglesby, L; et  al. (1983) Cell-mediated mutagenesis of Chinese hamster V79 cells and
Salmonella typhimurium. Ann NY Acad Sci 407:258-266.
                                             228            DRAFT - DO NOT CITE OR QUOTE

-------
Larsen, JC; Larsen, PB. (1998) Chemical carcinogens. In: Air pollution and health. Cambridge, UK: The Royal
Society of Chemistry, pp. 33-56.

Lavik, PS; Moore, PR; Rusch, HP; et al. (1942) Some additive effects of carcinogenic hydrocarbons.  Cancer Res
2:189-192.

LaVoie, EJ; Bedenko, V; Hirota, N; et al. (1979) A comparison of the mutagenicity, tumor-initiating activity and
complete carcinogenicity of polynuclear aromatic hydrocarbons.  In: Jones, PW; Leber, P, eds. Polynuclear aromatic
hydrocarbons. Ann Arbor, MI: Ann Arbor Science Publishers, pp. 705-721.

LaVoie, EJ; Tulley, L; Bedenko, V; et al. (1980) Mutagenicity, tumor initiating activity, and metabolism of tricyclic
polynuclear aromatic hydrocarbons.  In: Bjorseth, A; Dennis, AJ, eds. Polynuclear aromatic hydrocarbons:
chemistry and biological effects. Columbus, OH: Battelle Press, pp. 1041-1057.

LaVoie, EJ; Tulley, L; Bedenko, V; et al. (1981) Mutagenicity, tumor-initiating activity and metabolism of
methylphenanthrenes.  Cancer Res 41:3441-3447.

LaVoie, EJ; Amin, S; Hecht, SS; et al. (1982) Tumour initiating activity of dihydrodiols of benzo[b]fluoranthene,
benzo[j]fluoranthene, andbenzo[k]fluoranthene.  Carcinogenesis 3:49-52.

LaVoie, EJ; Coleman, DT; Tonne, RL; et al. (1983) Mutagenicity, tumor initiating activity and metabolism of
methylated anthracenes. In: Cooke, M; Dennis, AJ, eds. Proceedings of the seventh international symposium.
Columbus, OH: Battelle Press, pp. 785-798.

LaVoie, EJ; Coleman, DT; Rice, JE; et al. (1985) Tumor-initiating activity, mutagenicity, and metabolism of
methylated anthracenes. Carcinogenesis 6:1483-1488.

LaVoie, EJ; Braley, J; Rice, JE; et al. (1987) Tumorigenic activity of non-alternant polynuclear aromatic
hydrocarbons in newborn mice. Cancer Lett 34:15-20.

LaVoie, EJ; Cai, ZW; Meschter, CL; et al. (1994) Tumorigenic activity of fluoranthene, 2-methylfluoranthene and
3-methylfluoranthene in newborn CD-I mice.  Carcinogenesis 15:2131-2135.

Li, CS; Lin, RH. (1996) Evaluation of low-dosage environmental mutagens with a long-term, cultured epithelial cell
line. Bull Environ Contam Toxicol 56:919-925.

Li, KM; Todorovic, R; Rogan, EG; et al. (1995) Identification and quantitation of dibenzo[a,l]pyrene~DNA adducts
formed by rat liver microsomes in vitro: preponderance of depurinating adducts. Biochemistry 34(25):8043-8049.

Li, D; Wang, M; Firozi, PF; et al. (2002) Characterization of a major aromatic DNA adduct detected in human
breast tissues. Environ Mol Mutagen 39:193-200.

Lubet, RA; Kiss, E; Gallagher, MM; et al. (1983) Induction of neoplastic transformation and DNA single-strand
breaks in C3H/10T1/2 clone 8 cells by polycyclic hydrocarbons and alkylating agents.  J Natl Cancer Inst 71:991-
997.

Machala, M; Vondracek, J; Blaha, L; et al. (2001) Aryl hydrocarbon receptor-mediated activity of mutagenic
polycyclic aromatic hydrocarbons determined using in vitro reporter gene assay. Mutat Res 497:49-62.

MacLeod, MC; Cohen, GM; Selkirk, JK.  (1979) Metabolism and macromolecular binding of the carcinogen
benzo[a]pyrene and its relatively inert isomer benzo(e)pyrene by hamster embryo cells. Cancer Res 39:3463-3470.

Malcolm, HM; Dobson, S. (1994) The calculation of an environmental assessment level (EAL) for atmospheric
PAHs using relative potencies.  Department of the Environment, London, England; Report No.
DoE/HMIP/RR/94/041.

Mamber, SW; Bryson, V; Katz, SE. (1983) The Esherichia coli WP2/WP100 rec assay for detection of potential
chemical carcinogens.  Mutat  Res 119:135-144.
                                             229            DRAFT - DO NOT CITE OR QUOTE

-------
Mane, SS; Purnell, DM; Hsu, 1C. (1990) Genotoxic effects of five polycyclic aromatic hydrocarbons in human and
rat mammary epithelial cells. Environ Mol Mutagen 15:78-82.

Marshall, CJ; Vousden, KH; Phillips, DH. (1984) Activation of c-Ha-ras-1 proto-oncogene by in vitro modification
with a chemical carcinogen, benzo[a]pyrene diol-epoxide. Nature 310(5978):586-589.

Martin, CN; McDermid, AC. (1981) Testing of 42 coded compounds for their ability to induce unscheduled DNA
repair synthesis in HeLa cells.  Prog Mutat Res 1:533-537.

Martin, CN; McDermid, AC; Garner, RC. (1978) Testing of known carcinogens and noncarcinogens for their ability
to induce unscheduled DNA synthesis in HeLa cells. Cancer Res 38:2621-2627.

Mass, MJ; Jeffers, AJ; Ross, JA; et al. (1993) Ki-ras oncogene mutations in tumors and DNA adducts formed by
benz[j]aceanthrylene and benzo[a]pyrene in the lungs of strain A/J mice. Mol Carcinog 8:186-192.

Masuda, Y; Kagawa, R. (1972) A novel synthesis and carcinogenicity of dibenzo[a,l]pyrene. Chem Pharm Bull
20:2736-2737.

Matsuoka, A; Hayashi, M; Ishidate, MJ. (1979) Chromosomal aberration tests on 29 chemicals combined with S9
mix in vitro. Mutat Res 66:277-290.

Matthews, EJ; Kruhlak, NL; Cimino, MC; et al. (2006a) An analysis of genetic toxicity, reproductive and
developmental toxicity, and carcinogenicity data: I. Identification of carcinogens using surrogate endpoints. Regul
Toxicol Pharmacol 44:83-96.

Matthews, EJ; Kruhlak, NL; Cimino, MC; et al. (2006b) An analysis of genetic toxicity, reproductive and
developmental toxicity, and carcinogenicity data: II. Identification of genotoxicants, reprotoxicants, and carcinogens
using in silico methods.  Regul Toxicol Pharmacol 44:97-110.

McCann, J; Choi, E; Yamasaki, E; et al. (1975) Detection of carcinogens as mutagens in the Salmonella/microsome
test: assay of 300 chemicals. Proc Natl Acad Sci USA 72:5135-5139.

McCarroll, NE; Piper, CE; Keech, BH. (1981) An E coli microsuspension assay for the detection of DNA damage
induced by direct-acting agents and promutagens.  Environ Mutagen 3:429-444.

McClure, PR. (1996) Evaluation of a component-based relative potency approach to cancer risk  assessment for
exposure  to PAHs.  Toxicologist 30(1, Part 2):8.

McCoull, KD; Rindgen, D; Blair, IA; et al. (1999) Synthesis and characterization of polycyclic aromatic
hydrocarbon o-quinone depurinating N7-guanine adducts. Chem Res Toxicol 12:237-246.

Meek, ME; Chan, PKL; Bartlett, S. (1994) Polycyclic aromatic hydrocarbons: evaluation of risks to health from
environmental exposures in Canada. Environ Carcinog Ecotoxicol Rev C 12:443-452.

Melendez-Colon, VJ;  Luch, A; Seidel, A; et al. (2000) Formation of stable DNA adducts  and apurinic sites upon
metabolic activation of bay and fjord region polycyclic aromatic hydrocarbons in human cell cultures. Chem Res
Toxicol 13:10-17.

Mersch-Sundermann,  V; Mochayedi, S; Kevekordes, S.  (1992) Genotoxicity of polycyclic aromatic hydrocarbons in
Escherichia coli PQ37.  Mutat Res 278:1-9.

Miller, KP; Ramos, KS. (2001) Impact of cellular metabolism on the biological effects of benzo[a]pyrene and
related hydrocarbons.  Drug Metab Rev 33:1-35.

Milo, GE; Blakeslee, J; Yohn, DS; et al. (1978) Biochemical activation of aryl hydrocarbon hydroxylase activity,
cellular distribution of polynuclear hydrocarbon metabolites, and DNA damage by polynuclear hydrocarbon
products in human cells in vitro.  Cancer Res 38:1638-1644.
                                             230            DRAFT - DO NOT CITE OR QUOTE

-------
Mishra, NK; Wilson, CM; Pant, KJ; et al. (1978) Simultaneous determination of cellular mutagenesis and
transformation by chemical carcinogens in Fischer rat embryo cells.  J Toxicol Environ Health 4:79-91.

Mohapatra, N; MacNair, P; Bryant, BJ; et al. (1987) Morphological transforming activity and metabolism of
cyclopenta-fused isomers of benz[a]anthracene in mammalian cells.  Mutat Res 188:323-334.

Muller, P; Leece, B; Raha, D. (1997) Scientific criteria document for multimedia standards development. Polycyclic
aromatic hydrocarbons (PAHs). Part 1: Hazard identification and dose-response assessment. Ontario Ministry of the
Environment, Standards Development Branch.

Murison, GL. (1988) Induction of sister-chromatid exchanges by direct and indirect agents in a human teratoma cell
line. Mutat Res 203:347-354.

Myhr, BC; Caspary, WJ. (1988) Evaluation of the L5178Y mouse lymphoma cell mutagenesis assay: intralaboratory
results for sixty-three coded chemicals tested at Litton Bionetics, Inc. Environ Mol Mutag 12 (Suppl 13): 103-194.

Nagabhushan, M; Hussong, J; Polverini, PJ; et al. (1990) Inhibition of hamster buccal pouch epithelial cell
replication during in vitro  exposure to polycyclic aromatic hydrocarbons. Proc Am Assoc Cancer Res 31:86.

Nakatsuru, Y; Wakabayashi, K; Fujii-Kuriyama, Y; et al. (2004) Dibenzo[a,l]pyrene-induced genotoxic and
carcinogenic responses are dramatically suppressed in aryl hydrocarbon receptor-deficient mice.  Int J Cancer
112:179-183.

Neal, J; Rigdon, RH. (1967) Gastric tumors in mice fed benzo[a]pyrene: a quantitative study. Tex Rep Biol Med
25:553-557.

Nesnow, S; Triplett, LL; Slaga, TJ. (1983) Mouse skin tumor initiation-promotion and complete carcinogenesis
bioassays: mechanisms and biological activities of emission samples. Environ Health Perspect 47:255-268.

Nesnow, S; Gold, A; Sangaiah, R; et al. (1984) Mouse skin tumor-initiating activity of benz[e]aceanthrylene and
benz[l]aceanthrylene in Sencar mice.  Cancer Lett 22:263-268.

Nesnow, S; Milo, G; Kurian, P; et al. (1990) Induction of anchorage-independent growth in human diploid
fibroblasts by the cyclopenta-polycyclic aromatic hydrocarbon, benz[l]aceanthrylene. Mutat Res 244:221-225.

Nesnow, S; Ross, J; Mohapatra, N; et al. (1991) Genotoxicity and  identification of the major DNA-adducts of
aceanthrylene. In: Cooke M LKMJe, eds. Polynuclear aromatic hydrocarbons: measurements, means, and
metabolism. Columbus, OH: Battelle Press, pp. 629-639.

Nesnow, S; Beck, S; Ball,  LM; et al. (1993a) Morphological transformation of C3H10T1/2CL8 cells by cyclopenta-
fused derivatives of benzo[a]pyrene and benzo[e]pyrene. Cancer Lett 74:25-30.

Nesnow, S; Ross, J; Nelson, G; et al. (1993b) Quantitative and temporal relationships between DNA adduct
formation in target and surrogate tissues: implications for biomonitoring. Environ Health Perspect 101 (Suppl 6)
3:37-42.

Nesnow, S; Ross, J; Beck, S; et al. (1994) Morphological transformation and DNA adduct formation by
dibenz[a,h]anthracene and its metabolites in C3H10T1/2CL8 cells. Carcinogenesis 15:2225-2231.

Nesnow, S; Ross, JA; Stoner, GD; et al. (1995) Mechanistic linkage between DNA adducts, mutations in oncogenes
and tumorigenesis of carcinogenic environmental polycyclic aromatic hydrocarbons in strain A/J mice.  Toxicology
105:403^13.

Nesnow, S; Ross, JA; Stoner, GD; et al. (1996) Tumorigenesis of carcinogenic environmental polycyclic aromatic
hydrocarbons in strain A/J mice: linkage to DNA adducts and mutations in oncogenes. Polycyclic Aromatic
Hydrocarbons 10:259-266.
                                              231            DRAFT - DO NOT CITE OR QUOTE

-------
Nesnow, S; Davis, C; Nelson, G; et al. (1997) Comparison of the morphological transforming activities of
dibenzo[a,l]pyrene and benzo[a]pyrene in C3H10T1/2CL8 cells and characterization of the dibenzo[a,l]pyrene-DNA
adducts. Carcinogenesis 18:1973-1978.

Nesnow, S; Mass, MJ; Ross, JA; et al. (1998a) Lung tumorigenic interactions in strain A/J mice of five
environmental polycyclic aromatic hydrocarbons. Environ Health Perspect 106(Suppl 6): 1337-1346.

Nesnow, S; Ross, JA; Mass, MJ; et al. (1998b) Mechanistic relationships between DNA adducts, oncogene
mutations, and lung tumorigenesis in strain A mice. Exp Lung Res 24:395-405.

Nikonova, TV. (1977) Transplacental effect of benz[a]pyrene and pyrene.  Bull Exp Biol Med 84:1025-1027.

Nisbet, ICT; LaGoy, PK. (1992) Toxic equivalency factors (TEFs) for polycyclic aromatic hydrocarbons (PAHs).
Regul Toxicol Pharmacol 16:290-300.

Norpoth, K; Kemena, A; Jacob, J; et al. (1984) The influence of 18 environmentally relevant polycyclic aromatic
hydrocarbons and Clophen A50, as liver monooxygenase inducers, on the mutagenic activity of benz[a]anthracene
in the Ames test.  Carcinogenesis 5:747-752.

NTP (National Toxicology Program). (2000) Toxicology and Carcinogenesis studies of naphthalene (CAS no. 91-
20-3) in F344/N rats (inhalation studies). National Toxicology Program. U.S. Department of Health and Human
Services, National Institutes of Health, Rockville, MD. Technical report series no. 500.

Nyholm, SH;  Alexander, J;  Lundanes, E; et al. (1996) Biotransformation of the cyclopenta-fused polycyclic
aromatic hydrocarbon benz[j]aceanthrylene in isolated rat liver cells: identification of nine new metabolites.
Carcinogenesis 17(5): 111 1-1120.

Okey, AB; Riddick, DS; Harper, PA. (1994) Molecular biology of the aromatic hydrocarbon (dioxin) receptor.
Trends Pharmacol Sci 15(7):226-232.

Oshiro, Y; Balwierz, PS; Soelter, SG; et al. (1992) Evaluation of mouse peripheral blood micronucleus assay.
Environ Mol Mutag 19(Suppl 20):47.

Pahlman, R; Pelkonen,  O. (1987) Mutagenicity studies of different polycyclic aromatic hydrocarbons: the
significance of enzymatic factors and molecular structure.  Carcinogenesis 8:773-778.

Paika, IJ; Beauchesne, MT; Randall, M; et al. (1981) In vivo SCE analysis of 20 coded compounds. Prog Mutat Res
1:672-681.

Park, JH; Gopishetty, S; Szewczuk, LM; et al. (2005) Formation of 8-oxo-7,8-dihydro-2'-deoxyguanosine (8-oxo-
dGuo) by PAH o-quinones: involvement of reactive oxygen species and copper(II)/copper(I) redox cycling. Chem
Res Toxicol 18(6): 1026-1037.

Park, JH; Troxel, AB; Harvey, RG; et al. (2006) Polycyclic aromatic hydrocarbon (PAH)  o-quinones produced by
the aldo-keto-reductases (AKRs) generate abasic sites, oxidized pyrimidines, and 8-oxo-dGuo via reactive oxygen
species. Chem Res Toxicol 19(5):719-728.

Park, JH; Gelhaus, S; Vedantam, S; et al. (2008) The pattern of p53 mutations caused by PAH o-quinones is driven
by 8-oxo-dGuo formation while the spectrum of mutations is determined by biological selection for dominance.
Chem Res Toxicol 21(5):1039-1049.

Pataki, J; Huggins, C. (1969) Molecular site of substituents of benz[a]anthracene related to carcinogenicity. Cancer
Res 29:506-509.

Pavanello, S; Favretto, D; Brugnone, F; et al. (1999) HPLC/fluorescence determination of anti-BPDE-DNA adducts
in mononuclear white blood cells from PAH-exposed humans. Carcinogenesis 20(3):431^135.

Penman, BW; Kaden, DA; Liber, HL; et al. (1980) Perylene is a more potent mutagen thanbenzo[a]pyrene for
Salmonella typhimurium. Mutat Res 77:271-277'.
                                             232            DRAFT - DO NOT CITE OR QUOTE

-------
Penning, TM; Burczynski, ME; Hung, CF; et al. (1999) Dihydrodiol dehydrogenases and polycyclic aromatic
hydrocarbon activation: generation of reactive and redox active o-quinones. Chem Res Toxicol 12:1-18.

Perry, PE; Thomson, EJ. (1981) Evaluation of the sister chromatid exchange method in mammalian cells as a
screening system for carcinogens. Prog Mutat Res 1:560-569.

Petry, T; Schmid, P; Schlatter, C. (1996) The use of toxic equivalency factors in assessing occupational and
environmental health risk associated with exposure to airborne mixtures of polycyclic aromatic hydrocarbons
(PAHs). Chemosphere 32:639-648.

Pfeiffer, EH (1973) Investigations on the carcinogenic burden by air pollution in man. VII. Studies on the
oncogenetic interaction of polycyclic aromatic hydrocarbons. Zbl Bakt Hyg, J Abt Org B158:69-83.

Pfeiffer, EH. (1977) Oncogenic interaction of carcinogenic and non-carcinogenic polycyclic aromatic hydrocarbons
in mice. IARC Sci Publ 16:69-77.

Pfeiffer, CA; Allen, E. (1948) Attempts to produce cancer in rhesus monkeys with carcinogenic hydrocarbons and
estrogens.  Cancer Res 8:97-127.

Phillips, DH; Grover, PL; Sims, P. (1979) A quantitative determination of the covalent binding of a series of
polycyclic hydrocarbons to DNA in mouse skin.  Int J Cancer 23:201-208.

Phillipson, CE; loannides, C. (1989) Metabolic activation of polycyclic aromatic hydrocarbons to mutagens in the
Ames test by various animal species including man. Mutat Res Mar 211:147-151.

Pienta, RJ; Poiley, JA; Lebherz, WB 3rd. (1977) Morphological transformation of early passage golden Syrian
hamster embryo cells derived from cryopreserved primary cultures as a reliable in vitro bioassay for identifying
diverse carcinogens. Int J Cancer 19:642-655.

Platt, KL; Dienes, HP; Tommasone, M; et al. (2004) Tumor formation in the neonatal mouse bioassay indicates that
the potent carcinogen dibenzo[def,p]chrysene (dibenzo[a,l]pyrene) is activated in vivo via its trans-
11,12-dihydrodiol.  Chem Biol Interact 148:27-36.

Poncelet, F; Massanda, K; Fouassin, A; et al. (1978) Mutagenic study of some polycyclic aromatic hydrocarbons
present in smoked fishes from Africa.  Arch Int Phys Biochem 86:954-955.

Popescu, NC; Turnbull, D; DiPaolo, JA. (1977) Sister chromatid exchange and chromosome aberration analysis
with the use of several carcinogens and noncarcinogens: brief communication. J Natl Cancer Inst 59:289-293.

Prahalad, AK; Ross, JA; Nelson, GB; et al. (1997) Dibenzo[a,l]pyrene-induced DNA adduction, tumorigenicity, and
Ki-ras oncogene mutations in strain A/J mouse lung.  Carcinogenesis 18:1955-1963.

Probst, GS; McMahon, RE; Hill, LE; et al. (1981) Chemically-induced unscheduled DNA synthesis in primary rat
hepatocyte cultures: a comparison with bacterial mutagenicity using 218 compounds. Environ Mutagen 3:11-32.

Pufulete, M; Battershill, J; Boobis, A; et al. (2004) Approaches to carcinogenic risk assessment for polycylic
aromatic hydrocarbons: a UK perspective. Regul Toxicol Pharmacol 40:54-66.

Puisieux, A; Lim, S; Groopman, J; et al. (1991) Selective targeting of p53 gene mutational hotspots in human
cancers by etiologically defined carcinogens. Cancer Res 51 (22) :6185-6189.

Ramesh, A; Walker, SA; Hood, DB; et al. (2004) Bioavailability and risk assessment of orally ingested polycyclic
aromatic hydrocarbons. Int J Toxicol 23:301-333.

Rask-Nielson, R. (1950) The susceptibility of the thymus, lung, subcutaneous and mammary tissues in strain St mice
to direct application of small doses of four different carcinogenic hydrocarbons. Br J Cancer 4:108-116.
                                              233            DRAFT - DO NOT CITE OR QUOTE

-------
Raveh, D; Huberman, E. (1983) A microtiter plate assay for the selection of 6-thioguanine-resistant mutants in
Chinese hamster V79 cells in the presence of phorbol-12-myristate-13-acetate. MutatRes 113:499-506.

Raveh, D; Slaga, TJ; Huberman, E. (1982) Cell-mediated mutagenesis and tumor-initiating activity of the ubiquitous
polycyclic hydrocarbon, cyclopenta[c,d]pyrene. Carcinogenesis 3:763-766.

Reddy, MV; Gupta, RC; Randerath, E; et al. (1984) 32P-Postlabering test for covalent DNA binding of chemicals in
vivo: application to a variety of aromatic carcinogens and methylating agents. Carcinogenesis 5:231-243.

Rice, JE; Hosted, TJ, Jr.; LaVoie, EJ. (1984) Fluoranthene and pyrene enhance benzo[a]pyrene-DNA adduct
formation in vivo in mouse skin. Cancer Lett 24:3 27-3 3 3.

Rice, JE; Makowski, GS; Hosted, TJ, Jr.; et al. (1985) Methylene-bridged bay region chrysene and phenanthrene
derivatives and their keto-analogs: mutagenicity in Salmonella typhimurium and tumor-initiating activity on mouse
skin. Cancer Lett 27:199-206.

Rice, JE; Jordan, K; Little, P; et al.  (1988) Comparative tumor-initiating activity of methylene-bridged and bay-
region methylated derivatives of benz[a]anthracene and chrysene.  Carcinogenesis 9:2275-2278.

Rigdon, RH; Neal, J. (1966) Gastric carcinomas and pulmonary adenomas in mice fed benzo[a]pyrene.  Tex Rep
BiolMed 24:195-207.

Rigdon, RH; Neal, J. (1969) Relationship of leukemia to lung and stomach tumors in mice fed benzo[a]pyrene.  Proc
Soc Exp Biol Med 130:146-148.

Rigdon, RH; Benge, MC; Kirchoff, H; et al. (1969) Leukemia in mice fedbenzo[a]pyrene: a clinical, pathologic and
hematologic study.  Tex Rep Biol Med 27:803-820.

Robinson, DE; Mitchell, AD. (1981) Unscheduled DNA synthesis response of human fibroblasts, WI-38 cells, to
20 coded chemicals. Prog Mutat Res 1:517-527.

Roe, FJC. (1962) Effect of phenanthrene on tumour-initiation by 3,4-benzpyrene.  Br J Cancer 16:503-506.

Roe, FJ; Waters, MA. (1967) Induction of hepatoma in mice by carcinogens of the polycyclic hydrocarbon type.
Nature 214:299-300.

Rogan, EG; Cavalieri, EL; Ramakrishna, NVS; et al. (1993) Mechanisms of benzo[a]pyrene and
7,12-diemthylbenz[a]antrhacene activation: qualitative aspects of the stable and depurination DNA adducts obtained
from radical cations and diol epoxides. In: Polycyclic aromatic hydrocarbons: synthesis, properties,  analytical
measurements, occurrence and biological effects. Bordeaux, France: Gordon and Breach Science Publishers, pp.
733-740.

Rosenkranz, HS; Leifer, Z. (1980) Determining the DNA-modifying activity of chemicals using DNA polymerase-
deficient Escherichia coli. In:  Chemical mutagens: principles and methods for their detection. New York, NY:
Plenum Press, pp. 109-147.

Rosenkranz, HS; Poirier, LA. (1979) Evaluation of the mutagenicity and DNA-modifying activity of carcinogens
and noncarcinogens in microbial systems.  J Natl Cancer Inst 62:873-891.

Ross, JA; Nelson, GB; Wilson, KH; et al. (1995) Adenomas induced by polycyclic aromatic hydrocarbons in strain
A/J mouse lung  correlate with time-integrated DNA adduct levels.  Cancer Res 55:1039-1044.

Rossman, TG; Molina, M; Meyer, L; et al. (1991) Performance of 133 compounds in the lambda prophage induction
endpoint of the microscreen assay and a comparison with Salmonella typhimurium mutagenicity and rodent
carcinogenicity assays. Mutat Res 260:349-367.

Roszinsky-Kocher, G; Easier, A; Rohrborn, G. (1979) Mutagenicity of polycyclic hydrocarbons. V.  Induction of
sister-chromatid exchanges in vivo. Mutat Res 66:65-67.
                                              234            DRAFT - DO NOT CITE OR QUOTE

-------
Rugen, PJ; Stern, CD; Lamm, SH. (1989) Comparative carcinogenicity of the PAHs as a basis for acceptable
exposure levels (AELs) in drinking water.  Regul Toxicol Pharmacol 9:273-283.

Rummel, AM; Trosko, JE; Wilson, MR; et al. (1999) Poly cyclic aromatic hydrocarbons with bay-like regions
inhibited gap junctional intercellular communication and stimulated MAPK activity. Toxicol Sci 49:232-240.

Safe, S; Wormke, M. (2003) Inhibitory aryl hydrocarbon receptor-estrogen receptor alpha cross-talk and
mechanisms of action.  ChemRes Toxicol 16:807-816.

Sagredo, C; 0vrebe, S; Haugen, A; et al. (2006) Quantitative analysis of benzo[a]pyrene biotransformation and
adduct formation in AhR knockout mice. Toxicol Lett 167:173-182.

Sakai, M; Yoshida, D; Mizusaki, S. (1985) Mutagenicity of polycyclic hydrocarbons and quinones on Salmonella
thyphimuriumTA91. MutatRes 156:61-67.

Salaman, MH; Roe, FJC. (1956) Further tests for tumour-initiating activity: N,N-di(2-chloroethyl)-p-amino-
phenylbutic acid (CB1348) as an initiator of skin tumour formation in the mouse. Br J Cancer 10:363-378.

Salamone, MF; Heddle, JA; Katz, M. (1979a) The mutagenic activity of thirty polycyclic aromatic hydrocarbons
(PAH) and oxides in urban airborne particulates.  Environ Int 2:37-43.

Salamone, MF; Heddle, JA; Katz, M. (1979b) The use of the Salmonella/microsomal assay to determine
mutagenicity in paired chemical mixtures.  Can J Genet Cytol 21:101-107.

Salamone, MF; Heddle, JA; Katz, M. (1981) Mutagenic activity of 41 compounds in the in vivo micronucleus assay.
Prog Mutat Res 1:686-697.

Sangaiah, R; Gold, A; Toney, GE; et al. (1983) Benz[j]aceanthrylene: a novel polycyclic  aromatic hydrocarbon with
bacterial mutagenic activity.  MutatRes 119:259-266.

Sanner,  T; Dybing, E. (2005) Comparison of carcinogenic and in vivo genotoxic potency estimates. Basic Clin
Pharmacol Toxicol 96:131-139.

Schmahl, D; Schmidt, KG; Habs, M. (1977) Syncarcinogenic action of polycyclic hydrocarbons in automobile
exhaust gas condensates.  IARC Sci Publ 16:53-59.

Schmoldt, A; Jacob,  J; Grimmer, G. (1981) Dose-dependent induction of rat liver microsomal aryl hydrocarbon
monooxygenase by benzo[k]fluoranthene.  Cancer Lett 13:249-257.

Schneider, K; Roller, M; Kalberlah, F; et al. (2002)  Cancer risk assessment for oral exposure to PAH mixtures. J
Appl Toxicol 22:73-83.

Scribner, JD. (1973) Brief communication: tumor initiation by apparently noncarcinogenic polycyclic aromatic
hydrocarbons.  J Natl Cancer Inst 50:1717-1719.

Segerback, D; Vodicka, P. (1993) Recoveries of DNA adducts of polycyclic aromatic hydrocarbons in the
32P-postlabeling assay. Carcinogenesis 14:2463-2469.

Shah, GM; Bhattacharya, RK. (1989) Alteration in hepatic nuclear RNA polymerase activity following
benzo[a]pyrene administration in rat.  In Vivo 3:125-127.

Sharovskaia, I; Rokitskaia, TI; Kobliakov, VA. (2003) [Effect of some polycyclic aromatic hydrocarbons on gap
junction intercellular communication in hepatoma Hep G2 cell culture]. Tsitologiia 45:51-58.

Shen, YM; Troxel, AB; Vedantam, S; et al. (2006) Comparison of p53 mutations induced by PAH o-quinones with
those caused by anti-benzo[a]pyrene diol epoxide in vitro: role of reactive oxygen and biological selection. Chem
Res Toxicol 19(11):1441-1450.
                                             23 5            DRAFT - DO NOT CITE OR QUOTE

-------
Sheu, CW; Dobras, SN; Rodriguez, I; et al. (1994) Transforming activity of selected polycyclic aromatic
hydrocarbons and their nitro-derivatives in BALB/3T3 A31-1-1 cells. Food Chem Toxicol 32:611-615.

Shubik, P; Pietra, G; Delia Porta, G. (1960) Studies of skin carcinogenesis in the Syrian golden hamster. Cancer
Res 20:100-105.

Simmon, VF. (1979a) In vitro mutagenicity assays of chemical carcinogens and related compounds with Salmonella
typhimurium. J Natl Cancer Inst 62:893-899.

Simmon, VF. (1979b) In vitro assays for recombinogenic activity of chemical carcinogens and related compounds
with Saccharomyces cerevisiae D3. J Natl Cancer Inst 62:901-910.

Simmon, VF; Rosenkranz, HS; Zeiger, E; et al. (1979) Mutagenic activity of chemical carcinogens and related
compounds in the intraperitoneal host-mediated assay. J Natl Cancer Inst 62:911-918.

Sirianni, SR; Huang, CC. (1978) Sister chromatid exchange induced by promutagens/carcinogens in Chinese
hamster cells cultured in diffusion chambers in mice. Proc Soc Exp Biol Med 158:269-274.

Sjogren, M; Ehrenberg, L; Rannug, U. (1996) Relevance of different biological assays in assessing initiating and
promoting properties of polycyclic aromatic hydrocarbons with respect to carcinogenic potency. Mutat Res 358:97-
112.

Slaga, TJ; Fischer, SM. (1983) Strain differences and solvent effects in mouse skin carcinogenesis experiments
using carcinogens, tumor initiators and promoters. Prog Exp Tumor Res 26:85-109.

Slaga, TJ; Hubermann, E; Selkirk, JK; et al. (1978) Carcinogenicity and mutagenicity of benz[a]anthracene diols
and diol-epoxides. Cancer Res 38:1699-1704.

Slaga, TJ; Jecker, L; Bracken, WM; et al. (1979) The effects of weak or noncarcinogenic polycyclic hydrocarbons
on7,12-dimethylbenz[a]anthracene andbenzo[a]pyrene skin tumor-initiation. Cancer Lett 7:51-59.

Slaga, TJ; Iyer, RP; Lyga, W; etal.  (1980) Comparison of the skin tumor-initiating activities of dihydrodiols, diol-
epoxides, and methylated derivatives of various polycyclic aromatic hydrocarbons.  In: Bjorseth, A; Dennis, AJ, eds.
Polynuclear aromatic hydrocarbons: chemistry and biological effects. Columbus, OH: Battelle Press, pp. 753-769.

Slooff, W; Janus, JA; Matthijsen, AJCM; et al. (1989) Integrated criteria document PAHs (PDF includes addendum
by Montizaan). National Institute of Public Health and the Environment (RIVM), Bilthoven, The Netherlands.

Smith, LE; Denissenko, MF; Bennett, WP; et al. (2000) Targeting of lung cancer mutational hotspots by polycyclic
aromatic hydrocarbons. J Natl Cancer Inst 92:803-811.

Smolarek, TA; Baird, WM. (1984) Benzo[e]pyrene-induced alterations in the binding of benzo[a]pyrene to DNA  in
hamster embryo cell cultures. Carcinogenesis 5:1065-1069.

Smolarek, TA; Moynihan, CG; Salmon, CP; et al. (1986) Benz[a]anthracene-induced alterations in the metabolic
activation of benzo[a]pyrene by hamster embryo cell cultures.  Cancer Lett 30:243-249.

Smolarek, TA; Baird, WM; Fisher,  EP; et al. (1987) Benzo[e]pyrene-induced alterations in the binding of
benzo[a]pyrene and 7,12-dimethylbenz[a]anthracene to DNA in Sencar mouse epidermis.  Cancer Res 47:3701-
3706.

Snell, KC; Stewart, HL. (1962) Pulmonary adenomatosis induced in DBA/2 mice by oral administration of
dibenz[a,h]anthracene. J Natl Cancer Inst 28:1043-1049.

Solt, DB; Polverini, PJ; Calderon, L. (1987) Carcinogenic response of hamster buccal pouch epithelium to
4 polycyclic aromatic hydrocarbons. J Oral Pathol 16:294-302.

Staal, YCM; Hebels DGAJ; van Herwijnen, MHM; et al. (2007) Binary PAH-mixture cause additive or antagonistic
effects on gene expression but synergistic effects on DNA adduct formation.  Carcinogenesis 28:2632-2640.
                                             236            DRAFT - DO NOT CITE OR QUOTE

-------
Stanton, MF; Miller, E; Wrench, C; et al. (1972) Experimental induction of epidermoid carcinoma in the lungs of
rats by cigarette smoke condensate. J Natl Cancer Inst 49:867-877.

Steiner, PF. (1955) Carcinogenicity of multiple chemicals simultaneously administered. Cancer Res 15:632-635.

Steiner, PF; Falk, HL. (1951) Summation and inhibition effects of weak and strong carcinogenic hydrocarbons:
l:2-benzanthracene, chrysene, l:2:5:6-dibenzanthracene, and 20-methylcholanthrene.  Cancer Res 11:56-63.

Straif K; Baan, R; Grosse, Y; et al. (2005) Carcinogenicity of polycyclic aromatic hydrocarbons. Lancet 6:931-932.

Sugiyama, T. (1973) Chromosomal aberrations and carcinogenesis by various benz[a]anthracene derivatives. Gann
64:637-639.

Tannheimer, SL; Ethier, SP; Caldwell, KK; et al. (1998) Benzo[a]pyrene- and TCDD-induced alterations in tyrosine
phosphorylation and insulin-like growth factor signaling pathways in the MCF-10A human mammary epithelial cell
line. Carcinogenesis 19:1291-1297.

Teranishi, K; Hamada, K; Watanabe, H. (1975) Quantitative relationship between Carcinogenicity and mutagenicity
of polyaromatic hydrocarbons in Salmonella typhimurium mutants. Mutat Res 31:97-102.

Thyssen, J; Althoff, J; Kimmerle, G; et al. (1980) Investigations on the carcinogenic burden of air pollution in man.
XIX. Effect of inhaled benzo[a]pyrene in Syrian golden hamsters: a pilot study. Zentralbl Bakteriol Hyg I Abt Orig
B 171:441-444.

Thyssen, J; Althoff, J; Kimmerle, G; et al. (1981) Inhalation studies with benzo[a]pyrene in Syrian golden hamsters.
J Natl Cancer Inst 66:575-577.

Till, M; Riebniger, D' Schmitz,  HJ; et al. (1999) Potency of various polycyclic aromatic hydrocarbons as inducers of
CYP1 Al in rat hepatocyte cultures. Chem Biol Interact 117:135-150.

Tong, C; Brat, SV; Williams, GM. (1981a) Sister-chromatid exchange induction by polycyclic aromatic
hydrocarbons in an intact cell system of adult rat-liver epithelial cells. Mutat Res 91:467-473.

Tong, C; Laspia, MF; Telang, S; et al. (1981b) The use of adult rat liver cultures in the detection of the genotoxicity
of various polycyclic aromatic hydrocarbons. Environ Mutagen 3:477-487.

Tong, C; Brat, VS; Telang, S; et al. (1983) Effects of genotoxic polycyclic aromatic hydrocarbons in rat liver culture
systems.  In: Cooke, M; Dennis, AJ, eds.  Polynuclear aromatic hydrocarbons: formation; metabolism, and
measurement. Columbus, OH: Battelle Press, pp. 1189-1203.

Topping, DC; Martin, DH; Nettesheim, P. (1981) Determination of cocarcinogenic activity of benzo[e]pyrene for
respiratory tract mucosa.  Cancer Lett 11:315-321.

Travis, CC; Saulsbury, AW; Richter Pack, SA. (1990) Prediction of cancer potency using a battery of mutation and
toxicity data. Mutagenesis 5:213-219.

Tsuchimoto, T; Matter, BE. (1981) Activity of coded compounds in the micronucleus test. Prog Mutat Res 1:705-
711.

Tweats, DJ. (1981) Activity of 42 coded compounds in a differential killing test using Escherichia coli strains WP2,
WP67 (uvrA polA), and CM871 (uvrA lexA recA).  Prog Mutat Res 1:199-209.

Uno, S; Dalton, TP; Dragin, N; et al. (2006) Oral benzo[a]pyrene in Cypl knockout mouse lines: CYP1A1 important
in detoxication, CYP1B1 metabolism required for immune damage independent of total-body burden and clearance
rate. MolPharmacol 69:1103-1114.

U.S. EPA (U.S. Environmental Protection Agency). (1986) Guidelines forthe health risk assessment of chemical
mixtures. Federal Register 51(185):34014-34025.
                                              237            DRAFT - DO NOT CITE OR QUOTE

-------
U.S. EPA (U.S. Environmental Protection Agency). (1990) Drinking water criteria document for poly cyclic
aromatic hydrocarbons. Cincinnati, OH: Office of Health and Environmental Assessment, Environmental Criteria
and Assessment Office.

U.S. EPA (U.S. Environmental Protection Agency). (1993) Provisional guidance for quantitative risk assessment of
polycyclic aromatic hydrocarbons. Cincinnati, OH: Office of Health and Environmental Assessment,
Environmental Criteria and Assessment Office.

U.S. EPA (U.S. Environmental Protection Agency). (2000) Supplementary guidance for conducting health risk
assessment of chemical mixtures. EPA/630/R-00/002.

U.S. EPA (U.S. Environmental Protection Agency). (2002) Peer consultation workshop on approaches to polycyclic
aromatic hydrocarbon (PAH) health assessment.  Washington, DC: National Center for Environmental Assessment,
Office of Research and Development. EPA/635/R-02/005.

U.S. EPA (U.S. Environmental Protection Agency). (2004) An examination of EPA risk assessment principles and
practices.  Staff paper prepared for the U. S. EPA by members of the Risk Assessment Task Force at the request of
the EPA science advisor. Available online at http://www.epa.gov/osa/ratf.htm (accessed January 13, 2010).

U.S. EPA (U.S. Environmental Protection Agency). (2005a) Guidelines for carcinogen risk assessment. Risk
Assessment Forum, Washington, DC: EPA/630/P-03/001B. Available online at http://www.epa.gov/iris/backgr-
d.htm (accessed January 15, 2009).

U.S. EPA (U.S. Environmental Protection Agency). (2005b) Supplemental guidance for assessing susceptibility
from early-life exposure to carcinogens. Risk Assessment Forum, Washington, DC; EPA/630/R-03/003F. Available
online at http://www.epa.gov/iris/backgr-d.htm (accessed January 15, 2009).

U.S. EPA (U.S. Environmental Protection Agency). (2009) Integrated Risk Information System (IRIS). National
Center for Environmental Assessment, Washington, DC. Available online at http://www.epa.gov/iris (accessed
January 13,2010).

Utesch, D; Glatt, H; Oesch, F. (1987) Rat hepatocyte-mediated bacterial mutagenicity in relation to the carcinogenic
potency of benz[a]anthracene, benzo[a]pyrene, and twenty-five methylated derivatives. Cancer Res 47(6): 1509-
1515.

Vaca, C; Tornqvist, M; Rannug, U; et al. (1992) On the bioactivation and genotoxic action of fluoranthene. Arch
Toxicol 66:538-545.

Valencia, R; Houtchens, K. (1981) Mutagenic activity of 10 coded compounds in the Drosophila sex-linked
recessive lethal test.  Prog Mutat Res 1:652-659.

Van Duuren, BL; Goldschmidt, BM. (1976) Cocarcinogenic and tumor-promoting agents in tobacco carcinogenesis.
J Natl Cancer Inst 56:1237-1242.

Van Duuren, BL; Sivak, A; Segal, A; et al. (1966) The tumor-promoting agents of tobacco leaf and tobacco smoke
condensate.  J Natl Cancer Inst 37:519-526.

Van Duuren, BL; Sivak, A; Langseth, L; et al. (1968) Initiators and promoters in tobacco carcinogenesis.  Natl
Cancer Inst Monogr 28:173-80.

Van Duuren, BL; Sivak, A; Goldschmidt, BM; et al.  (1970) Initiating activity of aromatic hydrocarbons in two-stage
carcinogenesis. J Natl  Cancer Inst 44:1167-1173.

Van Duuren, BL; Katz, C; Goldschmidt, BM; et al. (1973) Brief communication: cocarcinogenic agents in tobacco
carcinogenesis. J Natl  Cancer Inst 51:703-705.

Vesselinovitch, SD; Kyriazis, AP; Mihailovich, N; et al. (1975) Factors influencing and/or acceleration of
ly mphoreticula tumors in mice by benzo [a]py rene treatment. Cancer Res35:1963-1969.
                                             23 8            DRAFT - DO NOT CITE OR QUOTE

-------
Vienneau, DS; DeBoni, U; Wells, PG. (1995) Potential genoprotective role for UDP-glucuronosyltransferases in
chemical carcinogenesis: initiation of micronuclei by benzo[a]pyrene and benzo[e]pyrene in UDP-
glucuronosyltransferase-deficient cultured rat skin fibroblasts. Cancer Res 55:1045-1051.

Vijayalakshmi, KP; Suresh, CH. (2008) Theoretical studies on the carcinogenicity of poly cyclic aromatic
hydrocarbons. J Comput Chem 29:1108-1117.

Wangenheim, J; Bolcsfoldi, G. (1988) Mouse lymphoma L5178Y thymidine kinase locus assay of 50 compounds.
Mutagenesis 3:193-205.

Warshawsky, D; Barkley, W. (1987) Comparative carcinogenic potencies of 7H-dibenzo[c,g]carbazole,
dibenz[aj]acridine and benzo[a]pyrene in mouse skin.  Cancer Lett 37:337-344.

Warshawsky, D; Barkley, W; Bingham, E. (1993) Factors affecting carcinogenic potential of mixtures.  Fundam
Appl Toxicol 20:376-382.

Warshawsky, D; Livingston, GK; Fonouni-Fard, M; et al. (1995) Induction of micronuclei and sister chromatid
exchanges by polycyclic and N-heterocyclic aromatic hydrocarbons in cultured human lymphocytes. Environ Mol
Mutagen26:109-118.

Weinstein, D; Katz, ML; Kazmer, S. (1977) Chromosomal effects of carcinogens and noncarcinogens on WI-38
after short term exposures with and without metabolic activation.  Mutat Res 46:297-304.

Wenzel-Hartung, R; Brune, H; Grimmer, G; et al. (1990) Evaluation of the carcinogenic potency of four
environmental polycyclic aromatic compounds following intrapulmonary application in rats. Exp Pathol 40:221-
227.

Weyand, EH; LaVoie, EJ. (1988) Comparison of PAH DNA adduct formation and tumor initiating activity in
newborn mice. Proc Am Assoc Cancer Res 29:98.

Weyand, EH; Wu, Y. (1995) Covalent binding of polycyclic aromatic hydrocarbon components of manufactured gas
plant residue to mouse lung and forestomach DNA.  Chem Res Toxicol 8:955-962.

Weyand, EH; He, ZM; Ghodrati, F; et al. (1992) Effect of fluorene substitution onbenzo[j]fluoranthene
genotoxicity. ChemBiol Interact 84:37-53.

Weyand, EH; Chen, YC; Wu,  Y; et al. (1995) Differences in the tumorigenic activity of a pure hydrocarbon and a
complex mixture following ingestion: benzo[a]pyrene vs. manufactured gas plant residue. Chem Res Toxicol
8:949-954.

WHO (World Health Organization). (1998) Selected non-heterocyclic polycyclic aromatic hydrocarbons
Environmental health criteria.  Vol. 202. International Programme on Chemical Safety, Geneva, Switzerland.

Willett, KL; Gardinali, PR; Sericano, JL; et al. (1997) Characterization of the H4IIE rat hepatoma cell bioassay for
evaluation of environmental samples containing polynuclear aromatic hydrocarbons (PAHs). Arch Environ Contam
Toxicol  32:442-448.

Williams, GM; Laspia, MF; Dunkel, VC. (1982) Reliability of the hepatocyte primary culture/DNA repair test in
testing of coded carcinogens and noncarcinogens. Mutat Res 97:359-370.

Wislocki, PG; Bagan, ES; Lu, AY; et al. (1986) Tumorigenicity of nitrated derivatives of pyrene, benz[a]anthracene,
chrysene and benzo[a]pyrene in the newborn mouse assay. Carcinogenesis 7:1317-1322.

Wood, AW; Chang, RL; Levin, W; et al. (1979) Mutagenicity and tumorigenicity of phenanthrene and chrysene
epoxides and diol epoxides. Cancer Res 39:4069-4077.

Wood, AW; Levin,  W; Chang, RL; et al. (1980) Mutagenicity and tumor-initiating activity of cyclopenta[c,d]pyrene
and structurally related compounds. Cancer Res 40:642-649.
                                             239            DRAFT - DO NOT CITE OR QUOTE

-------
Wynder, EL; Hoffmann, D. (1959a) The carcinogenicity of benzofluoranthenes. Cancer 12:1194-1199.

Wynder, EL; Hoffmann, D. (1959b) A study of tobacco carcinogenesis: VII. The role of higher poly cyclic
hydrocarbons. Cancer 12:1079-1086.

Wynder, EL; Hoffmann, D. (1961) Carcinogenicity of dibenzo[a,l]pyrene.  Nature 192:1092-1093.

Wynder, EL; Fritz, L; Furth, N. (1957) Effect of concentration of benzopyrene in skin carcinogenesis. J Natl Cancer
Inst 19:361-370.

Xu, D; Penning, TM; Blair, IA; et al. (2009) Synthesis of phenol and quinine metabolites of benzo[a]pyrene, a
carcinogenic component of tobacco smoke implicated in lung cancer.  J Org Chem 74:597-604.

Xue, W; Warshawsky, D. (2005) Metabolic activation of polycyclic and heterocyclic aromatic hydrocarbons and
DNA damage: a review. Toxicol Appl Pharmacol 206:73-93.

Yu, C; Xu, S; Chen, S; etal. (2002) Investigation of photobleaching of hypocrellinB in non-polar organic solvent
and in liposome suspension. J Photochem Photobiol B 68:73-78.

Zijlstra, JA; Vogel, EW. (1984) Mutagenicity of 7,12-dimethylbenz[a]anthracene and some other aromatic mutagens
in Drosophila melanogaster. Mutat Res 125:243-261.
                                             240            DRAFT - DO NOT CITE OR QUOTE

-------
   APPENDIX A.  SECONDARY SOURCES REVIEWED FOR IDENTIFICATION OF
                                    PRIMARY LITERATURE
ATSDR (Agency for Toxic Substances and Disease Registry). (1995) Toxicological profile for polycyclic aromatic
hydrocarbons (PAHs).  Atlanta, GA.

Bostrom, CC; Gerde, P; Hanberg, A; et al. (2002) Cancer risk assessment, indicators, and guidelines for polycyclic
aromatic hydrocarbons in the ambient air.  Environ Health Perspect 110(Suppl. 3):451-488.

California EPA (California Environmental Protection Agency). (2002) Air toxics hot spots program risk assessment
guidelines Part I. Technical support document for describing available cancer potency factors.  Office of
Environmental Health Hazard Assessment.

California EPA (California Environmental Protection Agency). (2004) No Significant Risk Levels (NSRLs) for the
Proposition 65 carcinogens benzo[b]fluoranthene, benzo[j]fluoranthene, chrysene, dibenzo[a,h]pyrene,
dibenzo[a,i]pyrene, and 5-methyl chrysene by the oral route.  Office of Environmental Health Hazard Assessment,
Reproductive and Cancer Hazard Assessment Section.

CCME (Canadian Council of the Ministers of the Environment). (2003) Canadian soil quality guidelines for
potentially carcinogenic and higher molecular weight polycyclic aromatic hydrocarbons (environmental and human
health aspects).  Scientific Supporting Document. UMA Group Ltd. Victoria, British Columbia.

Clement Associates. (1988) Comparative potency approach for estimating the cancer risk associated with exposure
to mixtures of polycyclic aromatic hydrocarbons. Report No. 68-02-4403.

Clement Associates. (1990) Development of relative potency estimates for PAHs and hydrocarbon combustion
product fractions compared to benzo[a]pyrene and their use in carcinogenic risk assessments. Draft Report,
prepared for the  U.S. EPA. September 30, 1990.

Collins, JF; Brown, JP; Alexeeff, GV; et al. (1998) Potency equivalency factors for some polycyclic aromatic
hydrocarbons and polycyclic aromatic hydrocarbon derivatives. Regul Toxicol Pharmacol 28:45-54.

Health Canada. (1994) Canadian Environmental Protection Act; Priority substances list assessment report:
polycyclic aromatic hydrocarbons. Government of Canada, Environment Canada.

IARC (International Agency for Research on Cancer). (1983) Polynuclear aromatic compounds. Part 1. Chemical,
environmental and experimental data. In: IARC monographs on the evaluation of carcinogenic risk of chemicals to
humans. Vol. 32. Lyon, France.

IARC (International Agency for Research on Cancer). (1984a) Polynuclear aromatic compounds. Part 2. Carbon
black, mineral oils (lubricant base oils and derived products) and some nitroarenes. In: IARC monographs on the
evaluation of carcinogenic risk of chemicals to humans. Lyon, France, pp. 87-168.

IARC (International Agency for Research on Cancer). (1984b) Polynuclear aromatic compounds. Part 3. Industrial
exposures in aluminum production, coal gasification, coke production, and iron and steel founding. In: IARC
monographs on the evaluation of carcinogenic risk of chemicals to humans. Lyon, France, pp. 37-111.

IARC (International Agency for Research on Cancer). (1985) Polynuclear aromatic compounds. Part 4. Bitumens,
coal-tars and derived products, shale-oils and soots.  In: IARC monographs on the evaluation of carcinogenic risk of
chemicals to humans. Lyon, France, pp. 65-159.

IARC (International Agency for Research on Cancer). (1989) Occupational exposures in petroleum refining; crude
oil and major petroleum fuels. In:  IARC monographs on the evaluation of carcinogenic risk of chemicals to
humans. Vol. 45. Lyon, France, pp. 239-270.
                                             A-1            DRAFT - DO NOT CITE OR QUOTE

-------
IARC (International Agency for Research on Cancer). (1996) Printing processes and printing inks, carbon black and
some nitro compounds. In: IARC monographs on the evaluation of carcinogenic risks to humans. Vol. 65. Lyon,
France.

IPCS/WHO (International Programme on Chemical Safety /World Health Organization). (1998) Selected non-
heterocyclic polycyclic aromatic hydrocarbons. Environmental health criteria 202. International Programme on
Chemical Safety.

Krewski, D; Thorslund, T; Withey, J. (1989) Carcinogenic risk assessment of complex mixtures. Toxicol Ind Health
5:851-867.

Larsen, JC; Larsen, PB. (1998) Chemical carcinogens. Air pollution and health. Cambridge, UK: The Royal Society
of Chemistry, pp. 33-56.

Malcolm, HM; Dobson, S. (1994) The calculation of an environmental assessment level (EAL) for atmospheric
PAHs using relative potencies. Report No. DoE/HMIP/RR/94/041. London, Department of the Environment.

McClure, P; Shoeny, R. (1995) Evaluation of a component-based relative potency approach to cancer risk
assessment for exposure to PAH.  In: Fifteenth international symposium on polycyclic aromatic compounds:
chemistry, biology and environmental impact. Belgirate, Italy, 19-22 September 1995. Ispra, Joint Research Centre
European Commission, pp. 161.

Meek, ME; Chan, PKL; Bartlett, S. (1994) Polycyclic aromatic hydrocarbons: evaluation of risks to health from
environmental exposures in Canada.  Environ Carcinog Ecotoxicol Rev C 12(2):443-452.

Muller, P.  (1997) Scientific criteria document for multimedia standards development polycyclic aromatic
hydrocarbons (PAH). Part I. Hazard identification and dose-response assessment.  Ontario: Ontario Ministry of
Environment and Energy, Standards Development Branch.

Nisbet, ICT; LaGoy, PK. (1992) Toxic equivalency factors (TEFs) for polycyclic aromatic hydrocarbons (PAHs).
Regul Toxicol Pharmacol 16:290-300.

Petry, T; Schmid, P; Schlatter, C. (1996) The use of toxic equivalency factors in assessing occupational and
environmental health risk associated with exposure to airborne mixtures of polycyclic aromatic hydrocarbons
(PAHs). Chemosphere 32(4):639-648.

Rugen, PJ; Stern, CD; Lamm, SH. (1989) Comparative carcinogenicity  of the PAHs as abasis for acceptable
exposure levels (AELs) in drinking water. Regul Toxicol Pharmacol 9(3):273-283.

Schneider, K; Roller, M; Kalberlah, F; et al. (2002) Cancer risk assessment for oral exposure to PAH mixtures. J
Appl Toxicol 22(l):73-83.

Sjogren, M; Ehrenberg, L; Rannug, U. (1996) Relevance of different biological assays in assessing initiating and
promoting properties of polycyclic aromatic hydrocarbons with respect  to carcinogenic potency.  Mutat Res
358(1):97-112.

Slooff, W; Janus, JA; Matthijsen, AJCM; et al. (1989) Integrated criteria document PAHs (PDF includes addendum
by Montizaan). Bilthoven, The Netherlands. National Institute of Public Health and the Environment (RIVM).

SRC (Syracuse Research Corporation). (1993) Estimating cancer risk from exposure to PAHs: a relative potency
approach.  SRC TR-93-045. Draft report prepared for U.S. EPA, Environmental Criteria and Assessment Office,
Cincinnati, OH.

U.S. EPA (Environmental Protection Agency). (1990) Drinking water criteria document for polycyclic aromatic
hydrocarbons. Cincinnati, OH: Office of Health and Environmental Assessment, Environmental Criteria and
Assessment Office.
                                             A-2           DRAFT - DO NOT CITE OR QUOTE

-------
U.S. EPA (Environmental Protection Agency). (1993) Provisional guidance for quantitative risk assessment of
polycyclic aromatic hydrocarbons.  Cincinnati, OH: Office of Health and Environmental Assessment,
Environmental Criteria and Assessment Office.
                                            A-3           DRAFT - DO NOT CITE OR QUOTE

-------
1    APPENDIX B. BIBLIOGRAPHY OF STUDIES WITHOUT BENZO[A]PYRENE AS A

2                         REFERENCE COMPOUND
3
4
5
6
                                B-1       DRAFT - DO NOT CITE OR QUOTE

-------
Table B-l. Bioassays with and without benzo[a]pyrene by PAH

PAH"
Aceanthrylene
Acenaphthene
Acenaphthylene
Acephenanthrylene
Acepyrene, 2,3-
Anthanthrene
Anthracene
Benz[a]anthracene
Benz[b]anthracene
Benz[b,c]aceanthrylene, 11H-
Benz[e]aceanthrylene
Benz[j]aceanthrylene
Benz[l]aceanthrylene
Benzacenaphthylene
Benzo [a] fluoranthene
Benzo [a] fluorene
Benzo[a]perylene
Benzo [b] chrysene
Benzo[b]fluoranthene
HH-Benzo[b]fluorene
Benzo[b]perylene
Benzo[c]chrysene
Benzo [c] fluorene
Benzo [c] phenanthrene
Benzo[e]pyrene
Benzo[g]chrysene
Benzo[g,h,i]fluoranthene
Benzo[g,h,i]perylene
Benzo[j]fluoranthene
Benzo[k]fluoranthene
Benzophenanthrene
Chrysene
Coronene
Cyclopenta[c,d]pyrene
Cyclopenta[d,e,f]chrysene, 4H-
Cyclopenta[d,e,f]phenanthrene, 4H-
Cyclopenta[h,i]acephenanthrylene
CASRN
202-03-09
83-32-9
208-96-8
201-06-9
25732-74-5
191-26-4
120-12-7
56-55-3
92-24-9
202-94-8
199-54-2
202-33-5
211-91-6
76774-50-0
203-33-8
238-84-6 or
30777-18-5
191-85-5
214-17-5
205-99-2
243- 17-4 or
30777-19-6
197-70-6
194-69-4
205-12-9 or
30777-20-9
195-19-7
192-97-2
196-78-1
203-12-3
191-24-2
205-82-3
207-08-9
65777-08-4
218-01-9
191-07-1
27208-37-3
202-98-2
203-64-5
114959-37-4
Bioassays with benzo[a]pyrene
Dermal
Initiation




X
X
X
X

X


X





X





X

X
X
X
X

X

X
X


Complete




X
X
X
X










X





X

X
X
X
X

X
X
X



Intra-
peritoneal







X



X






X









X
X

X

X



Sub-
cutaneous






X
X























X





Oral







X





























Other





X

X










X





X


X
X
X

X






Bioassays without benzo[a]pyrene
Dermal
Initiation





X
X
X



X


X
X

X
X
X


X
X
X


X
X
X

X
X




Complete





X
X
X















X







X





Intra-
peritoneal






X
X










X




X
X



X


X

X
X


Sub-
cutaneous






X
X















X







X





Oral






X
X





























Other






X
X










X









X








                                        B-2
DRAFT - DO NOT CITE OR QUOTE

-------
       Table B-l. Bioassays with and without benzo[a]pyrene by PAH

PAH"
Cyclopenta[h,i]aceanthrylene
Cyclopentaphenanthrene
Cyclopenteno-l,2-benzanthracene, 5,6-
Dibenz[a,c]anthracene
Dibenzo [a,e]fluoranthene
Dibenz[a,j]anthracene
Dibenzo [b,e]fluoranthene
Dibenzo [a,c]fluorene, 13H-
Dibenzo [a,e]pyrene
Dibenzo [a,f]fluoranthene
Dibenzo[a,g]fluorene, 13H-
Dibenz[a,h]anthracene
Dibenzo [a,h]pyrene
Dibenzo [a,i]pyrene
Dibenzo [a,l]pyrene
Dibenzo [e,l]pyrene
Dibenzo [h,rst]pentaphene
Dibenz[k,mno]acephenanthrylene
Dibenzo [j ,mno]acephenanthrylene
Dihydroaceanthrylene, 1,2-
Fluoranthene
Fluorene
Indeno [ 1 ,2,3-c,d]fluoranthene
Indeno [ 1 ,2,3-c,d]pyrene
Naphtho [1 ,2-b] fluoranthene
Naphtho[l,2,3,-nmo]acephenanthrylene
Naphtho [2,1 -a] fluoranthene
Naphtho[2,3-a]pyrene
Naphtho[2,3-e]pyrene
Pentacene
Pentaphene
Perylene
Phenanthrene
Picene
Pyrene
Tribenzofluoranthene 3,4-10, 11-12,13-
Triphenylene
CASRN
131581-33-4
219-08-9
7099-43-6
215-58-7
5385-75-1
224-41-9
2997-45-7
201-65-0
192-65-4
203-11-2
207-83-0
53-70-3
189-64-0
189-55-9
191-30-0
192-51-8
192-47-2
153043-81-3
153043-82-4
641-48-5
206-44-0
86-73-7
193-43-1
193-39-5
111189-32-3
113779-16-1
203-20-3
196-42-9
193-09-9
135-48-8
222-93-5
198-55-0
85-01-8
213-46-7
129-00-0
13579-05-0
217-59-4
Bioassays with benzo[a]pyrene
Dermal
Initiation



X
X



X
X

X
X
X
X
X




X


X




X


X
X

X


Complete



X
X



X
X

X
X
X
X
X




X


X




X


X
X

X

X
Intra-
peritoneal











X


X





X


X








X

X


Sub-
cutaneous











X




















X




Oral











X




















X




Other











X











X








X

X




Bioassays without benzo[a]pyrene
Dermal
Initiation



X
X
X


X
X

X
X
X
X






X

X
X

X




X
X
X
X


Complete



X





X
X
X

X
X






X










X
X



Intra-
peritoneal



X







X
X
X
X





X











X
X



Sub-
cutaneous


X
X







X

X
X




X












X
X



Oral











X


X






















Other











X

X






X











X

X


"PAHs in bold have at least one bioassay without benzo[a]pyrene and no bioassays with benzo[a]pyrene.
                                                         B-3
DRAFT - DO NOT CITE OR QUOTE

-------
B.I.  BIBLIOGRAPHY OF BIOASSAYS WITHOUT BENZO[A]PYRENE

Amin, S; Huie, K; Hecht, SS; (1985) Mutagenicity and tumor-initiating activity of methylated
benzo[b]fluoranthenes. Carcinogenesis 6:1023-1025.

Amin, S; Hussain, N; Balanikas, G; et al. (1985) Mutagenicity and tumor initiating activity of methylated benzo[k]
fluoranthenes.  Cancer Lett 26:343-347.

Amin, S; Misra, B; Braley, J; et al. (1991) Comparative tumorigenicity in newborn mice of chrysene and
5-alkylchrysene-l,2-diol-3,4-epoxides. Cancer Lett 58:115-118.

Amin, S; Weyand, EH; Huie, K; etal. (1991) Effects of fluorene substitution on benzo[b]fluoranthene tumorgenicity
and DNA adduct formation in mouse skin. In: Cooke, M; Loening, K; Merritt, J, eds. Polynuclear aromatic
hydrocarbons: measurements, means and metabolism.  Columbus, OH, Battelle Press, pp. 25-35.

Amin, S; Desai, D; Dai, W; et al. (1995) Tumorigenicity in newborn mice of fjord region and other sterically
hindered diol epoxides of benzo[g]chrysene, dibenzo[a,l]pyrene (dibenzo[def,p]chrysene),
4H-cyclopenta[def]chrysene  and fluoranthene.  Carcinogenesis 16:2813-2817.

Barry, G; Cook, CW; Amin,  S; et al. (1934) A comparison of the action of some polycyclic aromatic hydrocarbons
in producing tumours of connective tissue. Am J Cancer 20:58-69.

Bhatt, TS; Coombs, MM. (1990) The carcinogenicity of cyclopenta[a]phenanthrene and chrysene derivatives in the
Sencar mouse.  Polycycl Aromat Compd 1:51-58.

Bock, FG; King, DW. (1959) A study of the sensitivity of the mouse forestomach toward certain polycyclic
hydrocarbons.  J Natl Cancer Inst 23:833-838.

Bottomley, AC; Twort, CC. (1934) The carcinogenicity of chrysene andoleic acid.  Am J Cancer 21:781-786.

Boyland, E; Burrows, H. (1935) The experimental production of sarcoma in rats and mice by a colloidal aqueous
solution of l:2:5:6-dibenzanthracene. J Pathol Bacteriol 41:231-238.

Boyland, E; Sims, P. (1967) The carcinogenic activities in mice of compounds related to benz[a]anthracene.  Int J
Cancer 2:500-504.

Buening, MK; Levin, W; Karle, JM; et al. (1979) Tumorigenicity of bay-region epoxides and other derivatives of
chrysene and phenanthrene in newborn mice. Cancer Res 39:5063-5068.

Buening, MK; Levin, W; Wood, A; et al. (1979) Tumorigenicity of the dihydrodiols of dibenz[a,h]anthracene on
mouse skin and in newborn mice.  Cancer Res 39:1310-1314.

Buters, JT; Mahadevan, B; Quintanilla-Martinez, L; et al. (2002) Cytochrome P450 1B1 determines susceptibility to
dibenzo[a,l]pyrene-induced tumor formation. Chem Res Toxicol 15:1127-1135.

Cavalieri, EL; Rogan, EG; Higginbotham, S; et al. (1989) Tumor-initiating activity in mouse skin and
carcinogenicity in rat mammary gland of dibenzo[a]pyrenes: the very potent environmental carcinogen
dibenzo[a,l]pyrene. J Cancer Res ClinOncol 115:67-72.

Chang, RL; Levin, W; Wood, AW; et al.  (1981) Tumorigenicity of the diastereomeric bay-region benzo[e]pyrene
9,10-diol-ll,12-epoxides in newborn mice.  Cancer Res 41:915-918.

Chang, RL; Levin, W; Wood, AW; et al.  (1982) Tumorigenicity of bay-region diol-epoxides and other benzo-ring
derivatives of dibenzo[a,h]pyrene and dibenzo[a,i]pyrene on mouse skin and in newborn mice.  Cancer Res 42:25-
29.
                                             B-4            DRAFT - DO NOT CITE OR QUOTE

-------
Chang, RL; Levin, W; Wood, AW; et al. (1983) Tumorigenicity of enantiomers of chrysene 1,2-dihydrodiol and of
the diastereomeric bay-region chrysene l,2-diol-3,4-epoxides on mouse skin and in newborn mice. Cancer Res
43:192-196.

Chouroulinkov, I; Coulomb, H; MacNicoll, AD; et al. (1983) Tumour-initiating activities of dihydrodiols of
dibenz[a,c]anthracene. Cancer Lett 19:21-26.

Danz, M; Hartmann, A; Otto, M; et al. (1991) Hitherto unknown additive growth effects of fluorene and
2-acetylaminofluorene on bile duct epithelium and hepatocytes in rats. Arch Toxicol Suppl 14:71-74.

Flesher, JW; Horn, J; Lehner, AF. (2002) Comparative carcinogenicity of picene and dibenz[a,h]anthracene in the
rat. Biochem Biophys Res Commun 290:275-279.

Forbes, PD; Davies, RE; Urbach, F. (1976) Phototoxicity and photocarcinogenesis: comparative effects of
anthracene and 8-methoxypsoralen in the skin of mice. Food Cosmet Toxicol 14:303-306.

Geddie, JE; Amin, S; Huie, K; et al. (1987) Formation and tumorigenicity of benzo[b]fluoranthene metabolites in
mouse epidermis. Carcinogenesis 8:1579-1584.

Gill, HS; Kole, PL; Wiley, JC; et al. (1994) Synthesis and tumor-initiating activity in mouse skin of
dibenzo[a,l]pyrene syn- and anti-fjord-region diolepoxides.  Carcinogenesis 15:2455-2460.

Hecht, SS; LaVoie, E; Amin, S; et al. (1980) On the metabolic activation of the benzofluoranthenes.  In: Chemical
analysis and biological fate: polynuclear aromatic  hydrocarbons. Columbus, OH: Battelle Press, pp. 417-433.

Hecht, SS; LaVoie, EJ; Bedenko, V; et al. (1981a) On the metabolic activation of dibenzo[a,i]pyrene and
dibenzo[a,h]pyrene.  In: Chemical analysis and biological fate: polynuclear aromatic hydrocarbons. Columbus, OH:
Battelle Press, pp. 43-45.

Hecht, SS; LaVoie, EJ; Bedenko, V; etal. (1981b) Reduction of tumorigenicity and of dihydrodiol formation by
fluorene substitution in the angular rings of dibenzo[a,i]pyrene.  Cancer Res 41:4341-4345.

Hecht, SS; Amin, S;  Lin, JM; et al. (1995) Mammary carcinogenicity in female CD rats of a diol epoxide metabolite
of fluoranthene, a commonly occurring environmental pollutant.  Carcinogenesis 16:1433-1435.

Hecht, SS; Rivenson, A; Amin, S; et al. (1996) Mammary carcinogenicity of diol epoxide metabolites of
benzo[j]fluoranthene in female CD rats.  Cancer Lett 106:251-255.

Hermann,  M. (1981) Synergistic effects of individual polycyclic aromatic hydrocarbons on the mutagenicity of their
mixtures.  Mutat Res 90:399-409.

Hill, WT;  Stanger, DW; Pizzo, A; et al. (1951) Inhibition of 9,10-dimethyl-l,2-benzanthracene skin Carcinogenesis
in mice by polycyclic hydrocarbons. Cancer Res 11:892-897.

Homburger, F; Treger, A. (1970) Transplantation technique for acceleration of Carcinogenesis by benz[a]anthracene
or 3,4,9,10-dibenzpyrene. J Natl Cancer Inst 44:357-360.

Homburger, F; Treger, A; Boger, E. (1971) Inhibition of murine subcutaneous and intravenous
benzo[rst]pentaphene. Carcinogenesis by sweet orange oils and d-limonene. Oncology 25:1-10.

Jerina, DM; Sayer, JM; Yagi, H; et al.  (1981) Highly tumorigenic bay-region diol epoxides from the weak
carcinogenbenzo[c]phenanthrene.  Adv Exp MedBiol 136 Pt A:501-523.

Johnson, S. (1968) Effect of thymectomy on the induction of skin tumours by dibenzanthracene, and of breast
tumours by dimethylbenzanthracene in mice of the IF strain. Br J Cancer 22:755-761.

Klein, M.  (1952) Effect of croton oil on induction of tumors by 1,2-benzanthracene, deoxychloric or low doses of
20-methylcholanthrene in mice. J Natl Cancer Inst 13:333-341.
                                              B-5            DRAFT - DO NOT CITE OR QUOTE

-------
Klein, M. (1960) A comparison of the initiating and promoting actions of 9,10-dimethyl-l,2-benzanthracene and
1,2,5,6-dibenzanthracene in skin tumorigenesis. Cancer Res 20:1179-1183.

Klein, M. (1963) Susceptibility of strain B6AF1/J hybrid infant mice to tumorigenesis with 1,2-benzanthracene,
deoxycholic acid, and 3-methylcholanthrene. II. Tumours called forth by painting the skin with dibenzpyrene.
Cancer Res 23:1701-1707.

Kouri, RE; Connolly, GM; Nebert, DW; et al. (1983) Association between susceptibility to dibenzanthracene
induced fibrosarcoma formation and the Ah locus.  Int J Cancer 32:765-768.

Lacassagne, A; Buu-Hoi, NP; Zajdela, F; et al. (1968) The true dibenzo[a,l]pyrene, a new, potent carcinogen.
Naturwissenschaften 55:43.

LaVoie, EJ; Tulley L; Bedenko, V; et al. (1980) Mutagenicity, tumor initiating activity, and metabolism of tricyclic
polynuclear aromatic hydrocarbons.  In: Bjorseth, A; Dennis, AJ, eds.  Polynuclear aromatic hydrocarbons:
chemistry and biological effects. Columbus, OH: Battelle Press, pp. 1041-1057.

LaVoie, EJ; Tulley-Freiler, L; Bedenko, V; et al. (1981) Comparative studies on the tumor initiating activity and
metabolism of methylfluorenes and methylbenzofluorenes.  In: Cooke, M; Dennis, AJ, eds. Chemical analysis and
biological fate: polynuclear hydrocarbons. Columbus, OH: Battelle Press, pp. 417-427.

LaVoie, EJ; Coleman, DT; Tonne, RL; et al. (1983) Mutagenicity, tumor initiating activity and metabolism of
methylated anthracenes. In: Cooke, M;  Dennis, AJ, eds. Proceedings of the Seventh International Symposium.
Columbus, OH: Battelle Press, pp. 785-798.

LaVoie, EJ; Cai, ZW; Meegalla, RL; et  al. (1993a) Evaluation of the tumor-initiating activity of 4-, 5-, 6-, and
7-fluorobenzo[b]fluoranthene in mouse  skin.  Chem Bio Interact 89:129-139.

LaVoie, EJ; He, ZM; Meegalla, RL; et al. (1993b) Exceptional tumor-initiating activity of 4-fluorobenzo[j]-
fluoranthene on mouse skin: comparison with benzo[j]-fluoranthene, 10-fluoro-benzo[j]fluoranthene,
benzo[a]pyrene, dibenzo[a,l]pyrene and 7,12-dimethylbenz[a]anthracene.  Cancer Lett 70:7-14.

LaVoie, EJ; He, ZM; Wu, Y; et al. (1994) Tumorigenic activity of the 4,5- and 9,10-dihydrodiols of
benzo[j]fluoranthene and their syn- and anti-diol epoxides in newborn mice.  Cancer Res 54:962-968.

Levin, W; Wood, AW; Chang, RL; et al. (1978) Evidence for bay region activation of chrysene 1,2-dihydrodiol to
an ultimate carcinogen.  Cancer Res 38:1831-1834.

Levin, W; Wood, AW; Chang, RL; et al. (1980) Exceptionally high tumor-initiating activity of
benzo[c]phenanthrene bay-region diol-epoxides on mouse skin. Cancer Res 40:3910-3914.

Levin, W; Chang, RL; Wood, AW; et al. (1984) High stereoselectivity among the optical isomers of the
diastereomeric bay-region diolepoxides of benz [a] anthracene in the expression of tumorigenic activity in murine
tumor models.  Cancer Res 44:929-933.

Levin, W; Chang, RL; Wood, AW; et al. (1986) Tumorigenicity of optical isomers of the diastereomeric bay-region
3,4-diol-1,2-epoxides of benzo [c]phenanthrene in murine tumor models. Cancer Res 46:2257-2261.

Lijinsky, W; Garcia, H.  (1972) Skin carcinogenesis tests of hydrogenated derivatives of anthanthrene and other
polynuclear hydrocarbons. Z Krebsforsch 77:226-230.

Lijinsky, WH; Garcia, B; Terrracini, B.  (1965) Tumorigenic activity of hydrogenated derivatives of
dibenz[a,h]anthracene. J Natl Cancer Inst 34:1-6.

Lijinsky, W; Garcia, H;  Saffiotti, U. (1970) Structure-activity relationships among some polynuclear hydrocarbons
and their hydrogenated derivatives. J Natl Cancer Inst 44:641-649.

Lorenz, E; Stewart, HL. (1947) Tumors of the alimentary tract induced in mice by feeding olive oil emulsions
containing carcinogenic hydrocarbons.  J Natl Cancer Inst 7:227-238.
                                              B-6           DRAFT - DO NOT CITE OR QUOTE

-------
Lorenz, E; Stewart, HL. (1948) Tumors of alimentary tract in mice fed carcinogenic hydrocarbons in mineral-oil
emulsions.  J Natl Cancer Inst 9:173-180.

Lubet, RA; Connolly, GM; Nebert, DW; et al. (1983) Dibenz[a,h]anthracene-induced subcutaneous tumors in mice.
Strain sensitivity and the role of carcinogen metabolism. Carcinogenesis 4:513-517.

Malament, DS; Shklar, G. (1981) Inhibition of DMBA carcinogenesis of hamster buccal pouch by phenanthrene and
dimethylnaphthalene. Carcinogenesis 2:723-729.

Mass, MJ; Abu-Shakra, A; Roop, BC; et al. (1996) Benzo[b]fluoranthene: tumorigenicity in strain A/J mouse lungs,
DNA adducts and mutations in the Ki-ras oncogene. Carcinogenesis 17:1701-1704.

Nakatsuru, Y; Wakabayashi, K; Fujii-Kuriyama, Y; et al. (2004) Dibenzo[a,l]pyrene-induced genotoxic and
carcinogenic responses are dramatically suppressed in aryl hydrocarbon receptor-deficient mice.  Int J Cancer
112:179-183.

Nesnow, S; Gold, A; Sangaiah, R; et al. (1993) Mouse skin tumor-initiating activity of benz[j]aceanthrylene in
SENCAR mice.  Cancer Lett 73:73-76.

Nesnow, S; Ross, JA; Nelson, G; et al. (1994) Cyclopenta[cd]pyrene-induced tumorigenicity, Ki-ras codon 12
mutations and DNA adducts in strain A/J mouse lung. Carcinogenesis 15:601-606.

O'Gara, RW; Kelly, MG; Brown, J; et al. (1965) Induction of tumors in mice given a minute single dose of
dibenz[a,h]anthracene or 3-methylcholanthrene as newborns: a dose-response study.  J Natl Cancer Inst 35(6): 1027-
1042.

Platt, KL; Pfeiffer, EH;  Glatt, HR; et al. (1983) Bacterial mutagenicity and carcinogenicity of potential metabolites
of dibenz[a,h]anthracene.  J Cancer Res Clin Oncol 105: A23.

Platt, KL; Pfeiffer, E; Petrovic, P; et al.  (1990) Comparative tumorigenicity of picene and dibenz[a,h]anthracene in
the mouse.  Carcinogenesis 11:1721-1726.

Platt, KL; Dienes, HP; Tommasone, M; et al. (2004) Tumor formation in the neonatal mouse bioassay indicates that
the potent carcinogen dibenzo[def,p]chrysene (dibenzo[a,l]pyrene) is activated in vivo via its
trans-11,12-dihydrodiol. Chem Biol Interact  148:27-36.

Pollia, JA. (1939) Investigations on the possible carcinogenic effect of anthracene and chrysene and some of their
compounds. I. The effect of skin painting on the skin of mice.  J IndHyg Toxicol 21(8):219-220.

Pollia, JA. (1941) Investigation on the possible carcinogenic effect of anthracene and chrysene and some of their
compounds. II. The effect of subcutaneous injection in rats. J Ind Hyg Toxicol 23:449-451.

Prahalad, AK; Ross, JA; Nelson, GB; et al. (1997) Dibenzo[a,l]pyrene-induced DNA adduction, tumorigenicity, and
Ki-ras oncogene  mutations in strain A/J mouse lung. Carcinogenesis 18:1955-1963.

Ranadive, KJ; Karande, KA. (1963) Studies on 1,2,5,6-dibenzanthracene-induced mammary carcinogenesis in mice.
BrJ  Cancer 17:272-280.

Rice, JE;  Coleman, DT; Hosted, TJJ; et al.  (1985) On the metabolism, mutagenicity,  and tumor-initiating activity of
indeno[l,2,3-cd]pyrene.  In: Polynuclear aromatic hydrocarbons: mechanisms, methods and metabolism. Columbus,
OH:  Battelle Press, pp. 1097-1109.

Rice, JE; Hosted, TJ, Jr.; DeFloria, MC; et al. (1986) Tumor-initiating activity of major in vivo metabolites of
indeno[l,2,3-cd]pyrene  on mouse skin.  Carcinogenesis 7:1761-1764.

Rice, JE;  Weyand, EH; Geddie, NG; et al. (1987) Identification of tumorigenic metabolites of benzo|j]fluoranthene
formed in vivo in mouse skin. Cancer Res 47:6166-6170.
                                              B-7            DRAFT - DO NOT CITE OR QUOTE

-------
Rice, JE; Weyand, EH; Burrill, C; et al. (1990) Fluorene probes for investigating the mechanism of activation of
indeno[l,2,3-cd]pyrene to atumorigenic agent.  Carcinogenesis 11:1971-1974.

Riegel, B; Watman, WB; Hill, WT. (1951) Delay of methylcholanthrene skin carcinogenesis in mice by
1,2,5,6-dibenzofluorene.  Cancer Res 11:301-306.

Salaman, MH; Roe, FJC. (1956) Further tests for tumour-initiating activity: N,N-di(2-chloroethyl)-
paminophenylbutic acid (CB1348) as an initiator of skin tumour formation in the mouse.  Br J Cancer 10:363-378.

Sardella, DJ; Boger, E; Ghoshal, PK. (1981) Active sites in hexacyclic carcinogens probed by the fluorene
substitution methodology. In: Polynuclear aromatic hydrocarbons: chemical analysis and biological fate. Columbus,
OH: Battelle Press, pp. 529-538.

Schmahl, D. (1955) [Testing of naphthalene and anthracene for carcinogenic effects in rats.] Z Krebsforsch 60:697-
710. (German)

Schoental, R. (1959) Carcinogenic activity of 3:4:9:10-dibenzopyrene.  ActaUnio Int Contra Cancrum 15(1):216-
219.

Schoket, B; Hewer, A; Grover, PL; et al. (1988) Covalent binding of components of coal-tar, creosote and bitumen
to the DNA of the skin and lungs of mice following topical application. Carcinogenesis 9:1253-1258.

Scribner, JD. (1973) Brief communication: tumor initiation by apparently noncarcinogenic poly cyclic aromatic
hydrocarbons. J Natl Cancer Inst 50:1717-1719.

Sellakumar, A; Shubik, P. (1974) Carcinogenicity of different polycyclic hydrocarbons in the respiratory tract of
hamsters.  JNatl Cancer Inst 53:1713-1719.

Shear, MJ. (1938) Studies in carcinogenesis. V. Methyl derivatives of 1,2-benzanthracene. Am J Cancer 33:499-
537.

Shear, MJ; Leiter, J. (1941) Studies in carcinogenesis. XVI. Production of subcutaneous tumors in mice by
miscellaneous polycyclic compounds. J Natl Cancer Inst 2:241-259.

Siebert, D; Marquardt, H; Friesel, H;  et al. (1981) Polycyclic aromatic hydrocarbons and possible metabolites:
convertogenic activity in yeast and tumor initiating activity in mouse skin. J Cancer Res Clin Oncol 102:127-139.

Slaga, TJ; Gleason, GL; Mills, C; et al.  (1980) Comparison of the tumour-initiating activities of dihydrodiols and
diol-epoxides of various polycyclic aromatic hydrocarbons.  Cancer Res 40:1981-1984.

Snell, KC; Stewart, HL. (1962) Pulmonary adenomatosis induced in DBA/2 mice by oral administration of
dibenz[a,h]anthracene. J Natl Cancer Inst 28:1043-1049.

Snell, KC; Stewart, HL. (1963) Induction of pulmonary adenomatoses in DBA/2 mice by the oral administration of
dibenz[a,h]anthracene. ActaUnio Int Contra Cancrum  19:692-694.

Stanton, MF; Miller, E; Wrench,  C; et al. (1972) Experimental induction of epidermoid carcinoma in the lungs of
rats by cigarette smoke condensate. J Natl Cancer Inst 49:867-877.

Steiner, PE; Edgcomb, JH. (1952) Carcinogenicity  of 1,2-benzanthracene. Cancer Res  12:657-659.

Steiner, PF; Falk, HL. (1951) Summation and inhibition effects of weak and strong carcinogenic hydrocarbons:
l:2-Benzanthracene, chrysene, l:2:5:6-dibenzanthracene, and 20-methylcholanthrene. Cancer Res 11:56-63.

Stenbk, F; Sellakumar, A. (1974) Lung tumor induction by dibenz[a,i]pyrene in the  Syrian golden hamster.  Z
Krebsforsch 82:175-182.

Stevenson, JL; VonHaam, E. (1965)  Carcinogenicity of benz[a]anthracene andbenzo[c]phenanthrene derivatives.
Am Ind Hyg Assoc J 26:475-478.
                                              B-8            DRAFT - DO NOT CITE OR QUOTE

-------
Tawfic, HN. (1965) Studies on ear duct tumors in rats. Part II. Inhibitory effect of methylcholanthrene and
1,2-benzanthracene on tumor formation by 4-dimethylamino-stilbene.  ActaPathol Jpn 15:255-260.

VanDuuren, BL; Langseth, L; Goldschmidt, BM. (1967) Carcinogenicity of epoxides, lactones andperoxy
compounds. VI. Structure and carcinogenic activity. J Natl Cancer Inst 39:1217-1227.

Van Duuren, BL; Sivak, A; Langseth, L; et al. (1968) Initiators and promoters in tobacco carcinogenesis.  Natl
Cancer Inst Monogr 28:173-180.

VanDuuren, BL; Sivak, A; Goldschmidt, BM; et al. (1970) Initiating activity of aromatic hydrocarbons in two-stage
carcinogenesis.  JNatl Cancer Inst 44:1167-1173.

Vulimiri, SV; Baer-Dubowska, W; Harvey, RG; et al. (1999)  Analysis of highly polar DNA adducts formed in
SENCAR mouse epidermis following topical application of dibenz[a,j]anthracene. Chem Res Toxicol 12:60-67.

Wang, JS; Busby, WF, Jr. (1993) Induction of lung and liver tumors by fluoranthene in a preweanling CD-I mouse
bioassay. Carcinogenesis 14:1871-1874.

Waravdekar, SS; Ranadive, KJ. (1958) Biologic testing of 3,4,9,10-dibenzpyrene. J Natl Cancer Inst 21:1151-1159.

Weyand, EH; Amin, S; Huie, K; et al. (1989) Effects of fluorene substitution on the DNA binding and
tumorigenicity of benzo[b]fluoranthene in mouse epidermis.  ChemBiol Interact 71:279-290.

Weyand, EH; Patel, S; LaVoie, EJ; et al.  (1990) Relative tumor initiating activity of benzo[a]fluoranthene,
benzo[b]fluoranthene, naphtho[l,2-b]fluoranthene and naphtho[2,l-a]fluoranthene on mouse skin.  Cancer Lett
52:229-233.

Weyand, EH; Cai, ZW; Wu, Y; et al. (1993) Detection of the  major DNA adducts of benzo[b]fluoranthene in mouse
skin: role of phenolic dihydrodiols. Chem Res Toxicol 6:568-577.

White, FR;  Eschenbrenner, AB. (1945) Note on the occurrence of hepatomas in rats following the ingestion of
1,2-benzoanthracene. J Natl Cancer Inst 6:19-21.

Wislocki, PG; Buening, MK; Levin, W; et al. (1979) Tumorigenicity of the diastereomeric benz[a]anthracene
3,4-diol-l,2-epoxides and the (+)-and (-)-enantiomers of benz[a]anthracene 3,4-dihydrodiol in newborn mice.  J Natl
Cancer Inst 63:201-204.

Wodinsky, I; Helinski, A; Kensler, CJ. (1964) Susceptibility of Syrian hamsters to induction of fibrosarcomas with a
single injection of 3,4,9,10-dibenzpyrene. Nature 203:308-309.

Wood, AW; Chang, RL; Levin, W; et al. (1979) Mutagenicity and tumorigenicity of phenanthrene and chrysene
epoxides and diol epoxides.  Cancer Res  39:4069-4077.

Wynder, EL; Hoffmann, D. (1961) Carcinogenicity of dibenzo[a,l]pyrene.  Nature 192:1092-1093.

Zajdela, F; Perin-Roussel, O; Saguem, S. (1987) Marked differences between mutagenicity in Salmonella and
tumour-initiating activities of dibenzo[a,e]fluoranthene proximate metabolites; initiation inhibiting activity of
norharman. Carcinogenesis 8:461-464.
                                              B-9            DRAFT - DO NOT CITE OR QUOTE

-------
B.2.  BIBLIOGRAPHY OF STUDIES ON CANCER-RELATED ENDPOINTS

WITHOUT BENZO[A]PYRENE

Abe, S; Sasaki, M. (1977) Studies on chromosomal aberrations and sister chromatid exchanges induced by
chemicals. Proc Jpn Acad 53:46-49.

Agarwal, SK; Sayer, JM; Yeh, HJC; et al. (1987) Chemical characterization of DNA adducts derived from the
configurationally isomeric benzo[c]phenanthrene-3,4-diol 1,2-epoxides. J Am Chem Soc 109:2497-2504.

Agarwal, R; Canella, KA; Yagi, H; et al. (1996) Benzo[c]phenanthrene-DNA adducts in mouse epidermis in relation
to the tumorigenicities of four configurationally isomeric 3,4-dihydrodiol 1,2-epoxides. Chem Res Toxicol 9:586-
592.

Agarwal, R; Coffing, SL; Baird, WM; et al. (1997) Metabolic activation of benzo[g]chrysene in the human
mammary carcinoma cell line MCF-7. Cancer Res 57:415^4-19.

Amin, S; Desai, D; Hecht, SS. (1993) Tumor-initiating activity on mouse skin of bay region diol-epoxides of
5,6-dimethylchrysene andbenzo[c]phenanthrene. Carcinogenesis 14:2033-2037.

Arif, JM; Gupta, RC. (1997) Microsome-mediated bioactivation of dibenzo[a,l]pyrene and identification of DNA
adducts by 32P-postlabeling. Carcinogenesis 18:1999-2007.

Arif, JM; Smith, WA; Gupta, RC. (1999) DNA adduct formation and persistence in rat tissues  following exposure to
the mammary carcinogen dibenzo[a,l]pyrene. Carcinogenesis 20:1147-1150.

Ayrton, AD; McFarlane, M; Walker, R; et al. (1990) Induction of the P-450 I family of proteins by polycyclic
aromatic hydrocarbons: possible relationship to their carcinogenicity. Toxicology 60:173-186.

Babson, JR; Russo-Rodriguez, SE; Rastetter, WH; et al.  (1986) In vitro DNA-binding of microsomally-activated
fluoranthene: evidence that  the major product is a fluoranthene N2-deoxyguanosine adduct.  Carcinogenesis 7:859-
865.

Babson, JR; Russo-Rodriguez, SE; Wattley, RV; et al. (1986) Microsomal activation of fluoranthene to mutagenic
metabolites. Toxicol Appl Pharmacol 85:355-366.

Baer-Dubowska, W; Nair, RV;  Cortez, C; et al. (1995) Covalent DNA adducts formed in mouse epidermis from
dibenz[a,j]anthracene: evidence for the formation of polar adducts. Chem Res Toxicol 8:292-301.

Ball, LM; Warren, SH; Sangaiah, R; et al. (1989) Bacterial mutagenicity of new cyclopenta-fused cata-annelated
polycyclic aromatic hydrocarbons, and identification of the major metabolites of benz|j]acephenanthrylene formed
by aroclor-treated rat liver microsomes. Mutat Res 224:115-25.

Barfknecht, TR; Andon, BM; Thilly, WG; et al. (1981) Soot and mutation in bacteria and human cells. In: Cooke,
M; Dennis, AJ, eds. Chemical analysis and biological fate: polynuclear aromatic hydrocarbons, pp. 231-242.

Barrai, I; Barale, R; Scapoli, C; et al. (1992) The analysis of the joint effect of substances on reversion systems and
the assessment of antimutagenicity. Mutat Res 267:173-182.

Barratt, RW; Tatum, EL (1958) Carcinogenic mutagens. Ann NY Acad Sci 71:1072-1084.

Bartczak, AW; Sangaiah, S; Ball, LM; et al. (1987) Synthesis and bacterial mutagenicity of the cyclopenta oxides of
the four cyclopenta-fused isomers of benzanthracene.  Mutagenesis 2:101-105.

Easier, A; Herbold, B; Peter, S; et al. (1977) Mutagenicity  of polycyclic hydrocarbons. II. Monitoring genetical
hazards of chrysene in vitro and in vivo. Mutat Res 48:249-254.

Baum, M; Amin,  S; Guengerich, FP; et al. (2001) Metabolic activation of benzo[c]phenanthrene by cytochrome
P450 enzymes in human liver and lung.  Chem Res Toxicol 14:686-693.


                                            B-10           DRAFT - DO NOT  CITE OR QUOTE

-------
Beach, AC; Gupta, RC. (1991) Analysis of cyclopenta(CP)-fused and 'pseudo-CP' polycyclic aromatic hydrocarbon
(PAH)-DNA adducts by 32P-postlabeling. Proc Am Assoc Cancer Res 32:98.

Beach, AC; Gupta, RC. (1994) DNA adducts of the ubiquitous environmental contaminant cyclopenta[cd]pyrene.
Carcinogenesis 15:1065-1072.

Beach, AC; Agarwal, SC; Lamberg, GR; et al.  (1993) Reaction of cyclopenta[c,d]pyrene-3,4-epoxide with DNA and
desoxynucleotides. Carcinogenesis 14:767-771.

Bos, RP; Prinsen, WJ; van Rooy, JG; et al. (1987) Fluoranthene,  a volatile mutagenic compound, present in creosote
and coal tar. Mutat Res 187:119-125.

Boutwell, RK. (1989) Model systems for defining initiation, promotion, and progression of skin neoplasms.  Prog
ClinBiol Res 298:3-15.

Bryant, MF; Kwanyuen, P; Atwater, AL; et al.  (1991) Cytogenetic effects of benzo-b-fluoranthene in Sprague-
Dawley rat peripheral blood lymphocytes after in vivo exposure (Abstract 33). Environ Mol Mutag 17 (Suppl
19):13.

Bu-Abbas, A; loannides, C; Walker, R. (1994) Evaluation of the  antimutagenic potential of anthracene: in vitro and
ex vivo studies.  Mutat Res 309:101-107.

Budunova, IV; Mittleman, LA; Safaev, RD; et  al. (1993) The carcinogen benzo[e]pyrene is metabolized by DM15
cells without an uncoupling effect on their gap junctions. Cell Biol Toxicol 9:131-140.

Carmichael, PL; Platt, KL;  She, MN; et al. (1993) Evidence for the involvement of a bis-diol-epoxide in the
metabolic activation of dibenz[a,h]anthracene to DNA-binding species in mouse skin. Cancer Res 53:944-948.

Gary, PD; Turner, CH; Cooper, CS; et al. (1980) Metabolic activation of benz[a]anthracene in hamster embryo cells:
the structure of a guanosine-anti-B A-8,9-diol 10,11-oxide adduct. Carcinogenesis 1:505-512.

Casale, GP; Higginbotham, S; Johansson, SL; et al. (1997) Inflammatory response of mouse skin exposed to the
very potent carcinogen dibenzo[a,l]pyrene: a model for tumor promotion. Fundam Appl Toxicol 36:71-78.

Casto, BC. (1973) Enhancement of adenovirus transformation by treatment of hamsters with ultraviolet irradiation,
DNA base analogs, and dibenz[a,h]anthracene. Cancer Res 33:402-407.

Chakravarti, D; Mailander, P; Franzen, J; et al. (1998) Detection  of dibenzo[a,l]pyrene-induced H-ras codon 61
mutant genes in preneoplastic Sencar mouse skin using a new PCR-RFLP method.  Oncogene 16:3203-3210.

Chakravarti, D; Mailander, PC; Cavalieri, EL; et al. (2000) Evidence that error-prone DNA repair converts
dibenzo[a,l]pyrene-induced depurinating lesions into mutations: formation, clonal proliferation and regression of
initiated cells carrying H-ras oncogene mutations in early preneoplasia.  Mutat Res 456:17-32.

Chaloupka, K; Santostefano, M; Goldfarb, IS; et al. (1994) Aryl hydrocarbon (Ah) receptor-independent induction
of Cypla2 gene expression by acenaphthylene  and related compounds inB6C3Fl mice. Carcinogenesis 15:2835-
2840.

Chiarelli, MP; Chang, HF; Olsen, KW; et al. (2003) Structural differentiation of diastereomeric
benzo[ghi]fluoranthene adducts of deoxyadenosine by matrix-assisted laser desorption/ionization time-of-flight
mass spectrometry and postsource decay. Chem Res Toxicol 16:1236-1241.

Chroust, K; Jowett, T; Farid-Wajidi, MF; et al. (2001) Activation or detoxification of mutagenic and carcinogenic
compounds in transgenic Drosophila expressing human glutathione S-transferase.  Mutat Res 498:169-179.

Cizmas, L; Zhou, GD; Safe, SH; et al. (2004) Comparative in vitro and in vivo genotoxicities of
7H-benzo[c]fluorene, manufactured gas plant residue (MGP), and MGP fractions. Environ Mol Mutagen 43:159-
168.
                                             B-11           DRAFT - DO NOT CITE OR QUOTE

-------
Clayson, DB; Iverson, F; Nera, EA; et al. (1990) The significance of induced forestomach tumors.  Annu Rev
Pharmacol Toxicol 30:441^163.

Collin, G; H"ke, H. (1985) Anthracene.  In: Elvers, B; Hawkins, S; Schulz, G, eds. Ullmann's encyclopedia of
industrial chemistry. 5th ed.,Volume A2. Weinheim, Verlagsgesellschaft, pp. 343-345.

Collins, JF; Brown, JP; Alexeeff, GV; et al. (1998) Potency equivalency factors for some polycyclic aromatic
hydrocarbons and polycyclic aromatic hydrocarbon derivatives. Regul Toxicol Pharmacol 28:45-54.

Coombs, MM; Bhatt, TS, eds. (1987) Polycyclic aromatic compounds structurally related to steroids. In:
Cyclopenta[a]phenanthrenes. New York, NY: Cambridge University Press, pp. 132-210.

Coombs, MM; Dixon, C; Kissonerghis, AM; et al. (1976) Evaluation of the mutagenicity of compounds of known
carcinogenicity, belonging to the benz[a]anthracene, chrysene, and cyclopenta[a]phenanthrene series, using Ames'
test.  Cancer Res  36:4525-4529.

Dai, Q. (1980) Researches on chemical carcinogens and mechanism of chemical carcinogenesis. Dl-region theory: a
quantitative molecular orbital model of carcinogenic activity for polycyclic aromatic hydrocarbons. Sci Sin 23:453-
470.

Danz, M; Hartmann, A; Blaszyk, H.  (1998) Mitogenic short-term effects on hepatocytes and adrenocortical cells:
phenobarbital and reserpine compared to carcinogenic and non-carcinogenic fluorene derivatives. Exp Toxicol
Pathol 50:416-424.

Devanesan, P; Ariese, F; Jankowiak, R; et al. (1999) A novel method for the isolation and identification of stable
DNA adducts formed by dibenzo[a,l]pyrene and dibenzo[a,l]pyrene 11,12-dihydrodiol 13,14-epoxides in vitro.
ChemRes Toxicol 12:796-801.

DeVito, MJ; Maier, WE; Diliberto, JJ; et al. (1993) Comparative ability of various PCBs, PCDFs, and TCDD to
induce cytochrome P450 1A1 and 1A2 activity following 4 weeks of treatment. Fundam Appl Toxicol 20:125-130.

Diamond, L; Cherian, K; Harvey, RG; et al. (1984) Mutagenic activity of methyl- and fluoro-substituted derivatives
of polycyclic aromatic hydrocarbons in a human hepatoma (HepG2) cell-mediated assay.  Mutat Res 136:65-72.

Dipple, A; Pigott, MA; Agarwal, SK; et al. (1987) Optically active  benzo[c]phenanthrene diol epoxides bind
extensively to adenine in DNA.  Nature 327:535-536.

Dong, S; Fu, PP;  Shirsat, RN; et al. (2002) UVA light-induced DNA cleavage by isomeric
methylbenz[a]anthracenes.  Chem Res Toxicol 15:400-407.

Einolf, HJ; Amin, S; Yagi, H; et al. (1996) Benzo[c]phenanthrene is activated to DNA-binding diol epoxides in the
human mammary carcinoma cell line MCF-7 but only limited activation occurs in mouse skin.  Carcinogenesis
17:2237-2244.

Einolf, HJ; Story, WT; Marcus, CB;  et al. (1997) Role of cytochrome P450 enzyme induction in the metabolic
activation of benzo[c]phenanthrene in human cell lines and mouse epidermis. Chem Res Toxicol 10:609-617.

Ensell, MX; Hubbs, A; Zhou, G; et al. (1999) Neoplastic potential of rat tracheal epithelial cell lines induced by
1-nitropyrene and dibenzo[a,i]pyrene. Mutat Res 444:193-199.

Ensell, MX; Whong, WZ; Heng, ZC; et al. (1998) In vitro and in vivo transformation in rat tracheal epithelial cells
exposed to diesel emission particles and related compounds. Mutat Res 412:283-291.

Fahmy, OG; Fahmy, MJ. (1973) Oxidative activation of benz [a] anthracene and methylated derivatives in
mutagenesis and carcinogenesis.  Cancer Res 33:2354-2361.

Fuchs, J; Mlcoch, J; Platt, KL; et al. (1993) Characterization of highly polar bis-dihydrodiol epoxide~DNA adducts
formed after metabolic activation of dibenz[a,h]anthracene.  Carcinogenesis 14:863-867.
                                             B-12           DRAFT - DO NOT CITE OR QUOTE

-------
Gatehouse, D. (1980) Mutagenicity of 1,2 ring-fused acenaphthenes against S. typhimurium TA1537 and TA1538:
structure-activity relationship.  Mutat Res 78:121-135.

Giles, AS; Seidel, A; Phillips, DH. (1995) In vitro reaction with DN A of the fjord-region diolepoxides of
benzo[g]chrysene and benzo[c]phenanthrene as studied by 32P-postlabeling.  Chem Res Toxicol 8:591-599.

Giles, AS; Seidel, A; Phillips, DH. (1996) Covalent DNA adducts formed in mouse epidermis by benzo[g]chrysene.
Carcinogenesis 17:1331-1336.

Glatt, H; Seidel, A; Bochnitschek, W; et al. (1986) Mutagenic and cell-transforming activities of triol-epoxides as
compared to other chrysene metabolites.  Cancer Res 46:4556^565.

Glatt, H; Abu-Shqara, E; Harvey, RG; et al. (1994) Mutagenicity of K-region oxides and imines of chrysene,
benzo[c]phenanthrene and benzo[g]chrysene in Salmonella typhimurium. Mutat Res 308:135-141.

Gold, A; Nesnow, S; Moore, M; et al. (1980) Mutagenesis and morphological transformation of mammalian cells by
a non-bay-region polycyclic cyclopenta[cd]pyrene and its 3,4-oxide. Cancer Res 40:4482-4484.

Gorelick, NJ; Wogan, GN. (1989) Fluoranthene-DNA adducts: identification and quantification by an
HPLC-32P-postlabeling method. Carcinogenesis 10:1567-1577.

Gorelick, NJ; Hutchins, DA; Tannenbaum, SR; et al. (1989) Formation of DNA and hemoglobin adducts of
fluoranthene after single and multiple exposures. Carcinogenesis 10:1579-1587.

Goshman,  LM; Heidelberger, C. (1967) Binding of tritium-labeled polycyclic hydrocarbons to DNA of mouse skin.
Cancer Res 27:1678-1688.

Hall, M; Parker, DK; Hewer, AJ; et al. (1988) Further metabolism of diol-epoxides of chrysene and
dibenz[a,c]anthracene to DNA binding species as evidenced by 32P-postlabeling analysis.  Carcinogenesis 9:865-
868.

Harvey, RG. (1996) Mechanisms of Carcinogenesis of polycyclic aromatic compounds. Polycycl Aromat Compd
9:1-23.

Herner,  HA; Trosko, JE; Masten, SJ. (2001) The epigenetic toxicity of pyrene and related ozonation byproducts
containing an aldehyde functional group. Environ Sci Technol 35:3576-3583.

Hermann, M. (1981) Synergistic effects of individual polycyclic aromatic hydrocarbons on the mutagenicity of their
mixtures. Mutat Res 90:399-409.

Hewer, A; Cooper,  CS; Ribeiro, O;  et al.  (1981) The metabolic and activation of dibenz[a,c]anthracene.
Carcinogenesis 2:1345-1352.

Holme,  JA; Bjorge, C; Soderlund, EJ; et al. (1993) Genotoxic effects of cyclopenta-fused polycyclic aromatic
hydrocarbons in isolated rat hepatocytes and rabbit lung cells.  Carcinogenesis 14:1125-1131.

Huberman, E; Kuroki, T; Marquardt, H; et al. (1972) Transformation of hamster embryo cells by epoxides and other
derivatives of polycyclic hydrocarbons. Cancer Res 32:1391-1396.

Hughes, NC; Phillips, DH. (1993) 32P-postlabeling analysis of the covalent binding of benzo[ghi]perylene to DNA
in vivo and in vitro. Carcinogenesis 14:127-133.

Ishidate, M;  Odashima, S.  (1977) Chromosome tests with 134 compounds on Chinese hamster cells in vitro: a
screening for chemical carcinogens.  Mutat Res 48:337-354.

Isu, Y; Nagashima, U; Aoyama, T; et al. (1996) Development of neural network simulator for structure--activity
correlation of molecules (NECO). Prediction of endo/exo substitution of norbornane derivatives and of carcinogenic
activity  of PAHs from 13C-NMR shifts.  J Chem Inf Comput Sci 36:286-293.
                                             B-13           DRAFT - DO NOT CITE OR QUOTE

-------
Jankowiak, R; Ariese, F; Hewer, A; et al. (1998) Structure, conformations, and repair of DNA adducts from
dibenzo[a,l]pyrene: 32P-postlabeling and fluorescence studies. Chem Res Toxicol 11:674-685.

Jerina, DM; Lehr, RE. (1977) The bay-region theory: a quantum mechanical approach to aromatic hydrocarbon-
induced carcinogenicity, pp. 709-720.

Jerina, DM; Yagi, H; Lehr, RE; et al. (1978) The bay-region theory of carcinogenesis by polycyclic aromatic
hydrocarbons, pp. 173-188.

Juhasz, AL; Stanley, GA; Britz, ML. (2000) Microbial degradation and detoxification of high molecular weight
polycyclic aromatic hydrocarbons by Stenotrophomonas maltophilia strain VUN 10,003. Lett Appl Microbiol
30:396-401.

Kemena, A; Norpoth, KH; Jacob, J. (1988) Differential induction of the monooxygenase isoenzymes in mouse liver
microsomes by polycyclic aromatic hydrocarbons. In: Cooke, M; Dennis, AJ, eds. Polynuclear aromatic
hydrocarbons: a decade of progress. Columbus, OH: Battelle Press, pp. 449-460.

Keohavong, P; Melacrinos, A; Shukla, R. (1995) In vitro mutational spectrum of cyclopenta[cd]pyrene in the human
HPRT gene. Carcinogenesis 16:855-860.

King, LC; Adams, L; Allison, J; et al. (1999) A quantitative comparison of dibenzo[a,l]pyrene-DNA adduct
formation by recombinant human cytochrome P450 microsomes. Mol Carcinog 26:74-82.

Knaap, AGAC; Goze, C; Simons, JWIM (1981) Mutagenic activity of seven coded samples in V79 Chinese hamster
cells.  Prog Mutat Res 1:608-613.

Kumar,  S; Kole, PL; Sikka, HC. (1990) Mutagenicity of dibenz[a,c]anthracene and its derivatives in Salmonella
typhimurium TA100.  Mutat Res 242:337-343.

Laryea,  A; Cosman, M; Lin, JM; et al. (1995) Direct synthesis and characterization of site-specific adenosyl adducts
derived  from the binding of a 3,4-dihydroxy-l,2-epoxybenzo[c]phenanthrene stereoisomer to an 11-mer
oligodeoxyribonucleotide.  Chem Res Toxicol 8:444-454.

Lasley, J; Curti, S; Ross, J; et al. (1991) Morphological cell transformation and DNA adduction by
benz[j]aceanthrylene and its presumptive reaction metabolites in C3H10T1/2CL8 cells.  Adv Exp Med Biol
283:759-762.

LaVoie, EJ; Hecht, SS; Amin, S; et al. (1980) Identification of mutagenic dihydrodiols as metabolites of
benzo[j]fluoranthene and benzo[k]fluoranthene. Cancer Res 40:4528-4532.

LaVoie, EJ; Tulley, L; Bedenko, V; et al. (1981) Mutagenicity of methylated fluorenes and benzofluorenes. Mutat
Res 91:167-176.

LaVoie, EJ; Hecht, SS; Bedenko, V; et al. (1982) Identification of the mutagenic metabolites of fluoranthene,
2-methylfluoranthene, and 3-methylfluoranthene. Carcinogenesis 3:841-846.

Lecoq, S; Perm, F; Plessis, MJ; et al. (1989) Comparison of the in vitro metabolisms and mutagenicities of
dibenz[a,c]anthracene, dibenz[a,h]anthracene and dibenz[a,j]anthracene: influence of norharman.  Carcinogenesis
10:461-469.

Lecoq, S; Ni She, M; Grover, PL; et al. (1991a) The in vitro metabolic activation of dibenz[a,h]anthracene,
catalyzed by rat liver microsomes and examined by 32P-postlabeling. Cancer Lett 57:261-269.

Lecoq, S; Ni She, M; Hewer, A; et al. (1991b) The metabolic  activation of dibenz[a,h]anthracene in mouse skin
examined by 32P-postlabeling: minor contribution of the 3,4-diol 1,2-oxides to DNA binding. Carcinogenesis
12:1079-1083.
                                             B-14           DRAFT - DO NOT CITE OR QUOTE

-------
Lecoq, S; Pfau, W; Grover, PL; et al. (1992) HPLC separation of 32P-postlabelled DNA adducts formed from
dibenz[a,h]anthracene in skin.  Chem Biol Interact 85:173-185.

Levin, W; Wood, A; Chang, R; et al. (1982) Oxidative metabolism of polycyclic aromatic hydrocarbons to ultimate
carcinogens. Drug Metab Rev 13:555-580.

Lewtas, J. (1985) Development of a comparative potency method for cancer risk assessment of complex mixtures
using short-term in vivo and in vitro bioassays.  Toxicol Ind Health 1:193-203.

Lewtas, J. (1988) Genotoxicity of complex mixtures: strategies for the identification and comparative assessment of
airborne mutagens and carcinogens from combustion sources.  Fundam Appl Toxicol 10:571-589.

Li, KM; Todorovic, R; Rogan, EG; et al. (1995) Identification and quantitation of dibenzo[a,l]pyrene~DNA adducts
formed by rat liver microsomes in vitro: preponderance of depurinating adducts. Biochemistry 34:8043-8049.

Li, KM; Byun, J; Gross, ML; et al. (1999) Synthesis and structure determination of the adducts formed by
electrochemical oxidation of dibenzo[a,l]pyrene in the presence of adenine.  Chem Res Toxicol 12:749-757.

Lloyd, DR; Hanawalt, PC. (2002) p53 controls global nucleotide excision repair of low levels of structurally diverse
benzo[g]chrysene-DNA adducts in human fibroblasts.  Cancer Res 62:5288-5294.

Luch, A; Coffing, SL; Tang, YM; et al. (1998) Stable expression of human cytochrome P450 1B1 in V79 Chinese
hamster cells and metabolically catalyzed DNA adduct formation of dibenzo[a,l]pyrene.  Chem Res Toxicol
11:686-695.

Luch, A; Kishiyama,  S; Seidel, A; et al. (1999) The K-region trans-8,9-diol does not significantly contribute as an
intermediate in the metabolic activation of dibenzo[a,l]pyrene to DNA-binding metabolites by human cytochrome
P450 1A1 or 1B1.  Cancer Res 59:4603-4609.

Luch, A; Kudla, K; Seidel, A; et al. (1999)  The level of DNA modification by  (+)-syn-(HS,12R,13S,14R)- and
(-)-anti-(HR,12S,13S,14R)-dihydrodiol epoxides of dibenzo[a,l]pyrene determined the effect on the proteins p53
and p21WAFl in the  human mammary carcinoma cell line MCF-7. Carcinogenesis 20:859-865.

Lupp, A; Trails, M; Fuchs, U; et al. (1999)  Transplantation of fetal liver tissue suspension into the spleens of adult
syngenic rats: effects  of various mitogens and cytotoxins on cytochrome P450 (P450) isoforms expression and on
P450 mediated monooxygenase functions.  Exp Toxicol Pathol 51:375-388.

Malaveille, C; Hautefeuille, A; Bartsch, H; et al. (1980) Liver microsome-mediated mutagenicity of dihydrodiols
derived from dibenz[a,c]anthracene in S. typhimurium TA 100. Carcinogenesis 1:287-289.

Malaveille, C; Hautefeuille, A; Perin-Roussel, O; et al. (1984) Possible involvement of a vicinal, non-bay-region
dihydrodiol-epoxide in the activation of dibenzo[a,e]fluoranthene into bacterial mutagens.  Carcinogenesis 5:1263-
1266.

Marquardt, H; Heidelberger, C. (1972) Influence of 'feeder cells' and inducers  and inhibitors of microsomal mixed-
function oxidases on hydrocarbon-induced  malignant transformation of cells derived from C3H mouse prostate.
Cancer Res 32:721-725.

Marquardt, H; Kuroki, T; Huberman, E; et  al. (1972) Malignant transformation of cells derived from mouse prostate
by epoxides and other derivatives of polycyclic hydrocarbons.  Cancer Res 32:716-720.

Marrocchi, A; Minuti, L; Morozzi, G; et al. (1996) Synthesis and mutagenicity of some cyclopenta[c]phenanthrenes
and indeno[c]phenanthrenes. Carcinogenesis 17:2009-2012.

Marsch, GA; Jankowiak, R;  Small, GJ; et al. (1992)  Evidence of involvement of multiple sites of metabolism in the
in vivo covalent binding of dibenzo[a,h]pyrene to DNA. Chem Res Toxicol 5:765-772.
                                             B-15           DRAFT - DO NOT CITE OR QUOTE

-------
Marshall, MV; He, ZM; Weyand, EH; et al. (1993) Mutagenic activity of the 4,5- and 9,10-dihydrodiols of
benzo[j]fluoranthene and their syn- and anti-dihydrodiol epoxides in Salmonella typhimurium. Environ Mol
Mutagen 22:34-45.

Matijasevic, Z; Zeiger, E. (1985) Mutagenicity of pyrene in Salmonella.  MutatRes 142:149-152.

Matsuoka, A; Sofuni, T; Miyata, N; et al. (1991) Clastogenicity of 1-nitropyrene, dinitropyrenes, fluorene and
mononitrofluorenes in cultured Chinese hamster cells.  Mutat Res 259:103-110.

Meek, ME; Chan, PKL; Bartlett, S. (1994) Polycyclic aromatic hydrocarbons: evaluation of risks to health from
environmental exposures in Canada. Environ Carcinog Ecotoxicol Rev C 12:443-452.

Melendez-Colon, VJ; Smith, CA; Seidel, A; et al. (1997) Formation of stable adducts and absence of depurinating
DNA adducts in cells and DNA treated with the potent carcinogen dibenzo[a,l]pyrene or its diol epoxides. Proc Natl
Acad Sci U S A 94:13542-13547.

Melendez-Colon, VJ; Luch, A; Seidel, A; et al. (1999) Comparison of cytochrome P450- and peroxidase-dependent
metabolic activation of the potent carcinogen dibenzo[a,l]pyrene in human cell lines: formation of stable DNA
adducts and absence of a detectable increase in apurinic sites.  Cancer Res 59:1412-1416.

Mlcoch, J; Fuchs, J; Oesch, F; et al. (1993) Characterization of DNA adducts at the bay region of
dibenz[a,h]anthracene formed in vitro. Carcinogenesis  14:469-473.

Mori, Y; Goto, S; Onodera, S; et al. (1993) Changes in mutagenic properties and chemical fate of benz[a]anthracene
in chlorine-treated water with and without bromide ion. Chemosphere 27(11):2155-2162.

Moyer, SR; Jurs, PC.  (1990) An SRA study of the mutagenicity of PAH compounds in Salmonella typhimurium.
In: Mendelsohn, ML; Albertini, J, eds. Progress in clinical and biological research. Vol. 340. Mutation and the
environment. Part B. Metabolism, testing methods, and chromosomes. New York, NY: Wiley-Liss, pp. 1-10.

Nair, RV; Gill, RD; Nettikumara, AN; et al. (1991) Characterization of covalently modified deoxyribonucleosides
formed from dibenz[a,j]anthracene in primary cultures of mouse keratinocytes.  Chem Res Toxicol 4:115-122.

Nesnow, S; Leavitt, S; Easterling, R; et al. (1984) Mutagenicity of cyclopenta-fused isomers of benz[a]anthracene in
bacterial and rodent cells and identification of the major rat liver microsomal metabolites. Cancer Res 44:4993-
5003.

Nesnow, S; Lasley, J; Curti,  S; et al. (1991) Morphological transformation and DNA adduct formation by
benz[j]aceanthrylene and its  metabolites in C3H10T1/2CL8 cells: evidence for both cyclopenta-ring and bay-region
metabolic activation pathways.  Cancer Res 51:6163-6169.

Nesnow, S; Davis, C; Padgett, W; et al. (1998) Metabolic activation of racemic and enantiomeric trans-8,
9-dihydroxy-8,9-dihydrodibenzo[a,l]pyrene (dibenzo[def,p]chrysene) to dibenzo[a,l]pyrene-bis-dihydrodiols by
induced rat liver microsomes and a recombinant human P450 1A1 system: the role of the K-region-derived
metabolic intermediates in the formation of dibenzo[a,l]pyrene-DNA  adducts.  Chem Res Toxicol 11:1596-1607.

Newcomb, KO; Sangaiah, R; Gold, A; et al. (1993) Activation and metabolism of benz[j]aceanthrylene-
9,10-dihydrodiol, the precursor to bay-region metabolism of the genotoxic cyclopenta-PAH benz[j]aceanthrylene.
Mutat Res 287:181-190.

Oesch, F; Bticker, M; Glatt, HR. (1981) Activation of phenanthrene to mutagenic metabolites and evidence for at
least two different activation pathways. Mutat Res 81:1-10.

Oshiro, Y; Balwierz, PS. (1982) Morphological transformation of C3H/10T1/2 CL8 cells by procarcinogens.
Environ Mutagen 4:105-108.

Otero-Lobato, MJ; Jenneskens, LW; Seinen, W. (2004) Bacterial mutagenicity of the three isomeric dicyclopenta-
fused pyrenes: the effects of dicyclopenta topology.  Mutat Res 559:105-119.
                                             B-16           DRAFT - DO NOT CITE OR QUOTE

-------
Pal, K. (1981) The induction of sister-chromatid exchanges in Chinese hamster ovary cells by K-region epoxides
and some dihydrodiols derived from benz[a]anthracene, dibenz[a,c]anthracene and dibenz[a,h]anthracene. Mutat
Res 84:389-398.

Palitti, F; Cozzi, R; Fiore, M; et al. (1986) An in vitro and in vivo study on mutagenic activity of fluoranthene:
comparison between cytogenetic studies and HPLC analysis. Mutat Res 174:125-130.

Perin-Roussel, O; Saguem, S; Ekert, B; et al. (1983) Binding to DNA of bay region and pseudo bay region diol-
epoxides of dibenzo[a,e]fluoranthene and comparison with adducts obtained with dibenzo[a,e]fluoranthene or its
dihydrodiols in the presence of microsomes.  Carcinogenesis 4:27-32.

Perin-Roussel, O; Croisy, A; Ekert, B; et al. (1984a) The metabolic activation of dibenzo[a,e]fluoranthene in vitro.
Evidence that its bay-region and pseudo-bay-region diol-epoxides react preferentially with guanosine.  Cancer Lett
22:289-298.

Perin-Roussel, O; Ekert, B; Barat, N; et al. (1984b) DNA-protein crosslinks induced by exposure of cultured mouse
fibroblasts to dibenzo[a,e]fluoranthene and its bay- and pseudo-bay region dihydrodiols. Carcinogenesis 5:379-383.

Perin-Roussel, O; Barat, N; Zajdela, F. (1985) Formation and removal of dibenzo[a,e]fluoranthene-DNA adducts in
mouse embryo fibroblasts. Carcinogenesis 6:1791-1796.

Perin-Roussel, O; Barat, N; Zajdela, F. (1988) Non-random distribution of dibenzo[a,e]fluoranthene-induced DNA
adducts in DNA loops in mouse fibroblast nuclei.  Carcinogenesis 9:1383-1388.

Perin-Roussel, O; Perm, F; Zajdela, F. (1990) 32P-post-labeling analysis of DNA adducts in mouse embryo
fibroblasts treated with dibenzo[a,e]fluoranthene and its major metabolites. Carcinogenesis 11:301-306.

Peter, S; Palme, GE; Rohrborn, G. (1979) Mutagenicity of polycyclic hydrocarbons. III. Monitoring genetic hazards
of benz[a]anthracene. Acta Morphol Acad Sci Hung 27:199-204.

Pfau, W; Hughes, NC; Grover, PL; et al. (1992) HPLC separation of 32P-postlabelled benzo[b]fluoranthene-DNA
adducts. Cancer Lett 65:159-167.

Pfau, W; Lecoq, S; Hughes, NC; et al. (1993) Separation of 32P-labelled nucleoside 3',5'-bisphosphate adducts by
HPLC.  IARC SciPubl 124:233-242.

Phillips, DH. (1997) Detection of DNA modifications by the 32P-postlabeling assay. Mutat Res 378:1-12.

Platt, KL; Bucker, M; Golan, M; et al. (1982) The mutagenicity of dibenz[a,h] anthracene activated by
phenobarbital-inducible mouse-liver mono-oxygenase is potentiated by the presence of hydrophilic  residues at the
K-region of the molecule. Mutat Res 96:1-13.

Platt, KL;  Schollmeier, M; Frank, H; et al. (1990) Stereoselective metabolism of dibenz[a,h]anthracene to trans-
dihydrodiols and their activation to bacterial mutagens.  Environ  Health Perspect 88:37-41.

Polcaro, C; Nicoletti, I; Ossicini, L; et al. (1988) Chromatographic and cytogenetic analysis of in vivo metabolites of
fluoranthene. J Chromatogr 448:127-133.

Pruess-Schwartz, D; Baird, WM; Yagi, H; etal. (1987) Stereochemical specificity in the metabolic activation of
benzo[c]phenanthrene to metabolites that covalently bind to DNA in rodent embryo cell cultures. Cancer Res
47:4032^037.

Purchase, IFH; Longstaff, E; Ashby, J; et al. (1976) Evaluation of six short term tests for detecting organic chemical
carcinogens and recommendations for their use. Nature 264:624-627.

Ralston, SL; Lau, HH; Seidel,  A; et al. (1994) The potent carcinogen dibenzo[a,l]pyrene is metabolically activated
to fjord-region 11,12-diol 13,14-epoxides in human mammary  carcinoma MCF-7 cell cultures. Cancer Res 54:887-
890.
                                              B-17           DRAFT - DO NOT CITE OR QUOTE

-------
Ralston, SL; Seidel, A; Luch, A; et al. (1995) Stereoselective activation of dibenzo[a,l]pyrene to (-)-anti
(11R,12S,13S,14R)- and (+)-syn(HS,12R,13S,14R)-ll,12-diol-13,14-epoxides which bind extensively to
deoxyadenosine residues of DNA in the human mammary carcinoma cell line MCF-7. Carcinogenesis 16:2899-
2907.

Ralston, SL; Coffing, SL; Seidel, A; et al. (1997) Stereoselective activation of dibenzo[a,l]pyrene and its trans-
11,12-dihydrodiol to fjord region 11,12-diol 13,14-epoxides in a human mammary carcinoma MCF-7 cell-mediated
V79 cell mutation assay. Chem Res Toxicol 10:687-693.

RamaKrishna, NV; Padmavathi, NS; Cavalieri, EL; et al. (1993) Synthesis and structure determination of the
adducts formed by electrochemical oxidation of the potent carcinogen dibenzo[a,i]pyrene in the presence of
nucleosides. Chem Res Toxicol 6:554-560.

Rastetter, WH; Nachbar, RB; Russo-Rodriguez, S; et al. (1982) Fluoranthene: synthesis and mutagenicity of fluor
diol epoxides. J Org Chem 47:4873-4878.

Reznikoff, CA; Bertram, JS; Brankow, DW; et al. (1973) Quantitative and qualitative studies of chemical
transformation of cloned C3H mouse embryo cells sensitive to postconfluence inhibition of cell division. Cancer
Res 33:3239-3249.

Rice, JE; Coleman, DT; Hosted, TJ, Jr.; et al. (1985) Identification of mutagenic metabolites of indeno-
[ 1,2,3 -cd]pyrene formed in vitro with rat liver enzymes. Cancer Res 45:5421-5425.

Rice, JE; Geddie, NG; Defloria, MC; et al. (1988) Structural requirements favoring mutagenic activity among
methylated pyrenes in S. typhimurium.  In: Cooke, M; Dennis, AJ, eds. Polynuclear aromatic hydrocarbons: a
decade of progress. Columbus, OH: Battelle Press, pp. 773-785.

Ridler, P; Jennings, B. (1984) The binding of poly cyclic aromatic hydrocarbon diol-epoxides to DNA. Cancer Lett
22:95-98.

Ross, JA; Nelson,  GB; Holden, KL; et al. (1992)  DNA adducts and induction of sister chromatid exchanges in the
rat following benzo[b]fluoranthene administration.  Carcinogenesis 13:1731-1734.

Rugen, PJ; Stern, CD; Lamm, SH. (1989) Comparative carcinogenicity of the PAHs as a basis for acceptable
exposure levels (AELs)  in drinking water. Regul Toxicol Pharmacol 9:273-283.

Safe, S. (1990) Polychlorinated biphenyls (PCBs), dibenzo-p-dioxins (PCDDs), dibenzofurans (PCDFs), and related
compounds: environmental and mechanistic considerations which support the development of toxic equivalency
factors (TEFs).  Crit Rev Toxicol 21:51-88.

Saffiotti, U. (1969) Experimental respiratory tract Carcinogenesis. Prog Exp Tumor Res 11:302-333.

Sangaiah, R; Gold, A; Newcomb, KO; et al. (1991) Synthesis and biological activity of bay-region metabolites of a
cyclopenta-fused polycyclic aromatic hydrocarbon: benz[j]aceanthrylene.  J Med Chem 34:546-549.

Schneider, K; Roller, M; Kalberlah, F; et al. (2002) Cancer risk assessment for oral exposure to PAH mixtures. J
Appl Toxicol 22:73-83.

Slaga, TJ; Gleason, GL; Mills, C; et al. (1980) Comparison of the tumour-initiating activities of dihydrodiols and
diol-epoxides of various polycyclic aromatic hydrocarbons.  Cancer Res 40:1981-1984.

Snell, KC; Stewart, HL. (1962) Pulmonary adenomatosis induced in DBA/2 mice by oral administration of
dibenz[a,h]anthracene. J Natl Cancer Inst 28:1043-1049.

Snell, KC; Stewart, HL. (1963) Induction of pulmonary adenomatoses in DBA/2 mice by the oral administration of
dibenz[a,h]anthracene. Acta Unio Int Contra Cancrum 19:692-694.

Stanton, MF; Miller, E;  Wrench, C; et al. (1972) Experimental induction of epidermoid carcinoma in the lungs of
rats by cigarette smoke condensate. J Natl  Cancer Inst 49:867-877.
                                             B-18           DRAFT - DO NOT CITE OR QUOTE

-------
Stacker, KJ; Howard, WR; Statham, J; et al. (1996) Assessment of the potential in vivo genotoxicity of
fluoranthene. Mutagenesis 11:493-496.

Upham, BL; Weis, LM; Rummel, AM; et al. (1996) The effects of anthracene and methylated anthracenes on gap
junctional intercellular communication in rat liver epithelial cells.  Fundam Appl Toxicol 34:260-264.

U.S. EPA (Environmental Protection Agency). (1989) Health and environmental effects profile for
benzo[g,h,i]perylene. Cincinnati, OH: Environmental Criteria and Assessment Office. EPA-600-X-87-395.

Wang, JS; Busby, WF; Wogan, GN. (1995) Tissue distribution of DNA adducts in pre-weanling BLU:Ha mice
treated with a tumorigenic dose of fluoranthene.  Cancer Lett 92:9-19.

Wang, JS; Busby, WF, Jr.; Wogan,  GN. (1995) Formation and persistence of DNA adducts in organs of CD-I mice
treated with a tumorigenic dose of fluoranthene.  Carcinogenesis 16:2609-2616.

Weis, LM; Rummel, AM; Masten, SJ; et al. (1998) Bay orbaylike regions of polycyclic aromatic hydrocarbons
were potent inhibitors of gap junctional intercellular communication.  Environ Health Perspect 106:17-22.

Wester, PW; Kroes, R. (1988) Forestomach carcinogens: pathology and relevance to man. Toxicol Pathol 16:165-
171.

Weyand, EH; Rice, JE; Hussain, N; et al. (1987a) Detection of DNA adducts of tumorigenic nonalternant polycyclic
aromatic hydrocarbons by 32P-postlabeling. Proc Am Assoc Cancer Res 28:102.

Weyand, EH; Rice, JE; LaVoie, EJ. (1987b) 32P-postlabeling analysis of DNA adducts from non-alternant PAH
using thin-layer and high performance liquid chromatography. Cancer Lett 37:257-266.

Weyand, EH; Geddie, N; Rice, JE; et al. (1988) Metabolism and mutagenic activity of benzo[k]fluoranthene and 3-,
8- and 9-fluorobenzo[k]fluoranthene. Carcinogenesis 9:1277-1281.

Weyand, EH; Bryla,  P; Wu, Y; etal. (1993) Detection of the major DNA adducts of benzo|j]fluoranthene in mouse
skin: nonclassical dihydrodiol epoxides.  Chem Res Toxicol 6:117-124.

Whong, WZ; Stewart, JD; Cutler, D; et al. (1992) Comparative study of DNA adduct formation and cytogenic
effects of two constituents in coke oven emissions with an in vivo  rat lung cell system. Environ Mol Mutag
19(Suppl 20):70.

Whong, WZ; Stewart, JD; Cutler, D; et al. (1994) Induction of in vivo DNA adducts by 4 industrial by-products in
the rat-lung-cell system. Mutat Res 312:165-172.

Wigley, CB; Newbold, RF; Amos, J; et al. (1979) Cell-mediated mutagenesis in cultured Chinese hamster cells by
polycyclic hydrocarbons: mutagenicity and DNA reaction related to carcinogenicity in a series of compounds. Int J
Cancer 23, 691-696.

Willett, KL; Randerath, K; Zhou, GD; et al. (1998) Inhibition of CYP1 Al -dependent activity by the polynuclear
aromatic hydrocarbon (PAH) fluoranthene. Biochem Pharmacol 55:831-839.

Williams, GM. (1977) Detection  of chemical carcinogens by unscheduled DNA synthesis in rat liver primary cell
cultures.  Cancer Res 37:1845-1851.

Wood, AW; Levin, W; Ryan, D; et  al. (1977) High mutagenicity of metabolically activated chrysene
1,2-dihydrodiol: evidence for bay region activation of chrysene. Biochem Biophys Res Commun 78:847-854.

Wood, AW; Levin, W; Thomas, PE; et al. (1978) Metabolic activation of dibenz[a,h]anthracene and its dihydrodiols
to bacterial mutagens.  Cancer Res 38:1967-1973.

Wood, AW; Chang, RL; Huang, MT; et al. (1980a) Mutagenicity of benzo[e]pyrene and triphenylene
tetrahydroepoxides and diol-epoxides in bacterial and mammalian cells. Cancer Res 40:1985-1989.
                                             B-19           DRAFT - DO NOT CITE OR QUOTE

-------
Wood, AW; Chang, RL; Levin, W; et al. (1980b) Mutagenicity of the dihydrodiols and bay-region diol-epoxides of
benzo[c]phenanthrene in bacterial and mammalian cells.  Cancer Res 40:2876-2883.

Wood, AW; Chang, RL; Levin, W; et al. (1981) Mutagenicity of the bay-region diol-epoxides and other benzo-ring
derivatives of dibenzo[a,h]pyrene and dibenzo[a,i]pyrene. Cancer Res 41:2589-2597.

Wood, AW; Chang, RL; Levin, W; et al. (1983) Mutagenicity of the enantiomers of the diastereomeric bay-region
benz [a] anthracene 3,4-diol-l,2-epoxides in bacterial and mammalian cells.  Cancer Res 43:5821-5825.

Wu, J; Zhu, BB; Yu, J; et al. (2003) In vitro and in vivo modulations of benzo[c]phenanthrene-DNA adducts by
DNA mismatch repair system. Nucleic Acids Res 31:6428-6434.

Yamaguchi, K; Near, R; Shneider, A; et al. (1996) Fluoranthene-induced apoptosis in murine T cell hybridomas is
independent of the aromatic hydrocarbon receptor.  Toxicol Appl Pharmacol 139:144-152.

Zhong, BZ; Gu, ZW; Stewart, J; et al. (1995) Micronucleus formation induced by three polycyclic aromatic
hydrocarbons in rat bone marrow and spleen erythrocytes following intratracheal instillation. Mutat Res 326:147-
153.
                                            B-20           DRAFT - DO NOT CITE OR QUOTE

-------
APPENDIX C. DOSE-RESPONSE DATA FOR POTENCY CALCULATIONS
                         C-1       DRAFT - DO NOT CITE OR QUOTE

-------
Table C-l. Dermal bioassays: dose-response information for incidence data
Record
number
Reference
Study
type
Species
Tumor
type
PAH
Sex
Dose
of
PAH
Dose
units
Number of
animals
with tumors
Number
of
animals
in group
%
Tumor-
bearing
animals
Results of
authors'
statistical
analysis
(p-value)
Fisher's
exact
p-value
Cochran-
Armitage
trend test
p-value
Comments
Complete carcinogenicity studies
600





















13640


Habsetal.,
1980





















Cavalieri et
al., 1983


Complete





















Complete


Mice





















Mice


Sum of
Papilloma,
carcinoma,
sarcoma





















Papilloma,
adenoma,
carcinoma


Acetone
DMSO
BaP
BaP
BaP
BbF
BbF
BbF
BjF
BjF
BjF
BkF
BkF
BkF
CPcdP
CPcdP
CPcdP
IP
IP
IP
CO
CO
Acetone
BaP
BaP
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
0
0
1.7
2.8
4.6
3.4
5.6
9.2
3.4
5.6
9.2
3.4
5.6
9.2
1.7
6.5
27.2
3.4
5.6
9.2
5.6
15
0
2.2
6.6
Hg/animal
Hg/animal
Hg/animal
jig/animal
Hg/animal
Hg/animal
jig/animal
Hg/animal
Hg/animal
jig/animal
(ig/animal
Hg/animal
jig/animal
(ig/animal
Hg/animal
jig/animal
(ig/animal
Hg/animal
jig/animal
(ig/animal
Hg/animal
jig/animal
nmol
nmol
nmol
0
0
8
24
22
2
5
20
1
1
2
1
0
0
0
0
3
1
0
0
1
2
0
2
2
35
36
34
35
36
38
34
37
38
35
38
39
38
38
34
35
38
36
37
37
39
40
29
30
28
0
0
24
69
61
5
15
54
3
3
5
3
0
0
0
0
8
3
0
0
3
5
0
7
7



























1.92x 10'3
1.67 x 10'11
2.1 x 10'9
2.6 x 10'1
2.3 x 10'2
3.7 x 10'8
5.1 x 10'1
4.9 x 10'1
2.6 x 10'1
5.2 x 10'1




1.3 x 10'1
5 x 10'1


0.52
0.27

0.25
0.24




2.15 x 10'9


1.33 x 10'9


1.77x 10'1





6.36 x 10'2




1.83 x 10'1




























                                            C-2
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-l. Dermal bioassays: dose-response information for incidence data
Record
number




620










17660




Reference




Hoffmann
and Wynder
1966










Cavalieri et
al., 1977




Study
type




Complete










Complete




Species




Mice










Mice




Tumor
type




Papilloma










Papilloma,
kerato-
acanthoma,
carcinoma




PAH
BaP
CPcdP
CPcdP
CPcdP
Dioxane
BaP
BaP
DBaeP
DBaeP
DBahP
DBahP
DBaiP
DBaiP
DBaeF
DBaeF
Acetone
BaP
DBahP
AA
BaA
Sex
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
Dose
of
PAH
20
22.2
66.6
200
0
0.05
0.1
0.05
0.1
0.05
0.1
0.05
0.1
0.05
0.1
0
0.396
0.396
0.396
0.396
Dose
units
nmol
nmol
nmol
nmol
%
%
%
%
%
%
%
%
%
%
%
|imol/ap-
plication
|imol/ap-
plication
|imol/ap-
plication
|imol/ap-
plication
|imol/ap-
plication
Number of
animals
with tumors
17
2
2
24
0
17
19
16
9
16
15
16
16
17
18
0
30
35
18
1
Number
of
animals
in group
30
29
29
29
20
20
20
30
17
17
18
19
19
19
19
29
38
39
38
39
%
Tumor-
bearing
animals
57
7
7
83
0
85
95
53
53
94
83
84
84
89
95
0
79
90
47
3
Results of
authors'
statistical
analysis
(p-value)




















Fisher's
exact
p-value
4.32 x 10'7
0.25
0.25
9.25 x 10'12

1.28x 10'8
1.5 x 10'10
3.31 x 10'5
1.95 x 10'4
1.32x 10'9
5.27 x 10'8
2.58 x 10'9
2.58 x 10'9
3.35 x 10'9
3.05 x 10'10

4.9 x 10'12
2.98 x 10'15
3.59 x 10'6
0.66
Cochran-
Armitage
trend test
p-value
2.96 x 10'1


1.39 x 10'16


8.7 x 10'10

5.69 x 10'4

1.29x 10'7

9.81 x 10'8

1.13 x 10'9





Comments




















                                            C-3
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-l. Dermal bioassays: dose-response information for incidence data
Record
number
Reference
Study
type
Species
Tumor
type
PAH
Sex
Dose
of
PAH
Dose
units
Number of
animals
with tumors
Number
of
animals
in group
%
Tumor-
bearing
animals
Results of
authors'
statistical
analysis
(p-value)
Fisher's
exact
p-value
Cochran-
Armitage
trend test
p-value
Comments
Initiation studies
630










18570


24800







LaVoie et
al., 1982










Hecht et al.,
1974


Nesnow et
al., 1984







Initiation










Initiation


Initiation







Mice










Mice


Mice







Primarily
squamous
cell
papilloma










Unspeci-
fied


Papilloma







Acetone/
TPA
BaP
BbF
BbF
BbF
BjF
BjF
BjF
BkF
BkF
BkF
Acetone
BaP
CH
Acetone
Acetone
BaP
BaP
B1AC
B1AC
B1AC
B1AC
F
F
F
F
F
F
F
F
F
F
F
F
F
F
M
F
M
F
M
M
M
M
0
30
10
30
100
30
100
1,000
30
100
1,000
0
0.05
1
0
0
200
200
50
100
250
500
Hg/mouse
Hg/mouse
Hg/mouse
jig/mouse
Hg/mouse
Hg/mouse
jig/mouse
Hg/mouse
Hg/mouse
jig/mouse
Hg/mouse
mg/mouse
mg/mouse
mg/mouse
nmol
nmol
nmol
nmol
nmol
nmol
nmol
nmol
0
17
9
12
16
6
11
19
1
5
15
0
6
11
0
1
13
10
12
16
21
16
20
20
20
20
20
20
20
20
20
20
20
20
20
19
20
19
18
19
20
17
21
16
0
85
45
60
80
30
55
95
5
25
75
0
30
58
0
5
67
53
60
94
100
100
















<0.005
0.005
O.005
<0.005
0.005
O.005

1.28x 10'8
6.14 x 10'4
2.25 x 10'5
7.7 x 10'8
0.01
7.27 x 10'5
1.52x 10'10
0.01
0.02
3.85 x 10'7

0.01
4.51 x 10'5












1.46x 10'5


4.67 x 10'8


4.51 x lO'9






















Number of
surviving not
reported for
controls; initial
group size used
here


Data at 30 wks







                                            C-4
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-l. Dermal bioassays: dose-response information for incidence data
Record
number
















21420









Reference
















Slagaetal.,
1980









Study
type
















Initiation









Species
















Mouse









Tumor
type
















Papilloma









PAH
B1AC
B1AC
B1AC
B1AC
B1AC
B1AC
BeAC
BeAC
BeAC
BeAC
BeAC
BeAC
BeAC
BeAC
BeAC
BeAC
Control
Control
Control
Control
Control
pooled
BaP
BeP
CH
DBacA
DBahA
Sex
M
F
F
F
F
F
M
M
M
M
M
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
Dose
of
PAH
1,000
50
100
250
500
1,000
50
100
250
500
1,000
50
100
250
500
1,000
0
0
0
0
0
200
2,000
2,000
2,000
100
Dose
units
nmol
nmol
nmol
nmol
nmol
nmol
nmol
nmol
nmol
nmol
nmol
nmol
nmol
nmol
nmol
nmol
nmol
jimol
(imol
nmol
nmol
nmol
nmol
nmol
nmol
nmol
Number of
animals
with tumors
19
13
18
19
20
20
4
4
12
15
16
4
7
10
8
18
2
3
3
2
10
20
5
21
8
15
Number
of
animals
in group
20
20
19
21
21
20
20
20
20
20
18
20
19
19
18
20
30
30
30
29
119
30
29
29
28
29
%
Tumor-
bearing
animals
95
65
95
91
95
100
20
20
60
75
89
20
37
53
44
90
6
10
10
6
8
67
17
73
27
50
Results of
authors'
statistical
analysis
(p-value)
O.005
<0.005
<0.005
O.005
<0.005
<0.005


<0.005
0.005
<0.005

0.005
0.005
O.005
0.005










Fisher's
exact
p-value





















1.41 x 10'6
0.33
8.38 x 10'7
0.07
3.52 x 10'6
Cochran-
Armitage
trend test
p-value


























Comments
















Different
controls used for
each chemical
except DBacA
and BeP









                                            C-5
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-l. Dermal bioassays: dose-response information for incidence data
Record
number
15640








620










13650






Reference
Raveh et al.,
1982








Hoffmann
and Wynder
1966










Cavalieri et
al., 1981b






Study
type
Initiation








Initiation










Initiation






Species
Mice








Mice










Mice






Tumor
type
Papilloma








Papilloma










Papilloma






PAH
Control
BaP
BaP
BaP
BaP
BaP
CPcdP
CPcdP
CPcdP
Croton oil
control
BaP
DBaeF
DBaeP
DBelP
DBahP
DBaiP
AA
BghiP
N23eP
IP
Acetone/
TPA
BaP
CPcdP
CPcdP
CPcdP
ACEP
ACEP
Sex
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
Dose
of
PAH
0
10
25
50
100
200
10
100
200
0
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0.25
0
0.2
0.2
0.6
1.8
0.2
0.6
Dose
units
Hg
Hg
Hg
Hg
Hg
Hg
Hg
Hg
Hg
mg/mouse
mg/mouse
mg/mouse
mg/mouse
mg/mouse
mg/mouse
mg/mouse
mg/mouse
mg/mouse
mg/mouse
mg/mouse
(imol
(imol
(imol
jimol
(imol
(imol
jimol
Number of
animals
with tumors
3
17
21
24
27
26
3
11
16
2
24
18
10
0
21
12
2
2
9
5
3
12
1
9
6
0
1
Number
of
animals
in group
29
29
28
28
27
26
30
29
28
30
30
30
27
29
29
30
29
27
30
30
29
30
30
29
29
30
30
%
Tumor-
bearing
animals
10
58
76
87
100
100
11
39
57
7
80
60
37
0
72
40
7
7
30
17
10
40
3
31
21
0
3
Results of
authors'
statistical
analysis
(p-value)



























Fisher's
exact
p-value

1.11 x 10'4
5.96 x 10'7
5.43 x 10'9
5.50 x 10'13
1.03 x lO'12
0.65
0.01
1.90x 10'4

3.80 x 10'9
9.40 x 10'6
0.006
0.25
1.30x 10'7
0.002
0.68
0.65
0.02
0.21

0.009
0.29
0.05
0.24
0.11
0.29
Cochran-
Armitage
trend test
p-value





2.78 x 10'10


2.75 x lO'6















0.14


Comments



























                                            C-6
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-l. Dermal bioassays: dose-response information for incidence data
Record
number

15700










Reference

Rice et al,
1988










Study
type

Initiation










Species

Mice










Tumor
type

Unspeci-
fied










PAH
ACEP
Acetone
BaP
CH
CH
CH
CPdefC
(4,5-MC)
CPdefC
(4,5-MC)
CPdefC
(4,5-MC)
BbcAC
(1,12-
MBA)
BbcAC
(1,12-
MBA)
BbcAC
(1,12-
MBA)
Sex
F
F
F
F
F
F
F
F
F
F
F
F
Dose
of
PAH
1.8
0
0.1
0.15
0.5
1.5
0.15
0.5
1.5
0.5
2
4
Dose
units
(imol
(imol
jimol
(imol
(imol
(imol
(imol
(imol
(imol
(imol
(imol
(imol
Number of
animals
with tumors
4
1
17
5
18
19
13
19
19
15
18
18
Number
of
animals
in group
30
20
19
20
20
20
20
19
19
20
20
20
%
Tumor-
bearing
animals
13
5
89
25
90
95
65
100
100
75
90
90
Results of
authors'
statistical
analysis
(p-value)


<0.005
0.05
O.005
<0.005
0.005
0.005
O.005
O.005
0.005
0.005
Fisher's
exact
p-value
0.52











Cochran-
Armitage
trend test
p-value
0.18




6.39 x 10'9


1.90x 10'7


3.03 x 10'6
Comments












                                            C-7
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-2. Dermal bioassays: dose-response information for tumor multiplicity
Record
number
Reference
Study
type
Species
Tumor type
PAH
Sex
Dose of
PAH
Dose units
Number of
animals
with
tumors
Number
of
animals
in
group
%
Tumor-
bearing
animals
Results of
authors'
statistical
analysis
(p-value)
Results of
SRC
statistical
analysis
Fisher's
exact p-value
Mean
number
tumors/
animal
Comments
Complete carcinogenicity
13640






13650







Cavalierietal.,
1983






Cavalieri etal.,
1981b







Complete






Complete







Mice






Mice







Papilloma,
adenoma,
carcinoma






Primarily
squamous
cell
carcinoma







Acetone
BaP
BaP
BaP
CPcdP
CPcdP
CPcdP
Acetone
BaP
CPcdP
CPcdP
CPcdP
ACEP
ACEP
ACEP
F
F
F
F
F
F
F
US
US
US
US
US
US
US
US
0
2.2
6.6
20
22.2
66.6
200
0
0.2
0.2
0.6
1.8
0.2
0.6
1.8
nmol
nmol
nmol
nmol
nmol
nmol
nmol
|imol/
application
(imol/
application
(imol/
application
(imol/
application
|imol/
application
|imol/
application
|imol/
application
(imol/
application
0
2
2
17
2
2
24
0
30
17
11
7
0
1
1
29
30
28
30
29
29
29
30
30
30
30
30
30
30
30
0
7
7
57
7
7
83
0
100
57
37
23
0
3
3
















>0.05
>0.05
0.001
>0.05
>0.05
0.001

0.001
0.001
0.001
0.0053
>0.05
>0.05
>0.05
0
0.07
0.07
1.5
0.07
0.07
2.45
0
1.5
0.8
0.5
0.4
0
0.03
0.03
Number tumors per
animal at risk
calculated






Number tumors per
animal at risk
calculated







Initiation
630




LaVoie etal.,
1982




Initiation




Mice




Primarily
squamous
cell
papilloma




Acetone/
TPA
BaP
BbF
BbF
BbF
F
F
F
F
F
0
30
10
30
100
|xg/mouse
|ig/mouse
|xg/mouse
|ig/mouse
|ig/mouse
0
17
9
12
16
20
20
20
20
20
0
85
45
60
80






0.001
O.001
0.001
O.001
0
4.9
0.9
2.3
7.1





                                            C-8
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-2. Dermal bioassays: dose-response information for tumor multiplicity
Record
number






18570



21420









15640






Reference






Hecht et al,
1974



Slagaetal.,
1980









Raveh et al.,
1982






Study
type






Initiation



Initiation









Initiation






Species






Mice



Mouse









Mice






Tumor type






Unspecified



Papilloma









Papilloma






PAH
BjF
BjF
BjF
BkF
BkF
BkF
Acetone
BaP
CH

Control
Control
Control
Control
Control
pooled
BaP
BeP
CH
DBacA
DBahA
Control
BaP
BaP
BaP
BaP
BaP
CPcdP
Sex
F
F
F
F
F
F
F
F
F

F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
Dose of
PAH
30
100
1,000
30
100
1,000
0
0.05
1

0
0
0
0
0
200
2,000
2,000
2,000
100
0
10
25
50
100
200
10
Dose units
|ig/mouse
|ig/mouse
|ig/mouse
|xg/mouse
jig/mouse
|ig/mouse
mg/animal
mg/animal
mg/animal

nmol
nmol
nmol
nmol
nmol
nmol
nmol
nmol
nmol
nmol
re
re
re
re
re
re
re
Number of
animals
with
tumors
6
11
19
1
5
15
0
6
11

2
3
3
2
10
20
5
21
8
15
3
17
21
24
27
26
3
Number
of
animals
in
group
20
20
20
20
20
20
20
20
19

29
30
30
29
119
30
29
29
28
29
29
29
28
28
27
26
30
%
Tumor-
bearing
animals
30
55
95
5
25
75
0
30
61

6
10
10
6
8
67
17
73
27
50
10
58
76
87
100
100
11
Results of
authors'
statistical
analysis
(p-value)



























Results of
SRC
statistical
analysis
Fisher's
exact p-value
0.01
O.001
0.001
>0.05
0.02
0.001

0.01
0.001






O.001
>0.05
O.001
>0.05
0.001

O.001
O.001
0.001
O.001
0.001
>0.05
Mean
number
tumors/
animal
0.6
1.9
7.2
0.1
0.4
2.8
0
0.5
1

0.1
0.2
0.1
0.1
0.13
2.2
0.2
1.6
0.5
1.4
0.2
1.3
3.8
6.2
8.8
9
0.1
Comments






Number surviving
not reported for
controls; initial
group size used
here; number
tumors per animal
at risk calculated



Different controls
used for each
chemical except
DBacA and BeP
















                                            C-9
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-2. Dermal bioassays: dose-response information for tumor multiplicity
Record
number


13650







21410


16310




10200




24300


Reference


Cavalierietal.,
1981







Slagaetal.,
1978


Weyandetal,
1992




El-Bayoumy et
al., 1982




Rice et al.,
1985


Study
type


Initiation







Initiation


Initiation




Initiation




Initiation


Species


Mice







Mice


Mice




Mice




Mice


Tumor type


Papilloma







Papilloma


Unspecified




Primarily
squamous
cell
papilloma




Unspecified


PAH
CPcdP
CPcdP
Acetone/
TPA
BaP
CPcdP
CPcdP
CPcdP
ACEP
ACEP
ACEP
Acetone/
TPA
BaP
BaA
Acetone
BaP
BjF
BjF
BjF
Acetone
BaP
CH
Pery
Pyr
Acetone
BaP
CH
Sex
F
F
F
F
F
F
F
F
F
F
F
F
F
US
US
US
US
US
F
F
F
F
F
F
F
F
Dose of
PAH
100
200
0
0.2
0.2
0.6
1.8
0.2
0.6
1.8
0
0.2
2
0
0.01
0.3
1
2
0
0.05
1
1
1
0
0.3
1
Dose units
re
re
(imol
(imol
|imol
|imol
(imol
(xmol
jimol
|imol
(xmol
(xmol
(imol
(imol
(imol
(xmol
(imol
|imol
mg/mouse
mg/mouse
mg/mouse
mg/mouse
mg/mouse
mg/mouse
mg/mouse
mg/mouse
Number of
animals
with
tumors
11
16
3
12
1
9
6
0
1
4
2
27
17
1
24
11
21
24
1
18
20
1
4
2
24
23
Number
of
animals
in
group
29
28
29
30
30
29
29
30
30
30
29
29
30
21
24
20
24
24
20
20
20
20
20
25
25
25
%
Tumor-
bearing
animals
39
57
10
40
3
31
21
0
3
13
6
92
57
5
100
55
88
100
5
90
100
5
20
8
96
92
Results of
authors'
statistical
analysis
(p-value)














0.01
O.01
<0.01
0.01

O.01
O.01





Results of
SRC
statistical
analysis
Fisher's
exact p-value
0.01
O.001

0.009
>0.05
0.05
>0.05
>0.05
>0.05
>0.05

O.001
0.001











O.001
0.001
Mean
number
tumors/
animal
0.4
0.9
0.14
1.2
0.03
0.31
0.31
0
0.03
0.13
0.1
5.3
1.2
0.05
4.08
1.75
4.08
7.17
0.1
7.1
7.7
0.1
0.2
0.12
8.04
5
Comments























Mean number of
tumors/animal
digitally estimated
from Figure 2 and
rounded to even
number tumors


                                            C-10
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-2. Dermal bioassays: dose-response information for tumor multiplicity
Record
number

13660






13660






16440






18680


24800





Reference

Cavalierietal.,
1991






Cavalierietal.,
1991






Wood et al.,
1980






Hoffmann et
al., 1972


Nesnowet al.,
1984





Study
type

Initiation






Initiation






Initiation






Initiation


Initiation





Species

Mice






Mice






Mice






Mice


Mice





Tumor type

Primarily
papilloma






Primarily
papilloma






Papilloma






Papilloma


Papilloma





PAH
CPdefC
Acetone
BaP
BaP
BaP
DBalP
DBalP
DBalP
Acetone
BaP
BaP
BaP
DBalP
DBalP
DBalP
Acetone
BaP
BaP
Pyr
Pyr
CPcdP
CPcdP
Acetone
BaP
FA
Acetone
Acetone
BaP
BaP
BeAC
BeAC
Sex
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
M
F
M
F
M
F
Dose of
PAH
1
0
33.3
100
300
33.3
100
300
0
4
20
100
4
20
100
0
0.1
0.4
0.1
0.4
0.1
0.4
0
0.05
1
0
0
200
200
50
50
Dose units
mg/mouse
nmol
nmol
nmol
nmol
nmol
nmol
nmol
nmol
nmol
nmol
nmol
nmol
nmol
nmol
(xmol
jimol
|imol
|imol
(imol
(xmol
jimol
mg
mg
mg
nmol
nmol
nmol
nmol
nmol
nmol
Number of
animals
with
tumors
24
0
10
17
21
23
22
24
0
1
10
22
22
20
20
3
20
22
4
3
3
6
1
19
1
0
1
12
10
4
4
Number
of
animals
in
group
24
24
23
24
23
24
24
24
24
24
24
24
24
24
24
30
30
30
30
30
30
30
30
29
29
20
19
18
19
20
20
%
Tumor-
bearing
animals
100
0
43
71
91
96
92
100
0
4
42
92
92
83
83
10
68
73
14
10
10
21
3
66
3
0
5
67
53
20
20
Results of
authors'
statistical
analysis
(p-value)
















O.05
<0.05
>0.05
>0.05
>0.05
>0.05









Results of
SRC
statistical
analysis
Fisher's
exact p-value
0.001

0.001
0.001
O.001
0.001
O.001
0.001

>0.05
O.001
0.001
0.001
O.001
0.001








0.001
>0.05


0.001
0.0015
>0.05
>0.05
Mean
number
tumors/
animal
5.63
0
0.65
2.75
5.22
6.75
7.92
8.5
0
0.04
0.75
3.42
6.96
5.29
3.29
0.1
2
4.6
0.14
0.1
0.1
0.29
0.03
2.3
0.03
0
0.05
1.4
1.5
0.25
0.25
Comments
Number reported in
text
16-Wk experiment






27- Wk experiment






Number tumors per
animal at risk
calculated















                                            C-ll
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-2. Dermal bioassays: dose-response information for tumor multiplicity
Record
number


















Reference


















Study
type


















Species


















Tumor type


















PAH
BeAC
BeAC
BeAC
BeAC
BeAC
BeAC
BeAC
BeAC
B1AC
B1AC
B1AC
B1AC
B1AC
B1AC
B1AC
B1AC
B1AC
B1AC
Sex
M
F
M
F
M
F
M
F
M
F
M
F
M
F
M
F
M
F
Dose of
PAH
100
100
250
250
500
500
1,000
1,000
50
50
100
100
250
250
500
500
1,000
1,000
Dose units
nmol
nmol
nmol
nmol
nmol
nmol
nmol
nmol
nmol
nmol
nmol
nmol
nmol
nmol
nmol
nmol
nmol
nmol
Number of
animals
with
tumors
4
7
12
10
15
8
16
18
12
13
16
18
21
19
16
20
19
20
Number
of
animals
in
group
20
19
20
19
20
18
18
20
20
20
17
19
21
21
16
21
20
20
%
Tumor-
bearing
animals
20
37
60
53
75
44
89
90
60
65
94
95
100
91
100
95
95
100
Results of
authors'
statistical
analysis
(p-value)


















Results of
SRC
statistical
analysis
Fisher's
exact p-value
>0.05
0.02
0.001
O.001
O.001
0.007
O.001
0.001
O.001
O.001
0.001
O.001
0.001
O.001
O.001
0.001
O.001
0.001
Mean
number
tumors/
animal
0.4
0.53
1.3
1.1
1.9
1.2
3.1
2.2
1.4
1.1
2.3
3.1
8.4
4.7
10.8
6.6
8.7
10.8
Comments


















                                            C-12
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-3. Intraperitoneal bioassays:  dose-response information for incidence data
Record
number
17560









Reference
Busby etal.,
1989









Species
Mice









Expo-
sure
route
Intra-
periton-
eal









Target
organ
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Tumor type
Adenoma +
adeno-
carcinoma
Adenoma +
adeno-
carcinoma
Adenoma +
adeno-
carcinoma
Adenoma +
adeno-
carcinoma
Adenoma +
adeno-
carcinoma
Adenoma +
adeno-
carcinoma
Adenoma +
adeno-
carcinoma
Adenoma +
adeno-
carcinoma
Adenoma +
adeno-
carcinoma
Adenoma +
adeno-
carcinoma
PAH
DMSO
DMSO
BaP
BaP
Pyr
Pyr
Pyr
Pyr
FA
FA
Sex
M
F
M
F
M
F
M
F
M
F
Dose
0
0
59.5
59.5
86.1
86.1
1,750
1,750
257.6
257.6
Dose
units
Hg
(total)
Mg
(total)
Hg
(total)
Hg
(total)
Hg
(total)
Hg
(total)
Hg
(total)
Hg
(total)
Hg
(total)
Hg
(total)
Number of
animals with
tumors
13
7
13
19
4
1
2
3
5
9
Number of
animals in
group
91
101
28
27
23
28
27
26
23
29
% Tumor bear-
ing animals
0.14
0.07
0.46
0.70
0.17
0.04
0.07
0.12
0.22
0.31
Results of
authors'
statistical
analysis
(p-value)










SRC Statistical
Analysis
Fisher's
exact
p-value


7.2 x 10'4
3.96 x 10'11
4.60 x 10'1
4.50 x 10'1
2.80 x 10'1
3.30 x 10'1
2.80 x 10'4
1.65 x lO'3
Cochran-
Armitage
trend test
p-value






3.13X10'1
3.50 x 10'1


Comments
Stats reported for
combined M and F
only for each dose
and treatment
compared to control
not individual sexes









                                            C-13
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-3. Intraperitoneal bioassays:  dose-response information for incidence data
Record
number




640









Reference




LaVoie et
al., 1987









Species




Mice









Expo-
sure
route




Intra-
periton-
eal









Target
organ
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Tumor type
Adenoma +
adeno-
carcinoma
Adenoma +
adeno-
carcinoma
Adenoma +
adeno-
carcinoma
Adenoma +
adeno-
carcinoma
Adenoma
Adenoma
Adenoma
Adenoma
Adenoma
Adenoma
Adenoma
Adenoma
Adenoma
Adenoma
PAH
CH
CH
CH
CH
DMSO
DMSO
BaP
BaP
BbF
BbF
BjF
BjF
BkF
BkF
Sex
M
F
M
F
M
F
M
F
M
F
M
F
M
F
Dose
6.3
6.3
210
210
0
0
1.1
1.1
0.5
0.5
1.1
1.1
2.1
2.1
Dose
units
Mg
(total)
Hg
(total)
Hg
(total)
Hg
(total)
(imol/
mouse
(imol/
mouse
(imol/
mouse
(imol/
mouse
jimol/
mouse
jimol/
mouse
jimol/
mouse
(imol/
mouse
(imol/
mouse
(imol/
mouse
Number of
animals with
tumors
2
3
3
0
0
0
14
9
2
3
11
4
1
3
Number of
animals in
group
27
29
20
29
17
18
17
14
15
17
21
18
16
18
% Tumor bear-
ing animals
0.07
0.10
0.15
0.00
0
0
0.82
0.64
0.13
0.18
0.52
0.22
0.06
0.17
Results of
authors'
statistical
analysis
(p-value)






<0.005

>0.05
>0.05
<0.005
0.05
>0.05
>0.05
SRC Statistical
Analysis
Fisher's
exact
p-value
2.80 x 10'1
3.90 x 10'1
5.85 x 10'1
1.60 x lO'1










Cochran-
Armitage
trend test
p-value


8.03 x 10'1
1.28X10'1










Comments














                                            C-14
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-3. Intraperitoneal bioassays:  dose-response information for incidence data
Record
number















Reference















Species















Expo-
sure
route















Target
organ
Lung
Lung
Liver
Liver
Liver
Liver
Liver
Liver
Liver
Liver
Liver
Liver
Liver
Liver
Liver or
lung
Tumor type
Adenoma
Adenoma
Adenoma +
hepatoma
Adenoma +
hepatoma
Adenoma +
hepatoma
Adenoma +
hepatoma
Adenoma +
hepatoma
Adenoma +
hepatoma
Adenoma +
hepatoma
Adenoma +
hepatoma
Adenoma +
hepatoma
Adenoma +
hepatoma
Adenoma +
hepatoma
Adenoma +
hepatoma
Adenoma +
hepatoma
PAH
IP
IP
DMSO
DMSO
BaP
BaP
BbF
BbF
BjF
BjF
BkF
BkF
IP
IP
DMSO
Sex
M
F
M
F
M
F
M
F
M
F
M
F
M
F
M
Dose
2.1
2.1
0
0
1.1
1.1
0.5
0.5
1.1
1.1
2.1
2.1
2.1
2.1
0
Dose
units
(imol/
mouse
(imol/
mouse
(imol/
mouse
(imol/
mouse
jimol/
mouse
jimol/
mouse
(imol/
mouse
(imol/
mouse
(imol/
mouse
(imol/
mouse
(imol/
mouse
(imol/
mouse
(imol/
mouse
(imol/
mouse
jimol/
mouse
Number of
animals with
tumors
1
0
1
0
13
0
8
0
11
0
3
0
0
0
1
Number of
animals in
group
11
9
17
18
17
14
15
17
21
18
16
18
11
9
17
% Tumor bear-
ing animals
0.09
0
0.06
0
0.76
0
0.53
0
0.52
0
0.19
0
0
0
0.06
Results of
authors'
statistical
analysis
(p-value)




<0.005

<0.005

0.005

>0.05




SRC Statistical
Analysis
Fisher's
exact
p-value















Cochran-
Armitage
trend test
p-value















Comments


Adenoma and
hepatoma also
reported separately;
none of animals
surviving 35 wks












                                            C-15
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-3. Intraperitoneal bioassays:  dose-response information for incidence data
Record
number











7510



Reference











LaVoie et
al., 1994



Species











Mice



Expo-
sure
route











Intra-
periton-
eal



Target
organ
Liver or
lung
Liver or
lung
Liver or
lung
Liver or
lung
Liver or
lung
Liver or
lung
Liver or
lung
Liver or
lung
Liver or
lung
Liver or
lung
Liver or
lung
Lung
Lung
Lung
Lung
Tumor type
Adenoma +
hepatoma
Adenoma +
hepatoma
Adenoma +
hepatoma
Adenoma +
hepatoma
Adenoma +
hepatoma
Adenoma +
hepatoma
Adenoma +
hepatoma
Adenoma +
hepatoma
Adenoma +
hepatoma
Adenoma +
hepatoma
Adenoma +
hepatoma
Total
Total
Total
Total
PAH
DMSO
BaP
BaP
BbF
BbF
BjF
BjF
BkF
BkF
IP
IP
DMSO
DMSO
BaP
BaP
Sex
F
M
F
M
F
M
F
M
F
M
F
M
F
M
F
Dose
0
1.1
1.1
0.5
0.5
1.1
1.1
2.1
2.1
2.1
2.1
0
0
1.1
1.1
Dose
units
(imol/
mouse
(imol/
mouse
(imol/
mouse
(imol/
mouse
(imol/
mouse
(imol/
mouse
(imol/
mouse
jimol/
mouse
jimol/
mouse
jimol/
mouse
(imol/
mouse
(imol/
mouse
(imol/
mouse
(imol/
mouse
(imol/
mouse
Number of
animals with
tumors
0
13
9
8
3
17
4
3
3
1
0
5
4
24
17
Number of
animals in
group
18
17
14
15
17
21
18
16
18
11
9
29
34
32
20
% Tumor bear-
ing animals
0
0.76
0.64
0.53
0.18
0.81
0.22
0.19
0.17
0.09
0
0.17
0.12
0.75
0.85
Results of
authors'
statistical
analysis
(p-value)













0.001
O.001
SRC Statistical
Analysis
Fisher's
exact
p-value















Cochran-
Armitage
trend test
p-value















Comments











Survival to 1 yr



                                            C-16
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-3. Intraperitoneal bioassays:  dose-response information for incidence data
Record
number












Reference












Species












Expo-
sure
route












Target
organ
Lung
Lung
Lung
Lung
Liver
Liver
Liver
Liver
Liver
Liver
Liver
Liver
Tumor type
Total
Total
Total
Total
Foci +
adenoma +
carcinoma
Foci +
adenoma +
carcinoma
Foci +
adenoma +
carcinoma
Foci +
adenoma +
carcinoma
Foci +
adenoma +
carcinoma
Foci +
adenoma +
carcinoma
Foci +
adenoma +
carcinoma
Foci +
adenoma +
carcinoma
PAH
FA
FA
FA
FA
DMSO
DMSO
BaP
BaP
FA
FA
FA
FA
Sex
M
F
M
F
M
F
M
F
M
F
M
F
Dose
3.46
3.46
17.3
17.3
0
0
1.1
1.1
3.46
3.46
17.3
17.3
Dose
units
(imol/
mouse
(imol/
mouse
(imol/
mouse
(imol/
mouse
(imol/
mouse
(imol/
mouse
(imol/
mouse
(imol/
mouse
jimol/
mouse
jimol/
mouse
jimol/
mouse
(imol/
mouse
Number of
animals with
tumors
12
11
11
25
5
2
27
2
18
0
17
2
Number of
animals in
group
28
31
17
29
29
34
32
20
28
31
17
29
% Tumor bear-
ing animals
0.43
0.35
0.65
0.86
0.17
0.06
0.84
0.10
0.64
0
1.00
0.07
Results of
authors'
statistical
analysis
(p-value)
0.05
0.05
O.005
O.001


0.001
>0.05
O.001

O.001

SRC Statistical
Analysis
Fisher's
exact
p-value












Cochran-
Armitage
trend test
p-value


2.84 x 10'3
2.18 x 10'9






5.10 x 10'7
5.47 x 10'1
Comments




Foci, adenomas,
carcinomas also
reported separately







                                            C-17
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-3. Intraperitoneal bioassays:  dose-response information for incidence data
Record
number
24590

















24590




24801
Reference
Nesnow et
al., 1998b

















Nesnow et
al., 1998b




Weyand et
al., 2004
Species
Mice

















Mice




Mouse
Expo-
sure
route
Intra-
periton-
eal

















Intra-
periton-
eal




Intra-
periton-
eal
Target
organ
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Tumor type
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
Adenoma
PAH
Control
BaP
BaP
BaP
BaP
BaP
BbF
BbF
BbF
BbF
CPcdP
CPcdP
CPcdP
CPcdP
DBahA
DBahA
DBahA
DBahA
Control
DBalP
DBalP
DBalP
DBalP
Tri-
caprylin
Sex
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
F
Dose
0
5
10
50
100
200
10
50
100
200
10
50
100
200
1.25
2.5
5
10
0
0.3
1.5
3
6
0
Dose
units
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
Number of
animals with
tumors
6
6
7
19
16
24
9
16
20
19
8
20
19
19
12
18
20
19
15
13
33
35
30
14
Number of
animals in
group
20
20
17
19
16
24
18
20
20
19
20
20
19
19
18
19
20
19
30
33
34
35
30
29
% Tumor bear-
ing animals
0.30
0.30
0.41
1.00
1.00
1.00
0.50
0.80
1.00
1.00
0.40
1.00
1.00
1.00
0.67
0.95
1.00
1.00
0.50
0.39
0.97
1.00
1.00
0.48
Results of
authors'
statistical
analysis
(p-value)
























SRC Statistical
Analysis
Fisher's
exact
p-value

>0.05
>0.05
0.001
0.0018
0.001
>0.05
>0.05
0.001
0.001
>0.05
0.001
O.001
O.001
0.05
0.0053
O.001
0.001

>0.05
0.001
O.001
O.001

Cochran-
Armitage
trend test
p-value
























Comments
Data provided by S.
Nesnow

















Data provided by S.
Nesnow





                                            C-18
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-3. Intraperitoneal bioassays:  dose-response information for incidence data
Record
number


22510












Reference


Wislocki et
al., 1986












Species


Mice












Expo-
sure
route


Intra-
periton-
eal












Target
organ


Liver
Liver
Liver
Liver
Liver
Liver
Liver
Liver
Liver
Liver
Liver
Liver
Liver
Tumor type


Adenoma +
carcinoma
Adenoma +
carcinoma
Adenoma +
carcinoma
Adenoma +
carcinoma
Adenoma +
carcinoma
Adenoma +
carcinoma
Adenoma +
carcinoma
Adenoma +
carcinoma
Adenoma +
carcinoma
Adenoma +
carcinoma
Adenoma +
carcinoma
Adenoma +
carcinoma
Adenoma +
carcinoma
PAH
BaP
BcFE
DMSO
DMSO
DMSO
DMSO
DMSO
pooled
DMSO
pooled
BaP
BaP
CH
CH
CH
CH
BaA
Sex
F
F
M
F
M
F
M
F
M
F
M
F
M
F
M
Dose
100
100
0
0
0
0
0
0
560
560
700
700
2,800
2,800
2,800
Dose
units
mg/kg
mg/kg
nmol
nmol
nmol
nmol
nmol
nmol
nmol
nmol
nmol
nmol
nmol
nmol
nmol
Number of
animals with
tumors
27
26
2
0
5
0
7
0
18
0
10
0
14
0
31
Number of
animals in
group
30
28
28
31
45
34
73
65
37
27
35
33
34
24
39
% Tumor bear-
ing animals
0.90
0.92
0.07
0
0.11
0
0.09
0
0.49
0
0.29
0
0.41
0
0.79
Results of
authors'
statistical
analysis
(p-value)








<0.05

<0.05

<0.05

<0.05
SRC Statistical
Analysis
Fisher's
exact
p-value
0.0005
0.0002













Cochran-
Armitage
trend test
p-value












6 x 10'3
1

Comments


Animals surviving
through weaning
0
This group started
10 wks after other
groups
This group started
10 wks after other
groups




This group started
10 wks after other
groups
This group started
10 wks after other
groups



                                            C-19
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-3. Intraperitoneal bioassays:  dose-response information for incidence data
Record
number














Reference














Species














Expo-
sure
route














Target
organ
Liver
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Tumor type
Adenoma +
carcinoma
Adenoma +
carcinoma
Adenoma +
carcinoma
Adenoma +
carcinoma
Adenoma +
carcinoma
Adenoma +
carcinoma
Adenoma +
carcinoma
Adenoma +
carcinoma
Adenoma +
carcinoma
Adenoma +
carcinoma
Adenoma +
carcinoma
Adenoma +
carcinoma
Adenoma +
carcinoma
Adenoma +
carcinoma
PAH
BaA
DMSO
DMSO
DMSO
DMSO
DMSO
pooled
DMSO
pooled
BaP
BaP
CH
CH
CH
CH
BaA
Sex
F
M
F
M
F
M
F
M
F
M
F
M
F
M
Dose
2,800
0
0
0
0
0
0
560
560
700
700
2,800
2,800
2,800
Dose
units
nmol
nmol
nmol
nmol
nmol
nmol
nmol
nmol
nmol
nmol
nmol
nmol
nmol
nmol
Number of
animals with
tumors
0
1
0
4
2
5
2
13
13
6
2
7
1
6
Number of
animals in
group
32
28
31
45
34
73
65
37
27
35
33
34
24
39
% Tumor bear-
ing animals
0
0.04
0
0.09
0.06
0.07
0.03
0.35
0.48
0.17
0.06
0.21
0.04
0.15
Results of
authors'
statistical
analysis
(p-value)







O.05
O.05


O.05


SRC Statistical
Analysis
Fisher's
exact
p-value














Cochran-
Armitage
trend test
p-value











1.1 x 10'1
5.6 x 10'1

Comments



This group started
10 wks after other
groups
This group started
10 wks after other
groups




This group started
10 wks after other
groups
This group started
10 wks after other
groups



                                            C-20
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-3. Intraperitoneal bioassays:  dose-response information for incidence data
Record
number











Reference











Species











Expo-
sure
route











Target
organ
Lung
Lymph-
atic
system
Lymph-
atic
system
Lymph-
atic
system
Lymph-
atic
system
Lymph-
atic
system
Lymph-
atic
system
Lymph-
atic
system
Lymph-
atic
system
Lymph-
atic
system
Lymph-
atic
system
Tumor type
Adenoma +
carcinoma
Lymphoma
Lymphoma
Lymphoma
Lymphoma
Lymphoma
Lymphoma
Lymphoma
Lymphoma
Lymphoma
Lymphoma
PAH
BaA
DMSO
DMSO
DMSO
DMSO
BaP
BaP
CH
CH
CH
CH
Sex
F
M
F
M
F
M
F
M
F
M
F
Dose
2,800
0
0
0
0
560
560
700
700
2,800
2,800
Dose
units
nmol
nmol
nmol
nmol
nmol
nmol
nmol
nmol
nmol
nmol
nmol
Number of
animals with
tumors
6
1
1
0
0
2
4
3
1
0
0
Number of
animals in
group
32
28
31
45
34
37
27
35
33
34
24
% Tumor bear-
ing animals
0.19
0.04
0.03
0
0
0.05
0.15
0.09
0.03
0
0
Results of
authors'
statistical
analysis
(p-value)
0.05






O.05



SRC Statistical
Analysis
Fisher's
exact
p-value











Cochran-
Armitage
trend test
p-value









2.2 x 10'1
3.9 x 10'1
Comments



This group started
10 wks after other
groups
This group started
10 wks after other
groups


This group started
10 wks after other
groups
This group started
10 wks after other
groups


                                            C-21
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-3. Intraperitoneal bioassays:  dose-response information for incidence data
Record
number


Reference


Species


Expo-
sure
route


Target
organ
Lymph-
atic
system
Lymph-
atic
system
Tumor type
Adenoma +
carcinoma
Adenoma +
carcinoma
PAH
BaA
BaA
Sex
M
F
Dose
2,800
2,800
Dose
units
nmol
nmol
Number of
animals with
tumors
1
3
Number of
animals in
group
39
32
% Tumor bear-
ing animals
0.03
0.09
Results of
authors'
statistical
analysis
(p-value)


SRC Statistical
Analysis
Fisher's
exact
p-value


Cochran-
Armitage
trend test
p-value


Comments


                                            C-22
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-4.  Intraperitoneal bioassays: dose-response information for tumor multiplicity
Record
number
17560













7510




Reference
Busby etal., 1989













LaVoieetal., 1994




Species
Mice













Mice




Exposure
route
Intra-
peritoneal













Intra-
peritoneal




Target
organ
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Tumor type
Adenoma+
adeno-
carcinoma
Adenoma+
adeno-
carcinoma
Adenoma+
adeno-
carcinoma
Adenoma+
adeno-
carcinoma
Adenoma+
adeno-
carcinoma
Adenoma+
adeno-
carcinoma
Adenoma+
adeno-
carcinoma
Adenoma+
adeno-
carcinoma
Adenoma+
adeno-
carcinoma
Adenoma+
adenocarcin
oma
Adenoma+
adeno-
carcinoma
Adenoma+
adeno-
carcinoma
Adenoma+
adeno-
carcinoma
Adenoma+
adeno-
carcinoma
Total
Total
Total
Total
Total
PAH
DMSO
DMSO
BaP
BaP
Pyr
Pyr
Pyr
Pyr
FA
FA
CH
CH
CH
CH
DMSO
DMSO
BaP
BaP
FA
Sex
M
F
M
F
M
F
M
F
M
F
M
F
M
F
M
F
M
F
M
Dose
0
0
59.5
59.5
86.1
86.1
1,750
1,750
257.6
257.6
6.3
6.3
210
210
0
0
1.1
1.1
3.46
Dose units
Hg (total)
Hg (total)
ug (total)
ug (total)
Ug (total)
ug (total)
ug (total)
ug (total)
ug (total)
ug (total)
ug (total)
ug (total)
ug (total)
ug (total)
umol/mouse
umol/mouse
umol/mouse
umol/mouse
umol/mouse
Number
of animals
with
tumors
13
7
13
19
4
1
2
-*
5
9
2
3
-^
0
5
4
24
17
12
Number
of animals
in group
91
101
28
27
23
28
27
26
23
29
27
29
20
29
29
34
32
20
28
%
Tumor-
bearing
animals
0.14
0.07
0.46
0.70
0.17
0.04
0.07
0.12
0.22
0.31
0.07
0.10
0.15
0.00
0.17
0.12
0.75
0.85
0.43
Results of
authors'
statistical
analysis
(p-value)
















<0.001
<0.001
<0.05
Results of
SRC
statistical
analysis
(Fisher's
exact
p-value)


<0.001
<0.001
>0.05
>0.05
>0.05
>0.05
>0.05
0.00165
>0.05
>0.05
>0.05
>0.05





Mean
number
tumors/
animal
0.15
0.08
0.71
1.19
0.17
0.04
0.07
0.12
0.22
0.41
0.07
0.1
0.15
0
0.17
0.15
4.3
3.55
0.64
SDof
mean
0.38
0.30
1.01
1.09
0.38
0.21
0.26
0.31
0.43
0.70
0.26
0.32
0.36
0.00





Results of
SRC
statistical
analysis
(t-test
p-value)


<0.001
<0.001
>0.05
>0.05
>0.05
>0.05
>0.05
<0.0001
>0.05
>0.05
>0.05
>0.05





Comments
Stats
reported for
combined M
and F













Survived to
lyr




                                             C-23
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-4.  Intraperitoneal bioassays: dose-response information for tumor multiplicity
Record
number











22510






Reference











Wislocki et al., 1986






Species











Mice






Exposure
route











Intra-
peritoneal






Target
organ
Lung
Lung
Lung
Liver
Liver
Liver
Liver
Liver
Liver
Liver
Liver
Liver
Liver
Liver
Liver
Liver
Liver
Liver
Tumor type
Total
Total
Total
FOCI +
adenoma +
carcinoma
Foci +
adenoma +
carcinoma
Foci +
adenoma +
carcinoma
Foci +
adenoma +
carcinoma
FOCI +
adenoma +
carcinoma
Foci +
adenoma +
carcinoma
Foci +
adenoma +
carcinoma
FOCI +
adenoma +
carcinoma
Adenoma +
carcinoma
Adenoma +
carcinoma
Adenoma +
carcinoma
Adenoma +
carcinoma
Adenoma +
carcinoma
Adenoma +
carcinoma
Adenoma +
carcinoma
PAH
FA
FA
FA
DMSO
DMSO
BaP
BaP
FA
FA
FA
FA
DMSO
DMSO
DMSO
DMSO
DMSO
pooled
DMSO
pooled
BaP
Sex
F
M
F
M
F
M
F
M
F
M
F
M
F
M
F
M
F
M
Dose
3.46
17.3
17.3
0
0
1.1
1.1
3.46
3.46
17.3
17.3
o
o
0
0
0
0
560
Dose units
umo I/mouse
umo I/mouse
umo I/mouse
umo I/mouse
umo I/mouse
umo I/mouse
umo I/mouse
umo I/mouse
umol/mouse
umo I/mouse
umol/mouse
nmol
nmol
nmol
nmol
nmol
nmol
nmol
Number
of animals
with
tumors
11
11
25
5
2
27
2
18
0
17
0
2
0
5
0
1
0
18
Number
of animals
in group
31
17
29
29
34
32
20
28
31
17
29
28
31
45
34
73
65
37
%
Tumor-
bearing
animals
0.35
0.65
0.86
0.17
0.06
0.84
0.10
0.64
0
1.00
0.07
0.07
0
0.11
0
0.09
0
0.49
Results of
authors'
statistical
analysis
(p-value)
<0.05
<0.005
<0.001


<0.001
>0.05
<0.001

<0.001







<0.05
Results of
SRC
statistical
analysis
(Fisher's
exact
p-value)


















Mean
number
tumors/
animal
0.35
1.12
2.45
0.41
0.06
4.53
0.3
1.86
0
7.53
0.07
0.07
0
0.11
0
0.096
0
1.46
SDof
mean


















Results of
SRC
statistical
analysis
(t-test
p-value)


















Comments




Tumor
count
appears to
be error in
publication






Animals
surviving
through
weaning

This group
started
lOwks after
other groups
This group
started
1 0 wks after
other groups



                                             C-24
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-4.  Intraperitoneal bioassays: dose-response information for tumor multiplicity
Record
number













13610







Reference













Busby etal., 1984







Species













Mice







Exposure
route













Intra-
peritoneal







Target
organ
Liver
Liver
Liver
Liver
Liver
Liver
Liver
Liver
Liver
Liver
Liver
Liver
Liver
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Tumor type
Adenoma +
carcinoma
Adenoma +
carcinoma
Adenoma +
carcinoma
Adenoma +
carcinoma
Adenoma +
carcinoma
Adenoma +
carcinoma
Adenoma +
carcinoma
Adenoma +
carcinoma
Adenoma +
carcinoma
Adenoma +
carcinoma
Adenoma +
carcinoma
Adenoma +
carcinoma
Adenoma +
carcinoma
Adenoma +
carcinoma
Adenoma +
carcinoma
Adenoma +
carcinoma
Adenoma +
carcinoma
Adenoma +
carcinoma
Adenoma +
carcinoma
Adenoma +
carcinoma
Adenoma +
carcinoma
PAH
BaP
Pyr
Pyr
Pyr
Pyr
Pyr
Pyr
CH
CH
CH
CH
BaA
BaA
DMSO
DMSO
BaP
BaP
BaP
BaP
FA
FA
Sex
F
M
F
M
F
M
F
M
F
M
F
M
F
M
F
M
F
M
F
M
F
Dose
560
200
200
700
700
2,800
2,800
700
700
2,800
2,800
2,800
2,800
0
0
0.28
0.28
1.4
1.4
0.7
0.7
Dose units
nmol
nmol
nmol
nmol
nmol
nmol
nmol
nmol
nmol
nmol
nmol
nmol
nmol
mg (total)
mg (total)
mg (total)
mg (total)
mg (total)
mg (total)
mg (total)
mg (total)
Number
of animals
with
tumors
0
0
0

0
3
0
10
0
14
0
31
0
1
4
24
25
16
21
7
3
Number
of animals
in group
27
29
31
25
49
14
18
35
33
34
24
39
32
27
28
25
27
20
24
31
20
%
Tumor-
bearing
animals
0
0
0
0.12
0
0.21
0
0.29
0
0.41
0
0.79
0
0.04
0.14
0.96
0.93
0.80
0.88
0.23
0.15
Results of
authors'
statistical
analysis
(p-value)
>0.05
>0.05
>0.05
>0.05
>0.05
>0.05
>0.05
<0.05
>0.05
<0.05
>0.05
<0.05
>0.05








Results of
SRC
statistical
analysis
(Fisher's
exact
p-value)















<0.001
<0.001
<0.001
<0.001
0.0412
>0.05
Mean
number
tumors/
animal
0
0
0
0.12
0
0.21
0
0.86
0
1.03
0
2.38
0
0.04
0.14
4.32
3.7
10.15
4.25
0.29
0.15
SDof
mean













0.21
0.37
3.5
3.10
13.0
4.70
0.84
0.49
Results of
SRC
statistical
analysis
(t-test
p-value)















<0.001
<0.001
<0.001
<0.001
>0.05
>0.05
Comments



This group
started
1 0 wks after
other groups
This group
started
1 0 wks after
other groups


This group
started
10 wks after
other groups
This group
started
10 wks after
other groups








No model fit
No model fit


                                             C-25
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-4.  Intraperitoneal bioassays: dose-response information for tumor multiplicity
Record
number


24590

















24590




11190






24801


Reference


Nesnowetal., 1998b

















Nesnowetal., 1998b




Massetal., 1993






Weyand et al., 2004


Species


Mice

















Mice




Mice






Mice


Exposure
route


Intra-
peritoneal

















Intra-
peritoneal




Intra-
peritoneal






Intra-
peritoneal


Target
organ
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Lung






Lung
Lung
Lung
Tumor type
Adenoma +
carcinoma
Adenoma +
carcinoma
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
Adenoma
Adenoma
Adenoma
PAH
FA
FA
Control
BaP
BaP
BaP
BaP
BaP
BbF
BbF
BbF
BbF
CPcdP
CPcdP
CPcdP
CPcdP
DBabA
DBahA
DBahA
DBahA
Control
DBalP
DBalP
DBalP
DBalP
Control
BaP
BaP
BaP
BjAC
BjAC
BjAC
Tn-
caprylin
BaP
BcFE
Sex
M
F
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
F
F
F
Dose
3.5
3.5
0
5
10
50
100
200
10
50
100
200
10
50
100
200
1.25
2.5
5
10
0
0.3
1.5
3
6
0
20
50
100
20
50
100
0
100
100
Dose units
mg (total)
mg (total)
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
Number
of animals
with
tumors
20
8
6
6
7
19
16
24
9
16
20
19
8
20
19
19
12
18
20
19
15
13
33
35
30
19
10
15
14
12
13
14
14
27
26
Number
of animals
in group
27
21
20
20
17
19
16
24
18
20
20
19
20
20
19
19
18
19
20
19
30
33
34
35
30
34
16
16
14
12
13
14
29
30
28
%
Tumor-
bearing
animals
0.74
0.38
0.30
0.30
0.41
1.00
1.00
1.00
0.50
0.80
1.00
1.00
0.40
1.00
1.00
1.00
0.67
0.95
1.00
1.00
0.50
0.39
0.97
1.00
1.00
0.56
0.63
0.94
1.00
1.00
1.00
1.00
0.48
0.9
0.92
Results of
authors'
statistical
analysis
(p-value)



































Results of
SRC
statistical
analysis
(Fisher's
exact
p-value)
<0.001
>0.05

>0.05
>0.05
<0.001
0.0018
<0.001
>0.05
<0.05
<0.001
<0.001
>0.05
<0.001
<0.001
<0.001
<0.05
0.0053
<0.001
<0.001

>0.05
<0.001
<0.001
<0.001

>0.05
0.0065
0.0017
0.0036
0.0025
0.0017

0.0005
0.0002
Mean
number
tumors/
animal
1.52
0.52
0.53
0.45
0.53
4.37
12.75
32.96
0.67
2.00
5.30
6.95
0.55
4.75
32.21
97.68
1.44
3.05
13.05
32.16
0.67
0.42
4.32
7.49
16.10
0.85
1
3.9
5.9
60.3
140.6
97.6
0.6
6.7
4
SDof
mean
1.66
0.82
0.72
0.80
0.78
2.74
4.28
10.23
0.75
1.82
3.21
3.52
0.80
2.12
15.15
28.68
1.46
1.90
5.99
10.78
0.80
0.56
2.86
3.79
7.26
0.9
1
2.9
3.3
14.6
21.5
28.2
0.75
5.26
2.8
Results of
SRC
statistical
analysis
(t-test
p-value)
<0.001
0.0343

>0.05
>0.05
<0.001
<0.001
<0.001
>0.05
0.0022
<0.001
<0.001
>0.05
<0.001
<0.001
<0.001
0.0229
<0.001
<0.001
<0.001

>0.05
<0.001
<0.001
<0.001

>0.05
<0.001
<0.001
<0.001
<0.001
<0.001

<0.01
<0.01
Comments
Nonconstant
variance
NS
incidence;
nonconstant
variance
Pooled
controls
from data
provided by
Nesnow
































                                             C-26
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-5. Lung implantation bioassays: dose-response information for incidence data
Record
number
17940


























Reference
Deutsch-Wenzel et
al., 1983


























Species
Rat


























Target
organ
Lung
























Lung

Tumor type
Epidermoid
carcinoma
























Pleomorphic
sarcoma

PAH
Untreated
control
Vehicle
control
BaP
BaP
BaP
BbF
BbF
BbF
BeP
BeP
BeP
BjF
BjF
BjF
BkF
BkF
BkF
IP
IP
IP
AA
AA
BghiP
BghiP
BghiP
Untreated
control
Vehicle
control
Dose
0
0
0.1
0.3
1
0.1
0.3
1
0.2
1
5
0.2
1
5
0.16
0.83
4.15
0.16
0.83
4.15
0.16
0.83
0.16
0.83
4.15
0
0
Dose
units
mg
mg
mg
mg
mg
mg
mg
mg
mg
mg
mg
mg
mg
mg
mg
mg
mg
mg
mg
mg
mg
mg
mg
mg
mg
mg
mg
Number
of
animals
with
tumors
0
0
4
21
33
0
1
9
0
0
1
1
3
18
0
3
12
3
8
21
1
19
0
1
4
0
0
Number
of
animals
in group
35
35
35
35
35
35
35
35
35
30
35
35
35
35
35
31
27
35
35
35
35
35
35
35
34
35
35
% Tumor-
bearing
animals
0.00
0.00
0.11
0.60
0.94
0.00
0.03
0.26
0.00
0.00
0.03
0.03
0.09
0.51
0.00
0.10
0.44
0.09
0.23
0.60
0.03
0.54
0.00
0.03
0.12
0.00
0.00
SRC statistical analysis
Fisher's
exact p-value


5.70 x 10'2
6.02 x 10'9
5.93 x 10'18

5 x 10'1
1 x 10'3


5 x 10'1
5 x 10'1
1.2 x 10'1
1.96x 10'7

1 x lO'1
8.05 x 10'6
1.20x 10'1
2 x lO'3
6.02 x 10'9
5 x 10'1
6.4 x 10'8

1.2 x 10'1
5.4 x 10'2


Cochran-
Armitage trend
testp-value




1.57 x 10'17


5.12 x 10'7


9.49 x 10'2


1.28 x 10'11


1.03 x 10'9


2.09 x 10'10

1.13 xlO'10


2.47 x 10'3


Comments



























                                            C-27
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-5. Lung implantation bioassays: dose-response information for incidence data
Record
number




























Reference




























Species




























Target
organ























Lung




Tumor type























Carcinoma+
sarcoma




PAH
BaP
BaP
BaP
BbF
BbF
BbF
BeP
BeP
BeP
BjF
BjF
BjF
BkF
BkF
BkF
IP
IP
IP
AA
AA
BghiP
BghiP
BghiP
Untreated
control
Vehicle
control
BaP
BaP
BaP
Dose
0.1
0.3
1
0.1
0.3
1
0.2
1
5
0.2
1
5
0.16
0.83
4.15
0.16
0.83
4.15
0.16
0.83
0.16
0.83
4.15
0
0
0.1
0.3
1
Dose
units
mg
mg
mg
mg
mg
mg
mg
mg
mg
mg
mg
mg
mg
mg
mg
mg
mg
mg
mg
mg
mg
mg
mg
mg
mg
mg
mg
mg
Number
of
animals
with
tumors
6
2
0
1
2
4
0
1
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
10
23
33
Number
of
animals
in group
35
35
35
35
35
35
35
30
35
35
35
35
35
31
27
35
35
35
35
35
35
35
34
35
35
35
35
35
% Tumor-
bearing
animals
0.17
0.06
0.00
0.03
0.06
0.11
0.00
0.03
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.03
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.29
0.66
0.94
SRC statistical analysis
Fisher's
exact p-value
1.2 x 10'2
2.5 x 10'1

1.2 x 10'1
2.5 x 10'1
6. x 10'2









1.2 x 10'1









4.63 x 10'4
4.7 x 10'10
5.9 x ID'19
Cochran-
Armitage trend
testp-value


1.36x 10'1


7.55 x 10'3





















3.66 x 10'9
Comments




























                                            C-28
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-5. Lung implantation bioassays: dose-response information for incidence data
Record
number




















22000




Reference




















Wenzel-Hartung et
al., 1990




Species




















Rat




Target
organ




















Lung




Tumor type




















Carcinoma




PAH
BbF
BbF
BbF
BeP
BeP
BeP
BjF
BjF
BjF
BkF
BkF
BkF
IP
IP
IP
AA
AA
BghiP
BghiP
BghiP
Untreated
control
Vehicle
control
BaP
BaP
BaP
Dose
0.1
0.3
1
0.2
1
5
0.2
1
5
0.16
0.83
4.15
0.16
0.83
4.15
0.16
0.83
0.16
0.83
4.15
0
0
0.03
0.1
0.3
Dose
units
mg
mg
mg
mg
mg
mg
mg
mg
mg
mg
mg
mg
mg
mg
mg
mg
mg
mg
mg
mg
mg/
animal
mg/
animal
mg/
animal
mg/
animal
mg/
animal
Number
of
animals
with
tumors
1
3
13
0
1
1
1
3
18
0
3
12
4
8
21
1
19
0
1
4
0
0
3
11
27
Number
of
animals
in group
35
35
35
35
30
35
35
35
35
35
31
27
35
35
35
35
35
35
35
34
35
35
35
35
35
% Tumor-
bearing
animals
0.03
0.09
0.37
0.00
0.03
0.03
0.03
0.09
0.51
0.00
0.10
0.44
0.11
0.23
0.60
0.03
0.54
0.00
0.03
0.12
0.00
0.00
0.09
0.31
0.77
SRC statistical analysis
Fisher's
exact p-value
1.2 x 10'1
1.2 x 10'1
3.1 x 10'5


1.2 x 10'1
1.2 x 10'1
1.20x 10'1
1.96x 10'7

1 x 10'1
8.05 x 10'4
6 x 10'2
2 x 10'3
6.02 x 10'9

6.4 x 10'8


5.4 x 10'2


1.2 x 10'1
1.93 x 10'4
1.29E x 10'12
Cochran-
Armitage trend
testp-value


9.63 x 10'8


3.23 x 10'1


1.28 x 10'11


1.03 x 10'9


7.56 x 10'10

1.13 x 10'10


2.47 x 10'3




8.85 x 10'15
Comments




















ED10, relative
potencies reported




                                            C-29
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-5. Lung implantation bioassays: dose-response information for incidence data
Record
number






Reference






Species






Target
organ






Tumor type






PAH
PH
PH
PH
CH
CH
DBahA
Dose
1
3
10
1
3
0.1
Dose
units
mg/
animal
mg/
animal
mg/
animal
mg/
animal
mg/
animal
mg/
animal
Number
of
animals
with
tumors
0
0
1
5
10
20
Number
of
animals
in group
35
35
35
35
35
35
% Tumor-
bearing
animals
0.00
0.00
0.03
0.14
0.29
0.57
SRC statistical analysis
Fisher's
exact p-value


5 x 10'1
2.7 x 10'2
4.63 x lO'4
2.01 x 10'8
Cochran-
Armitage trend
testp-value


1

7.96 x 10'4

Comments






                                           C-30
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-6.  Oral bioassays: dose-response information for incidence data
Record
number
24801







Reference
Weyand et al., 2004







Species
Mouse







Target
organ
Lung



Fore-
stomach



Tumor type
Adenoma



Squamous
cell
carcinoma



PAH
Control
BaP
BcFE
BcFE
Control
BaP
BcFE
BcFE
Dose
0
230
13.6
197
0
230
13.6
197
Dose units
Hg/mouse/
day
Hg/mouse/
day
Hg/mouse/
day
Hg/mouse/
day
Hg/mouse/
day
Hg/mouse/
day
Hg/mouse/
day
Hg/mouse/
day
Number
of
animals
with
tumors
7
21
13
29
0
10
0
0
Number
of
animals
in group
29
27
28
29
29
27
28
29
% Tumor-
bearing
animals
0.24
0.77
0.46
1
0
0.36
0
0
SRC statistical analysis
Fisher's
exact p-value

>0.0001
0.0684
>0.0001




Cochran-
Armitage trend
testp-value








Comments








Table C-7.  Oral bioassays: dose-response information for tumor multiplicity






Reference
24801













Species
Weyand et
al., 2004











Exposure
route
Mouse












Target
organ
Lung













Tumor type
Adenoma













PAH
Control

BaP

BcFE

BcFE







Sex
F

F

F

F







Dose
0

230

13.6

197







Dose units
ug/mouse/
day
ug/mouse/
day
ug/mouse/
day
ug/mouse/
day



Number
of animals
with
tumors
7

21

13

29





Number
of animals
in group
29

27

28

29





°/o Tumor-
bearing
animals
0.24

0.77

0.46

1



Results of
authors'
statistical
analysis
(p-value)








Results of
SRC
statistical
analysis
(Fisher's
exact
p-value)


>0.0001

0.0684

>0.0001




Mean
number
tumors/
animal
0.31

1.4

0.57

46






SDof
mean
0.59

1.14

0.69

15.1


Results of
SRC
statistical
analysis
(t-test
p-value)


>0.0001

0.13

>0.0001







Comments








                                             C-31
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-8. In vitro bacterial mutagenicity: data use
Record
number
17030
23830
23660
17380
17590
17630
9620
24030
18050
Reference
Andrews et
al., 1978
Baker et
al., 1980
Bartsch et
al., 1980
Bosetal.,
1988
Carver et
al., 1986
Cavalieri et
al., 1981a
Chang et
al., 2002
De Flora et
al., 1984
Eisenstadt
and Gold,
1978
Data
source
Figure 1
Table 2
Appendix
table
Table 1
Figure 1
Figure 1
Figure 7
Table 2
Figure 2B
Data points
Dose (ug) and number of
revertant colonies for DBacA,
DBajA, DBahA, AA, BghiP,
BeP, BaP
Use data for guinea pig-MC S9
only (column D); dose in
ug/plate and number of
revertant colonies; BaP,
DBaiP, BaA, DBacA, DBahA
Use data for BaA and BaP;
dose in umol/plate and
mutagenic activity in
revertants/umol
Use TA100 strain only; dose
(ug/plate) and number of
revertant colonies/plate for PH,
Pyr, BaP
Use curves for BaP, BaA,
BghiF, and Pery; use 400 uL
S9 per plate (last data point on
x-axis); each curve is different
dose in ug/plate, use hamster
data; revertants per plate is
y-axis
Dose-response curves for BaP,
CPcdP (CPEP in figure), and
ACEP (CPAP in figure); dose
as uM, response as mutant
fraction x 105
Dose-response curves for
BghiF, BcPH, and BaP; dose
(ug/plate) and revertants/plate
Table provides potency
estimates as revertants/nmol
for BaA, Pery, BaP, and BeP
Use TA100 data for BaP and
CPcdP (open circles); dose is
1 ug for CPcdP and 2 ug for
BaP (legend); use the same S9
concentration (20 uL/plate)
Basis for RPF approach
Point estimate
Point estimate Table 2
Point estimate
Derive point estimate for
BaP (use PH control as
background); continuous
model PH and Pyr using the
BaP response as the BMR
Point estimate; use highest
dose in hamster, except for
perylene (use 10 ug/plate);
this is maximal response in
hamsters
Model as quanta! data
(mutant fraction reported)
Point estimate; use
5 ug/plate dose for BghiF
and BaP; use 10 ug/plate for
BcPH
Calculate the RPF ratio
using the potency estimates
provided
Point estimate; single point
data (20 uL S9/plate)
Comments
TA100withArS9
TA100 with guinea
pig-MC S9; Table 1
data not used,
different S9 mix
used for each of
three experiments
TA100 rat MC S9
TA100 with rat Ar
S9
TA100 with hamster
Ar S9; multidose
data but not SD was
reported
TM677withArS9
TA100 with rat Ar
S9; SD not available
from graph (reported
for some data points,
but not all)
Determine strain
used to calculate
potencies; rat Ar S9
TA100 with rat Ar
S9; uL S9 that
maximizes the BaP
response does not
produce maximal
response for CPcdP
                            C-32
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-8. In vitro bacterial mutagenicity: data use
Record
number
18180
24080
14080
18650
10670
19000
24680
19320
Reference
Florin et
al., 1980
Gibson et
al., 1978
Gold and
Eisenstadt,
1980
Hermann,
1981
Johnsen et
al., 1997
Kaden et
al., 1979
Lafleur et
al., 1993
LaVoie et
al., 1979
Data
source
Table III
Table 1
(BaP)
Table 3
(PAHs)
Table 2
Table 1
Figure 2
Table 1
Figures 3
and 4
Table VI
Data points
Use TA100 data for BaA, CH,
and BaP, use TA98 data for
Pery, CO, and BaP; dose is
indicated as optimal dose
(umol/plate) and number
revertants/plate
Use data for TA98; in Table 1
use Expt. No.l for BaP; in
Table 3 use data for DBahA,
Tphen, BaA, BghiP, CH, FE,
Pyr; dose as ug/plate, response
as increase in revertants
Use data for 3-MC induction at
50 uL S9/plate; dose is 4 nmol
for BaP and CPcdP, results as
revertants/plate
Table provides potency
estimates as revertants/nmol
forBbA,BaA, CH, FA,
Tphen, BeP, DBacA, DBahA,
BbF, Pery, DBalP, DBaiP, AA,
CO; potency of BaP in legend
as 100 revertants/nmol
Use data for PCB microsomes
for BaP, BjAC, B1AC; dose as
ug/plate, response as revertants
RPFs calculated for AN, ANL,
Pyr, BbFE, CPcdP, BaA, CH,
Tphen, FA, BeP, Pery, BghiP,
AA, DBacA, DBahA, DBbeF
Use dose-response curves for
BaP, BghiF, CPcdP,
CPhiACEA (CPAA), ACEA
(AA), CPhiAPA (CPAP), APA
(AP); dose as ug/mL, response
as mutant fraction (* 105)
Use data for TA98 for BaP,
BeP, and Pery; 10 ug dose and
response as revertants/plate
Basis for RPF approach
Point estimate; please note
that reported response
includes subtraction of
spontaneous revertants
(control); need to use
formula for added risk; make
sure to flag in comments
Point estimate; use the dose
associated with the max-
imum response (if reported
as a range, do not use);
controls were reported as
negative (no mutagenic or
toxic response)
Point estimate
Calculate the RPF ratio
using the potency estimates
provided
Model to derive BMDsdl;
need to extract SDs from
graph; control response is
1 13 ± 9 revertants per plate
(see legend); add control
response to each response
for modeling (it was
subtracted prior to graphing)
Not applicable
Model as quanta! data
(mutant fraction reported)
Point estimate; use 20 ug for
BaP; 10 ug for BeP; and
20 ug for Pery
Comments
Note that data for
both TA100 and
TA98 strains were
used; BaP results
were provided for
each; rat MC S9
TA98 with non-
enzymatic induction
(gamma irradiation);
multidose data but
not SD reported
TA100 using 50 uL
ofratMCS9;
important to note
that maximal
response for CPcdP
occurred at much
lower dose of S9
(5 uL/plate)
TA98 with rat Ar
S9; potency
estimates were
calculated from the
linear portion of the
dose-response curve
TA98 with PCB
microsomes
TM677 with Ar S9
and PB S9
Forward mutation to
8-azaguanine
resistance in TM677
with rat AR S9
TA98 with rat Ar
S9; for BeP and
Pery the maximal
response was in
TA100
                            C-33
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-8. In vitro bacterial mutagenicity: data use
Record
number
23650
20220
20450
21000
11860
21360
21640
16180
16440
Reference
McCann et
al., 1975
Pahlman
and
Pelkonen,
1987
Phillipson
and
loannides,
1989
Sakaiet
al., 1985
Sangaiah et
al., 1983
Simmon,
1979a
Teranishi
etal., 1975
Utesch et
al., 1987
Wood et
al., 1980
Data
source
Table 1
Table 1
Figures 2
and 3
Table 3
Figure 2
Table 1
Table I
and
Figure 3
Figures 2
and 3
Chart 3A
Data points
Table provides potency
estimates as revertants/nmol
forDBaiP, BaP, BeP, DBacA,
DBahA, CH, BaA
Use data for rat-MC induced
(last column); potency
estimates are provided as
revertants/nmol for BaA, CH,
Tphen, DBacA, DBahA
Use the curve for hamster S9
(open triangles); data for BaP,
DBaiP, BaA, and DBahA, dose
as ug/plate, revertants/plate
Use data for TA97 +S9 for FE,
AC, PH, FA, Ch, Pyr, BaP,
BeP, Pery, BghiP, CO; dose
ug, response as revertants per
plate
Use data for Bj AC and BaP;
dose as ug/plate, response as
revertants/plate
Use data for TA100 for BaA,
BaP, BeP; dose as ug, response
as revertants/plate after
subtracting background
Use data for TA1538 for
DBaiP and BaP; use data in
Figure 3 for TA 1538, PB and
DBahA-induced S9 (open
circles) for DBaeP
Use data for homogenized
hepatocytes (open circles) for
BaA and BaP; dose as
ug/plate, response as
revertants/plates
Use dose-response curves for
BaP and CPcdP; dose as nmol,
response as revertants/plate
Basis for RPF approach
Calculate the RPF ratio
using the potency estimates
provided
Calculate the RPF ratio
using the potency estimates
provided
Point estimate; use
10 ug/plate for BaP,
DBahA; 20 ug/plate BaA,
DBaiP
Point estimate; use 10 ug for
AC, PH, FA, BaP, BeP; use
5 ug for FE; use 20 ug for
CH, Pyr, BghiP; use 4 ug for
Pery; use 100 ug for CO
Point estimate; use
10 ug/plate forBjAC; use
6 ug/plate for BaP
Point estimate
Point estimate
Point estimate; use
12.5 ug/plate for BaP; use
25 ug/plate for BaA
Point estimate; use 15 nmol
forBaPandCPcdP
Comments
Multiple strains, rat
ArS9
TA100 with rat MC
S9
TA100 with hamster
S9; multidose data
but not SD reported
TA97 with rat Ar
S9; multidose data
but not SD reported
TA98 with rat Ar
S9; multidose data
but not SD was
reported
TA100 with rat Ar
S9
TA1538 with rat PB
S9 for DBaiP;
TA1538withPB
and DBahA S9 for
DBaeP
TA100 with homo-
genized hepatocytes
from Ar-treated rats;
multidose data but
not SD reported
TA98 with purified
microsomal P450;
multidose data but
not SD reported
                            C-34
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-9. In vitro bacterial mutagenicity: dose-response data
Record
number
17030







23830





23660

17380

Reference
Andrews etal.,
1978







Baker etal., 1980





Bartsch et al, 1980

Bos etal., 1988

Cell type
TA100







TA100





TA100

TA100

Activation
system
ArS9







Guinea pig-
MC





Rat MC S9

RatArS9

PAH
Control
BaP
DBacA
DBajA
DBahA
AA
BghiP
BeP
Control
BaP
DBaiP
BaA
DBacA
DBahA
BaP
BaA
BaP
Control
Dose
0
250
10
10
25
250
100
1,000
0
2.5
5
10
2.5
5
0.027
0.067
7.5
0
Dose units
Hg
Mg
Hg
Hg
Hg
Hg
Hg
Mg
Hg/plate
Hg/plate
Hg/plate
Hg/plate
Hg/plate
Hg/plate
(imol/plate
(imol/plate
Hg/plate
Hg/plate
Response
150
1,681
2,957
843
617
1,796
793
643
134
1,278
737
947
1,738
1,331
29,000
6,000
824
85
Response
units
Revertant
colonies
Revertant
colonies
Revertant
colonies
Revertant
colonies
Revertant
colonies
Revertant
colonies
Revertant
colonies
Revertant
colonies
Revertant
colonies
Revertant
colonies
Revertant
colonies
Revertant
colonies
Revertant
colonies
Revertant
colonies
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
n









10
10
10
10
10


3
3
Units
















Replic-
ates
Replic-
ates
% Resp-
onse


















SD








18
97
73
47
88
98


21
12
SE
















12
7
Comments














Control response
subtracted
Control response
subtracted


                                         C-35
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-9. In vitro bacterial mutagenicity: dose-response data
Record
number







17590








Reference







Carver etal., 1986








Cell type







TA100








Activation
system







Hamster
ArS9








PAH
PH
PH
PH
Control
Pyr
Pyr
Pyr
Control
BaP
BaP
BaP
BaA
BaA
BaA
BghiF
BghiF
Dose
1
5
25
0
1
5
25
0
1
10
50
15
40
50
10
25
Dose units
Hg/plate
Hg/plate
Hg/plate
Hg/plate
Hg/plate
Hg/plate
Hg/plate
Hg/plate
Hg/plate
jig/plate
Hg/plate
Hg/plate
Hg/plate
Hg/plate
Hg/plate
Hg/plate
Response
108
167
240
86
93
164
279
140
141
482
1,035
346
892
1,263
333
727
Response
units
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
n
3
3
3
3
3
3
3









Units
Replic-
ates
Replic-
ates
Replic-
ates
Replic-
ates
Replic-
ates
Replic-
ates
Replic-
ates









% Resp-
onse
















SD
10
5
10
7
9
23
10









SE
6
3
6
4
5
13
6









Comments







Control curves
difficult to
digitize; control
value estimated
from BaP graph
and used for all
Continuous data,
noSD







                                         C-36
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-9. In vitro bacterial mutagenicity: dose-response data
Record
number




17630









9620


Reference




Cavalieri etal.,
1981a









Chang et al., 2002


Cell type




TM677









TA100


Activation
system




ArS9









RatArS9


PAH
BghiF
Perylene
Perylene
Perylene
Control
BaP
BaP
BaP
BaP
CPcdP
CPcdP
ACEP
ACEP
ACEP
Control
BaP
BghiF
Dose
50
5
10
15
0
10
20
40
60
20
40
10
40
120
0
5
5
Dose units
Mg/plate
Hg/plate
Hg/plate
Mg/plate
MM
MM
MM
MM
MM
MM
MM
MM
MM
MM
Mg/plate
Mg/plate
Mg/plate
Response
985
195
993
922
5
15
26
84
131
34
133
11
25
55
326
2,543
1,630
Response
units
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Mutants
Mutants
Mutants
Mutants
Mutants
Mutants
Mutants
Mutants
Mutants
Mutants
Revertants/
plate!
Revertants/
plate]
Revertants/
plate]
n




1 x 105
1 x 105
1 x IQ5
1 x IQ5
1 x IQ5
1 x IQ5
1 x IQ5
1 x IQ5
1 x IQ5
1 x IQ5



Units




Surviv-
ors
Surviv-
ors
Surviv-
ors
Surviv-
ors
Surviv-
ors
Surviv-
ors
Surviv-
ors
Surviv-
ors
Surviv-
ors
Surviv-
ors



% Resp-
onse




0.000050
0.000150
0.000256
0.000839
0.001308
0.000337
0.001330
0.000110
0.000248
0.000551



SD

















SE

















Comments




Control value
estimated









SDnot
consistently
plotted; extracted
only point estimate
data


                                         C-37
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-9. In vitro bacterial mutagenicity: dose-response data
Record
number

24030



18050

18180





24080
Reference

De Flora etal.,
1984



Eisenstadt and
Gold, 1978

Florin etal., 1980





Gibson etal., 1978
Cell type

RatAR
S9



TA100

TA100
TA100
TA100
TA98
TA98
TA98
TA98
Activation
system





RatArS9

RatMCS9





[60Co]
gamma
radiation,
for7d
(2.5 x
107rad)
PAH
BcPH
BaP
BaA
Pery
BeP
BaP
CPcdP
BaP
BaA
CH
BaP
Pery
CO
Control
Dose
10




2
1
0.0030
0.10
0.0050
0.0030
0.025
0.070
0
Dose units
Hg/plate




Mg
Mg
(imol/plate
(imol/plate
(imol/plate
(imol/plate
(imol/plate
(imol/plate
Hg/plate
Response
1,043
185
12
21
1.6
1,705
134
255
326
196
235
91
82
0
Response
units
Revertants/
plate!
Revertants/
nmol
(potency)
Revertants/
nmol
(potency)
Revertants/
nmol
(potency)
Revertants/
nmol
(potency)
Revertants/
plate!
Revertants/
plate!
Revertants/
plate!
Revertants/
plate!
Revertants/
plate!
Revertants/
plate!
Revertants/
plate!
Revertants/
plate!
Increase in
revertants
n














Units














% Resp-
onse














SD














SE














Comments





Background
subtracted from
data reported

Background
subtracted from
data reported
Only peak
response reported




Continuous data,
noSD
                                         C-38
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-9. In vitro bacterial mutagenicity: dose-response data
Record
number














14080

18650
Reference














Gold and
Eisenstadt, 1980

Hermann, 1981
Cell type














TA100

TA98
Activation
system














50 (iL rat
MCS9

RatArS9
PAH
BaP
BaP
BaP
BaP
BaP
BaP
BaA
BaA
BghiP
CH
CH
FE
FE
Pyr
BaP
CPcdP
BaP
Dose
10
20
50
100
200
300
150
250
400
500
1,000
200
360
160
4
4

Dose units
Hg/plate
Hg/plate
Hg/plate
Hg/plate
Hg/plate
Hg/plate
Hg/plate
Hg/plate
Hg/plate
Hg/plate
Hg/plate
Hg/plate
Hg/plate
Hg/plate
nmol
nmol

Response
1.5
3
10
15
21
35
1.8
6.4
4.2
6.1
6.7
1.1
2.2
28
1,103
281
100
Response
units
Increase in
revertants
Increase in
revertants
Increase in
revertants
Increase in
revertants
Increase in
revertants
Increase in
revertants
Increase in
revertants
Increase in
revertants
Increase in
revertants
Increase in
revertants
Increase in
revertants
Increase in
revertants
Increase in
revertants
Increase in
revertants
Revertants/
plate!
Revertants/
plate!
Revertants/
nmol
(potency)
n

















Units

















% Resp-
onse

















SD

















SE

















Comments














Background
subtracted from
data reported


                                         C-39
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-9. In vitro bacterial mutagenicity: dose-response data
Record
number













Reference













Cell type













Activation
system













PAH
BbA
BaA
CH
FA
Tphen
BeP
DBacA
DBahA
BbF
Pery
DBalP
DBaiP
AA
Dose













Dose units













Response
8
4
2
3
13
15
42
8
15
31
21
38
62
Response
units
Revertants/
nmol
(potency)
Revertants/
nmol
(potency)
Revertants/
nmol
(potency)
Revertants/
nmol
(potency)
Revertants/
nmol
(potency)
Revertants/
nmol
(potency)
Revertants/
nmol
(potency)
Revertants/
nmol
(potency)
Revertants/
nmol
(potency)
Revertants/
nmol
(potency)
Revertants/
nmol
(potency)
Revertants/
nmol
(potency)
Revertants/
nmol
(potency)
n













Units













% Resp-
onse













SD













SE













Comments













                                         C-40
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-9. In vitro bacterial mutagenicity: dose-response data
Record
number

10670






19000










Reference

Johnsen et al., 1997






Kadenetal., 1979










Cell type

TA98






TM677










Activation
system

PCB
micro-
somes






ArS9 and
PBS9










PAH
CO
Control
BaP
BaP
BjAC
BjAC
BIAC
BIAC
BaP
AN
ANL
Pyr
BbFE
CPcdP
BaA
CH
Tphen
FA
BeP
Dose

0
10
20
10
20
10
20

NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
Dose units

Hg/plate
Hg/plate
Hg/plate
Hg/plate
Hg/plate
Hg/plate
Hg/plate











Response
60
113
128
123
192
213
204
207
1
0.010
0.070
0.070
0.080
1.5
0.14
0.20
0.070
1.0
0.11
Response
units
Revertants/
nmol
(potency)
Revertants/
plate!
Revertants/
plate]
Revertants/
plate]
Revertants/
plate]
Revertants/
plate]
Revertants/
plate]
Revertants/
plate]
RPF
RPF
RPF
RPF
RPF
RPF
RPF
RPF
RPF
RPF
RPF
n

3
3
3
3
3
3
3











Units



















% Resp-
onse



















SD

8.54
3.66
13.41
10.98
9.76
13.41
43.90











SE



















Comments

Control response
added back to each
response for
modeling






Mutagenic activity
relative to that of
the 80 |imol BaP-
positive control
performed
simultaneously
with test
compound










                                         C-41
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-9. In vitro bacterial mutagenicity: dose-response data
Record
number






24680














Reference






Lafleuretal, 1993














Cell type






TM677














Activation
system






RatARS9














PAH
Pery
BghiP
AA
DBacA
DBahA
DBbeF
BaP
BaP
BaP
BaP
BaP
BaP
BghiF
BghiF
BghiF
BghiF
CPcdP
CPcdP
CPcdP
CPcdP
CPcdP
Dose
NA
NA
NA
NA
NA
NA
0
0.5
1
2
4
8
0
1
3
10
0
0.5
1
2
4
Dose units






Hg/mL
Hg/mL
Hg/mL
|ig/mL
Hg/mL
Hg/mL
|ig/mL
|ig/mL
|ig/mL
|ig/mL
|ig/mL
|ig/mL
|ig/mL
|ig/mL
|ig/mL
Response
6
0.080
0.080
0.77
0.080
0.88
7
8
10
18
22
33
11
10
14
55
12
15
13
17
27
Response
units
RPF
RPF
RPF
RPF
RPF
RPF
Mutants
Mutants
Mutants
Mutants
Mutants
Mutants
Mutants
Mutants
Mutants
Mutants
Mutants
Mutants
Mutants
Mutants
Mutants
n






100,000
100,000
100,000
100,000
100,000
100,000
100,000
100,000
100,000
100,000
100,000
100,000
100,000
100,000
100,000
Units






Surviv-
ors
Surviv-
ors
Surviv-
ors
Surviv-
ors
Surviv-
ors
Surviv-
ors
Surviv-
ors
Surviv-
ors
Surviv-
ors
Surviv-
ors
Surviv-
ors
Surviv-
ors
Surviv-
ors
Surviv-
ors
Surviv-
ors
% Resp-
onse






0.000070
0.000080
0.000101
0.000175
0.000220
0.000327
0.00011
0.00010
0.00014
0.00055
0.000120
0.000146
0.000130
0.000172
0.000274
SD





















SE





















Comments





















                                         C-42
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-9. In vitro bacterial mutagenicity: dose-response data
Record
number

















19320
Reference

















LaVoieetal., 1979
Cell type

















TA98
Activation
system

















RatArS9
PAH
CPcdP
CPhiACE
A
CPhiACE
A
CPhiACE
A
CPhiACE
A
CPhiACE
A
CPhiAPA
CPhiAPA
CPhiAPA
CPhiAPA
ACEA
ACEA
ACEA
APA
APA
APA
APA
BaP
Dose
8
0
0.5
1
2
4
0
10
30
100
0
10
35
0
10
30
100
10
Dose units
Hg/mL
Hg/mL
Hg/mL
Hg/mL
Hg/mL
|ig/mL
|ig/mL
|ig/mL
Hg/rnL
Hg/rnL
|ig/mL
Hg/mL
|ig/mL
Hg/mL
Hg/mL
|ig/mL
|ig/mL
Mg
Response
60
8
10
16
29
67
9
12
21
26
9
21
69
16
37
42
22
450
Response
units
Mutants
Mutants
Mutants
Mutants
Mutants
Mutants
Mutants
Mutants
Mutants
Mutants
Mutants
Mutants
Mutants
Mutants
Mutants
Mutants
Mutants
Revertants/
plate}
n
100,000
100,000
100,000
100,000
100,000
100,000
100,000
100,000
100,000
100,000
100,000
100,000
100,000
100,000
100,000
100,000
100,000

Units
Surviv-
ors
Surviv-
ors
Surviv-
ors
Surviv-
ors
Surviv-
ors
Surviv-
ors
Surviv-
ors
Surviv-
ors
Surviv-
ors
Surviv-
ors
Surviv-
ors
Surviv-
ors
Surviv-
ors
Surviv-
ors
Surviv-
ors
Surviv-
ors
Surviv-
ors

% Resp-
onse
0.000597
0.000084
0.000103
0.000157
0.000286
0.000670
0.000090
0.000117
0.000210
0.000263
0.000092
0.000214
0.000686
0.000160
0.000375
0.000416
0.000220

SD


















SE


















Comments

















Background
subtracted from
data reported
                                         C-43
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-9. In vitro bacterial mutagenicity: dose-response data
Record
number




23650






20220
Reference




McCannet al.,
1975






Pahlman and
Pelkonen, 1987
Cell type




Multiple
strains






TA100
Activation
system




RatArS9






Rat MC S9
PAH
BaP
BeP
BeP
Pery
BaP
DBaiP
BeP
DBacA
DBahA
CH
BaA
BaP
Dose
20
10
20
20
NA
NA
NA
NA
NA
NA
NA
NA
Dose units
Mg
US
US
W








Response
480
20
20
70
121
20
0.6
175
11
38
11
272
Response
units
Revertants/
plate!
Revertants/
plate!
Revertants/
plate!
Revertants/
plate!
Revertants/
nmol
(potency)
Revertants/
nmol
(potency)
Revertants/
nmol
(potency)
Revertants/
nmol
(potency)
Revertants/
nmol
(potency)
Revertants/
nmol
(potency)
Revertants/
nmol
(potency)
Revertants/
nmol
(potency)
n












Units












% Resp-
onse












SD












SE












Comments




Paper states that
comparison of
potency estimates
should be done
with caution (non-
linear dose-
response), see
table footnotes







                                         C-44
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-9. In vitro bacterial mutagenicity: dose-response data
Record
number





20450










Reference





Phillipson and
loannides, 1989










Cell type





TA100










Activation
system





Hamster S9










PAH
BaA
CH
Tphen
DBacA
DBahA
BaP
BaP
BaP
BaP
BaP
BaA
BaA
BaA
BaA
BaA
DBaiP
Dose
NA
NA
NA
NA
NA
0
5
10
15
20
0
20
40
60
100
0
Dose units





Hg/plate
Hg/plate
Hg/plate
Hg/plate
Hg/plate
Hg/plate
Hg/plate
Hg/plate
Hg/plate
Hg/plate
Hg/plate
Response
10.4
9.7
4
35
4.4
0.000
68.833
118.948
99.744
96.101
0.000
109.877
115.248
114.430
98.846
0.000
Response
units
Revertants/
nmol
(potency)
Revertants/
nmol
(potency)
Revertants/
nmol
(potency)
Revertants/
nmol
(potency)
Revertants/
nmol
(potency)
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
n
















Units
















% Resp-
onse
















SD
















SE
















Comments
















                                         C-45
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-9. In vitro bacterial mutagenicity: dose-response data
Record
number









21000








Reference









Sakaietal., 1985








Cell type









TA97








Activation
system









RatArS9








PAH
DBaiP
DBaiP
DBaiP
DBaiP
DBahA
DBahA
DBahA
DBahA
DBahA
Control
BaP
BaP
BaP
Control
FE
FE
FE
FE
Dose
20
40
60
100
0
10
20
30
50
0
1
5
10
0
5
10
50
250
Dose units
Hg/plate
Hg/plate
Hg/plate
Hg/plate
Hg/plate
Hg/plate
Hg/plate
Hg/plate
Hg/plate
Hg
Mg
Mg
Mg
Hg
Hg
Hg
Mg
Mg
Response
64.638
75.747
80.394
63.880
0.000
50.899
56.886
52.419
34.980
177
1,208
1,432
1,742
189
254
240
240
232
Response
units
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
n


















Units


















% Resp-
onse


















SD


















SE


















Comments


















                                         C-46
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-9. In vitro bacterial mutagenicity: dose-response data
Record
number


















Reference


















Cell type


















Activation
system


















PAH
Control
AC
AC
AC
AC
Control
PH
PH
PH
PH
Control
FA
FA
FA
FA
Control
CH
CH
Dose
0
5
10
50
250
0
5
10
50
250
0
5
10
50
250
0
5
10
Dose units
Hg
Mg
Hg
Hg
Hg
Hg
Hg
Hg
Hg
Hg
Hg
Hg
Hg
Hg
Hg
Hg
Mg
Mg
Response
189
360
509
293
279
189
454
534
321
T
177
652
1,012
1,042
518
177
640
815
Response
units
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
n


















Units


















% Resp-
onse


















SD


















SE


















Comments


















                                         C-47
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-9. In vitro bacterial mutagenicity: dose-response data
Record
number


















Reference


















Cell type


















Activation
system


















PAH
CH
CH
Control
Pyr
Pyr
Pyr
Pyr
Pyr
Pyr
Control
BeP
BeP
BeP
BeP
Control
Pery
Pery
Pery
Dose
20
50
0
2
4
6
10
20
50
0
5
10
50
250
0
1
2
4
Dose units
Hg
Mg
Hg
Hg
Hg
Hg
Hg
Hg
Hg
Hg
Hg
Mg
Mg
Hg
Mg
Hg
Mg
Mg
Response
888
723
177
929
1,582
2,057
2,577
2,832
2,296
177
944
1,100
606
640
177
1,516
2,236
2,784
Response
units
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
n


















Units


















% Resp-
onse


















SD


















SE


















Comments


















                                         C-48
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-9. In vitro bacterial mutagenicity: dose-response data
Record
number













11860




Reference













Sangaiahetal.,
1983




Cell type













TA98




Activation
system













RatArS9




PAH
Pery
Pery
Control
BghiP
BghiP
BghiP
BghiP
Control
CO
CO
CO
CO
CO
Control
BaP
BaP
BaP
BaP
Dose
10
50
0
10
20
50
250
0
5
10
50
100
200
0
2
3
6
10
Dose units
Hg
Mg
Hg
Hg
Hg
Hg
Hg
Hg
Hg
Hg
Hg
Mg
Mg
Hg/plate
Hg/plate
Hg/plate
Hg/plate
Hg/plate
Response
2,550
1,808
177
896
991
896
612
177
362
400
405
490
479
35.43
177.37
266.02
419.68
312.76
Response
units
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
n


















Units


















% Resp-
onse


















SD


















SE


















Comments


















                                         C-49
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-9. In vitro bacterial mutagenicity: dose-response data
Record
number











21360


21640

Reference











Simmon, 1979a


Teranishi et al,
1975

Cell type











TA100


TA1538

Activation
system











RatArS9


RatPBS9

PAH
BaP
BaP
BaP
Control
BjAC
BjAC
BjAC
BjAC
BjAC
BjAC
BjAC
BaP
BaA
BeP
Control
BaP
Dose
30
50
100
0
2
3
6
10
30
50
100
5
50
50
0
50
Dose units
Hg/plate
Hg/plate
Hg/plate
Hg/plate
Hg/plate
Hg/plate
Hg/plate
Hg/plate
Hg/plate
Hg/plate
Hg/plate
Mg
Mg
Mg
Hg/plate
Hg/plate
Response
358.41
350.92
323.12
53.15
124.15
331.10
674.11
993.21
1,027.06
883.45
1,021.36
1,141
280
57
38
77
Response
units
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertant
colonies/
plate
Revertant
colonies/
plate
n
















Units
















% Resp-
onse
















SD
















SE
















Comments











Background
subtracted from
data reported




                                         C-50
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-9. In vitro bacterial mutagenicity: dose-response data
Record
number




16180










Reference




Uteschetal., 1987










Cell type

TA1538


TA100










Activation
system

Rat PB and
DBahA S9


With
homogen-
ized
hepatocytes
from Ar-
treated rats










PAH
DBaiP
Control
BaP
DBaeP
Control
BaP
BaP
BaP
BaP
BaP
Control
BaA
BaA
BaA
BaA
Dose
50
0
50
50
0
6.3
12.5
25
50
100
0
6.3
12.5
25
50
Dose units
Hg/plate
Hg/plate
Hg/plate
Hg/plate
Hg/plate
Hg/plate
Hg/plate
Hg/plate
Hg/plate
Hg/plate
Hg/plate
Hg/plate
Hg/plate
Hg/plate
Hg/plate
Response
102
25
279
88
159
998
1,079
1,178
1,141
1,114
199
861
2,583
3,546
3,786
Response
units
Revertant
colonies/
plate
Revertant
colonies/
plate
Revertant
colonies/
plate
Revertant
colonies/
plate
Revertants/
plates
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
n















Units















% Resp-
onse















SD















SE















Comments















                                         C-51
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-9. In vitro bacterial mutagenicity: dose-response data
Record
number

16440









Reference

Wood etal., 1980









Cell type

TA98









Activation
system

Purified
microsomal
P450









PAH
BaA
Control
BaP
BaP
BaP
BaP
Control
CPcdP
CPcdP
CPcdP
CPcdP
Dose
100
0
3.75
7.5
15
30
0
3.75
7.5
15
30
Dose units
Hg/plate
nmol
nmol
nmol
nmol
nmol
nmol
nmol
nmol
nmol
nmol
Response
3,406
0
45
63
99
103
0
303
491
685
776
Response
units
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
Revertants/
plate
n











Units











% Resp-
onse











SD











SE











Comments

Background
subtracted from
data reported









                                         C-52
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-10. In vitro mammalian mutagenicity: data use
Record
number
16920
16940
16910
17140
14250
18740
18990
24720
Reference
Amacher and
Paillet, 1982
Amacher and
Turner, 1980
Amacher et
al., 1980
Barfknecht et
al., 1982
Hass et al.,
1982
Huberman
and Sachs,
1976
Jotz and
Mitchell,
1981
Kligerman et
al., 1986
Data
source
Figure 1
Figure 3
Table 3
Figure 2
(BaP, FA);
Figure 4
(BaA, CH,
Tphen);
Figure 6
(CPcdP)
Table 1
Table 2
Table 2
Figure 1
Data points
Use lines for BaP (open
circles) and BaA (closed
triangles; dose is ug/mL and
response is mutation
frequency (MF)/106 survivors
Use bars for SM2 S9
activation for BaP and BaA;
dose is 1.25 x 10"5MforBaP
and 3.22 x 10"5MforBaP;
response is IMF/104 survivors
Use dose-response data for
BaA and BaP; dose as
concentration (M), response
as mutants per 104 survivors
Dose is uM and mutant
fraction xlO6
Dose-response data for
DBaiP, DBahP, and BaP; dose
is ug/mL; use response data
for TG mutants only
(mutants/106 cells); control
value is 4 ± 1 mutants/
106 cells
Use data for BaP, DBacA,
DBahA; 8-azaguanine
resistance only; use 1 ug/mL
dose for all (*), response as
mutants per 105 survivors
Use data for BaP and Pyr with
metabolic activation; subtract
negative control, dose as
ug/mL, response as MF x 10"6
Use dose-response data for
BaP and B1AC; dose as
ug/mL, response as mutant
frequency/106 survivors;
average data from two
experiments
Basis for RPF
approach
Model;
quanta! data
Point estimate
Model;
quanta! data
Model;
quanta! data
Model;
quanta! data
Point estimate
Point estimate
Model;
quanta! data
Comments
Thymidine kinase assay
(resistance to trifluorothymi-
dine) in mouse lymphoma cells
(L5178Y) with Syrian golden
hamster S9 mix or cocultivated
hamster hepatocytes
Thymidine kinase assay
(resistance to trifluorothymi-
dine) in mouse lymphoma cells
(L5178Y) with mouse S9 mix
Thymidine kinase assay
(resistance to trifluorothymi-
dine) in mouse lymphoma cells
(L5178Y) with mouse S9 mix
Thymidine kinase assay
(resistance to trifluorothymi-
dine) in human lymphoblast
cells with rat Ar S9 mix
Hypoxanthine-guanine phos-
phoribosyl transferase assay
(resistance to 6-thioguanine) in
V79 Chinese hamster cells
with rat MC S9
Hypoxanthine-guanine phos-
phoribosyl transferase assay
(resistance to 8-azaguanine) in
V79 Chinese hamster cells
with hamster embryo cells
Thymidine kinase assay
(resistance to trifluorothymi-
dine) in mouse lymphoma cells
(L5178Y) with rat ArS9
Thymidine kinase assay
(resistance to trifluorothymi-
dine) in mouse lymphoma cells
(L5178Y) with rat ArS9
                           C-53
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-10. In vitro mammalian mutagenicity: data use
Record
number
19180
24680
7550
11450
15630
15640
21410
16190
21900
Reference
Krahn and
Heidelberger,
1977
Lafleur etal.,
1993
Li and Lin,
1996
Nesnow et
al., 1984
Raveh and
Huberman,
1983
Raveh etal.,
1982
Slagaetal.,
1978
Vacaetal.,
1992
Wangenheim
and
Bolcsfoldi,
1988
Data
source
Table II
Figures 5
and 6
Text
Chart 9
Table 1
Figure 4
Table 3
Figure 5
Table 1
Data points
Use data for BaP, DBahA,
DBacA, and BaA; cell
survival at 40% control
(column 3), controls are 100%
survival group (column 1);
use 3-MC S9 data only; dose
as nmol/mL, response as
6-TG/105 cells
Use dose-response curves for
BaP, CPcdP (CPP),
CPhiACEA (CPAA), ACEA
(AA); dose as ug/mL,
response as mutant fraction
(ppm)
Mutant frequency of controls
2 x lO'5; 10 ng/mL BaP = 5 x
10'5; BaA =5.6 x 10'5
Use data for BaP, B1AC,
BeAC, and BjAC; dose as
ug/mL, response as
6TG-resistant mutants/
106 survivors
Use data for CPcdP and BaP,
with PMA only; dose in
ug/mL, response in
mutants/105 cells
Use dose-response data for
CPcdP and BaP (ouabain
resistance only); dose in
ug/mL, response in
mutants/106 cells
Use dose-response data for
BaA and BaP; dose as uM,
response as ouabain resistant
mutants/104 survivors
Dose-response data for FA
and BaP; dose as uM,
response as 6-Tg resistant
cells/100,000
Use +S9 dose-response data
for Pyr, BaP, and FE; dose as
mol/L, response as mutation
frequency
Basis for RPF
approach
Point estimate
Model as
quanta! data
(mutant
fraction
reported)
Point estimate
Model;
quanta! data
Model;
quanta! data
Model;
quanta! data
Model;
quanta! data
Model;
quanta! data
Model;
quanta! data
Comments
Hypoxanthine-guanine
phosphoribosyl transferase
assay (resistance to 6-thio-
guanine) in V79 Chinese
hamster cells with hamster
embryo cells
Thymidine kinase assay
(resistance to trifluorothymi-
dine) in MCL-3 cells (human
B-lymphoblastoid cells)
Hypoxanthine-guanine phos-
phoribosyl transferase assay
(resistance to 6-thioguanine) in
HS1 HeLa cells (human
epithelial cells)
Hypoxanthine-guanine phos-
phoribosyl transferase assay
(resistance to 6-thioguanine) in
V79 Chinese hamster cells
with rat AR S9
Hypoxanthine-guanine
phosphoribosyl transferase
assay (resistance to 6-thio-
guanine) in V79 Chinese
hamster cells with hamster
embryo cells
Hypoxanthine-guanine phos-
phoribosyl transferase assay
(resistance to ouabain) in V79
Chinese hamster cells with
hamster embryo cells
Hypoxanthine-guanine phos-
phoribosyl transferase assay
(resistance to ouabain) in V79
Chinese hamster cells with
hamster embryo cells
Hypoxanthine-guanine
phosphoribosyl transferase
assay (resistance to 6-thio-
guanine) in UV-sensitive CHO
cells with rat Ar S9
Thymidine kinase assay
(resistance to trifluoro-
thymidine) in mouse lymph-
oma cells (L5 178Y) with rat
ArS9
                           C-54
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-10. In vitro mammalian mutagenicity: data use
Record
number
24670
Reference
Durantetal.,
1999
Data
source
Table 1
Data points
Use dose-response data for
BaPery, BbPery, DBaeF,
DBafF, DBahP, DBaiP,
DBelP, N23aP, N23eP;
positive control is reported as
1,000 ng/mL BaP (reported
separately for each PAH)
Basis for RPF
approach
Model;
quanta! data
Comments
Thymidine kinase assay
(resistance to trifluoro-
thymidine) in human hi Alv2
cells
                           C-55
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-ll. In vitro mammalian mutagenicity:  dose-response data
Record
number
16920









16940


16910
















Reference
Amacher and
Paillet, 1982









Amacher and
Turner, 1980


Amacher et al,
1980
















PAH
Control
BaP
BaP
BaP
BaP
Control
BaA
BaA
BaA
BaA
Control
BaP
BaA
Control
BaP
BaP
BaP
BaP
BaP
BaP
BaP
BaP
Control
BaA
BaA
BaA
BaA
BaA
BaA
BaA
Dose
0
2.5
5
7.5
10
0
2.5
5
10
15
0
1.25 x 10'5
3.22 x 10'5
0
5.30 x 10'6
7.00 x 10'6
9.40 x 10'6
1.25 x 10'5
1.67x 10'5
2.23 x 10'5
2.97 x 10'5
3.96 x 10'5
0
1.36x 10'5
1.81 x 10'5
2.42 x 10'5
3.22 x 10'5
4.30 x 10'5
5.47 x 10'5
7.65 x 10'5
Dose units
|ig/mL
|ig/mL
|ig/mL
|ig/mL
|ig/mL
|ig/mL
|ig/mL
|ig/mL
|ig/mL
|ig/mL
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
M
Mutants
39
119
170
196
267
20
65
62
88
89
0.4
2.85
3.12
0.680
1.360
1.790
1.470
1.870
2.600
2.490
2.650
3.970
0.770
0.810
0.840
1.000
1.230
1.470
NS
NS
In number
1 x 106
1 x 106
1 x 106
1 x IQ6
1 x IQ6
1 x 106
1 x IQ6
1 x IQ6
1 x 106
1 x 106
1 x IQ4
1 x IQ4
1 x IQ4
1 x IQ4
1 x IQ4
1 x IQ4
1 x IQ4
1 x IQ4
1 x IQ4
1 x IQ4
1 x IQ4
1 x IQ4
1 x IQ4
1 x IQ4
1 x IQ4
1 x IQ4
1 x IQ4
1 x IQ4
1 x IQ4
1 x IQ4
Units
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
% Response
0.000039
0.00012
0.00017
0.00020
0.00027
0.000020
0.000065
0.000062
0.000088
0.000089
0.000040
0.000285
0.000312
0.000068
0.000136
0.000179
0.000147
0.000187
0.000260
0.000249
0.000265
0.000397
0.000077
0.000081
0.000084
0.000100
0.000123
0.000147


Comments










Control without S9
treatment

















NS = no survivors

                                        C-56
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-ll. In vitro mammalian mutagenicity:  dose-response data
Record
number

17140
























24670




Reference

Barfknechtet al,
1982
























Durantetal., 1999




PAH
BaA
Control
BaP
BaP
BaP
Control
FA
FA
FA
Control
BaA
BaA
BaA
BaA
Control
CH
CH
CH
Control
Tphen
Tphen
Tphen
Control
CPcdP
CPcdP
CPcdP
BaP
BaP
BaP
BaP
BaP
Dose
1.02x 10'4
0
10
20
30
0
10
20
40
0
20
50
100
150
0
20
50
100
0
50
100
200
0
23
47
88
1,000
1,000
1,000
1,000
1,000
Dose units
M
MM
MM
MM
MM
MM
MM
MM
MM
MM
MM
MM
MM
MM
MM
MM
MM
MM
MM
MM
MM
MM
MM
MM
MM
MM
ng/mL
ng/mL
ng/mL
ng/mL
ng/mL
Mutants
NS
0
51
120
155
0
27
50
62
0
12
29
34
64
0
17
26
30
0
10
20
35
3
11
24
27
170
170
200
200
160
In number
1 x 104
1 x 106
1 x 106
1 x IQ6
1 x IQ6
1 x 106
1 x IQ6
1 x 1Q6
1 x 106
1 x IQ6
1 x 1Q6
1 x 106
1 x IQ6
1 x 1Q6
1 x 106
1 x IQ6
1 x IQ6
1 x 106
1 x IQ6
1 x IQ6
1 x 106
1 x IQ6
1 x IQ6
1 x IQ6
1 x IQ6
1 x IQ6
1 x 1Q6
1 x IQ6
1 x IQ6
1 x 1Q6
1 x 106
Units
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
% Response

0.000000
0.000051
0.000120
0.000155
0.000000
0.000027
0.000050
0.000062
0.000000
0.000012
0.000029
0.000034
0.000064
0.000000
0.000017
0.000026
0.000030
0.000000
0.000010
0.000020
0.000035
0.000003
0.000011
0.000024
0.000027
0.00017
0.00017
0.00020
0.00020
0.00016
Comments































                                        C-57
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-ll. In vitro mammalian mutagenicity:  dose-response data
Record
number






























Reference






























PAH
BaP
BaP
BaP
BaP
Averaged
BaP
Averaged
controls
Control
BaPery
BaPery
BaPery
BaPery
BaPery
BaPery
Control
BbPery
BbPery
BbPery
BbPery
Control
DBaeF
DBaeF
DBaeF
DBaeF
Control
DBafF
DBafF
DBafF
DBafF
Control
DBahP
Dose
1,000
1,000
1,000
1,000
1,000
0
0
0.1
0.3
1
3
10
100
0
1
3
10
100
0
1
10
100
1,000
0
1
10
100
1,000
0
0.1
Dose units
ng/mL
ng/mL
ng/mL
ng/mL
ng/mL
ng/mL
ng/mL
ng/mL
ng/mL
ng/mL
ng/mL
ng/mL
ng/mL
ng/mL
ng/mL
ng/mL
ng/mL
ng/mL
ng/mL
ng/mL
ng/mL
ng/mL
ng/mL
ng/mL
ng/mL
ng/mL
ng/mL
ng/mL
ng/mL
ng/mL
Mutants
170
190
200
210
186
20
18
21
23
28
50
82
200
18
19
22
32
54
21
29
72
190
np
21
21
37
81
190
19
24
In number
1 x 106
1 x 106
1 x 106
1 x 1Q6
1 x 106
1 x 106
1 x IQ6
1 x IQ6
1 x IQ6
1 x IQ6
1 x IQ6
1 x IQ6
1 x 106
1 x IQ6
1 x IQ6
1 x 106
1 x IQ6
1 x IQ6
1 x 106
1 x 106
1 x 106
1 x 106
1 x 106
1 x IQ6
1 x IQ6
1 x 106
1 x IQ6
1 x IQ6
1 x 106
1 x 106
Units
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
% Response
0.00017
0.00019
0.00020
0.00021
0.00019
0.00002
0.000018
0.000021
0.000023
0.000028
0.000050
0.000082
0.00020
0.000018
0.000019
0.000022
0.000032
0.000054
0.000021
0.000029
0.000072
0.00019

0.000021
0.000021
0.000037
0.000081
0.00019
0.000019
0.000024
Comments






















Not plated due to
excessive toxicity







                                        C-58
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-ll. In vitro mammalian mutagenicity:  dose-response data
Record
number























14250








Reference























Hassetal., 1982








PAH
DBahP
DBahP
DBahP
Control
DBaiP
DBaiP
DBaiP
DBaiP
Control
DBelP
DBelP
DBelP
Control
N23aP
N23aP
N23aP
N23aP
N23aP
Control
N23eP
N23eP
N23eP
N23eP
Control
BaP
BaP
DBaiP
DBaiP
DBaiP
DBahP
DBahP
DBahP
Dose
1
10
100
0
0.3
1
10
100
0
10
100
1,000
0
0.1
1
10
100
1,000
0
1
10
100
1,000
0
0.30
1.00
0.03
0.10
0.30
0.03
0.10
0.30
Dose units
ng/mL
ng/mL
ng/mL
ng/mL
ng/mL
ng/mL
ng/mL
ng/mL
ng/mL
ng/mL
ng/mL
ng/mL
ng/mL
ng/mL
ng/mL
ng/mL
ng/mL
ng/mL
ng/mL
ng/mL
ng/mL
ng/mL
ng/mL
Hg/mL
Hg/mL
Hg/mL
Hg/mL
Hg/mL
Hg/mL
Hg/mL
|ig/mL
Hg/rnL
Mutants
24
46
80
20
20
35
88
150
21
28
34
55
21
23
44
84
94
73
19
20
41
74
98
4
267
293
124
289
1211
110
264
668
In number
1 x 106
1 x 106
1 x 106
1 x 1Q6
1 x 106
1 x 106
1 x IQ6
1 x 106
1 x IQ6
1 x IQ6
1 x 106
1 x IQ6
1 x IQ6
1 x 1Q6
1 x IQ6
1 x IQ6
1 x 1Q6
1 x IQ6
1 x IQ6
1 x IQ6
1 x 106
1 x IQ6
1 x 1Q6
1 x 106
1 x IQ6
1 x 1Q6
1 x 106
1 x IQ6
1 x 1Q6
1 x 106
1 x IQ6
1 x IQ6
Units
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
CFC
CFC
CFC
CFC
CFC
CFC
CFC
CFC
CFC
% Response
0.000024
0.000046
0.000080
0.000020
0.000020
0.000035
0.000088
0.00015
0.000021
0.000028
0.000034
0.000055
0.000021
0.000023
0.000044
0.000084
0.000094
0.000073
0.000019
0.000020
0.000041
0.000074
0.00010
0.0000040
0.00027
0.00029
0.00012
0.00029
0.00121
0.00011
0.00026
0.00067
Comments
































                                        C-59
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-ll. In vitro mammalian mutagenicity:  dose-response data
Record
number
18740



18990



24720







19180



24680







Reference
Huberman and
Sachs, 1976



Jotz and Mitchell,
1981



Kligerman et al,
1986







Krahn and
Heidelberger, 1977



Lafleur et al., 1993







PAH
Control
BaP
DBacA
DBahA
Control
BaP
Control
Pyr
Control
BaP
BaP
BaP
Control
B1AC
B1AC
B1AC
Control
BaP
Control
BaA
Control
BaP
BaP
BaP
BaP
BaP
Control
ACEA
Dose
0
1
1
1
0
4.5
0
10.6
0
2.0
3.0
4.0
0
0.5
2.5
5.0
0
15.9
0
46.5
0
0.02
0.06
0.2
1
5
0
1
Dose units
Hg/mL
Hg/mL
Hg/mL
Hg/mL
|ig/mL
Hg/mL
Hg/mL
|ig/mL
nmol/mL
nmol/mL
nmol/mL
nmol/mL
nmol/mL
nmol/mL
nmol/mL
nmol/mL
nmol/mL
nmol/mL
nmol/mL
nmol/mL
|ig/mL
|ig/mL
|ig/mL
|ig/mL
|ig/mL
|ig/mL
|ig/mL
|ig/mL
Mutants
6
425
22
17
80
224
116
150
92
258
417
557
90
93
197
374
1.7
14
1.5
6.5
1.2
4.8
24
25
39
56
1.8
6.0
In number
1 x 105
1 x 105
1 x IQ5
1 x IQ5
1 x IQ6
1 x IQ6
1 x 106
1 x IQ6
1 x IQ6
1 x IQ6
1 x 1Q6
1 x IQ6
1 x IQ6
1 x IQ6
1 x 106
1 x IQ6
1 x IQ5
1 x IQ5
1 x IQ5
1 x IQ5
1 x 1Q6
1 x 106
1 x IQ6
1 x 1Q6
1 x 106
1 x IQ6
1 x 1Q6
1 x 106
Units
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
% Response
0.000060
0.00425
0.00022
0.00017
0.000080
0.00022
0.00012
0.00015
0.00009
0.00026
0.00042
0.00056
0.00009
0.00009
0.00020
0.00037
0.000017
0.000136
0.000015
0.000065
0.0000012
0.0000048
0.000024
0.000025
0.000039
0.000056
0.0000018
0.0000060
Comments





With metabolic
activation

With metabolic
activation
Average of two
experiments








3-MC S9; 40% survival

3-MC S9; 40% survival








                                        C-60
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-ll. In vitro mammalian mutagenicity:  dose-response data
Record
number












7550


11450
















Reference












Li and Lin, 1996


Nesnowetal., 1984
















PAH
ACEA
ACEA
Control
CPcdP
CPcdP
CPcdP
CPcdP
CPcdP
Control
CPhiACEA
CPhiACEA
CPhiACEA
Control
BaP
BaA
Control
BaP
BaP
BaP
BaP
BaP
BaP
BeAC
BeAC
BeAC
BeAC
BeAC
BjAC
BjAC
BjAC
BjAC
BjAC
Dose
3
8
0
0.03
0.06
0.2
0.6
2
0
0.1
0.3
0.8
0
10
10
0
0.5
1.0
2.5
5.0
10.0
20.0
1.0
2.5
5.0
10.0
20.0
1.0
2.5
5.0
10.0
20.0
Dose units
Hg/mL
Hg/mL
Hg/mL
Hg/mL
Hg/mL
|ig/mL
Hg/rnL
|ig/mL
|ig/mL
Hg/mL
|ig/mL
|ig/mL
ng/mL
ng/mL
ng/mL
|ig/mL
|ig/mL
|ig/mL
|ig/mL
|ig/mL
|ig/mL
|ig/mL
|ig/mL
|ig/mL
|ig/mL
|ig/mL
|ig/mL
|ig/mL
|ig/mL
|ig/mL
|ig/mL
|ig/mL
Mutants
15
21
2.5
4.2
4.9
5.9
10
17
2.8
12
25
31
2
5
5.6
16
10
46
72
206
215
293
17
53
435
235
349
24
94
268
225
215
In number
1 x 106
1 x 106
1 x 106
1 x 1Q6
1 x 106
1 x 106
1 x IQ6
1 x 106
1 x IQ6
1 x IQ6
1 x 106
1 x IQ6
1 x IQ5
1 x IQ5
1 x IQ5
1 x IQ6
1 x 1Q6
1 x IQ6
1 x IQ6
1 x IQ6
1 x 106
1 x IQ6
1 x 1Q6
1 x 106
1 x IQ6
1 x 1Q6
1 x 106
1 x IQ6
1 x 1Q6
1 x 106
1 x IQ6
1 x IQ6
Units
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
% Response
0.000015
0.000021
0.0000025
0.0000042
0.0000049
0.0000059
0.000010
0.000017
0.0000028
0.000012
0.000025
0.000031
0.000020
0.000050
0.000056
0.000016
0.000010
0.000046
0.000072
0.000206
0.000215
0.000293
0.000017
0.000053
0.000435
0.000235
0.000349
0.000024
0.000094
0.000268
0.000225
0.000215
Comments
































                                        C-61
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-ll. In vitro mammalian mutagenicity:  dose-response data
Record
number





15630




15640







21410





16190






Reference





Raveh and
Huberman, 1983




Raveh etal., 1982







Slagaetal., 1978





Vaca etal., 1992






PAH
B1AC
B1AC
B1AC
B1AC
B1AC
Control
BaP
BaP
CPcdP
CPcdP
BaP
BaP
BaP
BaP
CPcdP
CPcdP
CPcdP
CPcdP
Control
BaA
BaA
BaP
BaP
BaP
BaP
BaP
BaP
BaP
FA
FA
FA
Dose
1.0
2.5
5.0
10.0
20.0
0
0.3
1
0.3
1
0
0.3
1
3
0
0.3
1
3
0
4.4
44.0
0.4
1.3
4.0
0
2
4
10
0
5
7.5
Dose units
Mg/mL
Mg/mL
Hg/mL
Mg/mL
Mg/mL
Mg/mL
Mg/mL
Mg/mL
Mg/mL
Hg/mL
Mg/mL
Mg/mL
Mg/mL
Mg/mL
Mg/mL
Mg/mL
Mg/mL
Mg/mL
MM
MM
MM
MM
MM
MM
MM
MM
MM
MM
MM
MM
MM
Mutants
31
454
320
704
769
3
25
103
9
20
7
20
74
74
1
5
10
28
0.7
0.9
2.1
11.0
25.0
99.0
3
10
23
31
10
20
27
In number
1 x 106
1 x 106
1 x 106
1 x 1Q6
1 x 106
1 x IQ5
1 x IQ5
1 x IQ5
1 x IQ5
1 x IQ5
1 x 1Q6
1 x 106
1 x IQ6
1 x 1Q6
1 x 106
1 x IQ6
1 x IQ6
1 x 106
1 x IQ4
1 x IQ4
1 x IQ4
1 x IQ4
1 x IQ4
1 x IQ4
1 x IQ5
1 x IQ5
1 x IQ5
1 x IQ5
1 x IQ5
1 x IQ5
1 x IQ5
Units
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
CFC
CFC
CFC
CFC
CFC
CFC
CFC
CFC
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
% Response
0.000031
0.000454
0.000320
0.000704
0.000769
0.000030
0.00025
0.0010
0.000090
0.00020
0.0000070
0.000020
0.000074
0.000074
0.0000010
0.0000047
0.000010
0.000028
0.000070
0.000090
0.00021
0.0011
0.0025
0.0099
0.000032
0.000102
0.000229
0.000306
0.000105
0.000203
0.000274
Comments































                                        C-62
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-ll. In vitro mammalian mutagenicity:  dose-response data
Record
number

21900



















Reference

Wangenheim and
Bolcsfoldi, 1988



















PAH

Control
Control
Average
BaP
BaP
BaP
Control
FE
FE
FE
FE
FE
Control
Control
Average
Pyr
Pyr
Pyr
Pyr
Pyr
Dose
10
0
0
0
0.000001
0.000005
0.000010
0
0.0000195
0.0000389
0.0000681
0.000122
0.000170
0
0
0
0.0000101
0.0000151
0.0000202
0.0000252
0.0000302
Dose units
HM
mol/L
mol/L
mol/L
mol/L
mol/L
mol/L
mol/L
mol/L
mol/L
mol/L
mol/L
mol/L
mol/L
mol/L
mol/L
mol/L
mol/L
mol/L
mol/L
mol/L
Mutants
32
61
62
62
65
243
858
68
92
91
114
154
147
125
106
116
162
228
345
418
650
In number
1 x 105
1 x 106
1 x 106
1 x IQ6
1 x IQ6
1 x 106
1 x IQ6
1 x 1Q6
1 x 106
1 x IQ6
1 x 1Q6
1 x 106
1 x IQ6
1 x 1Q6
1 x 106
1 x IQ6
1 x IQ6
1 x 106
1 x IQ6
1 x IQ6
1 x 106
Units
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
Survivors
% Response
0.000318
0.000061
0.000062
0.000062
0.000065
0.000243
0.00086
0.00007
0.00009
0.00009
0.00011
0.00015
0.00015
0.00013
0.00011
0.00012
0.00016
0.00023
0.00035
0.00042
0.00065
Comments


Used average of controls


















                                        C-63
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-12. In vitro malignant/morphological cell transformation: data use
Record
number
17610
17970
18020
18080
14130
14640
14700
14850
24710
24700
7980
7990
Reference
Casto, 1979
DiPaolo et al.,
1969
Dunkel etal.,
1981
Emura et al.,
1980
Greb et al.,
1980
Krolewski et
al., 1986
Laaksonen et
al., 1983
Lubetetal.,
1983
Mohapatra et
al., 1987
Nesnowetal.,
1990
Nesnowetal.,
1997
Nesnowetal.,
1994
Page
54
871

153,
154
147
1,648
62
992
327
224
1,975
2,227
Table
number
I and IV
3

I and II
1
1
4
1
1
1
I
I
Figure
number












PAHs
BaP, DBahA
BaP, DBahA,
BaA, BeP,
DBacA

BaP, BbF,
BaA, IP
BaP, CH, BaA,
BbF, DBahA,
BeP
BaP, CPcdP
BaP, BaA
BaP, BeP
BaP, BeAC,
BjAC, B1AC
BaP, B1AC
BaP, DBalP
BaP, DBahA
Data to be extracted
TF in number foci per 105 surviving
cells and dose (jig/mL)
Total transformants, total number of
colonies, and dose (jig/mL)
Use data as reported in 23720 Pienta
1 977; report under that record
T, number of transformed
colonies/1,000 survivals in 10 dishes
and dose (jig/mL)
Relative transformation rate
(potency) in percent/mmol
Transformation frequency per viable
cell x 10"3; single dose (5 |iM)
Transformation frequency (number of
foci/105 surviving cells) and dose
(uM)
DwT-III/td (dishes with Type III foci/
total dishes) and dose (jig/mL)
Number of dishes scored and percent
of dishes with Type II or Type III
foci and dose (jig/mL)
Anchorage independent
colonies/50,000 cells and dose
(Hg/mL)
Type II and III foci/dish (mean and
SD) and dose (|iM)
Type II and III Foci/dish and dose;
use 1 ng/mL dose for DBahA and
mean foci/dish (in parentheses);
single dose for BaP
Basis for
RPF
Ratio of
slopes
Point
estimate

Ratio of
slopes
Ratio of
slopes
Point
estimate
Ratio of
slopes
Ratio of
slopes
Ratio of
slope to
BaP point
estimate
Ratio of
slopes
Ratio of
slopes
Point
estimate
Comment
Data on enhancement of
viral transformation not
used; no straightforward
way to model dose-
response







Use BaP incidence as
BMR



Notes
Model as incidence data
using multistage
Do not use percent
transformants; appears to be
error for DBahA

Model as incidence data
using multistage
Relative transformation
potency at LC50; slope
already calculated
Use only BaP and CPcdP
alone (not with IVA/AIA)
Inverse dose-response
relationship possible due to
cytotoxicity; use peak
Control data in caption (no
transformants); model as
incidence data
Convert percent into number
of dishes and model as
incidence data
Continuous data, no SD for
controls; use peak
Model as continuous data

                                         C-64
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-12. In vitro malignant/morphological cell transformation: data use
Record
number
8000
23720
Reference
Nesnowet al.,
1993a
Pienta et al.,
1977
Page
28
648
Table
number
I
IV
Figure
number


PAHs
DBkmnoAPH
BaP, BaA,
DBahA
Data to be extracted
Peak of Type II and III foci/dish; use
5 ng/mL dose for DBkmnoAPH and
3 ng/mL dose for BaP; average
number foci^dish across the two
experiments
Transformed colonies/surviving
colonies and dose (jig/mL, in row
across)
Basis for
RPF
Point
estimate
Ratio of
slopes
Comment


Notes
Peak transformation for each
compound; DBkmnoAPH
reported in paper as
CP(3,4)B[a]P
Model as incidence data
using multistage
                                         C-65
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-13. In vitro malignant/morphological cell transformation:  dose-response data
Record
number
17610




17970





18080








Reference
Casto, 1979




DiPaolo et al.,
1969





Emura et al.,
1980
Expt 1







PAH
Control
BaP
BaP
DBahA
DBahA
Control
BaP
DBahA
BaA
BeP
DBacA
Control
BaP
BaP
BaP
BaP
BaP
BbF
BbF
BbF
Dose
0
0.62
1.25
1.2
2.5
0
10
10
10
10
10
0
0.01
0.05
0.1
0.25
0.5
0.025
0.1
0.25
Dose
units
Hg/mL
Hg/mL
Hg/mL
|ig/mL
Hg/rnL
|ig/mL
Hg/mL
Hg/rnL
Hg/rnL
|ig/mL
|ig/mL
|ig/mL
|ig/mL
|ig/mL
|ig/mL
|ig/mL
|ig/mL
|ig/mL
|ig/mL
|ig/mL
Transformation measure
Mean
0
8
10
0.5
1
0
8
11
2
1
2
0
0
1.1
2.9
5.3
6.8
0
0.4
0.3
SD




















SE




















Units
Foci
Foci
Foci
Foci
Foci
Transformants
Transformants
Transformants
Transformants
Transformants
Transformants
Transformed
colonies
Transformed
colonies
Transformed
colonies
Transformed
colonies
Transformed
colonies
Transformed
colonies
Transformed
colonies
Transformed
colonies
Transformed
colonies
n
100,000
100,000
100,000
100,000
100,000
354
138
354
190
172
181
1,000
1,000
1,000
1,000
1,000
1,000
1,000
1,000
1,000
units
Surviving cells
Surviving cells
Surviving cells
Surviving cells
Surviving cells
Number of
surviving
Number of
surviving
Number of
surviving
Number of
surviving
Number of
surviving
Number of
surviving
Survivals
Survivals
Survivals
Survivals
Survivals
Survivals
Survivals
Survivals
Survivals
% Response
0
0.00008
0.0001
0.000005
0.00001
0
0.058
0.031
0.011
0.0058
0.011
0
0
0.0011
0.0029
0.0053
0.0068
0
0.00040
0.00030
Notes




















                                         C-66
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-13. In vitro malignant/morphological cell transformation:  dose-response data
Record
number


















Reference







Expt2










PAH
BbF
BbF
BaA
BaA
BaA
BaA
BaA
Control
BaP
BaP
BaP
BaP
BaP
IP
IP
IP
IP
IP
Dose
0.5
1
0.025
0.1
0.25
0.5
1
0
0.01
0.05
0.1
0.25
0.5
0.025
0.1
0.25
0.5
1
Dose
units
Hg/mL
Hg/mL
Hg/mL
Hg/mL
Hg/rnL
|ig/mL
|ig/mL
|ig/mL
|ig/mL
|ig/mL
|ig/mL
|ig/mL
|ig/mL
|ig/mL
|ig/mL
|ig/mL
|ig/mL
|ig/mL
Transformation measure
Mean
0.6
1.2
0
0.3
0.3
0.6
1
0
0.4
1
2.9
4.6
7.8
0
0.3
0.3
0.7
1
SD


















SE


















Units
Transformed
colonies
Transformed
colonies
Transformed
colonies
Transformed
colonies
Transformed
colonies
Transformed
colonies
Transformed
colonies
Transformed
colonies
Transformed
colonies
Transformed
colonies
Transformed
colonies
Transformed
colonies
Transformed
colonies
Transformed
colonies
Transformed
colonies
Transformed
colonies
Transformed
colonies
Transformed
colonies
n
1,000
1,000
1,000
1,000
1,000
1,000
1,000
1,000
1,000
1,000
1,000
1,000
1,000
1,000
1,000
1,000
1,000
1,000
units
Survivals
Survivals
Survivals
Survivals
Survivals
Survivals
Survivals
Survivals
Survivals
Survivals
Survivals
Survivals
Survivals
Survivals
Survivals
Survivals
Survivals
Survivals
% Response
0.00060
0.0012
0
0.00030
0.00030
0.00060
0.0010
0
0.00040
0.0010
0.0029
0.0046
0.0078
0
0.00030
0.00030
0.00070
0.0010
Notes


















                                         C-67
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-13. In vitro malignant/morphological cell transformation:  dose-response data
Record
number
14130





14640


14700









14850



Reference
Greb etal., 1980





Krolewski etal.,
1986


Laaksonen et al.,
1983









Lubet etal.,
1983



PAH
BaP
CH
BaA
BbF
DBahA
BeP
Control
BaP
CPcdP
Control
BaP
BaP
BaP
BaP
Control
BaA
BaA
BaA
BaA
Control
BaP
BaP
BaP
Dose
NA
NA
NA
NA
NA
NA
0
5
5
0
5
10
20
40
0
11
22
44
88
0
1
3
10
Dose
units






MM
MM
MM
MM
MM
MM
MM
MM

MM
MM
MM
MM
Mg/mL
Mg/mL
Mg/mL
Mg/mL
Transformation measure
Mean
277
37
13.9
11.5
0.3
3.1
0
5.5
1.7
0
0.8
0.9
0.3
0.4
0
1.8
1.5
1.1
0.8
0
1
4
5
SD







0.7
0.3














SE























Units
%/mmol
%/mmol
%/mmol
%/mmol
%/mmol
%/mmol
Transformation
frequency
Transformation
frequency
Transformation
frequency
Foci
Foci
Foci
Foci
Foci
Foci
Foci
Foci
Foci
Foci
Dishes with
Type III foci
Dishes with
Type III foci
Dishes with
Type III foci
Dishes with
Type III foci
n






1,000
1,000
1,000
1 x IQ5
1 x IQ5
1 x IQ5
1 x IQ5
1 x IQ5
1 x IQ5
1 x IQ5
1 x IQ5
1 x IQ5
1 x IQ5

15
15
15
units






Viable cells
Viable cells
Viable cells
Surviving cells
Surviving cells
Surviving cells
Surviving cells
Surviving cells
Surviving cells
Surviving cells
Surviving cells
Surviving cells
Surviving cells
Total dishes
Total dishes
Total dishes
Total dishes
% Response






0
0.0055
0.0017
0
0.0000080
0.0000090
0.0000030
0.0000040
0
0.000018
0.000015
0.000011
0.0000080
0
0.067
0.267
0.333
Notes










Inverse dose-response
relationship possible due
to cytotoxicity; use peak




Inverse dose-response
relationship possible due
to cytotoxicity; use peak







                                         C-68
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-13. In vitro malignant/morphological cell transformation:  dose-response data
Record
number



24710














Reference



Mohapatra et al.,
1987














PAH
BeP
BeP
BeP
Control
BaP
BjAC
BjAC
BjAC
BjAC
BjAC
Control
BaP
B1AC
B1AC
B1AC
B1AC
B1AC
Control
Dose
10
30
100
0
1
0.01
0.05
0.5
1
2
0
1
0.5
1
2.5
5
10
0
Dose
units
Hg/mL
Hg/mL
Hg/mL
Hg/rnL
Hg/rnL
|ig/mL
|ig/mL
|ig/mL
|ig/mL
|ig/mL
|ig/mL
|ig/mL
|ig/mL
|ig/mL
|ig/mL
|ig/mL
|ig/mL
|ig/mL
Transformation measure
Mean
0
1
7
0
44
2
5
34
45
48
0
50
8
14
31
42
51
0
SD


















SE


















Units
Dishes with
Type III foci
Dishes with
Type III foci
Dishes with
Type III foci
Dishes with
Type II or III foci
Dishes with
Type II or III foci
Dishes with
Type II or III foci
Dishes with
Type II or III foci
Dishes with
Type II or III foci
Dishes with
Type II or III foci
Dishes with
Type II or III foci
Dishes with
Type II or III foci
Dishes with
Type II or III foci
Dishes with
Type II or III foci
Dishes with
Type II or III foci
Dishes with
Type II or III foci
Dishes with
Type II or III foci
Dishes with
Type II or III foci
Dishes with
Type II or III foci
n
15
15
15
48
48
48
48
48
48
48
60
60
60
60
60
60
60
36
units
Total dishes
Total dishes
Total dishes
Dishes scored
Dishes scored
Dishes scored
Dishes scored
Dishes scored
Dishes scored
Dishes scored
Dishes scored
Dishes scored
Dishes scored
Dishes scored
Dishes scored
Dishes scored
Dishes scored
Dishes scored
% Response
0
0.067
0.467
0
0.92
0.04
0.1
0.71
0.94
1
0
0.83
0.13
0.26
0.52
0.7
0.85
0
Notes


















                                         C-69
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-13. In vitro malignant/morphological cell transformation:  dose-response data
Record
number






24700





Reference






Nesnowetal.,
1990





PAH
BaP
BeAC
BeAC
BeAC
BeAC
BeAC
Acetone
BaP
BaP
BaP
BaP
Acetone
Dose
1
0.5
1
2.5
5
10
0
0.1
0.5
2.5
10
0
Dose
units
Hg/mL
Hg/mL
Hg/mL
Hg/rnL
Hg/rnL
|ig/mL
|ig/mL
Hg/mL
Hg/rnL
|ig/mL
Hg/rnL
Hg/rnL
Transformation measure
Mean
31
4
6
13
15
21
25
43
42
39
72
30
SD







14.7
20.7
19.5
23.1

SE












Units
Dishes with
Type II or III foci
Dishes with
Type II or III foci
Dishes with
Type II or III foci
Dishes with
Type II or III foci
Dishes with
Type II or III foci
Dishes with
Type II or III foci
Anchorage
independent
colonies/
50,000 cells
Anchorage
independent
colonies/
50,000 cells
Anchorage
independent
colonies/
50,000 cells
Anchorage
independent
colonies/
50,000 cells
Anchorage
independent
colonies/
50,000 cells
Anchorage
independent
colonies/
50,000 cells
n
36
36
36
36
36
36






units
Dishes scored
Dishes scored
Dishes scored
Dishes scored
Dishes scored
Dishes scored






% Response
0.86
0.11
0.17
0.36
0.42
0.58






Notes












                                         C-70
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-13. In vitro malignant/morphological cell transformation:  dose-response data
Record
number




7980






7990



Reference




Nesnowetal.,
1997






Nesnowetal.,
1994



PAH
B1AC
B1AC
B1AC
B1AC
Control
BaP
BaP
BaP
DBalP
DBalP
DBalP
Control
BaP
DBahA
DBahA
Dose
0.1
0.5
2.5
10
0
0.4
1.2
4
0.0033
0.1
0.33
0
1
0.25
0.5
Dose
units
Hg/mL
Hg/mL
Mg/mL
Hg/mL
MM
MM
MM
MM
MM
MM
MM
Mg/mL
Mg/mL
Mg/mL
Mg/mL
Transformation measure
Mean
74
68
123
150
0
0.44
1.25
2.54
0.14
1
1.74
0.06
1
0.23
0.25
SD
5.2
14.4
15.6
16.8
0
0.24
0.15
0.56
0.35
0.24
0.78
0.10
0.43
0.21
0.33
SE















Units
Anchorage
independent
colonies/
50,000 cells
Anchorage
independent
colonies/
50,000 cells
Anchorage
independent
colonies/
50,000 cells
Anchorage
independent
colonies/
50,000 cells
Type II and III
foci/dish
Type II and III
foci/dish
Type II and III
foci/dish
Type II and III
foci/dish
Type II and III
foci/dish
Type II and III
foci/dish
Type II and III
foci/dish
Type II and III
foci/dish
Type II and III
foci/dish
Type II and III
foci/dish
Type II and III
foci/dish
n















units















% Response















Notes















                                         C-71
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-13. In vitro malignant/morphological cell transformation:  dose-response data
Record
number


8000








23720






Reference


Nesnowetal.,
1993a








Pienta et al,
1977






PAH
DBahA
DBahA
Control
BaP
BaP
BaP
Control
DBkmno
APH
DBkmno
APH
DBkmno
APH
DBkmno
APH
Control
BaP
BaP
BaP
BaP
BaP
Control
Dose
1
2.5
0
0.3
1
3
0
0.5
1
2.5
5
0
1
5
10
20
40
0
Dose
units
Hg/mL
Hg/mL
Hg/mL
Hg/mL
Hg/rnL
|ig/mL
|ig/mL
Hg/mL
Hg/rnL
Hg/rnL
|ig/mL
|ig/mL
|ig/mL
|ig/mL
|ig/mL
|ig/mL
|ig/mL
|ig/mL
Transformation measure
Mean
0.43
0.29
0
0.48
0.665
1.4
0
0.23
0.52
0.605
1.085
0
1
2
3
5
4
0
SD
0.11
0.085
















SE


















Units
Type II and III
foci/dish
Type II and III
foci/dish
Type II and III
foci/dish
Type II and III
foci/dish
Type II and III
foci/dish
Type II and III
foci/dish
Type II and III
foci/dish
Type II and III
foci/dish
Type II and III
foci/dish
Type II and III
foci/dish
Type II and III
foci/dish
Transformed
colonies
Transformed
colonies
Transformed
colonies
Transformed
colonies
Transformed
colonies
Transformed
colonies
Transformed
colonies
n











504
393
406
434
410
427
229
units











Surviving
colonies
Surviving
colonies
Surviving
colonies
Surviving
colonies
Surviving
colonies
Surviving
colonies
Surviving
colonies
% Response











0
0.0025
0.0049
0.0069
0.0122
0.0094
0
Notes











BaP and BaA data also
reported in Record 18020
Dunkel 1981






                                         C-72
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-13. In vitro malignant/morphological cell transformation:  dose-response data
Record
number











Reference











PAH
BaA
BaA
BaA
BaA
BaA
Control
DBahA
DBahA
DBahA
DBahA
DBahA
Dose
0.1
0.5
1
5
10
0
0.1
0.5
1
5
10
Dose
units
Hg/mL
Hg/mL
Hg/mL
Hg/rnL
Hg/rnL
|ig/mL
|ig/mL
|ig/mL
|ig/mL
|ig/mL
|ig/mL
Transformation measure
Mean
1
2
2
1
7
0
0
4
4
5
0
SD











SE











Units
Transformed
colonies
Transformed
colonies
Transformed
colonies
Transformed
colonies
Transformed
colonies
Transformed
colonies
Transformed
colonies
Transformed
colonies
Transformed
colonies
Transformed
colonies
Transformed
colonies
n
225
252
193
312
250
229
219
233
217
270
232
units
Surviving
colonies
Surviving
colonies
Surviving
colonies
Surviving
colonies
Surviving
colonies
Surviving
colonies
Surviving
colonies
Surviving
colonies
Surviving
colonies
Surviving
colonies
Surviving
colonies
% Response
0.0044
0.0079
0.0104
0.0032
0.028
0
0
0.0172
0.0184
0.0185
0
Notes











                                         C-73
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-14. In vitro DNA adducts: data use
Record
number
16890



6300

9510

22800



10660


10670

7870

21200



Reference
Allen and
Coombs,
1980



Binkova et
al., 2000

Bryla and
Weyand,
1992

Grover and
Sims, 1968


Johnsen et
al., 1998


Johnsen et
al., 1997

Melendez-
Colon et
al., 2000

Segerback
and
Vodicka,
1993



Page
245



62

39

160



80


196

13

2,465



Table
number
1





1

1






II







Figure
number




o
5







2




2

o
3



PAHs
BaP, BaA



BaP,
DBalP

BaP, BaA,
DBacA

BaP,
DBahA,
DBacA,
BaA, Pyr,
PH
BjAC,
B1AC, BaP


BjAC,
B1AC, BaP

BaP,
DBalP

Pyr, BghiP,
FA,
DBahA,
BbF, BaP,
BaA, CH


Data to be
extracted
umol com-
pound/mol
DNA PI



Adducts at
each dose
level

Adducts at
each dose
level

Reaction
with DNA


Total adduct
levels in
human
lymphocytes
andHL-60
cells
DNA adduct
levels in
PCB-treated
rat lung cells
Stable DNA
adducts at
each dose
level
Total adduct
levels



Basis for
RPF
Point
estimate



Ratio of
slopes

Ratio of
slopes

Point
estimate


Point
estimate


Point
estimate

Ratio of
slopes

Point
estimate



Comment
Adducts in
nuclear and
mitochondrial
DNA


Slope of adduct
versus dose
curve

Slope of adduct
versus dose
curve under
light conditions
(maximum
response for all
compounds)




Total adducts
formed in
human
lymphocytes or
HL-60 cells

Adducts in
PCB-treated rat
lung Clara and
Type 2 cells
Slope of adduct
versus dose
curve at two
doses
Total adduct
level in
optimized
nuclease PI
adduct
enrichment
procedure
Notes
Calculate
separate
RPFs for
nuclear and
mitochon-
drial DNA
May need to
drop high-
dose data
for adequate
fit






Calculate
RPFs
separately
by cell type

Calculate
RPFs
separately
by cell type






                           C-74
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-15. In vitro DNA adducts: dose-response data
Record
number
16890



6300













9510







Reference
Allen and
Coombs,
1980



Binkova et
al., 2000













Bryla and
Weyand,
1992







PAH
BaP
BaA
BaP
BaA
BaP






DBalP






BaP
BaP
BaP
BaP
BaA
BaA
BaA
BaA
Dose
0.235
0.644
0.235
0.644
0.010
0.10
0.40
1.0
4.0
10
40
0.010
0.020
0.040
0.080
0.10
0.40
1.0
0.12
12
120
600
0.12
12
120
600
Dose
units
Hg/mL
Hg/mL
Hg/mL
Mg/mL
MM
MM
MM
MM
MM
MM
MM
MM
MM
MM
MM
MM
MM
MM
nmol
nmol
nmol
nmol
nmol
nmol
nmol
nmol
DNA adducts
Mean
7.5
0.44
413
104
1.8
18
95
258
205
69
37
179
534
1,304
1,696
2,317
1,971
632
0.17
1.37
2.21
5.45
0.15
0.09
0.8
0.95
SD
1.9
0.11
164
40.2
1.16
7.18
39.4
115
81.9
21.9
10.8
55.3
52.6
375
644
774
729
170








Adduct units
Mmol/mol DNA
P
Mmol/mol DNA
P
Mmol/mol DNA
P
Mmol/mol DNA
P
Adducts
Adducts
Adducts
Adducts
Adducts
Adducts
Adducts
Adducts
Adducts
Adducts
Adducts
Adducts
Adducts
Adducts
Adducts
Adducts
Adducts
Adducts
Adducts
Adducts
Adducts
Adducts
n




1 x 108
1 x 108
1 x 108
IxlQ8
1 x 108
1 x 108
IxlQ8
1 x IQ8
1 x 108
IxlQ8
1 x IQ8
1 x 108
IxlQ8
1 x IQ8
1 x IQ7
1 x IQ7
1 x IQ7
1 x IQ7
1 x IQ7
1 x IQ7
1 x IQ7
1 x IQ7
Units




Nucleotides
Nucleotides
Nucleotides
Nucleotides
Nucleotides
Nucleotides
Nucleotides
Nucleotides
Nucleotides
Nucleotides
Nucleotides
Nucleotides
Nucleotides
Nucleotides
Nucleotides
Nucleotides
Nucleotides
Nucleotides
Nucleotides
Nucleotides
Nucleotides
Nucleotides
Notes
Nuclear DNA
Nuclear DNA
Mitochondrial DNA
Mitochondrial DNA














Light conditions; max for BaP and others







                                        C-75
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-15. In vitro DNA adducts: dose-response data
Record
number




22800





10670





10660



Reference




Grover and
Sims, 1968





Johnsen et
al, 1997





Johnsen et
al., 1998



PAH
DBacA
DBacA
DBacA
DBacA
BaP
DBahA
DBacA
BaA
Pyr
PH
BaP
BjAC
B1AC
BaP
BjAC
B1AC
BaP
BjAC
B1AC
BaP
Dose
0.12
12
120
600
5
5
5
5
5
5
30
30
30
30
30
30
30
30
30
30
Dose
units
nmol
nmol
nmol
nmol
Hg
Mg
Hg
Hg
Hg
Hg
Hg/mL
|ig/mL
Hg/mL
Hg/mL
|ig/mL
|ig/mL
|ig/mL
|ig/mL
|ig/mL
|ig/mL
DNA adducts
Mean
0
0.06
0.57
1.76
1.41
0.44
0.56
0.7
0.31
0.05
0.05
0.15
0.24
0.02
0.06
0.03
0.333
0.110
1.089
0.239
SD
















0.093
0.026
0.595
0.172
Adduct units
Adducts
Adducts
Adducts
Adducts
jimol/g-atom of
DNAP
|imol/g-atom of
DNAP
|imol/g-atom of
DNAP
limol/g-atom of
DNAP
limol/g-atom of
DNAP
limol/g-atom of
DNAP
fmol adducts/ng
DNA
fmol adducts/ng
DNA
fmol adducts/ng
DNA
fmol adducts/ng
DNA
fmol adducts/ng
DNA
fmol adducts/ng
DNA
fmol adducts/ng
DNA
fmol adducts/ng
DNA
fmol adducts/ng
DNA
fmol adducts/ng
DNA
n
1 x 107
1 x 107
1 x IQ7
1 x IQ7












3
3
3
3
Units
Nucleotides
Nucleotides
Nucleotides
Nucleotides
















Notes










Clara cells
Clara cells
Clara cells
Type 2 cells
Type 2 cells
Type 2 cells
Human lymphocytes
Human lymphocytes
Human lymphocytes
HL-60 cells
                                        C-76
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-15. In vitro DNA adducts: dose-response data
Record
number


7870



21200







Reference


Melendez-
Colonetal.,
2000



Segerback
and
Vodicka,
1993







PAH
BjAC
B1AC
BaP
BaP
DBalP
DBalP
BaP
Pyr
BghiP
FA
DBahA
BbF
BaA
CH
Dose
30
30
1
2
1
2
100
100
100
100
100
100
100
100
Dose
units
Hg/mL
Hg/mL
\im
\im
|im
|im
mM
mM
mM
mM
mM
mM
mM
mM
DNA adducts
Mean
0.149
0.942
18
34
254
348
15
0.14
0.50
1.5
2.8
3.7
30
50
SD
0.146
0.344
8.07
6.46
4.30
17.20








Adduct units
fmol adducts/jig
DNA
fmol adducts/jig
DNA
Stable adducts
Stable adducts
Stable adducts
Stable adducts
(imol adducts per
mol dNp
(imol adducts per
mol dNp
(imol adducts per
mol dNp
(imol adducts per
mol dNp
(imol adducts per
mol dNp
(imol adducts per
mol dNp
(imol adducts per
mol dNp
|imol adducts per
mol dNp
n
3
3
1 x 106
1 x 106
1 x 106
1 x 106








Units


Nucleotides
Nucleotides
Nucleotides
Nucleotides








Notes
HL-60 cells
HL-60 cells












                                        C-77
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-16. In vitro DNA damage:  data use
Record
number
16840
23790
10660
19740
19830
20810
20940
21730
Reference
Agrelo and
Amos, 1981
Ichinotsubo
etal., 1977
Johnsen et
al, 1998
Martinet al,
1978
Mersch-
Sundennann
etal., 1992
Robinson
and Mitchell,
1981
Rossman et
al., 1991
long etal.,
1981b
Page
531
56
82
2,624
3-6
520
354
480
Table
number
2
Table II

1
2
1
2
I
Figure
number


4





PAHs
BaP, Pyr
BaP, DBaiP, DBahA
BaP, BjAC, B1AC
BaP, BeP, BaA,
DBacA, DBahA
BaP, AA, BaA, BbF,
BghiF, BjF, BbFE,
BghiP, BeP, CH,
DBacA, DBahA,
DBalP, DBahP,
DBaiP, FA, IP, PH,
Tphen
BaP, Pyr
BaP, AC, DBacA,
DBahA, PH
BaP, BaA
Data to be extracted
Hydroxyurea inhibited [3H]-thymidine
incorporation into cells (dpm) and dose
(Hg/mL); use 10 ng/mL dose for BaP and
100 ng/mL dose for pyrene
Use column designated JC5519 +S9 for
BaP, DBaiP, and DBahA; dose as ng/well
and response as diameter of zone of
inhibition (mm); the control is wild type
strain AB 11 57
DNA damage (NAAC, 10'V), SDand
dose (ng/mL) for both human
lymphocytes and HL-60 cells; use 24 h +
1 h AraC/HU data (crosshatched bars)
Maximum dpm/jig DNA above
background and dose (M); dose is in
column marked "M"
SOS induction potential for assay (+S9)
for each compound (already incorporates
dose)
Maximum [3H]-TDR incorporation and
dose (test concentration in ng/rnL in
parentheses after maximum) for rows with
metabolic activation (+); use compound-
specific background [ HJ-TDR
incorporation in same row
Max enhancement of prophage induction
over background and dose (amount at
max, in jig/well) for those rows with
S9 (+ rows).
DNA repair grains/nucleus, SD, and dose
(M); four doses BaA, three doses BaP and
DMSO control
Basis for RPF
Point estimate
Point estimate
Ratio of slopes
(human lympho-
cytes); point esti-
mates (HL-60 cells)
Point estimate
Ratio of SOS
induction potentials
Point estimate
Point estimate
Ratio of slopes
Comment

E. coftRecBC, S9
identification
unknown

Background already
subtracted
SOSIP reported in text
as slope of steepest
portion of the
induction factor dose-
response curve

Background already
addressed

Notes


Model as
continuous
data

No modeling
necessary;
slopes
reported in
text


Model as
continuous
data
                                      C-78
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-17. In vitro DNA damage: dose-response data
Record
number
16840














23790




Reference
Agrelo and
Amos, 1981














Ichinotsubo et
al., 1977




PAH
Control
BaP
BaP
BaP
BaP
BaP
BaP
BaP
Control
Pyr
Pyr
Pyr
Pyr
Pyr
Pyr
Control
BaP
Control
DBaiP
Control
Dose
0
0.001
0.01
0.1
1
10
100
1,000
0
0.032
0.16
0.8
4
20
100
0
70
0
600
0
Dose
units
ug/mL
ug/mL
ug/mL
ug/mL
ug/mL
ug/mL
ug/mL
ug/mL
ug/mL
ug/mL
ug/mL
ug/mL
ug/mL
ug/mL
ug/mL

ug/well

ug/well

Endpoint
Unscheduled DNA synthesis
Unscheduled DNA synthesis
Unscheduled DNA synthesis
Unscheduled DNA synthesis
Unscheduled DNA synthesis
Unscheduled DNA synthesis
Unscheduled DNA synthesis
Unscheduled DNA synthesis
Unscheduled DNA synthesis
Unscheduled DNA synthesis
Unscheduled DNA synthesis
Unscheduled DNA synthesis
Unscheduled DNA synthesis
Unscheduled DNA synthesis
Unscheduled DNA synthesis
DNA damage
DNA damage
DNA damage
DNA damage
DNA damage
DNA damage
Mean
177
195
126
262
818
2,270
819
373
1,168
1,293
1,192
1,367
1,510
1,694
1,716
0
6
0
10
0
SD




















Units
dpm
dpm
dpm
dpm
dpm
dpm
dpm
dpm
dpm
dpm
dpm
dpm
dpm
dpm
dpm
Diameter of
zone of
inhibition mm
Diameter of
zone of
inhibition mm
Diameter of
zone of
inhibition mm
Diameter of
zone of
inhibition mm
Diameter of
zone of
inhibition mm
n




















Notes
HU inhibited
HU inhibited
HU inhibited
HU inhibited
HU inhibited
HU inhibited
HU inhibited
HU inhibited
HU inhibited
HU inhibited
HU inhibited
HU inhibited
HU inhibited
HU inhibited
HU inhibited





                                       C-79
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-17. In vitro DNA damage: dose-response data
Record
number

10660










19740




Reference

Johnsen et al,
1998










Martin etal.,
1978




PAH
DBahA
DMSO
BaP

BjAC

BIAC

DMSO
BaP
BjAC
BIAC
BaP
BeP
BaA
DBacA
DBahA
Dose
25
0
3
30
3
30
3
30
0
30
30
30
1 x lO'5
1 x lO'6
1 x lO'7
1 x lO'5
1 x lO'5
Dose
units
ug/well
ug/mL
ug/mL
ug/mL
ug/mL
ug/mL
ug/mL
ug/mL
ug/mL
ug/mL
ug/mL
ug/mL
M
M
M
M
M
Endpoint
DNA damage
DNA damage
DNA damage
DNA damage
DNA damage
DNA damage
DNA damage
DNA damage
DNA damage
DNA damage
DNA damage
DNA damage
Unscheduled DNA synthesis
Unscheduled DNA synthesis
Unscheduled DNA synthesis
Unscheduled DNA synthesis
Unscheduled DNA synthesis
DNA damage
Mean
10
4.4
12
15
6.0
9.4
8.2
9.3
7.8
13.2
9.6
11.6
210
256
59
97
96
SD

1.3
3.2
2.7
2.1
3.4
3.2
2.1
3.1
9.5
3.0
5.5





Units
Diameter of
zone of
inhibition mm
NAAC, 10'3 h'1
NAAC, 10'3 h'6
NAAC, 10'3 h'7
NAAC, 10'3 h'2
NAAC, 10'3 h'3
NAAC, 10'3 h'4
NAAC, 10"3 h"5
NAAC, 10"3 h"5
NAAC, 10"3 h"5
NAAC, 10'3 h'5
NAAC, 10'3 h'5
Maximum
dpm/ug DNA
Maximum
dpm/ug DNA
Maximum
dpm/ug DNA
Maximum
dpm/ug DNA
Maximum
dpm/ug DNA
n

3
3
3
3
3
3
3
3
3
3
3





Notes

Human lymphocytes with
AraC/HU
Human lymphocytes with
AraC/HU; no continuous linear
model fit
Human lymphocytes with
AraC/HU
Human lymphocytes with
AraC/HU
Human lymphocytes with
AraC/HU
Human lymphocytes with
AraC/HU; no continuous linear
model fit
Human lymphocytes with
AraC/HU
HL-60 cells with AraC/HU
HL-60 cells with AraC/HU
HL-60 cells with AraC/HU
HL-60 cells with AraC/HU
Increase above background
Increase above background
Increase above background
Increase above background
Increase above background
                                       C-80
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-17. In vitro DNA damage: dose-response data
Record
number
19830















Reference
Mersch-
Sundermann et
al., 1992















PAH
BaP
AA
BaA
BbF
BghiF
BjF
BbFE
BghiP
BeP
CH
DBacA
DBahA
DBalP
DBahP
DBaiP
FA
Dose
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
Dose
units
















Endpoint
SOS induction potential
SOS induction potential
SOS induction potential
SOS induction potential
SOS induction potential
SOS induction potential
SOS induction potential
SOS induction potential
SOS induction potential
SOS induction potential
SOS induction potential
SOS induction potential
SOS induction potential
SOS induction potential
SOS induction potential
SOS induction potential
DNA damage
Mean
0.605
0.142
0.1
0.045
0.34
0.254
0.024
0.033
0.032
0.221
0.104
0.039
2.1
0.117
0.174
0.412
SD
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
Units
















n
















Notes
Steepest slope of induction factor
dose-response curve; + S9
Steepest slope of induction factor
dose-response curve; + S9
Steepest slope of induction factor
dose-response curve; + S9
Steepest slope of induction factor
dose-response curve; + S9
Steepest slope of induction factor
dose-response curve; + S9
Steepest slope of induction factor
dose-response curve; + S9
Steepest slope of induction factor
dose-response curve; + S9
Steepest slope of induction factor
dose-response curve; + S9
Steepest slope of induction factor
dose-response curve; + S9
Steepest slope of induction factor
dose-response curve; + S9
Steepest slope of induction factor
dose-response curve; + S9
Steepest slope of induction factor
dose-response curve; + S9
Steepest slope of induction factor
dose-response curve; + S9
Steepest slope of induction factor
dose-response curve; + S9
Steepest slope of induction factor
dose-response curve; + S9
Steepest slope of induction factor
dose-response curve; + S9
                                       C-81
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-17. In vitro DNA damage: dose-response data
Record
number



20810



20940




Reference



Robinson and
Mitchell, 1981



Rossmanetal.,
1991




PAH
IP
PH
Tphen
Control
BaP
Control
Pyr
BaP
AC
DBacA
DBahA
PH
Dose
NA
NA
NA
0
10
0
7.2
12.5
12.5
1.44
2
25
Dose
units



ug/mL
ug/mL
ug/mL
ug/mL
ug/mL
ug/mL
ug/mL
ug/mL
ug/mL
Endpoint
SOS induction potential
SOS induction potential
SOS induction potential
Unscheduled DNA synthesis
Unscheduled DNA synthesis
Unscheduled DNA synthesis
Unscheduled DNA synthesis
DNA damage
DNA damage
DNA damage
DNA damage
DNA damage
DNA damage
Mean
0.036
0.053
0.26
53
142
52
115
10.4
4.8
8
4
4.5
SD
NA
NA
NA
4
7
2
9





Units



[3H]-TdR
incorporation
[3H]-TdR
incorporation
[3H]-TdR
incorporation
[3H]-TdR
incorporation
Lambda pro-
phage induction
Lambda pro-
phage induction
Lambda pro-
phage induction
Lambda pro-
phage induction
Lambda pro-
phage induction
n












Notes
Steepest slope of induction factor
dose-response curve; + S9
Steepest slope of induction factor
dose-response curve; + S9
Steepest slope of induction factor
dose-response curve; + S9
Maximum [3H]-TdR incorporation
Maximum [3H]-TdR incorporation
Maximum [3H]-TdR incorporation
Maximum [3H]-TdR incorporation
Maximum enhancement over
background
Maximum enhancement over
background
Maximum enhancement over
background
Maximum enhancement over
background
Maximum enhancement over
background
                                       C-82
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-17. In vitro DNA damage: dose-response data
Record
number
21730







Reference
Tong et al.,
1981b







PAH
Control
BaP
BaP
BaP
BaA
BaA
BaA
BaA
Dose
0
1 x lO'4
5 x 1Q-4
1 x ID'3
5 x ID'5
1 x ID'4
5 x 1Q-4
1 x ID'3
Dose
units
M
M
M
M
M
M
M
M
Endpoint
Unscheduled DNA synthesis
Unscheduled DNA synthesis
Unscheduled DNA synthesis
Unscheduled DNA synthesis
Unscheduled DNA synthesis
Unscheduled DNA synthesis
Unscheduled DNA synthesis
Unscheduled DNA synthesis
DNA damage
Mean
0.1
45.1
47.7
65.6
0.6
14.8
17.2
Toxic
SD
0.1
3.7
3.7
17.8

2.6
6

Units
Grains/nucleus
Grains/nucleus
Grains/nucleus
Grains/nucleus
Grains/nucleus
Grains/nucleus
Grains/nucleus
Grains/nucleus
n








Notes








                                       C-83
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-18. In vitro clastogenicity: data use
Record
number
14620
14640
19690
21710
Reference
Kochhar, 1982
Krolewski et al.,
1986
Maneetal., 1990
Tongetal.,
1981a
Page
846
1,648
81
469
Table
number
Not
numbered
II
III
1
PAHs
BaP,
BaA
BaP,
CPcdP
BaP,
BaA
BaP,
BaA
Data to be used
Percentage of cells
with aberrations and
dose (ug/mL)
Mean number sister
chromatid exchange/
chromosome, SD, and
dose (uM)
Sister chromatid
exchange frequencies/
for V79 cell + rat
MEC and dose
Sister chromatid
exchange/cell, SD, and
dose
Basis for
RPF
Ratio of
slopes
Ratio of
slopes
Point
estimates
Point
estimates
Comment
Model as incidence data
Use first column of
data; not data with AIA
or IVA; model as
continuous data
Use sister chromatid
exchange data for V79 +
rat MEC only
Continuous data, no n
provided in study
                             C-84
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-19. In vitro clastogenicity:  dose-response data
Record
number
14620








14640




19690


21710







Reference
Kochhar,
1982








Krolewski et
al., 1986




Mane etal.,
1990


Tong etal.,
1981a







PAH
Control
BaP
BaP
BaP
BaP
BaA
BaA
BaA
BaA
Control
BaP
BaP
CPcdP
CPcdP
Control
BaP
BaA
Control
BaP
BaP
BaP
Control
BaA
BaA
BaA
Dose
0
0.6
1.25
2.5
5
0.6
1.25
2.5
5
0
1
5
1
5
0
1
1
0
1 x 10'6
1 x 10'5
1 x 10'4
0
1 x 10'5
1 x 10'4
1 x 10'3
Dose
units
Hg/mL
Hg/mL
Mg/mL
Hg/mL
Mg/mL
Mg/mL
Mg/mL
Mg/mL
Mg/mL
MM
MM
MM
MM
MM
Mg/mL
Mg/mL
Mg/mL
M
M
M
M
M
M
M
M
n
100
100
100
100
100
100
100
100
100
30
30
30
30
30











Clastogenicity
Mean
0.06
0.23
0.32
0.45
0.56
0.17
0.23
0.3
0.38
0.147
0.874
0.932
0.348
0.432
0.3
3
0.7
11.15
16.15
59.75
103.3
15.75
21.2
29.15
26.2
SD









0.059
0.275
0.266
0.119
0.15
1
1
0.5
3.81
3.83
16.96
22.75
5.18
9.59
9.93
6.96
Units
Fraction cells with
aberrations
Fraction cells with
aberrations
Fraction cells with
aberrations
Fraction cells with
aberrations
Fraction cells with
aberrations
Fraction cells with
aberrations
Fraction cells with
aberrations
Fraction cells with
aberrations
Fraction cells with
aberrations
Sister chromatid
exchange
Sister chromatid
exchange
Sister chromatid
exchange
Sister chromatid
exchange
Sister chromatid
exchange
Sister chromatid
exchange frequency
Sister chromatid
exchange frequency
Sister chromatid
exchange frequency
Sister chromatid
exchange/cell
Sister chromatid
exchange/cell
Sister chromatid
exchange/cell
Sister chromatid
exchange/cell
Sister chromatid
exchange/cell
Sister chromatid
exchange/cell
Sister chromatid
exchange/cell
Sister chromatid
exchange/cell
Notes














For V79 cell + rat
MEC
For V79 cell + rat
MEC
For V79 cell + rat
MEC








                             C-85
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-20. In vivo DNA adducts: data use
Record
number
6210
17630
18810
11190
8010
24590/
20920
22810
24790
24801
Reference
Arif etal.,
1997
Cavalieri et
al., 1981a
Hughes
and
Phillips,
1990
Mass et al.,
1993
Nesnow et
al., 1993b
Nesnow et
al., 1998b;
Ross et al.,
1995
Phillips et
al., 1979
Kligerman
etal., 2002
Weyand et
al., 2004
Page
36
491
1,614
188
39
402
205
846
12,
14
Table
number

o
J

1

2
I
1

Figure
number
4

o
6

land 2



4 and 6
PAHs
DBalP and BaP
CPcdP, ACEP
(reported in
paper as CPAP),
BaP
DBalP, DBaeP,
DBahP, DBaiP,
BaP
BjAC, BaP
BbF, BaP
BaP, BbF,
DBahA, CPcdP,
DBalP
DBahA, DBacA,
BaP
BaA, BaP, BbF,
CH
BcFE, BaP
Data to be
extracted
Mean adduct levels
for heart, pancreas,
bladder, liver
Done
AUC for skin and
lung through 84 d
Done
AUC for lung, liver,
and PEL through
56 d
Done
Done
Done
Mean adduct levels
for lung and
forestomach
Basis for
RPF
Point
estimate
Point
estimate
Point
estimate
Ratio of
Slopes
Point
estimate
Ratio of
Slopes
Point
estimate
Point
estimate
Point
estimate
Comment
Mean adduct levels summed across
mammary epithelial, lung, heart,
pancreas, bladder, liver
DNA-bound PAH in mouse skin
after 4-hr or 24-hr treatment
Sum of AUCs for skin and lung
0-84 d
AUC (adduct-time curve) versus
dose for lung adducts 24-72 hr
Sum of AUCs for lung, liver, and
lymphocytes 0-56 d
Slope of TID AL/dose (slope reported
in Record 24590 based on data from
Record 20920); DBalP data reported
in separate study without BaP
concurrent
Peak binding in mouse skin; BaA
dropped; not clear if reported level is
peak
Adducts in mouse or rat PBLs at
single time point after either
intraperitoneal or gavage
administration
Adducts in mouse lung and
forestomach at single time point after
either intraperitoneal or dietary
administration
Notes

Calculate separate
RPFs for 4-hr and
24-hr treatment





Calculate separate
RPFs for
intraperitoneal and
gavage, rat and
mouse
Calculate separate
RPFs for lung and
forestomach after
oral exposure and
for lung after
intraperitoneal
exposure
                                       C-86
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-21. In vivo DNA adducts: dose-response data
Record
number
6210














17630



Reference
Arif etal,
1997














Cavalieri et
al., 1981a



PAH
Control
BaP
BaP
BaP
BaP
BaP
BaP

DBalP
DBalP
DBalP
DBalP
DBalP
DBalP

BaP
CPcdP
ACEP
BaP
Species
Rat
Rat
Rat
Rat
Rat
Rat
Rat

Rat
Rat
Rat
Rat
Rat
Rat





Dose
0
0.25
0.25
0.25
0.25
0.25
0.25

0.25
0.25
0.25
0.25
0.25
0.25

0.2
0.2
0.2
0.2
Dose units
(imol^mammary
gland
|imo I/mammary
gland
|imo I/mammary
gland
(imol/mammary
gland
(imol/mammary
gland
|imo I/mammary
gland
|imo I/mammary
gland

|imo I/mammary
gland
|imo I/mammary
gland
(imol/mammary
gland
(imol/mammary
gland
(imol/mammary
gland
|imo I/mammary
gland

(imol/mouse
(imol/mouse
(imol/mouse
(imol/mouse
Organ
Liver
Mammary
gland
Lung
Heart
Pancreas
Bladder
Liver
Sum
Mammary
gland
Lung
Heart
Pancreas
Bladder
Liver
Sum
Skin
Skin
Skin
Skin
Time















4hr
4hr
4hr
24 hr
DNA adducts
Mean
0
300
11
9.5
0
0
4.5
324.74
1,878
85
64
32
69
116
2,244.63
16.3
2.3
2.2
6.7
SD

45
1.3





378
24









SE















1
0.2
0.1
1.6
Adduct units
Adducts/109
nucleotides
Adducts/109
nucleotides
Adducts/109
nucleotides
Adducts/109
nucleotides
Adducts/109
nucleotides
Adducts/109
nucleotides
Adducts/109
nucleotides

Adducts/109
nucleotides
Adducts/109
nucleotides
Adducts/109
nucleotides
Adducts/109
nucleotides
Adducts/109
nucleotides
Adducts/109
nucleotides

(imol adduct/mol
DNA
(imol adduct/mol
DNA
(imol adduct/mol
DNA
(imol adduct/mol
DNA
Slope of
AUC
versus dose



















Comments



















                                       C-87
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-21. In vivo DNA adducts: dose-response data
Record
number


18810














Reference


Hughes
and
Phillips,
1990














PAH
CPcdP
ACEP
BaP
BaP
BaP
DBaeP
DBaeP
DBaeP
DBahP
DBahP
DBahP
DBaiP
DBaiP
DBaiP
DBaiP
DBaiP
DBaiP
Species

















Dose
0.2
0.2
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
Dose units
(imol/mouse
(imol/mouse
(imol
(imol
(imol
(imol
(imol
(imol
(imol
(imol
(imol
(imol
(imol
(imol
(imol
(imol
(imol
Organ
Skin
Skin
Skin
Lung
Sum skin
and lung
Skin
Lung
Sum skin
and lung
Skin
Lung
Sum skin
and lung
Skin
Lung
Sum skin
and lung
Skin
Lung
Sum skin
and lung
Time
24 hr
24 hr
Id
2d

2d
7d

2d
2d

2d
2d

Id
2d

DNA adducts
Mean
8.8
0.30
7.8
1.2
9.0
0.50
Cannot
determine
Cannot
determine
3.1
0.14
3.2
0.75
0.10
0.85
62
2.3
65
SD

















SE
1
0.1















Adduct units
(imol adduct/mol
DNA
(imol adduct/mol
DNA
fmol adducts/ jig
DNA
fmol adducts/ jig
DNA
fmol adducts/ jig
DNA
fmol adducts/ jig
DNA
fmol adducts/ jig
DNA
fmol adducts/ jig
DNA
fmol adducts/ jig
DNA
fmol adducts/ jig
DNA
fmol adducts/ jig
DNA
fmol adducts/ jig
DNA
fmol adducts/ jig
DNA
fmol adducts/ jig
DNA
fmol adducts/ jig
DNA
fmol adducts/ jig
DNA
fmol adducts/ jig
DNA
Slope of
AUC
versus dose

















Comments


Only peak extracted;
interrupted scale
precluded digitizing














                                       C-88
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-21. In vivo DNA adducts: dose-response data
Record
number
11190


















Reference
Mass et al.,
1993


















PAH
BaP
BaP
BaP
BaP
BaP
BaP
BaP
BaP
BaP
BaP
BaP
BaP
BjAC
BjAC
BjAC
BjAC
BjAC
BjAC
BjAC
Species



















Dose
20
20
20
50
50
50
100
100
100
20
50
100
20
20
20
50
50
50
100
Dose units
mg/kg bw
mg/kg bw
mg/kg bw
mg/kg bw
mg/kg bw
mg/kg bw
mg/kg bw
mg/kg bw
mg/kg bw
mg/kg bw
mg/kg bw
mg/kg bw
mg/kg bw
mg/kg bw
mg/kg bw
mg/kg bw
mg/kg bw
mg/kg bw
mg/kg bw
Organ
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Time
24 hr
48hr
72hr
24hr
48 hr
72 hr
24 hr
48 hr
72 hr
AUC
AUC
AUC
24hr
48 hr
72 hr
24 hr
48 hr
72 hr
24 hr
DNA adducts
Mean
116
122
181
120
201
432
427
407
2,004
7,884
12,888
44,064
63
97
255
116
402
1,954
180
SD
53
25
101
20
170
274
140
197
314



34
101
392
121
237
1,921
133
SE



















Adduct units
amol adducts/ jig
DNA
amol adducts/ jig
DNA
amol adducts/ jig
DNA
amol adducts/ jig
DNA
amol adducts/ jig
DNA
amol adducts/ jig
DNA
amol adducts/ jig
DNA
amol adducts/ jig
DNA
amol adducts/ jig
DNA



amol adducts/ jig
DNA
amol adducts/ jig
DNA
amol adducts/ jig
DNA
amol adducts/ jig
DNA
amol adducts/ jig
DNA
amol adducts/ jig
DNA
amol adducts/ jig
DNA
Slope of
AUC
versus dose









469.73









Comments
AUC calculated using
trapezoid rule











AUC calculated using
trapezoid rule






                                       C-89
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-21. In vivo DNA adducts: dose-response data
Record
number





8010






















Reference





Nesnow et
al., 1993b






















PAH
BjAC
BjAC
BjAC
BjAC
BjAC
BaP
BaP
BaP
BaP
BaP
BaP
BaP
BaP
BaP
BaP
BaP
BaP
BaP
BaP
BaP
BaP
BaP
BaP
BaP
BaP
BaP
BaP
BbF
Species




























Dose
100
100
20
50
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
Dose units
mg/kg bw
mg/kg bw
mg/kg bw
mg/kg bw
mg/kg bw
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
Organ
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Liver
Liver
Liver
Liver
Liver
Liver
Liver
PEL
PEL
PEL
PEL
PEL
PEL
PEL
Sum of
AUCs
Lung
Time
48 hr
72 hr
AUC
AUC
AUC
dl
d3
d7
d!4
d28
d56
AUC
dl
d3
d7
d!4
d28
d56
AUC
dl
d3
d7
d!4
d28
d56
AUC

dl
DNA adducts
Mean
532
2,439
6,900
35,880
46,356
453
1,001
574
386
381
143
20,892
398
1,317
931
537
394
116
25,207
158
273
162
187
72
41
5,985
52,084
21
SD
559
2,242


























SE




























Adduct units
amol adducts/jig
DNA
amol adducts/jig
DNA


























Slope of
AUC
versus dose


464.25

























Comments





AUC calculated using
trapezoid rule





















AUC calculated using
trapezoid rule
                                       C-90
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-21. In vivo DNA adducts: dose-response data
Record
number
























24590/
20920

Reference
























Nesnow et
al., 1998b;
Ross, 1995

PAH
BbF
BbF
BbF
BbF
BbF
BbF
BbF
BbF
BbF
BbF
BbF
BbF
BbF
BbF
BbF
BbF
BbF
BbF
BbF
BbF
BbF
BbF
BbF
BbF
BaP
BbF
Species


























Dose
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
NA
NA
Dose units
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg


Organ
Lung
Lung
Lung
Lung
Lung
Lung
Lung
Liver
Liver
Liver
Liver
Liver
Liver
Liver
Liver
PEL
PEL
PEL
PEL
PEL
PEL
PEL
PEL
Sum of
AUCs
Lung
Lung
Time
d3
d5
d7
d!4
d28
d56
AUC
dl
d3
d5
d7
d!4
d28
d56
AUC
dl
d3
d5
d7
d!4
d28
d56
AUC

>21 d
>21d
DNA adducts
Mean
184
233
211
229
145
106
8,763
12
35
51
61
21
15
12
1,173
12
29
59
57
40
15
13
1,378
11,314


SD


























SE
























3.9
5
Adduct units


























Slope of
AUC
versus dose
























113
37.5
Comments
























Slope of dose versus
TIDAL value (in fmol-
d/ng DNA)
Slope of dose versus
TIDAL value (in fmol-
d/jjg DNA)
                                       C-91
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-21. In vivo DNA adducts: dose-response data
Record
number



22810


24790









Reference



Phillips et
al., 1979


Kligerman
et al., 2002









PAH
CPcdP
DBahA
DBalP
BaP
DBacA
DBahA
BaP
BaA
BbF
CH
Control
BaP
BaA
BbF
CH
Control
Species






Mice
Mice
Mice
Mice
Mice
Mice
Mice
Mice
Mice
Mice
Dose
NA
NA
NA
1
1
1
100
100
100
100
0
100
100
100
100
0
Dose units



(imol/mouse
(imol/mouse
(imol/mouse
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
Organ
Lung
Lung
Lung
Skin
Skin
Skin
PEL
PEL
PEL
PEL
PEL
PEL
PEL
PEL
PEL
PEL
Time
>21 d
>21d
>21d
19hr
24 hr
72 hr
d7
d7
d7
d7
d7
d7
d7
d7
d7
d7
DNA adducts
Mean



27
10
15
4,186
93
516
81
0
143
32
39
37
0
SD






273
8
7
11

17
2
4
1

SE
3.69
19.1
267













Adduct units



pmol adducts/mg
DNA
pmol adducts/mg
DNA
pmol adducts/mg
DNA
amol adducts/ jig
DNA
amol adducts/ jig
DNA
amol adducts/ jig
DNA
amol adducts/ jig
DNA
amol adducts/ jig
DNA
amol adducts/ jig
DNA
amol adducts/ jig
DNA
amol adducts/ jig
DNA
amol adducts/ jig
DNA
amol adducts/ jig
DNA
Slope of
AUC
versus dose
148
219
1,390













Comments
Slope of dose versus
TIDAL value (in finol-
d/ng DNA)
Slope of dose versus
TIDAL value (in finol-
d/ng DNA)
Slope of dose versus
TIDAL value (in finol-
d/ng DNA)
peak
peak
peak
Intraperitoneal
Intraperitoneal
Intraperitoneal
Intraperitoneal
Intraperitoneal
Gavage
Gavage
Gavage
Gavage
Gavage
                                       C-92
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-21. In vivo DNA adducts: dose-response data
Record
number










24801






Reference










Weyand et
al., 2004






PAH
BaP
BaA
BbF
CH
Control
BaP
BaA
BbF
CH
Control
BaP
BcFE
BcFE
BaP
BcFE
BaP
BcFE
Species
Rat
Rat
Rat
Rat
Rat
Rat
Rat
Rat
Rat
Rat
Mice
Mice
Mice
Mice
Mice
Mice
Mice
Dose
100
100
100
100
0
100
100
100
100
0
230
13.6
197
230
197
230
13.6
Dose units
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg food
mg/kg food
mg/kg food
mg/kg food
mg/kg food
mg/kg food
mg/kg food
Organ
PEL
PEL
PEL
PEL
PEL
PEL
PEL
PEL
PEL
PEL
Lung
Lung
Lung
Forestomach
Forestomach
Sum of lung
and
forestomach
Sum of lung
and
forestomach
Time
d7
d7
d7
d7
d7
d7
d7
d7
d7
d7
d!4
d!4
d!4
d!4
d!4
d!4
d!4
DNA adducts
Mean
755
38
63
24
0
177
20
17
10
0
0.084
0.014
0.18
0.033
0.0092
0.117
0.014
SD
56
3
1
2

30
2
1
4








SE










0.009
0.002
0.023
0.005
0.001


Adduct units
amol adducts/ jig
DNA
amol adducts/ jig
DNA
amol adducts/ jig
DNA
amol adducts/ jig
DNA
amol adducts/ jig
DNA
amol adducts/ jig
DNA
amol adducts/ jig
DNA
amol adducts/ jig
DNA
amol adducts/ jig
DNA
amol adducts/ jig
DNA
pmol adducts/mg
DNA
pmol adducts/mg
DNA
pmol adducts/mg
DNA
pmol adducts/mg
DNA
pmol adducts/mg
DNA
pmol adducts/mg
DNA
pmol adducts/mg
DNA
Slope of
AUC
versus dose

















Comments
Intraperitoneal
Intraperitoneal
Intraperitoneal
Intraperitoneal
Intraperitoneal
Gavage
Gavage
Gavage
Gavage
Gavage
Diet
Diet
Diet
Diet
Diet
Diet
Diet
                                       C-93
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-21. In vivo DNA adducts: dose-response data
Record
number



Reference



PAH
BcFE
BaP
BcFE
Species
Mice
Mice
Mice
Dose
197
100
100
Dose units
mg/kg food
mg/kg bw
mg/kg bw
Organ
Sum of lung
and
forestomach
Lung
Lung
Time
d!4
24 h
24 h
DNA adducts
Mean
0.19
0.78
0.33
SD



SE

0.13
0.030
Adduct units
pmol adducts/mg
DNA
pmol adducts/mg
DNA
pmol adducts/mg
DNA
Slope of
AUC
versus dose



Comments
Diet
Intraperitoneal
Intraperitoneal
                                       C-94
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-22. In vivo clastogenicity:  data use
Record
number
24740

14270

17190

20950





24720


24790

Reference
Allen et
al., 1999

He and
Baker,
1991
Bayer,
1978

Roszinsky-
Kocher et
al., 1979





Kligerman
etal., 1986


Kligerman
etal., 2002

Page


166

426

66





129


846

Table
number
I and III

1

o
3

i





3


1

Figure
number

















PAHs
BaP,
DBalP

BaP,
CH
BaP,
PH

BaP,
DBah
A,
CH,
PH,
BeP,
BbF,
BaA
BaP,
B1AC


BaP,
BaA,
BbF,
CH
Data to be extracted
Total micronuleated poly-
chromatic erythrocytes (MN-
PCEs) and dose (mg/kg);
extract data for bone marrow
and peripheral blood for both
A/J mice (Table 1) and p53+/+
(wild type) mice (Table III)
MN cells/ 1,000 binucleated
and dose (ug/mouse)
Sister chromatid exchange/cells
and dose (mg/kg)

Sister chromatid exchanges/
metaphase and dose (mg/kg)





Sister chromatid exchanges/
metaphase and dose (mg/kg)


Sister chromatid exchanges/
metaphase, intraperitoneal, for
BaP, BaA, BbF, and CH; sister
chromatid exchanges, gavage,
for BaP and BaA (use 17.91
value for BaP); also use MN
bn/1,000 bn, gavage, for BaP
and BbF; dose in mg/kg
Basis for
RPF
Point
estimate

Ratio of
slopes
Point
estimate

Point
estimate





Point
estimate


Point
estimates

Comment
Incidence data;
single dose
BaP

Incidence data

Continuous
data; only one
dose PH
significant;
BaP given as
3,4-BaP






Continuous
data, no SD for
control; use
lowest dose
approaching
peak
Separate RPFs
for sister
chromatid
exchanges and
micronuclei,
oral and
intraperitoneal
                             C-95
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-23. In vivo clastogenicity: dose-response data
Record
number
24740

















Reference
Allen etal., 1999

















PAH
Tri-
caprylin
BaP
DBalP
DBalP
DBalP
DBalP
Tri-
caprylin
BaP
DBalP
DBalP
DBalP
DBalP
Tri-
caprylin
BaP
DBalP
DBalP
DBalP
Tri-
caprylin
Route of
admini-
stration
Intra-
peritoneal
Intra-
peritoneal
Intra-
peritoneal
Intra-
peritoneal
Intra-
peritoneal
Intra-
peritoneal
Intra-
peritoneal
Intra-
peritoneal
Intra-
peritoneal
Intra-
peritoneal
Intra-
peritoneal
Intra-
peritoneal
Intra-
peritoneal
Intra-
peritoneal
Intra-
peritoneal
Intra-
peritoneal
Intra-
peritoneal
Intra-
peritoneal
Clastogenicity
Dose
0
200
0.3
1.5
3
6
0
200
0.3
1.5
3
6
0
200
9
12
18
0
Dose units
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
Mean
2.6
11.2
2
3.9
3.4
3.8
2.8
9.5
2.8
2.9
4
4.3
3.2
5.1
4.3
7.4
6.1
3.5
SD


















Units
MN-PCEs
MN-PCEs
MN-PCEs
MN-PCEs
MN-PCEs
MN-PCEs
MN-PCEs
MN-PCEs
MN-PCEs
MN-PCEs
MN-PCEs
MN-PCEs
MN-PCEs
MN-PCEs
MN-PCEs
MN-PCEs
MN-PCEs
MN-PCEs
n
1,000
1,000
1,000
1,000
1,000
1,000
1,000
1,000
1,000
1,000
1,000
1,000
1,000
1,000
1,000
1,000
1,000
1,000
%
Response
0.0026
0.0112
0.0020
0.0039
0.0034
0.0038
0.0028
0.0095
0.0028
0.0029
0.0040
0.0043
0.0032
0.0051
0.0043
0.0074
0.0061
0.0035
Units
PCEs
PCEs
PCEs
PCEs
PCEs
PCEs
PCEs
PCEs
PCEs
PCEs
PCEs
PCEs
PCEs
PCEs
PCEs
PCEs
PCEs
PCEs
P<
0.05

X

X



X



X

X

X
X

Notes
A/J mice, bone marrow





A/J mice, peripheral
blood





p53 +/+ wt mice, bone
marrow




p53 +/+ wt mice, peri-
pheral blood
                                         C-96
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-23. In vivo clastogenicity: dose-response data
Record
number




14270










17190



Reference




He and Baker, 1991










Bayer, 1978



PAH
BaP
DBalP
DBalP
DBalP
Control
BaP
BaP
BaP
BaP
BaP
Control
CH
CH
CH
CH
Pooled
controls
BaP
BaP
BaP
Route of
admini-
stration
Intra-
peritoneal
Intra-
peritoneal
Intra-
peritoneal
Intra-
peritoneal
Dermal
Dermal
Dermal
Dermal
Dermal
Dermal
Dermal
Dermal
Dermal
Dermal
Dermal
Intra-
peritoneal
Intra-
peritoneal
Intra-
peritoneal
Intra-
peritoneal
Clastogenicity
Dose
200
9
12
18
0
0.5
5
50
100
500
0
50
100
500
1,000
0
2.5
25
40
Dose units
mg/kg
mg/kg
mg/kg
mg/kg
Hg/mouse
Hg/mouse
Hg/mouse
Hg/mouse
Hg/mouse
Hg/mouse
Hg/mouse
Hg/mouse
Hg/mouse
Hg/mouse
Hg/mouse
mg/kg
mg/kg
mg/kg
mg/kg
Mean
5.7
3.1
3.1
4.6
13.3
50.5
66.8
76
64.3
55.8
12.8
43.3
56
62
47.3
3.2
3.4
3.5
3.9
SD




2.8
11.5
4.1
2.8
5.4
13
2.2
2.2
4.9
8.6
3.8
0.07
0.8
0.2
0.2
Units
MN-PCEs
MN-PCEs
MN-PCEs
MN-PCEs
MN cells
MN cells
MN cells
MN cells
MN cells
MN cells
MN cells
MN cells
MN cells
MN cells
MN cells
Sister
chromatid
exchange/
cells
Sister
chromatid
exchange/
cells
Sister
chromatid
exchange/
cells
Sister
chromatid
exchange/
cells
n
1,000
1,000
1,000
1,000
1,000
1,000
1,000
1,000
1,000
1,000
1,000
1,000
1,000
1,000
1,000




%
Response
0.0057
0.0031
0.0031
0.0046
0.013
0.051
0.067
0.076
0.064
0.056
0.013
0.043
0.056
0.062
0.047




Units
PCEs
PCEs
PCEs
PCEs
Binucleated
Binucleated
Binucleated
Binucleated
Binucleated
Binucleated
Binucleated
Binucleated
Binucleated
Binucleated
Binucleated




P<
0.05
X




X
X
X
X
X

X
X
X
X



X
Notes



















                                         C-97
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-23. In vivo clastogenicity: dose-response data
Record
number







20950

Reference







Roszinsky-Kocher et
al., 1979

PAH
BaP
BaP
BaP
PH
PH
PH
PH
Control
BaP
Route of
admini-
stration
Intra-
peritoneal
Intra-
peritoneal
Intra-
peritoneal
Intra-
peritoneal
Intra-
peritoneal
Intra-
peritoneal
Intra-
peritoneal
Intra-
peritoneal
Intra-
peritoneal
Clastogenicity
Dose
50
75
100
25
50
75
100
0
900
Dose units
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
Mean
6.4
6.4
7.4
3.5
3.4
3.5
4.1
3.9
10.6
SD
0.2
0.3
0.2
0.2
0.2
0.2
0.2
0.9
1.6
Units
Sister
chromatid
exchange/
cells
Sister
chromatid
exchange/
cells
Sister
chromatid
exchange/
cells
Sister
chromatid
exchange/
cells
Sister
chromatid
exchange/
cells
Sister
chromatid
exchange/
cells
Sister
chromatid
exchange/
cells
Sister
chromatid
exchanges/
meta-phase
Sister
chromatid
exchanges/
meta-phase
n









%
Response









Units









P<
0.05
X
X
X



X

X
Notes



Only one dose significant





                                         C-98
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-23. In vivo clastogenicity: dose-response data
Record
number






24720




Reference






Kligerman et al,
1986




PAH
DBahA
CH
PH
BeP
BbF
BaA
Control
BaP
BaP
BaP
Control
Route of
admini-
stration
Intra-
peritoneal
Intra-
peritoneal
Intra-
peritoneal
Intra-
peritoneal
Intra-
peritoneal
Intra-
peritoneal
Gavage
Gavage
Gavage
Gavage
Gavage
Clastogenicity
Dose
900
900
900
900
900
900
0
63
252
504
0
Dose units
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
Mean
4.9
5.1
5.5
5.5
5.6
6.1
11.9
19.4
21.5
21.7
11.0
SD
0.7
1
0.7
0.7
0.5
0.4

0.0
1.4
1.4

Units
Sister
chromatid
exchanges
Sister
chromatid
exchanges
Sister
chromatid
exchanges
Sister
chromatid
exchanges
Sister
chromatid
exchanges
Sister
chromatid
exchanges
Sister
chromatid
exchanges/
meta-phase
Sister
chromatid
exchanges/
meta-phase
Sister
chromatid
exchanges/
meta-phase
Sister
chromatid
exchanges/
meta-phase
Sister
chromatid
exchanges/
meta-phase
n











%
Response











Units











P<
0.05
X
X
X
X
X
X





Notes











                                         C-99
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-23. In vivo clastogenicity: dose-response data
Record
number



24790








Reference



Kligerman et al.,
2002








PAH
B1AC
B1AC
B1AC
Control
BaP
BaA
BbF
CH
Control
BaP
BaA
Control
Route of
admini-
stration
Gavage
Gavage
Gavage
Intra-
peritoneal
Intra-
peritoneal
Intra-
peritoneal
Intra-
peritoneal
Intra-
peritoneal
Gavage
Gavage
Gavage
Gavage
Clastogenicity
Dose
32
63
126
0
100
100
100
100
0
100
100
0
Dose units
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
mg/kg
Mean
16.5
20.5
27.8
8.79
21.21
14.8
22.25
11.96
11.12
17.91
13.38
6.6
SD
3.6
1.6
2.6
1.26
2.93
3.16
1.45
1.8
1.5
1.49
1.53
0.9
Units
Sister
chromatid
exchanges/
meta-phase
Sister
chromatid
exchanges/
meta-phase
Sister
chromatid
exchanges/
meta-phase
Sister
chromatid
exchanges
Sister
chromatid
exchanges
Sister
chromatid
exchanges
Sister
chromatid
exchanges
Sister
chromatid
exchanges
Sister
chromatid
exchanges
Sister
chromatid
exchanges
Sister
chromatid
exchanges
MNbn
n











1,000
%
Response











0.007
Units











Binucleated
P<
0.05




X
X
X
X

X
X

Notes












                                         C-100
DRAFT - DO NOT CITE OR QUOTE

-------
Table C-23. In vivo clastogenicity: dose-response data
Record
number


Reference


PAH
BaP
BbF
Route of
admini-
stration
Gavage
Gavage
Clastogenicity
Dose
100
100
Dose units
mg/kg
mg/kg
Mean
9.1
8.3
SD
1.8
0.9
Units
MNbn
MNbn
n
1,000
1,000
%
Response
0.009
0.008
Units
Binucleated
Binucleated
P<
0.05
X
X
Notes


                                         C-101
DRAFT - DO NOT CITE OR QUOTE

-------
 1

 2
 o
 J

 4

 5

 6
            APPENDIX D. BENCHMARK DOSE MODELING OUTPUTS
D.I.  DERMAL BIOASSAYS
      o
      I
      C
      o
      •*=
      o
      (0
        0.8


        0.7


        0.6


        0.5


        0.4


        0.3


        0.2


        0.1

                              Multistage Cancer Model with 0.95 Confidence Level
                                      Multistage Cancer
                                     Linear extrapolation
                        BMDL
                               BMD
                                                10
                                               dose
                                                          15
20
       11:1412/282009
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
Cav 1983 bap dermal.out.txt
        Multistage  Cancer Model.  (Version: 1.7;  Date:  05/16/2008)
        Input Data  File:
C:\USEPA\IRIS\PAH\dermal\complete\Cavalieril983\BaP\msc_CavalieriBaP_MS_2.(d)
        Gnuplot  Plotting  File:
C:\USEPA\IRIS\PAH\dermal\complete\Cavalieril983\BaP\msc_CavalieriBaP_MS_2.pit
                                           Tue Dec 22  14:50:32  2009
 BMDS Model Run
   The form of the  probability function is:

   P[response] = background + (1-background)*[1-EXP(
                 -betal*doseAl-beta2*doseA2)]
                                        D-l
                                                DRAFT - DO NOT CITE OR QUOTE

-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60

The parameter betas are restricted


Dependent variable = incidence
Independent variable = dose

Total number of observations = 4
Total number of records with missing
Total number of parameters in model
Total number of specified parameters
Degree of polynomial = 2


Maximum number of iterations = 250

to be positive






values = 0
= 3
= 0




Relative Function Convergence has been set to: 2.22045e-016
Parameter Convergence has been set to: 1.49012e-008


**** We are sorry but Relative Function and Parameter Convergence ****
**** are currently unavailable in this model. Please keep checking ****
**** the web sight for model updates
which will eventually ****
**** incorporate these convergence criterion. Default values used. ****






Default Initial Parameter Values
Background =
Beta(l) =
Beta(2) = 0


0.0155298
0
.00204447


Asymptotic Correlation Matrix of Parameter Estimates

( *** The model parameter (
have been estimated
specified by the user,
and do not appear in

Beta(l) Beta(2)

Beta(l) 1 -0.96

Beta(2) -0.96 1




s) -Background
at a boundary point, or have been

the correlation matrix )









Parameter Estimates


Confidence Interval
Variable Estimate
Upper Conf. Limit
Background 0
*
Beta(l) 0.0126577
*
Beta(2) 0.00134916
*

* - Indicates that this value is not

95.0% Wald

Std. Err. Lower Conf. Limit

* *

* *

* *


calculated.
D-2       DRAFT - DO NOT CITE OR QUOTE

-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
i «
lo
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48




Model
Full model
Fitted model
0.3878
Reduced model

AIC:




Dose Est


0.0000 0.
2.2000 0.
6.6000 0.
20.0000 0.

ChiA2 = 1.95


Benchmark Dose

Specified effect

Risk Type

Confidence level

BMD

BMDL

BMDU

Taken together, (
interval for the

Multistage Cancer






Analysis of Deviance Table

Log (likelihood) # Param's Deviance Test d.f. P-value
-35.0798 4
-36.0272 2 1.89478 2

-55.062 1 39.9644 3 <.0001

76.0543


Goodness of Fit
Scaled
. Prob. Expected Observed Size Residual


0000 0.000 0.000 29 0.000
0338 1.014 2.000 30 0.996
1326 3.714 2.000 28 -0.955
5474 16.423 17.000 30 0.212

d.f. = 2 P-value = 0.3772


Computation

0.1

= Extra risk

0.95

5.31398

2.86439

8.84432

2.86439, 8.84432) is a 90 % two-sided confidence
BMD

Slope Factor = 0.0349115




D-3
DRAFT - DO NOT CITE OR QUOTE

-------
                              Multistage Cancer Model with 0.95 Confidence Level
 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
      T3
      
-------
 1
 2     Total number of observations = 4
 3     Total number of records with missing values = 0
 4     Total number of parameters in model = 3
 5     Total number of specified parameters = 0
 6     Degree of polynomial = 2
 7
 8
 9     Maximum number of iterations = 250
10     Relative Function Convergence has been set to: 2.22045e-016
11     Parameter Convergence has been set to: 1.49012e-008
12
13    ****   we are sorry but Relative Function and Parameter Convergence    ****
14    ****   are currently unavailable in this model.  Please keep checking  ****
15    ****   the web sight for model updates which will eventually           ****
15    ****   incorporate these convergence criterion.  Default values used.  ****
17
18
19
20                      Default Initial Parameter Values
21                         Background =            0
22                            Beta(l) =            0
23                            Beta(2) = 4.42193e-005
24
25
26               Asymptotic Correlation Matrix of Parameter Estimates
27
28               ( *** The model parameter(s)  -Background
29                     have been estimated at a boundary point, or have been
30    specified by the user,
31                     and do not appear in the correlation matrix )
32
33                    Beta(l)       Beta(2)
34
35       Beta(l)             1        -0.93
36
37       Beta(2)         -0.93            1
38
39
40
41                                     Parameter Estimates
42
43                                                             95.0% Wald
44    Confidence Interval
45           Variable         Estimate        Std. Err.     Lower Conf. Limit
46    Upper Conf.  Limit
47         Background                0            *                *
48    *
49            Beta(l)       0.000525847            *                *
50    *
51            Beta(2)      3.60995e-005            *                *
52    *
53
54    * - Indicates that this value is not calculated.
55
56
57
58                            Analysis of Deviance Table
59
60           Model      Log(likelihood)  # Param's  Deviance  Test d.f.   P-value


                                       D-5          DRAFT - DO NOT CITE OR QUOTE

-------
1
2
3
4
5
6
7
9
10
11
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
Full model
Fitted model
0.1115
Reduced model

AIC:



Dose Est
0.0000 0.
22.2000 0.
66.6000 0.
200.0000 0.

ChiA2 = 4.25


Benchmark Dose

Specified effect

Risk Type

Confidence level

BMD

BMDL

BMDU

Taken together, (
interval for the

Multistage Cancer




-27.8865 4
-30.0799 2 4.38685 2

-64.1091 1 72.4452 3 <.0

64.1598

Goodness of Fit
Scaled
. Prob. Expected Observed Size Residual
0000 0.000 0.000 29 0.000
0290 0.842 2.000 29 1.281
1773 5.141 2.000 29 -1.527
7876 22.840 24.000 29 0.527

d.f. = 2 P-value = 0.1194


Computation

0.1

= Extra risk

0.95

47.2296

30.0553

62.746

30.0553, 62.746 ) is a 90 % two-sided confidence
BMD

Slope Factor = 0.00332721




D-6
DRAFT - DO NOT CITE OR QUOTE

-------
                              Multistage Cancer Model with 0.95 Confidence Level
      T3
      0)
      •5

      I
      o
      13
      ro
             0.7
             0.6
             0.5
        0.4
        0.3
             0.2
             0.1
                                       Multistage Cancer
                                      Linear extrapolation
                                                BMDL
                                                     BMD
                                               dose
 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
       12:3912/282009
HABS1980BBF.OUT.txt
        Multistage  Cancer Model.  (Version:  1.7;   Date:  05/16/2008)
        Input  Data  File:
C:\USEPA\IRIS\PAH\dermal\complete\Habsl980\BbF\msc_HabsBbF_MS_2_10.(d)
        Gnuplot  Plotting  File:
C:\USEPA\IRIS\PAH\dermal\complete\Habsl980\BbF\msc_HabsBbF_MS_2_10.plt
                                           Thu  Dec 24  10:03:13 2009
 BMDS Model Run
   The form of  the  probability function is:

   P[response]  =  background + (1-background)*[1-EXP(
                  -betal*doseAl-beta2*doseA2)]

   The parameter  betas  are restricted to be  positive
   Dependent variable  = incidence
   Independent variable = dose
 Total number  of  observations = 4
 Total number  of  records with missing values
= 0
                                        D-7
                                                DRAFT - DO NOT CITE OR QUOTE

-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
 Total number of parameters in model = 3
 Total number of specified parameters = 0
 Degree of polynomial = 2
 Maximum number of iterations = 250
 Relative Function Convergence has been set to: 2.22045e-016
 Parameter Convergence has been set to: 1.49012e-008

****  We are sorry but Relative Function and Parameter Convergence
****  are currently unavailable in this model.  Please keep checking
****  the web sight for model updates which will eventually
****  incorporate these convergence criterion.  Default values used.
                  Default Initial Parameter Values
                     Background =            0
                        Beta(l) =            0
                        Beta(2) =   0.00945627
           Asymptotic Correlation Matrix of Parameter Estimates

            ( *** The model parameter(s)  -Background    -Beta(l)
                 have been estimated at a boundary point,  or have  been
specified by the user,
                 and do not appear in the correlation matrix )
   Beta(2;
Beta (2)

      1
Confidence Interval
       Variable
Upper Conf. Limit
     Background
*
        Beta (1)
*
        Beta(2)
        Estimate

               0

               0

      0.00748156
Parameter Estimates

                        95.0% Wald

       Std.  Err.     Lower Conf. Limit
  - Indicates that this value is not calculated.
       Model
     Full model
   Fitted model
0.5447
  Reduced model
        Analysis of Deviance Table

  Log(likelihood)  # Param's  Deviance  Test d.f.
       -47.5575         4
       -48.6255         1       2.13602      3
       -69.4912
       1
43.8674
                                   P-value
<.0001
                                       D-8
                                               DRAFT - DO NOT CITE OR QUOTE

-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
           AIC:
                         99.251
Goodness of Fit

Dose
0.0000
3.4000
5.6000
9.2000

Est. Prob.
0.0000
0.0829
0.2091
0.4691

Expected
0.000
3.148
7.110
17.358

Observed
0.000
2.000
5.000
20.000

Size
35
38
34
37
Scaled
Residual
0.000
-0.676
-0.890
0.870
 ChiA2 =2.01
d.f. = 3
P-value = 0.5711
   Benchmark Dose Computation
Specified effect =

Risk Type

Confidence level =

             BMD =

            BMDL =

            BMDU =
          0.24

     Extra risk

          0.95

       6.05655

       5.19938

       7.17099
Taken together,  (5.19938, 7.17099) is  a  90
interval for the BMD
                            % two-sided confidence
Multistage Cancer Slope Factor =
                  0.0461594
                                       D-9
                                               DRAFT - DO NOT CITE OR QUOTE

-------
                              Multistage Cancer Model with 0.95 Confidence Level
                                       Multistage Cancer
                                      Linear 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
      T3
      0)
      •5

      I
      o
      13
      ro
             0.8
        0.6
        0.4
             0.2
               0          0.02

   12:4412/282009

HOFFMANWYNDER966DBAIP.OUT.txt
                                          0.04         0.06
                                               dose
0.08
        Multistage  Cancer Model.  (Version:  1.7;   Date:  05/16/2008)
        Input  Data  File:
C:\USEPA\IRIS\PAH\dermal\complete\HoffWyndl966\DBaiP\msc_HoffWynDBaiP_MS_l.(d
)
        Gnuplot  Plotting File:
C:\USEPA\IRIS\PAH\dermal\complete\HoffWyndl966\DBaiP\msc_HoffWynDBaiP_MS_l.pl
t
                                            Tue  Dec 22 14:50:33 2009
 BMDS Model Run
   The form of  the  probability function  is:

   P[response]  =  background +  (1-background)*[1-EXP(
                  -betal*doseAl)]

   The parameter  betas are restricted to be  positive
   Dependent variable = incidence
   Independent  variable = dose

 Total number of  observations = 3
                                        D-10
                                                DRAFT - DO NOT CITE OR QUOTE

-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
 Total number of records with missing values = 0
 Total number of parameters in model = 2
 Total number of specified parameters = 0
 Degree of polynomial = 1
 Maximum number of iterations = 250
 Relative Function Convergence has been set to: 2.22045e-016
 Parameter Convergence has been set to: 1.49012e-008

****  We are sorry but Relative Function and Parameter Convergence    ****
****  are currently unavailable in this model.  Please keep checking  ****
****  the web sight for model updates which will eventually           ****
****  incorporate these convergence criterion.  Default values used.  ****
                  Default Initial Parameter Values
                     Background =     0.264818
                        Beta(l) =      18.4583
           Asymptotic Correlation Matrix of Parameter Estimates

           (  *** The model parameter(s)  -Background
                 have been estimated at a boundary point, or have been
specified by the user,
                 and do not appear in the correlation matrix )
   Beta (1)
                Beta (1)

                      1
                                 Parameter Estimates
                                        Std. Err.
Confidence Interval
       Variable         Estimate
Upper Conf. Limit
     Background                0
*
        Beta(l)          25.3832
* - Indicates that this value is not calculated.
                 95.0% Wald

              Lower Conf. Limit
       Model
     Full model
   Fitted model
0.2358
  Reduced model

           AIC:
                        Analysis of Deviance Table

                  Log(likelihood)  # Param's  Deviance  Test d.f.
                       -16.5742         3
                        -18.019         1       2.88957      2
                       -39.8916

                        38.0379
1
46.6349
2
                            P-value
<.0001
                                       D-ll
                                               DRAFT - DO NOT CITE OR QUOTE

-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
Goodness of Fit
Dose Est. Prob. Expected Observed

0.0000 0.0000 0.000 0.000
0.0500 0.7189 13.660 16.000
0.1000 0.9210 17.499 16.000

ChiA2 = 3.05 d.f. = 2 P-value = 0.2174


Benchmark Dose Computation

Specified effect = 0.1

Risk Type = Extra risk

Confidence level = 0.95

BMD = 0.00415079

BMDL = 0.00298234

BMDU = 0. 00587793

Taken together, (0.00298234, 0.00587793) is a 90
interval for the BMD

Multistage Cancer Slope Factor = 33.5308




Scaled
Size Residual

20 0.000
19 1.194
19 -1.275


















% two-sided confidence







D-12        DRAFT - DO NOT CITE OR QUOTE

-------
                              Multistage Cancer Model with 0.95 Confidence Level
                                       Multistage Cancer
                                      Linear 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
      T3
      0)
      •5

      I
      o
      13
      ro
             0.8
        0.6
        0.4
             0.2  -
               0         0.02

   12:4512/282009

HOFFMANWYNDER1966BAP.OUT.txt
                                          0.04        0.06
                                               dose
              0.08
0.1
        Multistage  Cancer Model.  (Version:  1.7;   Date:  05/16/2008)
        Input  Data  File:
C:\USEPA\IRIS\PAH\dermal\complete\HoffWyndl966\BaP\msc_HoffWynBaP_MS_l.(d)
        Gnuplot  Plotting  File:
C:\USEPA\IRIS\PAH\dermal\complete\HoffWyndl966\BaP\msc_HoffWynBaP_MS_l.plt
                                           Tue  Dec 22 14:50:32 2009
 BMDS Model Run
   The form of  the  probability function is:

   P[response]  =  background + (1-background)*[1-EXP(
                  -betal*doseAl)]

   The parameter  betas  are restricted to be  positive
   Dependent variable = incidence
   Independent  variable = dose
 Total number  of  observations = 3
 Total number  of  records with missing values
 Total number  of  parameters in model = 2
= 0
                                        D-13
                                                DRAFT - DO NOT CITE OR QUOTE

-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
 Total number of specified parameters = 0
 Degree of polynomial = 1
 Maximum number of iterations = 250
 Relative Function Convergence has been set to: 2.22045e-016
 Parameter Convergence has been set to: 1.49012e-008

****  We are sorry but Relative Function and Parameter Convergence     ****
****  are currently unavailable in this model.  Please keep checking   ****
****  the web sight for model updates which will eventually            ****
****  incorporate these convergence criterion.  Default values used.   ****
                  Default Initial Parameter Values
                     Background =     0.124609
                        Beta(l) =      29.9573
           Asymptotic Correlation Matrix of Parameter Estimates

            ( *** The model parameter(s)  -Background
                 have been estimated at a boundary point, or have been
specified by the user,
                 and do not appear in the correlation matrix )
   Beta (1)
Beta(l)

      1
                                 Parameter Estimates
Confidence Interval
       Variable
Upper Conf. Limit
     Background
*
        Beta (1)
        Estimate

               0

         34.3074
Std.  Err.
         95.0% Wald

      Lower Conf. Limit
* - Indicates that this value is not calculated.
       Model
     Full model
   Fitted model
0.8616
  Reduced model

           AIC:
        Analysis of Deviance Table

  Log(likelihood)  # Param's  Deviance  Test d.f.
       -12.4245         3
       -12.5735         1      0.297928      2
       -40.3807

        27.1469
1
55.9124
2
                            P-value
<.0001
                                  Goodness  of  Fit
                                       D-14
                                               DRAFT - DO NOT CITE OR QUOTE

-------
 1                                                                     Scaled
 2         Dose     Est.  Prob.     Expected    Observed     Size       Residual
 3      	1	
 4        0.0000     0.0000         0.000     0.000          20        0.000
 5        0.0500     0.8201        16.402    17.000          20        0.348
 6        0.1000     0.9676        19.353    19.000          20       -0.446
 7
 8     ChiA2  = 0.32      d.f.  = 2        P-value = 0.8522
 9
10
11       Benchmark Dose Computation
12
13    Specified effect =             0.1
14
15    Risk Type        =       Extra risk
16
17    Confidence level =            0.95
18
19                 BMD =      0.00307107
20
21                BMDL =      0.00215021
22
23                BMDU =      0.00440601
24
25    Taken together,  (0.00215021,  0.00440601)  is a 90     % two-sided confidence
26    interval for the BMD
27
28    Multistage Cancer Slope  Factor =       46.5071
29
30
31
                                       D-15        DRAFT - DO NOT CITE OR QUOTE

-------
                              Multistage Cancer Model with 0.95 Confidence Level
                                       Multistage Cancer
                                      Linear 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
      T3
      0)
      •5

      I
      o
      13
      ro
             0.8
        0.6
        0.4
             0.2  -
               0          0.02

   12:4712/282009

HOFFMANWYNDER1966DBAEF.OUT.txt
                                          0.04         0.06
                                               dose
0.08
0.1
        Multistage  Cancer Model.  (Version:  1.7;   Date: 05/16/2008)
        Input  Data  File:
C:\USEPA\IRIS\PAH\dermal\complete\HoffWyndl966\DBaeF\msc_HoffWynDBaeF_MS_l.(d
)
        Gnuplot  Plotting File:
C:\USEPA\IRIS\PAH\dermal\complete\HoffWyndl966\DBaeF\msc_HoffWynDBaeF_MS_l.pl
t
                                            Tue  Dec 22 14:50:34 2009
 BMDS Model Run
   The form of  the  probability function  is:

   P[response]  =  background +  (1-background)*[1-EXP(
                  -betal*doseAl)]

   The parameter  betas are restricted to be  positive
   Dependent variable = incidence
   Independent  variable = dose

 Total number of  observations = 3
                                        D-16
                                                DRAFT - DO NOT CITE OR QUOTE

-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
 Total number of records with missing values = 0
 Total number of parameters in model = 2
 Total number of specified parameters = 0
 Degree of polynomial = 1
 Maximum number of iterations = 250
 Relative Function Convergence has been set to: 2.22045e-016
 Parameter Convergence has been set to: 1.49012e-008

****  We are sorry but Relative Function and Parameter Convergence    ****
****  are currently unavailable in this model.  Please keep checking  ****
****  the web sight for model updates which will eventually           ****
****  incorporate these convergence criterion.  Default values used.  ****
                  Default Initial Parameter Values
                     Background =      0.22871
                        Beta(l) =      29.4444
           Asymptotic Correlation Matrix of Parameter Estimates

           (  *** The model parameter(s)  -Background
                 have been estimated at a boundary point, or have been
specified by the user,
                 and do not appear in the correlation matrix )
   Beta (1)
                Beta (1)

                      1
                                 Parameter Estimates
                                        Std. Err.
Confidence Interval
       Variable         Estimate
Upper Conf. Limit
     Background                0
*
        Beta(l)          37.3037
* - Indicates that this value is not calculated.
                 95.0% Wald

              Lower Conf. Limit
       Model
     Full model
   Fitted model
0.6395
  Reduced model

           AIC:
                        Analysis of Deviance Table

                  Log(likelihood)  # Param's  Deviance  Test d.f.
                       -10.3111         3
                       -10.7582         1      0.894194      2
                       -38.9521

                        23.5163
1
57.2822
2
                            P-value
<.0001
                                       D-17
                                               DRAFT - DO NOT CITE OR QUOTE

-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
Goodness of Fit
Dose Est. Prob. Expected Observed

0.0000 0.0000 0.000 0.000
0.0500 0.8451 16.058 17.000
0.1000 0.9760 18.544 18.000

ChiA2 = 1.02 d.f. = 2 P-value = 0.5995


Benchmark Dose Computation

Specified effect = 0.1

Risk Type = Extra risk

Confidence level = 0.95

BMD = 0.0028244

BMDL = 0.00193834

BMDU = 0. 00411821

Taken together, (0.00193834, 0.00411821) is a 90
interval for the BMD

Multistage Cancer Slope Factor = 51.5905




Scaled
Size Residual

20 0.000
19 0.598
19 -0.816


















% two-sided confidence







D-18       DRAFT - DO NOT CITE OR QUOTE

-------
Multistage Cancer Model with 0.95 Confidence Level
0.8
0.7
0.6
T3 0.5
0)
•5
^ 0.4
0
1 0.3
U-
0.2
0.1
0

: Multistage Ca
Linear extrapole
-
-
- //
//
• /
:T /
V
< •
BMDL BMD
0 0.02 0.04
1PPT
ILrCI
tjnn
LIUII
^-^^
'/^





0.06 0.08 0
-;
:
:
-

:


.1








dose
1 12:4812/282009
2
3 HOFFMANWYNDER1996DBAEP.OUT.txt
4
c
J 	
6 Multistage Cancer Model. (Version:
7 Input Data File:




1.7; Date: 05/16/2008)






8 C:\USEPA\IRIS\PAH\dermal\complete\HoffWyndl966\DBaeP\msc HoffWynDBaeP MS l.(d
9 )
10 Gnuplot Plotting File:




11 C:\USEPA\IRIS\PAH\dermal\complete\HoffWyndl966\DBaeP\msc HoffWynDBaeP MS l.pl
12 t


13 Tue Dec 22 14:50:32 2009
1 A
14 	
15
16 BMDS Model Run
1 7
18
19 The form of the probability function is :
20









21 P [response] = background + ( 1-background) * [ 1-EXP (
22 -betal*doseAl) ]
23




24 The parameter betas are restricted to be positive
25
26
27 Dependent variable = incidence
28 Independent variable = dose
29
30 Total number of observations = 3












           D-19
DRAFT - DO NOT CITE OR QUOTE

-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
 Total number of records with missing values = 0
 Total number of parameters in model = 2
 Total number of specified parameters = 0
 Degree of polynomial = 1
 Maximum number of iterations = 250
 Relative Function Convergence has been set to: 2.22045e-016
 Parameter Convergence has been set to: 1.49012e-008

****  We are sorry but Relative Function and Parameter Convergence     ****
****  are currently unavailable in this model.  Please keep checking   ****
****  the web sight for model updates which will eventually            ****
****  incorporate these convergence criterion.  Default values used.   ****
                  Default Initial Parameter Values
                     Background =     0.120514
                        Beta(l) =      7.53772
           Asymptotic Correlation Matrix of Parameter Estimates

           (  *** The model parameter(s)  -Background
                 have been estimated at a boundary point, or have been
specified by the user,
                 and do not appear in the correlation matrix )
   Beta (1)
                Beta (1)

                      1
                                 Parameter Estimates
                                        Std. Err.
Confidence Interval
       Variable         Estimate
Upper Conf. Limit
     Background                0
*
        Beta(l)          11.2084
* - Indicates that this value is not calculated.
                 95.0% Wald

              Lower Conf. Limit
       Model
     Full model
   Fitted model
0.2414
  Reduced model

           AIC:
                        Analysis of Deviance Table

                  Log(likelihood)  # Param's  Deviance  Test d.f.
                       -32.4818         3
                        -33.903         1       2.84251      2
                       -44.2604

                        69.8061
1
23.5572
2
                            P-value
<.0001
                                       D-20
                                               DRAFT - DO NOT CITE OR QUOTE

-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
Goodness of Fit
Dose Est. Prob. Expected Observed

0.0000 0.0000 0.000 0.000
0.0500 0.4290 12.871 16.000
0.1000 0.6740 11.458 9.000

ChiA2 = 2.95 d.f. = 2 P-value = 0.2288


Benchmark Dose Computation

Specified effect = 0.1

Risk Type = Extra risk

Confidence level = 0.95

BMD = 0.00940018

BMDL = 0.00681373

BMDU = 0. 0134192

Taken together, (0.00681373, 0.0134192) is a 90
interval for the BMD

Multistage Cancer Slope Factor = 14.6763



Scaled
Size Residual

20 0.000
30 1.154
17 -1.272


















% two-sided confidence






D-21        DRAFT - DO NOT CITE OR QUOTE

-------
                              Multistage Cancer Model with 0.95 Confidence Level
 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
      T3
      0)
      •5

      I
      o
      13
      ro
             0.8
             0.6
        0.4
             0.2
                                       Multistage Cancer
                                     Linear extrapolation
                                                       BMDL
                                                                         BMi)
                             200
                                  400
  600
dose
800
1000
1200
   12:5012/282009

LAVOIE1982BkF.OUT.txt
        Multistage  Cancer Model.  (Version: 1.7;   Date:  05/16/2008)
        Input Data  File:
C:\USEPA\IRIS\PAH\dermal\initiation\LaVoiel982\BkF\msc_LaVoieBkF_MS_2_85.(d)
        Gnuplot  Plotting  File:
C:\USEPA\IRIS\PAH\dermal\initiation\LaVoiel982\BkF\msc_LaVoieBkF_MS_2_85.plt
                                           Thu Dec 24  10:09:52 2009
 BMDS Model Run
   The form of the  probability function is:

   P[response] = background + (1-background)*[1-EXP(
                 -betal*doseAl-beta2*doseA2)]

   The parameter betas  are restricted to be positive
   Dependent variable  = incidence
   Independent variable = dose
 Total number of  observations = 4
 Total number of  records  with missing values
    = 0
                                        D-22
                                                DRAFT - DO NOT CITE OR QUOTE

-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
 Total number of parameters in model = 3
 Total number of specified parameters = 0
 Degree of polynomial = 2
 Maximum number of iterations = 250
 Relative Function Convergence has been set to: 2.22045e-016
 Parameter Convergence has been set to: 1.49012e-008

****  We are sorry but Relative Function and Parameter Convergence
****  are currently unavailable in this model.  Please keep checking
****  the web sight for model updates which will eventually
****  incorporate these convergence criterion.  Default values  used.
                  Default Initial Parameter Values
                     Background =    0.0504814
                        Beta(l) =   0.00134342
                        Beta(2) =            0
           Asymptotic Correlation Matrix of Parameter Estimates

            ( *** The model parameter(s)  -Background    -Beta(2)
                 have been estimated at a boundary point,  or have been
specified by the user,
                 and do not appear in the correlation matrix )
   Beta (1)
Beta (1)

      1
Confidence Interval
       Variable
Upper Conf. Limit
     Background
*
        Beta (1)
*
        Beta(2)
        Estimate

               0

      0.00163117

               0
Parameter Estimates

                        95.0% Wald

       Std.  Err.     Lower Conf. Limit
  - Indicates that this value is not calculated.
       Model
     Full model
   Fitted model
0.6388
  Reduced model
        Analysis of Deviance Table

  Log(likelihood)  # Param's  Deviance  Test d.f.
       -26.4637         4
       -27.3094         1       1.69146      3
       -46.0525
       1
39.1775
                                   P-value
<.0001
                                       D-23
                                               DRAFT - DO NOT CITE OR QUOTE

-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
           AIC:
                        56.6189
Goodness of Fit

Dose
0.0000
30.0000
100.0000
1000.0000

Est. Prob.
0.0000
0.0478
0.1505
0.8043

Expected
0.000
0.955
3.010
16.086

Observed
0.000
1.000
5.000
15.000

Size
20
20
20
20
Scaled
Residual
0.000
0.047
1.244
-0.612
 ChiA2 = 1.93
d.f. = 3
P-value = 0.5881
   Benchmark Dose Computation
Specified effect =

Risk Type

Confidence level =

             BMD =

            BMDL =

            BMDU =
          0.85

     Extra risk

          0.95

       1163.04

       802.998

       1836.46
Taken together,  (802.998, 1836.46) is  a  90
interval for the BMD
                            % two-sided confidence
Multistage Cancer Slope Factor =
                 0.00105853
                                       D-24
                                               DRAFT - DO NOT CITE OR QUOTE

-------
                              Multistage Cancer Model with 0.95 Confidence Level
 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
      T3
      0)
      •5

      I
      o
      13
      ro
        0.8


        0.7


        0.6


        0.5


        0.4


        0.3


        0.2


        0.1
                                       Multistage Cancer
                                      Linear extrapolation
                     BMDL   BMD
                                  50
                                           100
                                          dose
           150
200
   12:51 12/282009

RAVEH1982CPCDP.OUT.txt
        Multistage  Cancer Model. (Version:  1.7;   Date:  05/16/2008)
        Input Data  File:
C:\USEPA\IRIS\PAH\dermal\initiation\Ravehl982\CPcdP\msc_RavehCPcdP_MS_2.(d)
        Gnuplot  Plotting  File:
C:\USEPA\IRIS\PAH\dermal\initiation\Ravehl982\CPcdP\msc_RavehCPcdP_MS_2.plt
                                           Tue  Dec 22  14:50:35 2009
 HMDS Model Run
   The form of the  probability function is:

   P[response] = background + (1-background)*[1-EXP(
                 -betal*doseAl-beta2*doseA2)]

   The parameter betas  are restricted to be  positive
   Dependent variable  = incidence
   Independent variable = dose
 Total number of  observations = 4
 Total number of  records  with missing values
= 0
                                        D-25
                                                DRAFT - DO NOT CITE OR QUOTE

-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
 Total number of parameters in model = 3
 Total number of specified parameters = 0
 Degree of polynomial = 2
 Maximum number of iterations = 250
 Relative Function Convergence has been set to: 2.22045e-016
 Parameter Convergence has been set to: 1.49012e-008

****  We are sorry but Relative Function and Parameter  Convergence
****  are currently unavailable in this model.  Please  keep checking
****  the web sight for model updates which will eventually
****  incorporate these convergence criterion.  Default values  used.
                  Default Initial Parameter Values
                     Background =     0.086614
                        Beta(l) =   0.00379482
                        Beta(2) =             0
           Asymptotic Correlation Matrix of Parameter  Estimates

             Background      Beta(l)      Beta(2)

Background            1        -0.51          0.37

   Beta(l)        -0.51             1        -0.96

   Beta(2)         0.37        -0.96             1
                                 Parameter Estimates
Confidence Interval
       Variable
Upper Conf. Limit
     Background
*
        Beta(l)
*
        Beta (2)
      Estimate

     0.0898027

     0.0034393

  1.91358e-006
Std.  Err.
         95.0% Wald

      Lower Conf. Limit
* - Indicates that this value is not calculated.
       Model
     Full model
   Fitted model
0.6443
  Reduced model

           AIC:
      Analysis of Deviance Table

Log(likelihood)  # Param's  Deviance  Test d.f.
     -57.7672         4
     -57.8738         3      0.213129      1
     -69.2679

      121.748
1
23.0015
                            P-value
<.0001
                                       D-26
                                               DRAFT - DO NOT CITE OR QUOTE

-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
Goodness of Fit

Dose
0.0000
10.0000
100.0000
200.0000

Est. Prob.
0.0898
0.1207
0.3669
0.5762

Expected
2.604
3.622
10.641
16.134

Observed
3.000
3.000
11.000
16.000

Size
29
30
29
28
Scaled
Residual
0.257
-0.349
0.138
-0.051
 ChiA2 = 0.21
d.f.  =1
P-value = 0.6472
   Benchmark Dose Computation
Specified effect =

Risk Type

Confidence level =

             BMD =

            BMDL =

            BMDU =
           0.1

     Extra risk

          0.95

       30.1292

       19.4197

       83.2495
Taken together,  (19.4197, 83.2495) is  a  90
interval for the BMD
                            % two-sided confidence
Multistage Cancer Slope Factor =
                 0.00514942
                                       D-27
                                               DRAFT - DO NOT CITE OR QUOTE

-------
                              Multistage Cancer Model with 0.95 Confidence Level
 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
      T3
      0)
      •5

      I
      o
      13
      ro
             0.8
        0.6
             0.4
             0.2
                                       Multistage Cancer
                                     Linear extrapolation
                BMDLBMD
                                  50
                                           100
                                          dose
           150
200
       12:5212/282009
RAVEH 1982BaP.OUT.txt
        Multistage  Cancer Model.  (Version: 1.7;  Date:  05/16/2008)
        Input Data  File:
C:\USEPA\IRIS\PAH\dermal\initiation\Ravehl982\BaP\msc_RavehBaP_MS_4.(d)
        Gnuplot  Plotting  File:
C:\USEPA\IRIS\PAH\dermal\initiation\Ravehl982\BaP\msc_RavehBaP_MS_4.plt
                                           Tue Dec 22  14:50:34  2009
 BMDS Model Run
   The form of the probability function is:

   P[response] = background + (1-background)*[1-EXP(
                 -betal*doseAl-beta2*doseA2-beta3*doseA3-beta4*doseA4)

   The parameter betas  are restricted to be positive
   Dependent variable  =  incidence
   Independent variable  = dose
 Total number of observations = 6
 Total number of records  with missing values
= 0
                                        D-28
                                                DRAFT - DO NOT CITE OR QUOTE

-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
H
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
 Total number of parameters in model = 5
 Total number of specified parameters = 0
 Degree of polynomial = 4
 Maximum number of iterations = 250
 Relative Function Convergence has been set to: 2.22045e-016
 Parameter Convergence has been set to: 1.49012e-008

****  we are sorry but Relative Function and Parameter Convergence
****  are currently unavailable in this model.  Please keep checking
****  the web sight for model updates which will eventually
****  incorporate these convergence criterion.  Default values used.
                  Default Initial Parameter Values
                     Background =            0
                        Beta(l) = 6.01899e+017
                        Beta (2) =            0
                        Beta(3) =            0
                        Beta (4) =            0
           Asymptotic Correlation Matrix of Parameter Estimates

           (  *** The model parameter (s)  -Beta (2)    -Beta (3)
                 have been estimated at a boundary point, or have  been
specified by the user,
                 and do not appear in the correlation matrix )

             Background

Background            1

   Beta(l)        -0.66

   Beta(4)         0.27
Beta
-0

-0
(1)
.66
1
.52
Beta
0
-0

(4)
.27
.52
1
                                 Parameter Estimates
Confidence Interval
       Variable
Upper Conf. Limit
     Background
*
        Beta(l)
*
        Beta (2)
*
        Beta (3)
*
        Beta(4)
    Estimate

    0.132052

   0.0479561

           0

           0

4.58928e-009
      Std. Err.
                                                          95.0% Wald

                                                      Lower Conf.  Limit
  - Indicates that this value is not calculated.
D-29
                                               DRAFT - DO NOT CITE OR QUOTE

-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16

18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49

Model
Full model
Fitted model
0.2996
Reduced model

AIC:




Dose Est

0.0000 0.
10.0000 0.
25.0000 0.
50.0000 0.
100.0000 0.
200.0000 1.

ChiA2 = 3.86


Benchmark Dose

Specified effect

Risk Type

Confidence level

BMD

BMDL

BMDU

Taken together, (
interval for the

Multistage Cancer




Analysis of Deviance Table
Log (likelihood) # Param's Deviance Test d.f. P-value
-56.5419 6
-58.376 3 3.66814 3

-101.065 1 89.0461 5 <.0001

122.752


Goodness of Fit
Scaled
. Prob. Expected Observed Size Residual

1321 3.829 3.000 29 -0.455
4627 13.419 17.000 29 1.334
7388 20.685 21.000 28 0.135
9233 25.853 24.000 28 -1.316
9955 26.878 27.000 27 0.351
0000 26.000 26.000 26 0.001

d.f. = 3 P-value = 0.2771


Computation

0.1

= Extra risk

0.95

2.19702

1.66278

3.30927

1.66278, 3.30927) is a 90 % two-sided confidence
BMD

Slope Factor = 0.0601403




D-30
DRAFT - DO NOT CITE OR QUOTE

-------
                              Multistage Cancer Model with 0.95 Confidence Level
 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
      T3
      0)
      •5

      I
      o
      13
      ro
             0.8
        0.6
        0.4
             0.2
                                       Multistage Cancer
                                      Linear extrapolation
                      BMDL  BMD
               0      0.2

   12:5312/282009

RICE CPDEFC.OUT.txt
                                   0.4
0.6      0.8
     dose
1.2
1.4
        Multistage  Cancer Model.  (Version:  1.7;   Date:  05/16/2008)
        Input Data  File:
C:\USEPA\IRIS\PAH\dermal\initiation\Rice\CPdefC\msc_RiceCPdefC_MS_2
        Gnuplot  Plotting  File:
C:\USEPA\IRIS\PAH\dermal\initiation\Rice\CPdefC\msc_RiceCPdefC_MS_2
                                           Tue  Dec 22  16:05:10 2009
 HMDS Model Run
   The form of  the  probability function is:

   P[response]  = background + (1-background)*[1-EXP(
                 -betal*doseAl-beta2*doseA2)]

   The parameter betas  are restricted to be  positive
   Dependent variable  = incidence
   Independent variable = dose
                                   !. (d)

                                   ! .pit
 Total number of  observations = 4
 Total number of  records  with missing values
 Total number of  parameters in model = 3
         = 0
                                        D-31
                                                DRAFT - DO NOT CITE OR QUOTE

-------
 1     Total number of specified parameters = 0
 2     Degree of polynomial = 2
 3
 4
 5     Maximum number of iterations = 250
 6     Relative Function Convergence has been set to: 2.22045e-016
 7     Parameter Convergence has been set to: 1.49012e-008
 8
 9    ****   we are sorry but Relative Function and Parameter Convergence    ****
10    ****   are currently unavailable in this model.  Please keep checking  ****
H    ****   the wek sight for model updates which will eventually           ****
12    ****   incorporate these convergence criterion.  Default values used.  ****
13
14
15
16                      Default Initial Parameter Values
17                         Background =            1
18                            Beta(l) = 6.76726e+019
19                            Beta (2) =            0
20
21
22               Asymptotic Correlation Matrix of Parameter Estimates
23
24               ( *** The model parameter(s)  -Beta(l)
25                     have been estimated at a boundary point, or have been
26    specified by the user,
27                     and do not appear in the correlation matrix )
28
29                 Background      Beta(2)
30
31    Background            1        -0.52
32
33       Beta(2)         -0.52            1
34
35
36
37                                     Parameter Estimates
38
39                                                             95.0% Wald
40    Confidence Interval
41           Variable         Estimate        Std. Err.      Lower Conf. Limit
42    Upper Conf.  Limit
43         Background        0.0499931            *                *
44    *
45            Beta(l)                 0            *                *
46    *
47            Beta(2)           44.3919            *                *
48    *
49
50    * - Indicates that this value is not calculated.
51
52
53
54                            Analysis of Deviance Table
55
56           Model      Log(likelihood)  # Param's  Deviance  Test d.f.   P-value
57         Full model        -16.9192         4
58       Fitted model        -16.9195         2   0.000547543      2
59    0.9997
60      Reduced model        -49.6481         1       65.4577      3         <.0001
                                       D-32        DRAFT - DO NOT CITE OR QUOTE

-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
           AIC:
                         37.839
                                  Goodness   of   Fit

Dose
0.0000
0.1500
0.5000
1.5000

Est. Prob.
0.0500
0.6501
1.0000
1.0000

Expected
1.000
13.002
19.000
19.000

Observed
1.000
13.000
19.000
19.000

Size
20
20
19
19
Scaled
Residual
0.000
-0.001
0.017
0.000
 ChiA2 = 0.00
d.f.  =2
P-value = 0.9999
   Benchmark Dose Computation
Specified effect =

Risk Type

Confidence level =

             BMD =

            BMDL =

            BMDU =
          0.88

     Extra risk

          0.95

      0.218546

      0.172781

      0.384831
Taken together,  (0.172781, 0.384831)  is  a  90
interval for the BMD
                              % two-sided  confidence
Multistage Cancer Slope Factor =
                    5.09315
                                       D-33
                                               DRAFT - DO NOT CITE OR QUOTE

-------
                              Multistage Cancer Model with 0.95 Confidence Level
 1
 2
 o
 6
 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
      T3
      0)
      •5

      I
      o
      13
      ro
             0.8
        0.6
        0.4
             0.2
                                       Multistage Cancer
                                     Linear extrapolation
                                    BMDL
                                            BMD
               0          20

   12:5612/282009

NESNOW 1984 DERMAL  BLAC MALE.txt
                                           40
60
80
100
                                               dose
        Multistage  Cancer Model.  (Version:  1.7;   Date:  05/16/2008)
        Input  Data  File:
C:\USEPA\IRIS\PAH\dermal\initiation\Nesnowl984\BIACmale\msc_NesnowBAICmale3HD
D_MS_1.(d)
        Gnuplot  Plotting  File:
C:\USEPA\IRIS\PAH\dermal\initiation\Nesnowl984\BIACmale\msc_NesnowBAICmale3HD
D_MS_l.plt
                                           Tue  Dec 22  16:05:10 2009
 BMDS Model Run
   The form of  the  probability function is:

   P[response]  =  background + (1-background)*[1-EXP(
                  -betal*doseAl)]

   The parameter  betas  are restricted to be  positive
   Dependent variable = incidence
   Independent  variable = dose
                                        D-34
                                                DRAFT - DO NOT CITE OR QUOTE

-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
 Total number of observations = 3
 Total number of records with missing values = 0
 Total number of parameters in model = 2
 Total number of specified parameters = 0
 Degree of polynomial = 1
 Maximum number of iterations = 250
 Relative Function Convergence has been set to: 2.22045e-016
 Parameter Convergence has been set to: 1.49012e-008

****  We are sorry but Relative Function and Parameter Convergence     ****
****  are currently unavailable in this model.  Please keep checking   ****
****  the web sight for model updates which will eventually            ****
****  incorporate these convergence criterion.  Default values used.   ****
                  Default Initial Parameter Values
                     Background =            0
                        Beta(l) =    0.0283321
           Asymptotic Correlation Matrix of Parameter Estimates

            ( *** The model parameter(s)  -Background
                 have been estimated at a boundary point, or have been
specified by the user,
                 and do not appear in the correlation matrix )
                Beta(l)
   Beta(l)
                                 Parameter Estimates
Confidence Interval
       Variable         Estimate        Std. Err.
Upper Conf. Limit
     Background                0            *
*
        Beta(l)        0.0219722            *
* - Indicates that this value is not calculated.
                                       95.0% Wald

                                    Lower Conf. Limit
       Model
     Full model
   Fitted model
0.6233
  Reduced model

           AIC:
      Analysis of Deviance Table

Log(likelihood)   # Param's  Deviance  Test d.f.
     -17.2634         3
     -17.7362         1      0.945584      2
     -39.5006

      37.4725
1
44.4744
2
                            P-value
<.0001
                                       D-35
                                               DRAFT - DO NOT CITE OR QUOTE

-------
Dose
0.0000
50.0000
100.0000
Est. Prob.
0.0000
0.6667
0.8889
Expected
0.000
13.333
15.111
Observed
0.000
12.000
16.000
Size
20
20
17
Scaled
Residual
0.000
-0.632
0.686
 1
 2
 3                                      Goodness  of  Fit
 4
 5
 6
 7
 8
 9
10
11     ChiA2 = 0.87      d.f.  = 2        P-value = 0.6471
12
13
14       Benchmark Dose Computation
15
16    Specified effect =           0.67
17
18    Risk Type        =      Extra risk
19
20    Confidence level =           0.95
21
22                 BMD =        50.4574
23
24                BMDL =        35.8134
25
26                BMDU =        72.6771
27
28    Taken together,  (35.8134, 72.6771) is a 90     % two-sided confidence
29    interval for the BMD
30
31    Multistage Cancer Slope Factor =     0.0187081
32
33
34
35
36
                                       D-36         DRAFT - DO NOT CITE OR QUOTE

-------
                              Multistage Cancer Model with 0.95 Confidence Level
 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
      T3
      0)
      •5

      I
      o
      13
      ro
             0.8
        0.6
        0.4
             0.2
                                       Multistage Cancer
                                     Linear extrapolation
                            BMDL
                                 BMD
               0          20          40

   13:4612/282009

NESNOW 1984 DERMAL  BLAC FEMALE.txt
                                                      60
80
100
                                               dose
        Multistage  Cancer Model. (Version:  1.7;   Date:  05/16/2008)
        Input  Data  File:
C:\USEPA\IRIS\PAH\dermal\initiation\Nesnowl984\BIACfemale\msc_NesnowBlaCfemal
e3HDD_MS_4.(d)
        Gnuplot  Plotting  File:
C:\USEPA\IRIS\PAH\dermal\initiation\Nesnowl984\BIACfemale\msc_NesnowBlaCfemal
e3HDD_MS_4.pit
                                           Mon  Dec 28  13:46:08 2009
 BMDS Model Run
   The form of  the  probability function is:

   P[response]  =  background + (1-background)*[1-EXP(
                  -betal*doseAl)]

   The parameter  betas  are restricted to be positive
   Dependent variable  = incidence
   Independent variable = dose

 Total number of  observations = 3
                                        D-37
                                                DRAFT - DO NOT CITE OR QUOTE

-------
 1     Total number of records with missing values = 0
 2     Total number of parameters in model = 2
 3     Total number of specified parameters = 0
 4     Degree of polynomial = 1
 5
 6
 7     Maximum number of iterations = 250
 8     Relative Function Convergence has been set to: 2.22045e-016
 9     Parameter Convergence has been set to: 1.49012e-008
10
11    ****   We are Sorry but Relative Function and Parameter Convergence    ****
12    ****   are currently unavailable in this model.  Please keep checking  ****
13    ****   the wek sight for model updates which will eventually           ****
14    ****   incorporate these convergence criterion.  Default values used.  ****
15
16
17
18                      Default Initial Parameter Values
19                         Background =            0
20                            Beta(l) =    0.0289037
21
22
23               Asymptotic Correlation Matrix of Parameter Estimates
24
25                 Background      Beta(l)
26
27    Background            1        -0.49
28
29       Beta(l)         -0.49            1
30
31
32
33                                     Parameter Estimates
34
35                                                             95.0% Wald
36    Confidence Interval
37           Variable         Estimate        Std. Err.     Lower Conf. Limit
38    Upper Conf.  Limit
39         Background        0.0505105            *                *
40    *
41            Beta(l)         0.0234713            *                *
42    *
43
44    * - Indicates that this value is not calculated.
45
46
47
48                            Analysis of Deviance Table
49
50           Model      Log(likelihood)  # Param's  Deviance  Test d.f.   P-value
51         Full model        -20.7842         3
52       Fitted model        -21.1281         2      0.687832      1
53    0.4069
54      Reduced model        -39.8916         1       38.2148      2         <.0001
55
56               AIC:          46.2563
57
58
59                                      Goodness  of  Fit
60                                                                     Scaled
                                       D-3 8        DRAFT - DO NOT CITE OR QUOTE

-------
0.0000
50.0000
100.0000
0.0505
0.7064
0.9092
0.960
14.127
17.275
1.000
13.000
18.000
19
20
19
0.042
-0.553
0.579
 1         Dose     Est.  Prob.     Expected    Observed     Size       Residual
 2
 3
 4
 5
 6
 7     ChiA2 = 0.64      d.f.  = 1        P-value = 0.4224
 8
 9
10       Benchmark Dose Computation
11
12    Specified effect =           0.51
13
14    Risk Type        =      Extra risk
15
16    Confidence level =           0.95
17
18                 BMD =        30.3924
19
20                BMDL =        21.4681
21
22                BMDU =        44.3165
23
24    Taken together,  (21.4681, 44.3165)  is a 90     % two-sided confidence
25    interval for the BMD
26
27    Multistage Cancer Slope  Factor =     0.0237562
28
29
30
31
                                       D-39        DRAFT - DO NOT CITE OR QUOTE

-------
                              Multistage Cancer Model with 0.95 Confidence Level
 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
      T3
      
-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
 Total number of observations = 4
 Total number of records with missing values = 0
 Total number of parameters in model = 3
 Total number of specified parameters = 0
 Degree of polynomial = 2
 Maximum number of iterations = 250
 Relative Function Convergence has been set to: 2.22045e-016
 Parameter Convergence has been set to: 1.49012e-008

****  We are sorry but Relative Function and Parameter Convergence     ****
****  are currently unavailable in this model.  Please keep checking   ****
****  the web sight for model updates which will eventually            ****
****  incorporate these convergence criterion.  Default values used.   ****
                  Default Initial Parameter Values
                     Background =    0.0934237
                        Beta(l) =   0.00272909
                        Beta(2) =            0
           Asymptotic Correlation Matrix of Parameter Estimates

           (  *** The model parameter(s)  -Beta(2)
                 have been estimated at a boundary point, or have been
specified by the user,
                 and do not appear in the correlation matrix )

             Background      Beta(l)

Background            1         -0.7

   Beta(l)         -0.7            1
Confidence Interval
       Variable
Upper Conf. Limit
     Background
*
        Beta(l)
*
        Beta(2)
  Estimate

 0.0601262

0.00312448

         0
                                 Parameter Estimates
Std.  Err.
                                                         95.0% Wald

                                                      Lower Conf. Limit
  - Indicates that this value is not calculated.



                        Analysis of Deviance Table

       Model      Log(likelihood)  # Param's  Deviance  Test d.f.    P-value
                                       D-41
                                               DRAFT - DO NOT CITE OR QUOTE

-------
1
2
3
4
5
6
7
9
10
11
12
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
Full model -39.5733 4
Fitted model -39.7914 2 0.436272
0.804
Reduced model -46.0668 1 12.987
0.004665

AIC: 83.5828

Goodness of Fit

Dose Est. Prob. Expected Observed Size
0.0000 0.0601 1.142 1.000 19
50.0000 0.1961 3.921 4.000 20
100.0000 0.3123 5.934 7.000 19
250.0000 0.5696 10.823 10.000 19

ChiA2 = 0.44 d.f. = 2 P-value = 0.8007


Benchmark Dose Computation

Specified effect = 0.51

Risk Type = Extra risk

Confidence level = 0.95

BMD = 228.31

BMDL = 149. 811

BMDU = 436.477

Taken together, (149.811, 436.477) is a 90 % two-sided
interval for the BMD

Multistage Cancer Slope Factor = 0.00340429




2

3





Scaled
Residual
-0.137
0.044
0.527
-0.381


















confidence






43
                                  D-42        DRAFT - DO NOT CITE OR QUOTE

-------
     D.2. INTRAPERITONEAL BIOASSAYS
      T3
      
-------
 1
 2
 3       Dependent variable = incidence
 4       Independent variable = dose
 5
 6     Total number of observations = 3
 7     Total number of records with missing values = 0
 8     Total number of parameters in model = 2
 9     Total number of specified parameters = 0
10     Degree of polynomial = 1
11
12
13     Maximum number of iterations = 250
14     Relative Function Convergence has been set to: 2.22045e-016
15     Parameter Convergence has been set to: 1.49012e-008
16
YJ    ****   ^6 are sorry but Relative Function and Parameter Convergence    ****
18    ****   are currently unavailable in this model.  Please keep checking  ****
19    ****   the web sight for model updates which will eventually           ****
20    ****   incorporate these convergence criterion.  Default values used.  ****
21
22
23
24                      Default Initial Parameter Values
25                         Background =    0.0929049
26                            Beta(l) =     0.108473
27
28
29               Asymptotic Correlation Matrix of Parameter Estimates
30
31                 Background      Beta(l)
32
33    Background            1        -0.48
34
35       Beta(l)         -0.48            1
36
37
38
39                                     Parameter Estimates
40
41                                                             95.0% Wald
42    Confidence Interval
43           Variable         Estimate        Std. Err.     Lower Conf. Limit
44    Upper Conf.  Limit
45         Background         0.112498            *                *
46    *
47            Beta(l)          0.103015            *                *
48    *
49
50    * - Indicates that this value is not calculated.
51
52
53
54                            Analysis of Deviance Table
55
56           Model      Log(likelihood)  # Param's  Deviance  Test d.f.   P-value
57         Full model        -44.1118         3
58       Fitted model        -44.1689         2      0.114322      1
59    0.7353
60      Reduced model        -64.1094         1       39.9952      2         <.0001
                                       D-44        DRAFT - DO NOT CITE OR QUOTE

-------
 1
 2               AIC:          92.3379
 3
 4
 5                                      Goodness  of  Fit
 6                                                                     Scaled
 7         Dose     Est.  Prob.     Expected    Observed     Size       Residual
 8      	1	
 9        0.0000     0.1125         3.825     4.000          34        0.095
10        3.4600     0.3786        11.737    11.000          31       -0.273
11       17.3000     0.8507        24.669    25.000          29        0.172
12
13     ChiA2 = 0.11      d.f.  = 1        P-value = 0.7366
14
15
16       Benchmark Dose Computation
17
18    Specified effect =            0.83
19
20    Risk Type        =       Extra risk
21
22    Confidence level =            0.95
23
24                 BMD =          17.201
25
26                BMDL =         12.2186
27
28                BMDU =         25.6067
29
30    Taken together,  (12.2186, 25.6067)  is a 90     % two-sided confidence
31    interval for the BMD
32
33    Multistage Cancer Slope  Factor =      0.067929
34
35
36
37
                                       D-45        DRAFT - DO NOT CITE OR QUOTE

-------
                              Multistage Cancer Model with 0.95 Confidence Level
 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
      T3
      
-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
 Total number of records with missing values = 0
 Total number of parameters in model = 2
 Total number of specified parameters = 0
 Degree of polynomial = 1
 Maximum number of iterations = 250
 Relative Function Convergence has been set to: 2.22045e-016
 Parameter Convergence has been set to: 1.49012e-008

****  We are sorry but Relative Function and Parameter Convergence     ****
****  are currently unavailable in this model.  Please keep checking   ****
****  the web sight for model updates which will eventually            ****
****  incorporate these convergence criterion.  Default values used.   ****
                  Default Initial Parameter Values
                     Background =            0
                        Beta(l) = 6.19323e+018
           Asymptotic Correlation Matrix of Parameter Estimates

             Background      Beta(l)

Background            1        -0.47

   Beta(l)        -0.47            1
                                 Parameter Estimates
                                        Std. Err.
Confidence Interval
       Variable         Estimate
Upper Conf. Limit
     Background         0.168707
*
        Beta(l)         0.259821
* - Indicates that this value is not calculated.
                 95.0% Wald

              Lower Conf. Limit
       Model
     Full model
   Fitted model
0.5464
  Reduced model

           AIC:
                        Analysis of Deviance Table

                  Log(likelihood)  # Param's  Deviance  Test d.f.
                       -31.5803         3
                       -31.7622         2       0.363803      1
                       -51.0494

                        67.5244
1
38.9382
                            P-value
<.0001
                                  Goodness  of  Fit
                                                                  Scaled
                                       D-47
                                               DRAFT - DO NOT CITE OR QUOTE

-------
0.0000
3.4600
17.3000
0.1687
0.6617
0.9907
4.893
18.527
16.842
5.000
18.000
17.000
29
28
17
0.053
-0.210
0.399
 1         Dose     Est.  Prob.     Expected    Observed     Size       Residual
 2
 3
 4
 5
 6
 7     ChiA2 = 0.21      d.f.  = 1        P-value = 0.6496
 8
 9
10       Benchmark Dose Computation
11
12    Specified effect =           0.81
13
14    Risk Type        =      Extra risk
15
16    Confidence level =           0.95
17
18                 BMD =        6.39184
19
20                BMDL =        4.18834
21
22                BMDU =        10.3811
23
24    Taken together,  (4.18834, 10.3811)  is a 90     % two-sided confidence
25    interval for the BMD
26
27    Multistage Cancer Slope  Factor =      0.193394
28
29
30
                                       D-48        DRAFT - DO NOT CITE OR QUOTE

-------
                              Multistage Cancer Model with 0.95 Confidence Level
 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
             0.6
             0.5
      -o      0.4
      t>
      C
      o
      '
         0.3
             0.2
             0.1
                                  Multistage Cancer
                                 Linear extrapolation
                                                    BMDL
                                                                         BMJ)
                            500
                               1000
       07:4712/282009
1500      2000
   dose
2500
3000
3500
WISLOCKI CHRYSENE MALE LIVER.OUT.txt
        Multistage Cancer Model.  (Version:  1.7;   Date: 05/16/2008)
        Input  Data File:
C:\USEPA\IRIS\PAH\IP\Wislockil986\CH\msc_WislockiCHliver_MS_l_44.(d)
        Gnuplot  Plotting File:
C:\USEPA\IRIS\PAH\IP\Wislockil986\CH\msc_WislockiCHliver_MS_l_44.plt
                                            Wed Dec 23 11:10:41  2009
 HMDS Model  Run
   The form  of  the probability function  is:

   P[response]  = background +  (1-background)*[1-EXP(
                  -betal*doseAl)]

   The parameter betas are restricted  to be  positive
   Dependent  variable = incidence
   Independent  variable = dose
                                        D-49
                                                DRAFT - DO NOT CITE OR QUOTE

-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
 Total number of observations = 3
 Total number of records with missing values = 0
 Total number of parameters in model = 2
 Total number of specified parameters = 0
 Degree of polynomial = 1
 Maximum number of iterations = 250
 Relative Function Convergence has been set to: 2.22045e-016
 Parameter Convergence has been set to: 1.49012e-008

****  We are sorry but Relative Function and Parameter Convergence     ****
****  are currently unavailable in this model.  Please keep checking   ****
****  the web sight for model updates which will eventually            ****
****  incorporate these convergence criterion.  Default values used.   ****
                  Default Initial Parameter Values
                     Background =     0.147839
                        Beta(l) =  0.000139419
           Asymptotic Correlation Matrix of Parameter Estimates

             Background      Beta(l)

Background            1        -0.57

   Beta(l)        -0.57             1
                                 Parameter Estimates
Confidence Interval
       Variable
Upper Conf. Limit
     Background
*
        Beta (1)
      Estimate

      0.109703

    0.00017367
Std.  Err.
         95.0% Wald

      Lower Conf. Limit
* - Indicates that this value is not calculated.
       Model
     Full model
   Fitted model
0.229
  Reduced model
0.0005661

           AIC:
      Analysis of Deviance Table

Log(likelihood)   # Param's  Deviance  Test d.f.
     -67.0392         3
     -67.7628         2       1.44719      1
                            P-value
      -74.516
      139.526
1
14.9536
2
                                       D-50
                                               DRAFT - DO NOT CITE OR QUOTE

-------
1
2
3
A
4-
5
6
7
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
Goodness of Fit
Dose Est. Prob. Expected Observed Size
0.0000 0.1097 8.008 7.000 73
700.0000 0.2116 7.407 10.000 35
2800.0000 0.4525 15.387 14.000 34
ChiA2 = 1.52 d.f. = 1 P-value = 0.2172


Benchmark Dose Computation

Specified effect = 0.44

Risk Type = Extra risk

Confidence level = 0.95

BMD = 3338.63

BMDL = 2098.51

BMDU = 6591.77

Taken together, (2098.51, 6591.77) is a 90 % two-sided
interval for the BMD

Multistage Cancer Slope Factor = 0.000209673




Scaled
Residual
-0.378
1.073
-0.478

















confidence







D-51        DRAFT - DO NOT CITE OR QUOTE

-------
 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
                              Multistage Cancer Model with 0.95 Confidence Level
      O

      I
      O
      13
      ro
          1


        0.9


        0.8


        0.7


        0.6


        0.5


        0.4


        0.3


        0.2
                                       Multistage Cancer
                                      Linear extrapolation
               0        0.5         1         1.5        2
                                           dose
   07:5812/282009

Nesnow et al.  1998b  i.p DBalP male lung  High dose dropped
        Multistage Cancer Model.  (Version:  1.7;   Date: 05/16/2008)
        Input  Data File:
C:\USEPA\IRIS\PAH\IP\Nesnowl998b\DBalP\msc_NesnowDBalPHDD_MS_2_10.(d)
        Gnuplot  Plotting File:
C:\USEPA\IRIS\PAH\IP\Nesnowl998b\DBalP\msc_NesnowDBalPHDD_MS_2_10.plt
                                           Wed Dec 23 14:50:54 2009
 BMDS Model  Run
   The form  of  the probability function  is:

   P[response]  =  background +  (1-background)*[1-EXP(
                  -betal*doseAl-beta2*doseA2)]

   The parameter  betas are restricted  to be  positive
   Dependent  variable = incidence
                                        D-52
                                                DRAFT - DO NOT CITE OR QUOTE

-------
 1       Independent variable = dose
 2
 3     Total number of observations = 4
 4     Total number of records with missing values = 0
 5     Total number of parameters in model = 3
 6     Total number of specified parameters = 0
 7     Degree of polynomial = 2
 8
 9
10     Maximum number of iterations = 250
11     Relative Function Convergence has been set to: 2.22045e-016
12     Parameter Convergence has been set to: 1.49012e-008
13
14    ****   ^6 are sorry but Relative Function and Parameter Convergence    ****
15    ****   are currently unavailable in this model.  Please keep checking  ****
16    ****   the web sight for model updates which will eventually           ****
Y1    ****   incorporate these convergence criterion.  Default values used.  ****
18
19
20
21                      Default Initial Parameter Values
22                         Background =            0
23                            Beta(l) =            0
24                            Beta(2) = 1.14332e+019
25
26
27               Asymptotic Correlation Matrix of Parameter Estimates
28
29               ( *** The model parameter(s)  -Beta(l)
30                     have been estimated at a boundary point, or have been
31    specified by the user,
32                     and do not appear in the correlation matrix )
33
34                 Background      Beta(2)
35
36    Background            1        -0.27
37
38       Beta(2)         -0.27            1
39
40
41
42                                     Parameter Estimates
43
44                                                             95.0% Wald
45    Confidence Interval
46           Variable         Estimate        Std. Err.      Lower Conf. Limit
47    Upper Conf.  Limit
48         Background         0.419864            *                *
49    *
50            Beta(l)                 0            *                *
51    *
52            Beta(2)           1.23372            *                *
53    *
54
55    * - Indicates that this value is not calculated.
56
57
58
59                            Analysis of Deviance Table
60
                                       D-53        DRAFT - DO NOT CITE OR QUOTE

-------
1
2
3
4
5
6
7
9
10
11
12
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
Model
Full model
Fitted model
0.3993
Reduced model

AIC:



Dose Est
0.0000 0.
0.3000 0.
1.5000 0.
3.0000 1.

ChiA2 = 1.83


Benchmark Dose

Specified effect

Risk Type

Confidence level

BMD

BMDL

BMDU

Taken together, (
interval for the

Multistage Cancer



Log (likelihood) # Param's Deviance Test d.f. P-value
-47.4317 4
-48.3498 2 1.83615 2

-77.3457 1 59.8281 3 <.0001

100.7

Goodness of Fit
Scaled
. Prob. Expected Observed Size Residual
4199 12.596 15.000 30 0.889
4808 15.867 13.000 33 -0.999
9639 32.771 33.000 34 0.210
0000 35.000 35.000 35 0.017

d.f. = 2 P-value = 0.3998


Computation

0.1

= Extra risk

0.95

0.292233

0.125394

0.383954

0.125394, 0.383954) is a 90 % two-sided confidence
BMD

Slope Factor = 0.797488



D-54
DRAFT - DO NOT CITE OR QUOTE

-------
                              Multistage Cancer Model with 0.95 Confidence Level
 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
      T3
      0)
      •5

      I
      o
      13
      ro
             0.8
        0.6
             0.4
             0.2
                                       Multistage Cancer
                                      Linear extrapolation
                BVIDL   BMD
                                  50
                                           100
                                          dose
150
200
       08:01 12/282009
Nesnow et al.  1998b   i.p BaP male lung
        Multistage  Cancer Model.  (Version:  1.7;   Date:  05/16/2008)
        Input  Data  File:
C:\USEPA\IRIS\PAH\IP\Nesnowl998b\BaP\msc_NesnowBaP_MS_4_10.(d)
        Gnuplot  Plotting File:
C:\USEPA\IRIS\PAH\IP\Nesnowl998b\BaP\msc_NesnowBaP_MS_4_10.plt
                                           Wed  Dec 23 14:46:42 2009
 BMDS Model Run
   The form of  the  probability function is:

   P[response]  =  background + (1-background)*[1-EXP(
                  -betal*doseAl-beta2*doseA2-beta3*doseA3-beta4*doseA4)

   The parameter  betas  are restricted to be  positive
   Dependent variable = incidence
   Independent  variable = dose
                                        D-55
                                                DRAFT - DO NOT CITE OR QUOTE

-------
 1     Total number of observations = 6
 2     Total number of records with missing values = 0
 3     Total number of parameters in model = 5
 4     Total number of specified parameters = 0
 5     Degree of polynomial = 4
 6
 7
 8     Maximum number of iterations = 250
 9     Relative Function Convergence has been set to: 2.22045e-016
10     Parameter Convergence has been set to: 1.49012e-008
11
12    ****  we are sorry but Relative Function and Parameter Convergence    ****
13    ****  are currently unavailable in this model.  Please keep checking  ****
14    ****  the web sight for model updates which will eventually           ****
15    ****  incorporate these convergence criterion.  Default values used.  ****
16
17
18
19                      Default Initial Parameter Values
20                         Background =            1
21                            Beta(l) =  5.5061e+017
22                            Beta (2) =            0
23                            Beta (3) =            0
24                            Beta (4) =            0
25
26
27               Asymptotic Correlation Matrix of Parameter Estimates
28
29               ( *** The model parameter(s)  -Beta(l)     -Beta (2)
30                     have been estimated at a boundary point, or have been
31    specified by the user,
32                     and do not appear in the correlation matrix )
33
34                 Background      Beta (3)      Beta (4)
35
36    Background            1        -0.67         0.64
37
38       Beta(3)         -0.67            1           -1
39
40       Beta(4)          0.64           -1            1
41
42
43
44                                     Parameter Estimates
45
46                                                             95.0% Wald
47    Confidence Interval
48           Variable         Estimate        Std. Err.      Lower Conf. Limit
49    Upper Conf.  Limit
50         Background          0.29287            *                *
51    *
52            Beta(l)                 0            *                *
53    *
54            Beta(2)                 0            *                *
55    *
56            Beta(3)       0.000178164            *                *
57    *
58            Beta(4)      3.09556e-007            *                *
59    *
60
                                       D-56         DRAFT - DO NOT CITE OR QUOTE

-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
1 Q
1 y
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
* - Indicates tha





Model
Full model
Fitted model
0.9997
Reduced model

AIC:




Dose Est

0.0000 0.
5.0000 0.
10.0000 0.
50.0000 1.
100.0000 1.
200.0000 1.

ChiA2 = 0.01


Benchmark Dose

Specified effect

Risk Type

Confidence level

BMD

BMDL

BMDU

Taken together, (
interval for the

Multistage Cancer



t this value is not calculated.



Analysis of Deviance Table

Log (likelihood) # Param's Deviance Test d.f. P-value
-35.952 6
-35.958 3 0.0120148 3

-73.3649 1 74.8258 5 <.0001

77.916


Goodness of Fit
Scaled
. Prob. Expected Observed Size Residual

2929 5.857 6.000 20 0.070
3086 6.172 6.000 20 -0.083
4101 6.972 7.000 17 0.014
0000 19.000 19.000 19 0.000
0000 16.000 16.000 16 0.000
0000 24.000 24.000 24 0.000

d.f. = 3 P-value = 0.9997


Computation

0.1

= Extra risk

0.95

8.35346

2.00564

22.6111

2.00564, 22.6111) is a 90 % two-sided confidence
BMD

Slope Factor = 0.0498594



D-57
DRAFT - DO NOT CITE OR QUOTE

-------
                              Multistage Cancer Model with 0.95 Confidence Level
 2
 o
 6
 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
      O

      I
      O
      13
      (0
             0.8
        0.6
             0.4
             0.2
                                       Multistage Cancer
                                     Linear extrapolation
                EMDL BMD
                                  50
   08:0412/282009
Nesnow et al.  1998b   i.p BbF male lung
                                           100
                                          dose
150
200
        Multistage  Cancer Model.  (Version:  1.7;   Date:  05/16/2008)
        Input  Data  File:
C:\USEPA\IRIS\PAH\IP\Nesnowl998b\BbF\msc_NesnowBbF_MS_3.(d)
        Gnuplot  Plotting  File:
C:\USEPA\IRIS\PAH\IP\Nesnowl998b\BbF\msc_NesnowBbF_MS_3.plt
                                           Wed  Dec 23 14:46:42 2009
 BMDS Model Run
   The form of  the  probability function is:

   P[response]  =  background + (1-background)*[1-EXP(
                  -betal*doseAl-beta2*doseA2-beta3*doseA3)

   The parameter  betas  are restricted to be  positive
   Dependent variable = incidence
   Independent  variable = dose

 Total number of  observations = 5
                                        D-58
                                                DRAFT - DO NOT CITE OR QUOTE

-------
 1     Total number of records with missing values = 0
 2     Total number of parameters in model = 4
 3     Total number of specified parameters = 0
 4     Degree of polynomial = 3
 5
 6
 7     Maximum number of iterations = 250
 8     Relative Function Convergence has been set to: 2.22045e-016
 9     Parameter Convergence has been set to: 1.49012e-008
10
11    ****  We are Sorry but Relative Function and Parameter Convergence    ****
12    ****  are currently unavailable in this model.  Please keep checking  ****
13    ****  the wek sight for model updates which will eventually           ****
14    ****  incorporate these convergence criterion.  Default values used.  ****
15
16
17
18                      Default Initial Parameter Values
19                         Background =            0
20                            Beta(l) = 5.84708e+017
21                            Beta(2) =            0
22                            Beta(3) =            0
23
24
25               Asymptotic Correlation Matrix of Parameter Estimates
26
27               ( *** The model parameter(s)  -Beta (2)
28                     have been estimated at a boundary point, or have been
29    specified by the user,
30                     and do not appear in the correlation matrix )
31
32                 Background      Beta(l)      Beta (3)
33
34    Background            1        -0.56         0.31
35
36       Beta(l)         -0.56            1         -0.8
37
38       Beta(3)          0.31         -0.8            1
39
40
41
42                                     Parameter Estimates
43
44                                                             95.0% Wald
45    Confidence Interval
46           Variable         Estimate        Std. Err.      Lower Conf. Limit
47    Upper Conf.  Limit
48         Background         0.328834            *                *
49    *
50            Beta(l)         0.0184355            *                *
51    *
52            Beta(2)                 0            *                *
53    *
54            Beta(3)      3.37339e-006            *                *
55    *
56
57    * - Indicates that this value is not calculated.
58
59
60
                                       D-59        DRAFT - DO NOT CITE OR QUOTE

-------
1
2
3
4
5
6
7
9
10
11
12
13
14
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44


Model
Full model
Fitted model
0.7654
Reduced model
AIC:




Dose Est
0.0000 0.
10.0000 0.
50.0000 0.
100.0000 0.
200.0000 1.

ChiA2 = 0.47


Benchmark Dose

Specified effect

Risk Type

Confidence level

BMD

BMDL

BMDU

Taken together, (
interval for the

Multistage Cancer


Analysis of Deviance Table

Log (likelihood) # Param's Deviance Test d.f. P-value
-34.702 5
-34.9693 3 0.53462 2

-57.3647 1 45.3254 4 <.0001
75.9386


Goodness of Fit
Scaled
. Prob. Expected Observed Size Residual
3288 6.577 6.000 20 -0.274
4437 7.987 9.000 18 0.481
8249 16.497 16.000 20 -0.293
9964 19.927 20.000 20 0.270
0000 19.000 19.000 19 0.000

d.f. = 2 P-value = 0.7925


Computation

0.1

= Extra risk

0.95

5.68153

2.40867

28.009

2.40867, 28.009 ) is a 90 % two-sided confidence
BMD

Slope Factor = 0.0415166


D-60
DRAFT - DO NOT CITE OR QUOTE

-------
                              Multistage Cancer Model with 0.95 Confidence Level
 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
      T3
      0)
      •5

      I
      o
      13
      ro
             0.8
        0.6
             0.4
             0.2
                                       Multistage Cancer
                                      Linear extrapolation
                BVIDL  BMD
                                  50
                                           100
                                          dose
150
200
       08:0512/282009
Nesnow et al.  1998b   i.p CPcdP male lung
        Multistage  Cancer Model.  (Version:  1.7;   Date:  05/16/2008)
        Input  Data  File:
C:\USEPA\IRIS\PAH\IP\Nesnowl998b\CPcdP\msc_NesnowCPcdP_MS_3.(d)
        Gnuplot  Plotting File:
C:\USEPA\IRIS\PAH\IP\Nesnowl998b\CPcdP\msc_NesnowCPcdP_MS_3.plt
                                           Wed  Dec 23 14:46:43 2009
 HMDS Model Run
   The form of  the  probability function is:

   P[response]  =  background + (1-background)*[1-EXP(
                  -betal*doseAl-beta2*doseA2-beta3*doseA3)

   The parameter  betas  are restricted to be  positive
   Dependent variable = incidence
   Independent  variable = dose
                                        D-61
                                                DRAFT - DO NOT CITE OR QUOTE

-------
 1     Total number of observations = 5
 2     Total number of records with missing values = 0
 3     Total number of parameters in model = 4
 4     Total number of specified parameters = 0
 5     Degree of polynomial = 3
 6
 7
 8     Maximum number of iterations = 250
 9     Relative Function Convergence has been set to: 2.22045e-016
10     Parameter Convergence has been set to: 1.49012e-008
11
12    ****  we are sorry but Relative Function and Parameter Convergence    ****
13    ****  are currently unavailable in this model.  Please keep checking  ****
14    ****  the web sight for model updates which will eventually           ****
15    ****  incorporate these convergence criterion.  Default values used.  ****
16
17
18
19                      Default Initial Parameter Values
20                         Background =            1
21                            Beta(l) = 5.02249e+017
22                            Beta (2) =            0
23                            Beta (3) =            0
24
25
26               Asymptotic Correlation Matrix of Parameter Estimates
27
28               ( *** The model parameter(s)  -Beta(l)
29                     have been estimated at a boundary point, or have been
30    specified by the user,
31                     and do not appear in the correlation matrix )
32
33                 Background      Beta(2)      Beta(3)
34
35    Background            1        -0.13        0.025
36
37       Beta(2)         -0.13            1        -0.99
38
39       Beta(3)         0.025        -0.99            1
40
41
42
43                                     Parameter Estimates
44
45                                                             95.0% Wald
46    Confidence Interval
47           Variable         Estimate        Std. Err.      Lower Conf. Limit
48    Upper Conf.  Limit
49         Background         0.299994            *                *
50    *
51            Beta(l)                 0            *                *
52    *
53            Beta(2)       0.000554719            *                *
54    *
55            Beta(3)      9.86997e-005            *                *
56    *
57
58    * - Indicates that this value is not calculated.
59
60
                                       D-62        DRAFT - DO NOT CITE OR QUOTE

-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
1 A
ID
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46

Model
Full model
Fitted model
1
Reduced model

AIC:




Dose Est


0.0000 0.
10.0000 0.
50.0000 1.
100.0000 1.
200.0000 1.

ChiA2 = 0.00


Benchmark Dose

Specified effect

Risk Type

Confidence level

BMD

BMDL

BMDU

Taken together, (
interval for the

Multistage Cancer



Analysis of Deviance Table
Log (likelihood) # Param's Deviance Test d.f. P-value
-25.6775 5
-25.6775 3 3.06836e-005 2

-56.6963 1 62.0376 4 <.0001

57.3551


Goodness of Fit
Scaled
. Prob. Expected Observed Size Residual


3000 6.000 6.000 20 0.000
4000 8.000 8.000 20 -0.000
0000 20.000 20.000 20 0.004
0000 19.000 19.000 19 0.000
0000 19.000 19.000 19 0.000

d.f. = 2 P-value = 1.0000


Computation

0.1

= Extra risk

0.95

8.64922

1.95607

17.5713

1.95607, 17.5713) is a 90 % two-sided confidence
BMD

Slope Factor = 0.0511229



D-63
DRAFT - DO NOT CITE OR QUOTE

-------
                              Multistage Cancer Model with 0.95 Confidence Level
 1
 2
 o
 6
 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
      T3
      0)
      •5

      I
      o
      13
      ro
             0.8
        0.6
             0.4
             0.2
                                       Multistage Cancer
                                     Linear extrapolation
                BVIDL BMD
               024

   08:0512/282009

Nesnow et al.  1998b   i.p DBahA male lung
                                                                             10
                                               dose
        Multistage  Cancer Model.  (Version:  1.7;   Date:  05/16/2008)
        Input  Data  File:
C:\USEPA\IRIS\PAH\IP\Nesnowl998b\DBahA\msc_NesnowDBahA_MS_3.(d)
        Gnuplot  Plotting  File:
C:\USEPA\IRIS\PAH\IP\Nesnowl998b\DBahA\msc_NesnowDBahA_MS_3.plt
                                           Wed  Dec 23  14:46:43 2009
 BMDS Model Run
   The form of  the  probability function is:

   P[response]  =  background + (1-background)*[1-EXP(
                  -betal*doseAl-beta2*doseA2-beta3*doseA3)

   The parameter  betas  are restricted to be  positive
   Dependent variable = incidence
   Independent  variable = dose

 Total number of  observations = 5
                                        D-64
                                                DRAFT - DO NOT CITE OR QUOTE

-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Total number of records with missing values = 0
Total number of parameters in model = 4
Total number of specified parameters =
Degree of polynomial = 3


Maximum number of iterations = 250
Relative Function Convergence has been
Parameter Convergence has been set to:

**** We are sorry but Relative Function
**** are currently unavailable in this

0




set to: 2.22045e-016
1.49012e-008

and Parameter Convergence ****
model. Please keep checking ****
**** the web sight for model updates which will eventually ****
**** incorporate these convergence criterion. Default values used. ****






Default Initial Parameter Values
Background =
Beta(l) = 1.
Beta(2) =
Beta(3) =


Asymptotic Correlation Matrix

( *** The model parameter (s)
have been estimated at
specified by the user,
0
2e+019
0
0


of Parameter Estimates

-Beta (2)
a boundary point, or have been

and do not appear in the correlation matrix )

Background Beta(l)

Background 1 -0.48

Beta(l) -0.48 1

Beta(3) 0.2 -0.81




Beta (3)

0.2

-0.81

1



Parameter Estimates


Confidence Interval
Variable Estimate
Upper Conf. Limit
Background 0.300001
*
Beta(l) 0.446326
*
Beta(2) 0
*
Beta(3) 0.0942115
*


95. 0% Wald

Std. Err. Lower Conf. Limit

* *

* *

* *

* *


* - Indicates that this value is not calculated.






D-65        DRAFT - DO NOT CITE OR QUOTE

-------
1
2
3
4
5
6
7
9
10
11
12
13
14
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45

Model
Full model
Fitted model
1
Reduced model
AIC:




Dose Est
0.0000 0.
1.2500 0.
2.5000 0.
5.0000 1.
10.0000 1.

ChiA2 = 0.00


Benchmark Dose

Specified effect

Risk Type

Confidence level

BMD

BMDL

BMDU

Taken together, (
interval for the

Multistage Cancer



Analysis of Deviance Table
Log (likelihood) # Param's Deviance Test d.f. P-value
-27.5922 5
-27.5922 3 2.31121e-005 2

-50.4308 1 45.6773 4 <.0001
61.1844


Goodness of Fit
Scaled
. Prob. Expected Observed Size Residual
3000 6.000 6.000 20 -0.000
6667 12.000 12.000 18 0.000
9474 18.000 18.000 19 -0.000
0000 20.000 20.000 20 0.003
0000 19.000 19.000 19 0.000

d.f. = 2 P-value = 1.0000


Computation

0.1

= Extra risk

0.95

0.233378

0.0933198

0.955315

0.0933198, 0.955315) is a 90 % two-sided confidence
BMD

Slope Factor = 1.07158



D-66
DRAFT - DO NOT CITE OR QUOTE

-------
 1
 2
 3
 4
 5
 6
 7
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
Busby 1984 i.p. multiplicity
FA male
Linear
Nonconstant variance
BMR = lowest statistically  significant response in BaP treated  animals  (after
control subtracted)


                           Linear Model with 0.95 Confidence Level
      c
      o
      Q.
      o:
      c
      (0
      OJ
                     Linear
                                                          BMDL
                                                                        BME)
                                                                            10
                                             dose
       08:1212/282009
        Polynomial Model.  (Version:  2.13;  Date: 04/08/2008)
        Input Data File:
C:\IPmult\Busbyl984\FAmale\lin_BusbyFAM_linear_4_28.(d)
        Gnuplot Plotting  File:
C:\IPmult\Busbyl984\FAmale\lin_BusbyFAM_linear_4_28.plt
                                           Wed Dec 23 15:26:52 2009
 BMDS Model Run


   The form of the  response  function is:

   Y[dose] = beta_0  + beta_l*dose + beta_2*doseA2


   Dependent variable = mean
                                       D-67
                                               DRAFT - DO NOT CITE OR QUOTE

-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
56
57
58
59
60
Independent variable = dose
The polynomial coefficients are restricted to be positive
The variance is to be modeled as Var(i) = exp (lalpha + log (mean (i) ) *

Total number of dose groups = 3
Total number of records with missing values = 0
Maximum number of iterations = 250
Relative Function Convergence has been set to: le-008
Parameter Convergence has been set to: le-008



Default Initial Parameter Values
lalpha = 0.136152
rho = 0
beta 0 = 0.0180952
beta 1 = 0.427551


Asymptotic Correlation Matrix of Parameter Estimates

lalpha rho beta 0 beta 1

lalpha 1 0.65 0.015 0.00041

rho 0.65 1 0.22 -0.061

beta 0 0.015 0.22 1 -0.24

beta 1 0.00041 -0.061 -0.24 1



Parameter Estimates

95.0% Wald
Confidence Interval
Variable Estimate Std. Err. Lower Conf. Limit
Upper Conf. Limit
lalpha 0.634298 0.204652 0.233188
1.03541
rho 0.923372 0.0876305 0.751619
1.09512
beta 0 0.0170376 0.0434041 -0.0680328
0.102108
beta 1 0.426604 0.0861283 0.257796
0.595413



Table of Data and Estimated Values of Interest

Dose N Obs Mean Est Mean Obs Std Dev Est Std Dev Seal
Res .
_

0 27 0.04 0.017 0.21 0.21
0.7 31 0.29 0.316 0.84 0.806 -0
3.5 27 1.52 1.51 1.66 1.66 0.


rho)

















































ed



0.57
.177
0308
D-68
DRAFT - DO NOT CITE OR QUOTE

-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60



Model Descriptions for


Model Al: Yij =
Var{e(ij) } =

Model A2: Yij =
Var{e(ij) } =

Model A3: Yij =
Var{e(ij) } =
Model A3 uses any
were specified by

Model R: Yi =
Var{e(i) } =




Model
Al
A2
A3
fitted
R





likelihoods calculated


Mu (i) + e (ij )
SigmaA2

Mu (i) + e (ij )
Sigma (i) A2

Mu (i) + e (ij )
exp(lalpha + rho*ln (Mu (i) ) )
fixed variance parameters that
the user

Mu + e ( i )
S i gma A 2


Likelihoods of Interest

Log (likelihood) # Param's AIC
-46.759351 4 101.518703
-7.114400 6 26.228800
-7.317284 5 24.634569
-7.329046 4 22.658093
-59.984569 2 123.969139


Explanation of Tests

Test 1: Do responses
(A2 vs. R)
Test 2: Are Variances
Test 3: Are variances

and/or variances differ among Dose levels?

Homogeneous? (Al vs A2 )
adequately modeled? (A2 vs. A3)
Test 4: Does the Model for the Mean Fit? (A3 vs. fitted)
(Note: When rho=0 the

results of Test 3 and Test 2 will be the same.)

Tests of Interest


Test -2*log (Likelihood Ratio) Test df p-value

Test 1
Test 2
Test 3 0
Test 4 0.

The p-value for Test 1

105.74 4 <.0001
79.2899 2 <.0001
.405769 1 0.5241
0235238 1 0.8781

is less than .05. There appears to be a
difference between response and/or variances among the dose levels
It seems appropriate to

The p-value for Test 2
model the data

is less than .1. A non-homogeneous variance
model appears to be appropriate

The p-value for Test 3
to be appropriate here

The p-value for Test 4

is greater than .1. The modeled variance appears


is greater than .1. The model chosen seems
D-69
DRAFT - DO NOT CITE OR QUOTE

-------
 1    to  adequately describe  the data
 2
 3
 4                Benchmark  Dose Computation
 5
 6    Specified  effect =          4.28
 7
 8    Risk  Type        =      Point  risk
 9
10    Confidence level =          0.95
11
12                BMD =        9.99278
13
14
15                BMDL =        7.55762
16
17
18
19
                                      D-70         DRAFT - DO NOT CITE OR QUOTE

-------
1
2
3
4
5
6
7
Busby 1984 i.p. multiplicity
FA female
Linear
Nonconstant variance
BMR = lowest statistically  significant  response in BaP treated animals  (after
control subtracted)
                                 Linear Model with 0.95 Confidence Level









9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
3.5
3
2.5
en
1
en
0)
o:
i 1-5
0)
1
0.5
0


-
-
-
~
-
~~/


0

08:1412/282009



Polynomial
Input Data
inr*ir



/
X^
s/
/
-^—
BMDL

X^
,/
^












BMI
5 10 15 20 25 30
dose




Model. (Version: 2.13; Date: 04/08/2008)

File:
C:\IPmult\Busbyl984\FAfemale\lin BusbyFAF linear 3 56. (d)
Gnuplot Plotting File:
C:\IPmult\Busbyl984\FAfemale\lin BusbyFAF linear 3 56. pit


BMDS Model Run

The form of the

Y[dose] = beta 0


Wed Dec 23 15:26:52 2009





response function is:

+ beta l*dose + beta 2*doseA2 + . . .


                                      D-71
                                               DRAFT - DO NOT CITE OR QUOTE

-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
   Dependent variable = mean
   Independent variable = dose
   The polynomial coefficients are restricted to be positive
   The variance is to be modeled as Var(i) = exp(lalpha + log(mean(i))  *  rho)

   Total number of dose groups = 3
   Total number of records with missing values = 0
   Maximum number of iterations = 250
   Relative Function Convergence has been set to: le-008
   Parameter Convergence has been set to: le-008
                  Default Initial Parameter Values
                         lalpha =     -1.11206
                            rho =            0
                         beta_0 =     0.108571
                         beta 1 =     0.115306
           Asymptotic Correlation Matrix of Parameter Estimates

                 lalpha          rho       beta_0       beta_l

    lalpha            1         0.94        0.036       -0.047

       rho         0.94            1         0.04       -0.052

    beta_0        0.036         0.04            1        -0.46

    beta 1       -0.047       -0.052        -0.46             1
Confidence Interval
       Variable         Estimate
Upper Conf. Limit
         lalpha         0.353344
1.29466
            rho           1.1315
1.70558
         beta_0         0.123135
0.24438
         beta_l         0.106469
0.211399
                        Parameter Estimates



                               Std. Err.

                                0.480274

                                0.292904

                               0.0618608

                               0.0535364
                                       95.0% Wald

                                    Lower Conf. Limit

                                          -0.587974

                                           0.557421

                                         0.00189039

                                         0.00153987
 Dose
Res .
     Table of Data and Estimated Values of Interest

            N    Obs Mean     Est Mean   Obs Std Dev  Est Std Dev    Scaled
    0
  0.7
28
20
0.14
0.15
0.123
0.198
0.37
0.49
0.365
0.477
 0.245
-0.447
                                       D-72
                                               DRAFT - DO NOT CITE OR QUOTE

-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3.5 21 0.52



Model Descriptions for


Model Al: Yij =
Var{e(ij) } =

Model A2: Yij =
Var{e(ij) } =

Model A3: Yij =
Var{e(ij) } =
Model A3 uses any
were specified by

Model R: Yi =
Var{e(i) } =




Model
Al
A2
A3
fitted
R


0.496 0.82 0.802 0



likelihoods calculated


Mu (i) + e (ij )
S i gma A 2

Mu (i) + e (ij )
Sigma (i) A2

Mu (i) + e (ij )
exp(lalpha + rho*ln (Mu (i) ) )
fixed variance parameters that
the user

Mu + e ( i )
SigmaA2


Likelihoods of Interest

Log (likelihood) # Param's AIC
5.399546 4 -2.799091
13.307908 6 -14.615816
13.189903 5 -16.379806
13.167852 4 -18.335705
2.264796 2 -0.529591


Explanation of Tests

Test 1: Do responses
(A2 vs. R)
Test 2 : Are Variances
Test 3: Are variances

and/or variances differ among Dose levels?

Homogeneous? (Al vs A2 )
adequately modeled? (A2 vs. A3)
Test 4: Does the Model for the Mean Fit? (A3 vs. fitted)
(Note: When rho=0 the

results of Test 3 and Test 2 will be the same.)

Tests of Interest


Test -2*log (Likelihood Ratio) Test df p-value

Test 1
Test 2
Test 3
Test 4 0.

The p-value for Test 1

22.0862 4 0.0001927
15.8167 2 0.0003677
0.23601 1 0.6271
0441012 1 0.8337

is less than .05. There appears to be a
difference between response and/or variances among the dose levels
It seems appropriate to

The p-value for Test 2
model the data

is less than .1. A non-homogeneous variance
model appears to be appropriate

The p-value for Test 3
to be appropriate here


is greater than .1. The modeled variance appears


                                 0.138
D-73
DRAFT - DO NOT CITE OR QUOTE

-------
 1    The p-value  for  Test  4  is  greater than  .1.  The model chosen seems
 2    to adequately  describe  the data
 3
 4
 5                Benchmark  Dose  Computation
 6
 7    Specified  effect =           3.56
 8
 9    Risk  Type        =      Point risk
10
11    Confidence level =           0.95
12
13                BMD =         32.2804
14
15
16                BMDL =         18.094
17
18
19
                                      D-74         DRAFT - DO NOT CITE OR QUOTE

-------
 1    Nesnow 1998b i.p. multiplicity
 2    BbF
 3    Drop 2 high doses
 4    Linear
 5    Nonconstant variance
 6    BMR = lowest statistically  significant response in BaP treated animals (after
 7    control subtracted)
                                  Linear Model with 0.95 Confidence Level
10
11
12
13
14
15
16
17
18
19
20

22
23
24
25
26
27
28
29
30
             3.5
             2.5
      c
      o
      Q.
      
-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
   Dependent variable = mean
   Independent variable = dose
   The polynomial coefficients are restricted to be positive
   The variance is to be modeled as Var(i) = exp(lalpha + log(mean(i))  *  rho)

   Total number of dose groups = 3
   Total number of records with missing values = 0
   Maximum number of iterations = 250
   Relative Function Convergence has been set to: le-008
   Parameter Convergence has been set to: le-008
                  Default Initial Parameter Values
                         lalpha =     0.403617
                            rho =            0
                         beta_0 =     0.456667
                         beta 1 =       0.0305
           Asymptotic Correlation Matrix of Parameter Estimates

                 lalpha          rho       beta_0       beta_l

    lalpha            1         0.15        0.059        -0.07

       rho         0.15            1       -0.059        0.006

    beta_0        0.059       -0.059            1        -0.49

    beta 1        -0.07        0.006        -0.49             1
Confidence Interval
       Variable         Estimate
Upper Conf. Limit
         lalpha         0.123284
0.492576
            rho          1.49465
2.12253
         beta_0         0.511616
0.771396
         beta_l        0.0272932
0.0435087
                        Parameter Estimates



                               Std. Err.

                                0.188418

                                0.320356

                                0.132543

                              0.00827339
                                       95.0% Wald

                                    Lower Conf. Limit

                                          -0.246009

                                           0.866761

                                           0.251836

                                          0.0110776
 Dose
Res .
     Table of Data and Estimated Values of Interest

            N    Obs Mean     Est Mean   Obs Std Dev  Est Std Dev    Scaled
    0
   10
20
18
0.53
0.67
0.512
0.785
0.72
0.75
0.645
0.887
 0.128
-0.548
                                       D-76
                                               DRAFT - DO NOT CITE OR QUOTE

-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
50 20 2



Model Descriptions for


Model Al: Yij =
Var{e(ij) } =

Model A2: Yij =
Var{e(ij) } =

Model A3: Yij =
Var{e(ij) } =
Model A3 uses any
were specified by

Model R: Yi =
Var{e(i) } =




Model
Al
A2
A3
fitted
R


1.88 1.82 1.7 0



likelihoods calculated


Mu (i) + e (ij )
S i gma A 2

Mu (i) + e (ij )
Sigma (i) A2

Mu (i) + e (ij )
exp(lalpha + rho*ln (Mu (i) ) )
fixed variance parameters that
the user

Mu + e ( i )
SigmaA2


Likelihoods of Interest

Log (likelihood) # Param's AIC
-39.164718 4 86.329436
-27.688080 6 67.376160
-27.755992 5 65.511983
-28.699972 4 65.399945
-47.123187 2 98.246375


Explanation of Tests

Test 1: Do responses
(A2 vs. R)
Test 2 : Are Variances
Test 3: Are variances

and/or variances differ among Dose levels?

Homogeneous? (Al vs A2 )
adequately modeled? (A2 vs. A3)
Test 4: Does the Model for the Mean Fit? (A3 vs. fitted)
(Note: When rho=0 the

results of Test 3 and Test 2 will be the same.)

Tests of Interest


Test -2*log (Likelihood Ratio) Test df p-value

Test 1
Test 2
Test 3 0
Test 4

The p-value for Test 1

38.8702 4 <.0001
22.9533 2 <.0001
.135824 1 0.7125
1.88796 1 0.1694

is less than .05. There appears to be a
difference between response and/or variances among the dose levels
It seems appropriate to

The p-value for Test 2
model the data

is less than .1. A non-homogeneous variance
model appears to be appropriate

The p-value for Test 3
to be appropriate here


is greater than .1. The modeled variance appears


                                 0.325
D-77
DRAFT - DO NOT CITE OR QUOTE

-------
 1    The p-value  for Test  4  is  greater  than  .1.   The  model  chosen  seems
 2    to adequately describe  the data
 3
 4
 5                 Benchmark  Dose Computation
 6
 7    Specified effect =          3.85
 8
 9    Risk Type        =     Point risk
10
11    Confidence level =          0.95
12
13                 BMD =        122.316
14
15
16                BMDL =        84.0259
17
18
19
20
                                       D-78         DRAFT - DO NOT CITE OR QUOTE

-------
 1    Nesnow 1998b i.p. multiplicity
 2    DBahA
 3    Drop 2 high doses
 4    Linear
 5    Nonconstant variance
 6    BMR = lowest statistically  significant response in BaP treated animals  (after
 7    control subtracted)
                                  Linear Model with 0.95 Confidence Level
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
             3.5
      c
      o
      Q.
      (/)
      0)
      o:
      c
      (0
      0)
             0.5
       08:1712/282009
                                                                             BM1)
                                                                           3.5
        Polynomial Model.  (Version:  2.13;   Date: 04/08/2008)
        Input Data File:
C:\IPmult\Nesnowl998b\DBahA\lin_NesnowDBahA_linear_3_85.(d)
        Gnuplot Plotting  File:
C:\IPmult\Nesnowl998b\DBahA\lin_NesnowDBahA_linear_3_85.plt
                                           Wed Dec 23 15:26:52 2009
 BMDS Model Run


   The form of the response  function is:

   Y[dose] = beta_0 + beta_l*dose + beta_2*doseA2


   Dependent variable = mean
                                       D-79
                                               DRAFT - DO NOT CITE OR QUOTE

-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
   Independent variable = dose
   The polynomial coefficients are restricted to be positive
   The variance is to be modeled as Var(i) = exp(lalpha + log(mean(i))  *  rho)

   Total number of dose groups = 3
   Total number of records with missing values = 0
   Maximum number of iterations = 250
   Relative Function Convergence has been set to: le-008
   Parameter Convergence has been set to: le-008
                  Default Initial Parameter Values
                         lalpha =     0.721148
                            rho =            0
                         beta_0 =     0.413333
                         beta 1 =        1.008
           Asymptotic Correlation Matrix of Parameter Estimates

                 lalpha          rho       beta 0       beta  1

    lalpha            1        -0.35       -0.035         0.037

       rho        -0.35            1        0.073       -0.083

    beta_0       -0.035        0.073            1         -0.49

    beta 1        0.037       -0.083        -0.49             1
Confidence Interval
       Variable         Estimate
Upper Conf. Limit
         lalpha        0.0932028
0.484496
            rho          1.12871
1.63166
         beta_0         0.498826
0.803442
         beta_l         0.941334
1.26796
                            Parameter Estimates



                                   Std. Err.

                                    0.199643

                                    0.256611

                                    0.155419

                                    0.166649
                                       95.0% Wald

                                    Lower Conf. Limit

                                           -0.29809

                                           0.625764

                                            0.19421

                                           0.614709
 Dose
Res .
    0
 1.25
  2.5
Table of Data and Estimated Values of Interest

       N    Obs Mean     Est Mean   Obs Std Dev  Est Std Dev
                                                                     Scaled
    20
    18
    19
0.53
1.44
3.05
0.499
 1.68
 2.85
0.72
1.46
 1.9
0.708
  1.4
 1.89
 0.197
-0.713
 0.456
                                       D-80
                                               DRAFT - DO NOT CITE OR QUOTE

-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
 Model Descriptions for likelihoods  calculated
 Model Al:        Yij = Mu(i) + e(ij)
           Var{e(ij)} = SigmaA2

 Model A2:        Yij = Mu(i) + e(ij)
           Var{e(ij)} = Sigma(i)A2

 Model A3:        Yij = Mu(i) + e(ij)
           Var{e(ij)} = exp(lalpha  +  rho*ln(Mu(i)))
     Model A3 uses any fixed variance parameters  that
     were specified by the user

 Model  R:         Yi = Mu + e(i)
            Var{e(i)} = SigmaA2
                       Likelihoods  of  Interest
Log (likelihood)
-47.511796
-39.396001
-39.581359
-39.787219
-60.336483
# Param' s
4
6
5
4
2
AIC
103.023592
90.792002
89.162719
87.574439
124.672966
            Model
             Al
             A2
             A3
         fitted
              R
                   Explanation of Tests

 Test 1:  Do responses and/or variances  differ  among Dose levels?
          (A2 vs. R)
 Test 2:  Are Variances Homogeneous?  (Al vs A2)
 Test 3:  Are variances adequately modeled?  (A2  vs.  A3)
 Test 4:  Does the Model for the Mean  Fit?  (A3  vs.  fitted)
 (Note:  When rho=0 the results of Test  3 and Test  2 will be the same.
   Test

   Test 1
   Test 2
   Test 3
   Test 4
                     Tests of Interest
-2*log(Likelihood Ratio)  Test df
             41.881
            16.2316
           0.370717
            0.41172
4
2
1
1
    p-value

   <.0001
0.0002988
   0.5426
   0.5211
The p-value for Test 1 is less than  .05.   There  appears  to be a
difference between response and/or variances  among  the dose levels
It seems appropriate to model the data
The p-value for Test 2 is less than  .1.
model appears to be appropriate

The p-value for Test 3 is greater  than
 to be appropriate here
                              A non-homogeneous  variance


                             ,1.  The modeled variance  appears
The p-value for Test 4 is greater  than  .1.   The  model  chosen seems
                                       D-81
                                               DRAFT - DO NOT CITE OR QUOTE

-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19

20
21
22

23
24
25
26
27
to adequately describe  the  data


              Benchmark  Dose Computation

Specified  effect =           3.85

Risk Type         =      Point risk

Confidence level =           0.95

              BMD =         3.56003
             BMDL =
                            2.81986
D.3. LUNG IMPLANTATION BIOASSAYS
                               Multistage Cancer Model with 0.95 Confidence Level
      O

      I
      O
      13
      ro
              0.7
              0.6
              0.5
         0.4
         0.3
              0.2
              0.1
                                        Multistage Cancer
                                       Linear extrapolation
                       BMDL
                          BMD
                           0.1     0.2     0.3     0.4     0.5
                                                dose
                                                          0.6     0.7
0.8
        10:4912/282009
DEUTSCH-WENZEL1983AA.OUT.txt
        Multistage Cancer  Model.  (Version:  1.7;   Date:  05/16/2008;
                                         D-82
                                                 DRAFT - DO NOT CITE OR QUOTE

-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
        Input Data File:
C:\USEPA\IRIS\PAH\lungimplant\Deutschl983\AA\msc_DeutschAA_MS_l_10.(d)
        Gnuplot Plotting File:
C:\USEPA\IRIS\PAH\lungimplant\Deutschl983\AA\msc_DeutschAA_MS_l_10.plt
                                          Wed Dec 23 11:48:09 2009
 HMDS Model Run
   The form of the probability function is:

   P[response] = background +  (1-background)*[1-EXP(
                 -betal*doseAl)]

   The parameter betas are restricted to be positive
   Dependent variable = incidence
   Independent variable = dose

 Total number of observations = 3
 Total number of records with missing values = 0
 Total number of parameters in model = 2
 Total number of specified parameters = 0
 Degree of polynomial = 1
 Maximum number of iterations = 250
 Relative Function Convergence has been set to: 2.22045e-016
 Parameter Convergence has been set to: 1.49012e-008

****  We are sorry but Relative Function and Parameter Convergence    ****
****  are currently unavailable in this model.  Please keep checking  ****
****  the web sight for model updates which will eventually
****  incorporate these convergence criterion.  Default values used.
                                                                      ****
                  Default Initial Parameter Values
                     Background =            0
                        Beta(l) =     0.996523
           Asymptotic Correlation Matrix of Parameter Estimates

           (  *** The model parameter(s)  -Background
                 have been estimated at a boundary point, or have been
specified by the user,
                 and do not appear in the correlation matrix )
                Beta (1)
   Beta(l)
                                 Parameter Estimates
                                       D-83
                                               DRAFT - DO NOT CITE OR QUOTE

-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56

95 . 0% Wald
Confidence Interval
Variable
Upper Conf. Limit
Background
*
Beta (1)
*

Estimate Std. Err. Lower Conf. Limit

0 * *

0.773841 * *


* - Indicates that this value is not calculated.





Model
Full model
Fitted model
0.1162
Reduced model

AIC:




Dose Est


0.0000 0.
0.1600 0.
0.8300 0.

ChiA2 = 3.29


Benchmark Dose

Specified effect

Risk Type

Confidence level

BMD

BMDL

BMDU

Taken together, (
interval for the

Multistage Cancer






Analysis of Deviance Table

Log (likelihood) # Param's Deviance Test d.f. P-value
-28.6723 3
-30.8245 1 4.30422 2

-51.1258 1 44.907 2 <.0001

63.6489


Goodness of Fit
Scaled
. Prob. Expected Observed Size Residual


0000 0.000 0.000 35 0.000
1165 4.076 1.000 35 -1.621
4739 16.587 19.000 35 0.817

d.f. = 2 P-value = 0.1926


Computation

0.1

= Extra risk

0.95

0.136153

0.0956191

0.202527

0.0956191, 0.202527) is a 90 % two-sided confidence
BMD

Slope Factor = 1.04582



D-84
DRAFT - DO NOT CITE OR QUOTE

-------
                              Multistage Cancer Model with 0.95 Confidence Level
 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
      T3
      0)
      •5

      I
      o
      13
      ro
             0.8
        0.6
        0.4
             0.2
                                       Multistage Cancer
                                      Linear extrapolation
                 3MDLBMD
                               0.2
                                     0.4         0.6
                                          dose
0.8
       10:5012/282009
DEUTSCH-WENZEL1983BaP.OUT.txt
        Multistage  Cancer Model.  (Version:  1.7;   Date:  05/16/2008)
        Input Data  File:
C:\USEPA\IRIS\PAH\lungimplant\Deutschl983\BaP\msc_DeutschBaP_MS_2_10.(d)
        Gnuplot  Plotting  File:
C:\USEPA\IRIS\PAH\lungimplant\Deutschl983\BaP\msc_DeutschBaP_MS_2_10.plt
                                           Wed  Dec 23 11:48:08 2009
 BMDS Model Run
   The form of  the  probability function is:

   P[response]  =  background + (1-background)*[1-EXP(
                  -betal*doseAl-beta2*doseA2)]

   The parameter  betas  are restricted to be  positive
   Dependent variable  = incidence
   Independent variable = dose

 Total number of  observations = 4
                                        D-85
                                                DRAFT - DO NOT CITE OR QUOTE

-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
 Total number of records with missing values = 0
 Total number of parameters in model = 3
 Total number of specified parameters = 0
 Degree of polynomial = 2
 Maximum number of iterations = 250
 Relative Function Convergence has been set to: 2.22045e-016
 Parameter Convergence has been set to: 1.49012e-008

****  We are sorry but Relative Function and Parameter Convergence     ****
****  are currently unavailable in this model.  Please keep checking   ****
****  the web sight for model updates which will eventually            ****
****  incorporate these convergence criterion.  Default values used.   ****
                  Default Initial Parameter Values
                     Background =    0.0757681
                        Beta(l) =      2.82425
                        Beta(2) =            0
           Asymptotic Correlation Matrix of Parameter Estimates

            ( *** The model parameter(s)  -Background    -Beta(2)
                 have been estimated at a boundary point, or have  been
specified by the user,
                 and do not appear in the correlation matrix )
                Beta(l)
   Beta(l)
Confidence Interval
       Variable
Upper Conf. Limit
     Background
*
        Beta (1)
*
        Beta (2)
      Estimate

             0

       3.25323

             0
Parameter Estimates

                        95.0% Wald

       Std.  Err.     Lower Conf. Limit
* - Indicates that this value is not calculated.
       Model
     Full model
   Fitted model
0.926
  Reduced model
      Analysis of Deviance Table

Log(likelihood)   # Param's  Deviance  Test d.f.
     -51.1075         4
     -51.3412         1      0.467435      3
     -96.8119
       1
91.4088
                                   P-value
<.0001
                                       D-86
                                               DRAFT - DO NOT CITE OR QUOTE

-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
           AIC:
                        104.682
                                  Goodness   of   Fit

Dose
0.0000
0.1000
0.3000
1.0000

Est. Prob.
0.0000
0.2777
0.6232
0.9614

Expected
0.000
9.720
21.811
33.647

Observed
0.000
10.000
23.000
33.000

Size
35
35
35
35
Scaled
Residual
0.000
0.106
0.415
-0.568
 ChiA2 = 0.51
d.f.  =3
P-value = 0.9177
   Benchmark Dose Computation
Specified effect =

Risk Type

Confidence level =

             BMD =

            BMDL =

            BMDU =
           0.1

     Extra risk

          0.95

     0.0323864

     0.0255063

     0.0445507
Taken together,  (0.0255063, 0.0445507)  is  a  90
interval for the BMD
                                % two-sided  confidence
Multistage Cancer Slope Factor =
                     3.9206
                                       D-87
                                               DRAFT - DO NOT CITE OR QUOTE

-------
                              Multistage Cancer Model with 0.95 Confidence Level
 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
      T3
      0)
      •5

      I
      o
      13
      ro
             0.6
             0.5
             0.4
        0.3
             0.2
             0.1
                                       Multistage Cancer
                                      Linear extrapolation
               0          0.2

   10:51 12/282009

DEUTSCH-WENZEL1983BbF.OUT.txt
                                          0.4
  0.6
0.8
                                               dose
        Multistage  Cancer Model.  (Version:  1.7;   Date:  05/16/2008)
        Input Data  File:
C:\USEPA\IRIS\PAH\lungimplant\Deutschl983\BbF\msc_DeutschBbF_MS_2_10.(d)
        Gnuplot  Plotting  File:
C:\USEPA\IRIS\PAH\lungimplant\Deutschl983\BbF\msc_DeutschBbF_MS_2_10.plt
                                           Wed  Dec 23 11:48:08 2009
 BMDS Model Run
   The form of  the  probability function is:

   P[response]  =  background + (1-background)*[1-EXP(
                  -betal*doseAl-beta2*doseA2)]

   The parameter  betas  are restricted to be  positive
   Dependent variable  = incidence
   Independent variable = dose
 Total number of  observations = 4
 Total number of  records with missing values
 Total number of  parameters in model = 3
= 0
                                        D-88
                                                DRAFT - DO NOT CITE OR QUOTE

-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
 Total number of specified parameters = 0
 Degree of polynomial = 2
 Maximum number of iterations = 250
 Relative Function Convergence has been set to: 2.22045e-016
 Parameter Convergence has been set to: 1.49012e-008

****  We are sorry but Relative Function and Parameter Convergence     ****
****  are currently unavailable in this model.  Please keep checking   ****
****  the web sight for model updates which will eventually            ****
****  incorporate these convergence criterion.  Default values  used.   ****
                  Default Initial Parameter Values
                     Background =   0.00149382
                        Beta(l) =     0.226374
                        Beta(2) =     0.236366
           Asymptotic Correlation Matrix of Parameter Estimates

            ( *** The model parameter(s)  -Background
                 have been estimated at a boundary point,  or have  been
specified by the user,
                 and do not appear in the correlation matrix )
   Beta(l)

   Beta(2)
Beta (1)

      1

  -0.97
Beta (2)

  -0.97

      1
Confidence Interval
       Variable
Upper Conf. Limit
     Background
*
        Beta (1)
*
        Beta (2)
        Estimate

               0

         0.24518

        0.217701
    Parameter Estimates

                            95.0% Wald

           Std. Err.     Lower Conf. Limit
* - Indicates that this value is not calculated.
       Model
     Full model
   Fitted model
0.9944
  Reduced model
        Analysis of Deviance Table

  Log(likelihood)  # Param's  Deviance  Test d.f.
       -37.8686         4
       -37.8743         2     0.0112712      2
       -51.7666
           1
27.796
                                       P-value
<.0001
                                       D-89
                                               DRAFT - DO NOT CITE OR QUOTE

-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
           AIC:
                        79.7485
                                  Goodness   of   Fit

Dose
0.0000
0.1000
0.3000
1.0000

Est. Prob.
0.0000
0.0263
0.0889
0.3705

Expected
0.000
0.922
3.113
12.969

Observed
0.000
1.000
3.000
13.000

Size
35
35
35
35
Scaled
Residual
0.000
0.082
-0.067
0.011
 ChiA2 = 0.01
d.f.  =2
P-value = 0.9943
   Benchmark Dose Computation
Specified effect =

Risk Type

Confidence level =

             BMD =

            BMDL =

            BMDU =
           0.1

     Extra risk

          0.95

       0.33191

      0.184961

      0.544229
Taken together,  (0.184961, 0.544229) is  a  90
interval for the BMD
                              % two-sided confidence
Multistage Cancer Slope Factor =
                   0.540655
                                       D-90
                                               DRAFT - DO NOT CITE OR QUOTE

-------
                               Multistage Cancer Model with 0.95 Confidence Level
 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
      T3
      0)
      •5

      I
      o
      13
      ro
               0.3
              0.25
               0.2
         0.15
               0.1
              0.05
                                   Multistage Cancer
                                  Linear extrapolation
                                          BMDL
                                                                BMD
                            0.5
                                    1.5
 2
dose
2.5
3.5
       10:5212/282009
DEUTSCH-WENZEL1983BghiP.OUT.txt
        Multistage  Cancer Model. (Version:  1.7;   Date:  05/16/2008)
        Input Data  File:
C:\USEPA\IRIS\PAH\lungimplant\Deutschl983\BghiP\msc_DeutschBghiP_MS_2_10.(d)
        Gnuplot  Plotting  File:
C:\USEPA\IRIS\PAH\lungimplant\Deutschl983\BghiP\msc_DeutschBghiP_MS_2_10.plt
                                           Wed  Dec 23  11:48:09 2009
 BMDS Model Run
   The form of the  probability function is:

   P[response] = background + (1-background)*[1-EXP(
                 -betal*doseAl-beta2*doseA2)]

   The parameter betas  are restricted to be  positive
   Dependent variable  = incidence
   Independent variable = dose
 Total number of  observations = 4
 Total number of  records  with missing values
   =  0
                                        D-91
                                                DRAFT - DO NOT CITE OR QUOTE

-------
 1     Total number of parameters in model = 3
 2     Total number of specified parameters = 0
 3     Degree of polynomial = 2
 4
 5
 6     Maximum number of iterations = 250
 7     Relative Function Convergence has been set to: 2.22045e-016
 8     Parameter Convergence has been set to: 1.49012e-008
 9
10    ****   we are sorry but Relative Function and Parameter Convergence    ****
H    ****   are currently unavailable in this model.  Please keep checking  ****
12    ****   the web sight for model updates which will eventually           ****
13    ****   incorporate these convergence criterion.  Default values used.  ****
14
15
16
17                      Default Initial Parameter Values
18                         Background =            0
19                            Beta(l) =    0.0304801
20                            Beta (2) =            0
21
22
23               Asymptotic Correlation Matrix of Parameter Estimates
24
25               ( *** The model parameter(s)  -Background
26                     have been estimated at a boundary point, or have been
27    specified by the user,
28                     and do not appear in the correlation matrix )
29
30                    Beta(l)       Beta(2)
31
32       Beta(l)             1        -0.98
33
34       Beta(2)         -0.98            1
35
36
37
38                                     Parameter Estimates
39
40                                                             95.0% Wald
41    Confidence Interval
42           Variable         Estimate        Std. Err.     Lower Conf. Limit
43    Upper Conf.  Limit
44         Background                0            *                *
45    *
46            Beta(l)         0.0277423            *                *
47    *
48            Beta(2)       0.000645059            *                *
49    *
50
51    * - Indicates that this value is not calculated.
52
53
54
55                            Analysis of Deviance Table
56
57           Model      Log(likelihood)  # Param's  Deviance  Test d.f.   P-value
58         Full model        -16.8561         4
59       Fitted model         -17.033         2      0.353756      2
60    0.8379
                                       D-92        DRAFT - DO NOT CITE OR QUOTE

-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
  Reduced model
0.02491

           AIC:
    -21.5342
     38.0659
             9.35614
                                  Goodness   of   Fit

Dose
0.0000
0.1600
0.8300
4.1500

Est. Prob.
0.0000
0.0044
0.0232
0.1186

Expected
0.000
0.156
0.812
4.032

Observed
0.000
0.000
1.000
4.000

Size
35
35
35
34
Scaled
Residual
0.000
-0.395
0.211
-0.017
 ChiA2 = 0.20
d.f. = 2
P-value = 0.9043
   Benchmark Dose Computation
Specified effect =

Risk Type

Confidence level =

             BMD =

            BMDL =

            BMDU =
           0.1

     Extra risk

          0.95

       3.51117

       1.82558

       8.33008
Taken together,  (1.82558, 8.33008) is  a  90
interval for the BMD
                            % two-sided  confidence
Multistage Cancer Slope Factor =
                  0.0547771
                                       D-93
                                               DRAFT - DO NOT CITE OR QUOTE

-------
                              Multistage Cancer Model with 0.95 Confidence Level
 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
      T3
      0)
      •5

      I
      o
      13
      ro
             0.7
             0.6
             0.5
        0.4
        0.3
             0.2
             0.1
                                  Multistage Cancer
                                 Linear extrapolation
       10:5312/282009
DEUTSCH-WENZEL1983BJ F.OUT.txt
        Multistage  Cancer Model. (Version:  1.7;   Date:  05/16/2008)
        Input Data  File:
C:\USEPA\IRIS\PAH\lungimplant\Deutschl983\BjF\msc_DeutschBjF_MS_2_10.(d)
        Gnuplot  Plotting  File:
C:\USEPA\IRIS\PAH\lungimplant\Deutschl983\BjF\msc_DeutschBjF_MS_2_10.plt
                                           Wed  Dec 23  11:48:08 2009
 BMDS Model Run
   The form of  the  probability function is:

   P[response]  = background + (1-background)*[1-EXP(
                 -betal*doseAl-beta2*doseA2)]

   The parameter betas  are restricted to be  positive
   Dependent variable  = incidence
   Independent variable = dose
 Total number of  observations = 4
 Total number of  records  with missing values
= 0
                                        D-94
                                                DRAFT - DO NOT CITE OR QUOTE

-------
 1     Total number of parameters in model = 3
 2     Total number of specified parameters = 0
 3     Degree of polynomial = 2
 4
 5
 6     Maximum number of iterations = 250
 7     Relative Function Convergence has been set to: 2.22045e-016
 8     Parameter Convergence has been set to: 1.49012e-008
 9
10    ****   we are sorry but Relative Function and Parameter Convergence    ****
H    ****   are currently unavailable in this model.  Please keep checking  ****
12    ****   the web sight for model updates which will eventually           ****
13    ****   incorporate these convergence criterion.  Default values used.  ****
14
15
16
17                      Default Initial Parameter Values
18                         Background =   0.00616121
19                            Beta(l) =    0.0709095
20                            Beta(2) =    0.0144537
21
22
23               Asymptotic Correlation Matrix of Parameter Estimates
24
25               ( *** The model parameter(s)  -Background
26                     have been estimated at a boundary point, or have been
27    specified by the user,
28                     and do not appear in the correlation matrix )
29
30                    Beta(l)       Beta(2)
31
32       Beta(l)             1        -0.98
33
34       Beta(2)         -0.98            1
35
36
37
38                                     Parameter Estimates
39
40                                                             95.0% Wald
41    Confidence Interval
42           Variable         Estimate        Std. Err.     Lower Conf. Limit
43    Upper Conf.  Limit
44         Background                0            *                *
45    *
46            Beta(l)         0.0929144            *                *
47    *
48            Beta(2)         0.0101278            *                *
49    *
50
51    * - Indicates that this value is not calculated.
52
53
54
55                            Analysis of Deviance Table
56
57           Model      Log(likelihood)  # Param's  Deviance  Test d.f.   P-value
58         Full model        -39.0246         4
59       Fitted model        -39.1336         2      0.218103      2
60    0.8967
                                       D-95        DRAFT - DO NOT CITE OR QUOTE

-------
1
2
3
4
5
6
7
8
Q
y
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
Reduced model

AIC:




Dose Est

0.0000 0.
0.2000 0.
1.0000 0.
5.0000 0.

ChiA2 = 0.24


Benchmark Dose

Specified effect

Risk Type

Confidence level

BMD

BMDL

BMDU

Taken together, (
interval for the

Multistage Cancer




-60.8862 1 43.7233 3 <.0

82.2673


Goodness of Fit
Scaled
. Prob. Expected Observed Size Residual

0000 0.000 0.000 35 0.000
0188 0.658 1.000 35 0.425
0979 3.427 3.000 35 -0.243
5122 17.926 18.000 35 0.025

d.f. = 2 P-value = 0.8868


Computation

0.1

= Extra risk

0.95

1.02045

0.580958

2.07945

0.580958, 2.07945) is a 90 % two-sided confidence
BMD

Slope Factor = 0.172129




D-96        DRAFT - DO NOT CITE OR QUOTE

-------
                              Multistage Cancer Model with 0.95 Confidence Level
 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
      T3
      0)
      •5

      I
      o
      13
      ro
             0.7
             0.6
             0.5
        0.4
        0.3
             0.2
             0.1
                                       Multistage Cancer
                                     Linear extrapolation
                       BMDL
                              BMD
               0     0.5      1

   10:5412/282009

DEUTSCH-WENZEL1983BkF.OUT.txt
                                        1.5
2
dose
2.5
3.5
        Multistage  Cancer Model. (Version:  1.7;   Date:  05/16/2008)
        Input Data  File:
C:\USEPA\IRIS\PAH\lungimplant\Deutschl983\BkF\msc_DeutschBkF_MS_2_10.(d)
        Gnuplot  Plotting  File:
C:\USEPA\IRIS\PAH\lungimplant\Deutschl983\BkF\msc_DeutschBkF_MS_2_10.plt
                                           Wed  Dec 23  11:48:09 2009
 BMDS Model Run
   The form of the  probability function is:

   P[response] = background + (1-background)*[1-EXP(
                 -betal*doseAl-beta2*doseA2)]

   The parameter betas  are restricted to be  positive
   Dependent variable  = incidence
   Independent variable = dose
 Total number of  observations = 4
 Total number of  records  with missing values
 Total number of  parameters in model = 3
    = 0
                                        D-97
                                                DRAFT - DO NOT CITE OR QUOTE

-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
 Total number of specified parameters = 0
 Degree of polynomial = 2
 Maximum number of iterations = 250
 Relative Function Convergence has been set to: 2.22045e-016
 Parameter Convergence has been set to: 1.49012e-008

****  We are sorry but Relative Function and Parameter  Convergence     ****
****  are currently unavailable in this model.  Please  keep checking   ****
****  the web sight for model updates which will eventually            ****
****  incorporate these convergence criterion.  Default values  used.   ****
                  Default Initial Parameter Values
                     Background =             0
                        Beta(l) =     0.126747
                        Beta(2) =   0.00410997
           Asymptotic Correlation Matrix of Parameter Estimates

            ( *** The model parameter(s)  -Background
                 have been estimated at a boundary point,  or  have  been
specified by the user,
                 and do not appear in the correlation matrix  )
   Beta(l)

   Beta(2)
Beta (1)

      1

  -0.97
Beta (2)

  -0.97

      1
Confidence Interval
       Variable         Estimate
Upper Conf. Limit
     Background                0
*

        Beta(l)        0.0842968
*
        Beta(2)        0.0142917
* - Indicates that this value is not calculated.
                 Parameter Estimates

                                         95.0% Wald

                        Std. Err.     Lower Conf. Limit
       Model
     Full model
   Fitted model
0.5667
  Reduced model
        Analysis of Deviance Table

  Log(likelihood)  # Param's  Deviance  Test d.f.
        -28.404         4
       -28.9719         2        1.1357      2
       -46.2443
           1
35.6806
                                       P-value
<.0001
                                       D-98
                                               DRAFT - DO NOT CITE OR QUOTE

-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
           AIC:
                        61.9437
                                  Goodness   of   Fit

Dose
0.0000
0.1600
0.8300
4.1500

Est. Prob.
0.0000
0.0138
0.0767
0.4490

Expected
0.000
0.482
2.378
12.122

Observed
0.000
0.000
3.000
12.000

Size
35
35
31
27
Scaled
Residual
0.000
-0.699
0.420
-0.047
 ChiA2 =0.67
d.f.  =2
P-value = 0.7165
   Benchmark Dose Computation
Specified effect =

Risk Type

Confidence level =

             BMD =

            BMDL =

            BMDU =
           0.1

     Extra risk

          0.95

       1.05954

      0.557079

       1.79525
Taken together,  (0.557079, 1.79525)  is  a  90
interval for the BMD
                             % two-sided  confidence
Multistage Cancer Slope Factor =
                   0.179508
                                       D-99
                                               DRAFT - DO NOT CITE OR QUOTE

-------
                              Multistage Cancer Model with 0.95 Confidence Level
 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
      T3
      0)
      •5

      I
      o
      13
      ro
        0.8


        0.7


        0.6


        0.5


        0.4


        0.3


        0.2


        0.1
                                       Multistage Cancer
                                      Linear extrapolation
                    BMDL
                     BMD
               0     0.5      1

   10:5512/282009

DEUTSCH-WENZEL1983IP.OUT.txt
                                        1.5
2
dose
2.5
3.5
        Multistage  Cancer Model.  (Version:  1.7;   Date:  05/16/2008)
        Input Data  File:
C:\USEPA\IRIS\PAH\lungimplant\Deutschl983\IP\msc_DeutschIP_MS_2_10.(d)
        Gnuplot  Plotting  File:
C:\USEPA\IRIS\PAH\lungimplant\Deutschl983\IP\msc_DeutschIP_MS_2_10.plt
                                           Wed  Dec 23 11:48:09 2009
 HMDS Model Run
   The form of  the  probability function is:

   P[response]  =  background + (1-background)*[1-EXP(
                  -betal*doseAl-beta2*doseA2)]

   The parameter  betas  are restricted to be  positive
   Dependent variable  = incidence
   Independent variable = dose
 Total number of  observations = 4
 Total number of  records with missing values
    = 0
                                       D-100
                                                DRAFT - DO NOT CITE OR QUOTE

-------
 1     Total number of parameters in model = 3
 2     Total number of specified parameters = 0
 3     Degree of polynomial = 2
 4
 5
 6     Maximum number of iterations = 250
 7     Relative Function Convergence has been set to: 2.22045e-016
 8     Parameter Convergence has been set to: 1.49012e-008
 9
10    ****   we are sorry but Relative Function and Parameter Convergence    ****
H    ****   are currently unavailable in this model.  Please keep checking  ****
12    ****   the web sight for model updates which will eventually           ****
13    ****   incorporate these convergence criterion.  Default values used.  ****
14
15
16
17                      Default Initial Parameter Values
18                         Background =    0.0539703
19                            Beta(l) =      0.20919
20                            Beta (2) =            0
21
22
23               Asymptotic Correlation Matrix of Parameter Estimates
24
25               ( *** The model parameter(s)  -Beta (2)
26                     have been estimated at a boundary point, or have been
27    specified by the user,
28                     and do not appear in the correlation matrix )
29
30                 Background      Beta(l)
31
32    Background            1        -0.55
33
34       Beta(l)         -0.55            1
35
36
37
38                                     Parameter Estimates
39
40                                                             95.0% Wald
41    Confidence Interval
42           Variable         Estimate        Std. Err.      Lower Conf. Limit
43    Upper Conf.  Limit
44         Background        0.0224449            *                *
45    *
46            Beta(l)          0.241452            *                *
47    *
48            Beta(2)                 0            *                *
49    *
50
51    * - Indicates that this value is not calculated.
52
53
54
55                            Analysis of Deviance Table
56
57           Model      Log(likelihood)  # Param's  Deviance  Test d.f.   P-value
58         Full model        -54.8079         4
59       Fitted model        -56.5662         2        3.5166      2
60    0.1723
                                      D-101        DRAFT - DO NOT CITE OR QUOTE

-------
1
2
3
4
5
6
7
8
Q
y
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
Reduced model

AIC:




Dose Est

0.0000 0.
0.1600 0.
0.8300 0.
4.1500 0.

ChiA2 =3.12


Benchmark Dose

Specified effect

Risk Type

Confidence level

BMD

BMDL

BMDU

Taken together, (
interval for the

Multistage Cancer




-76.4525 1 43.2893 3 <.0

117.132


Goodness of Fit
Scaled
. Prob. Expected Observed Size Residual

0224 0.786 0.000 35 -0.896
0595 2.082 4.000 35 1.370
2000 6.999 8.000 35 0.423
6411 22.439 21.000 35 -0.507

d.f. = 2 P-value = 0.2104


Computation

0.1

= Extra risk

0.95

0.436361

0.309504

0.819969

0.309504, 0.819969) is a 90 % two-sided confidence
BMD

Slope Factor = 0.323098




D-102       DRAFT - DO NOT CITE OR QUOTE

-------
                              Multistage Cancer Model with 0.95 Confidence Level
 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
      T3
      0)
      •5

      I
      o
      13
      ro
             0.8
             0.6
         0.4
             0.2
                                       Multistage Cancer
                                      Linear extrapolation
               0        0.05

   10:5612/282009

WENZEL-HARTUNG1990BaP.OUT.txt
                                       0.1
0.15
dose
0.2
0.25
0.3
        Multistage Cancer Model.  (Version:  1.7;   Date: 05/16/2008)
        Input  Data File:
C:\USEPA\IRIS\PAH\lungimplant\Wenzell990\BaP\msc_WenzelBaP_MS_2_10.(d)
        Gnuplot  Plotting File:
C:\USEPA\IRIS\PAH\lungimplant\Wenzell990\BaP\msc_WenzelBaP_MS_2_10.plt
                                           Wed Dec 23 11:48:09 2009
 BMDS Model  Run
   The form  of  the probability function  is:

   P[response]  =  background +  (1-background)*[1-EXP(
                  -betal*doseAl-beta2*doseA2)]

   The parameter  betas are restricted to be  positive
   Dependent  variable = incidence
   Independent  variable = dose
 Total number  of  observations = 4
 Total number  of  records with missing values
    = 0
                                       D-103
                                                DRAFT - DO NOT CITE OR QUOTE

-------
 1     Total number of parameters in model = 3
 2     Total number of specified parameters = 0
 3     Degree of polynomial = 2
 4
 5
 6     Maximum number of iterations = 250
 7     Relative Function Convergence has been set to: 2.22045e-016
 8     Parameter Convergence has been set to: 1.49012e-008
 9
10    ****   we are sorry but Relative Function and Parameter Convergence    ****
H    ****   are currently unavailable in this model.  Please keep checking  ****
12    ****   the web sight for model updates which will eventually           ****
13    ****   incorporate these convergence criterion.  Default values used.  ****
14
15
16
17                      Default Initial Parameter Values
18                         Background =            0
19                            Beta(l) =      3.21631
20                            Beta(2) =       5.7325
21
22
23               Asymptotic Correlation Matrix of Parameter Estimates
24
25               ( *** The model parameter(s)  -Background
26                     have been estimated at a boundary point, or have been
27    specified by the user,
28                     and do not appear in the correlation matrix )
29
30                    Beta(l)       Beta(2)
31
32       Beta(l)             1        -0.93
33
34       Beta(2)         -0.93            1
35
36
37
38                                     Parameter Estimates
39
40                                                             95.0% Wald
41    Confidence Interval
42           Variable         Estimate        Std. Err.     Lower Conf. Limit
43    Upper Conf.  Limit
44         Background                0            *                *
45    *
46            Beta(l)           3.01149            *                *
47    *
48            Beta(2)           6.44644            *                *
49    *
50
51    * - Indicates that this value is not calculated.
52
53
54
55                            Analysis of Deviance Table
56
57           Model      Log(likelihood)  # Param's  Deviance  Test d.f.   P-value
58         Full model        -50.8389         4
59       Fitted model        -50.8521         2     0.0264626      2
60    0.9869
                                      D-104        DRAFT - DO NOT CITE OR QUOTE

-------
1
2
3
4
5
6
7
8
Q
y
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
Reduced model

AIC:




Dose Est

0.0000 0.
0.0300 0.
0.1000 0.
0.3000 0.

ChiA2 = 0.03


Benchmark Dose

Specified effect

Risk Type

Confidence level

BMD

BMDL

BMDU

Taken together, (
interval for the

Multistage Cancer




-84.6566 1 67.6355 3 <.0

105.704


Goodness of Fit
Scaled
. Prob. Expected Observed Size Residual

0000 0.000 0.000 35 0.000
0917 3.208 3.000 35 -0.122
3062 10.718 11.000 35 0.103
7732 27.062 27.000 35 -0.025

d.f. = 2 P-value = 0.9870


Computation

0.1

= Extra risk

0.95

0.0326976

0.0198862

0.0559366

0.0198862, 0.0559366) is a 90 % two-sided confidence
BMD

Slope Factor = 5.02861




D-105       DRAFT - DO NOT CITE OR QUOTE

-------
 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
                              Multistage Cancer Model with 0.95 Confidence Level
      T3
      
-------
 1
 2     Total number of observations = 4
 3     Total number of records with missing values = 0
 4     Total number of parameters in model = 3
 5     Total number of specified parameters = 0
 6     Degree of polynomial = 2
 7
 8
 9     Maximum number of iterations = 250
10     Relative Function Convergence has been set to: 2.22045e-016
11     Parameter Convergence has been set to: 1.49012e-008
12
13    ****   we are sorry but Relative Function and Parameter Convergence    ****
14    ****   are currently unavailable in this model.  Please keep checking  ****
15    ****   the web sight for model updates which will eventually           ****
15    ****   incorporate these convergence criterion.  Default values used.  ****
17
18
19
20                      Default Initial Parameter Values
21                         Background =            0
22                            Beta(l) =      3.21631
23                            Beta(2) =       5.7325
24
25
26               Asymptotic Correlation Matrix of Parameter Estimates
27
28               ( *** The model parameter(s)  -Background
29                     have been estimated at a boundary point, or have been
30    specified by the user,
31                     and do not appear in the correlation matrix )
32
33                    Beta(l)       Beta(2)
34
35       Beta(l)             1        -0.93
36
37       Beta(2)         -0.93            1
38
39
40
41                                     Parameter Estimates
42
43                                                             95.0% Wald
44    Confidence Interval
45           Variable         Estimate        Std. Err.     Lower Conf. Limit
46    Upper Conf.  Limit
47         Background                0            *                *
48    *
49            Beta(l)           3.01149            *                *
50    *
51            Beta(2)           6.44644            *                *
52    *
53
54    * - Indicates that this value is not calculated.
55
56
57
58                            Analysis of Deviance Table
59
60           Model      Log(likelihood)  # Param's  Deviance  Test d.f.   P-value


                                      D-107        DRAFT - DO NOT CITE OR QUOTE

-------
1
2
3
4
5
6
7
9
10
11
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
Full model
Fitted model
0.9869
Reduced model

AIC:



Dose Est
0.0000 0.
0.0300 0.
0.1000 0.
0.3000 0.

ChiA2 = 0.03


Benchmark Dose

Specified effect

Risk Type

Confidence level

BMD

BMDL

BMDU

Taken together, (
interval for the

Multistage Cancer



-50.8389 4
-50.8521 2 0.0264626 2

-84.6566 1 67.6355 3 <.0

105.704

Goodness of Fit
Scaled
. Prob. Expected Observed Size Residual
0000 0.000 0.000 35 0.000
0917 3.208 3.000 35 -0.122
3062 10.718 11.000 35 0.103
7732 27.062 27.000 35 -0.025

d.f. = 2 P-value = 0.9870


Computation

0.57

= Extra risk

0.95

0.197095

0.157781

0.247357

0.157781, 0.247357) is a 90 % two-sided confidence
BMD

Slope Factor = 3.6126



D-108
DRAFT - DO NOT CITE OR QUOTE

-------
 1    WENZEL-HARTUNG1990BaPforDBahA.OUT.txt
 2     ====================================================================
 3            Multistage Model.  $Revision: 2.1 $ $Date: 2000/08/21 03:38:21 $
 4            Input Data File:  C:\PAH\BMD ANALYSIS\BIOASSAY\OTHER
 5    ROUTE\SETS\WENZEL-HARTUNG1990.(d)
 6            Gnuplot Plotting File:   C:\PAH\BMD ANALYSIS\BIOASSAY\OTHER
 7    ROUTE\SETS\WENZEL-HARTUNG1990.plt
 8                                              Thu Jun 02 09:02:58 2005
 9     ====================================================================
10
11     BMDS MODEL RUN
19    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
13
14       The form of the probability function is:
15
16       P[response]  = background + (1-background)*[1-EXP(
17    -betal*doseAl-beta2*doseA2)]
18
19       The parameter betas are restricted to be positive
20
21
22       Dependent variable = responseBaP
23       Independent variable = doseBaP
24
25     Total number of observations = 4
26     Total number of records with missing values = 0
27     Total number of parameters in model = 3
28     Total number of specified parameters = 0
29     Degree of polynomial = 2
30
31
32     Maximum number of iterations = 250
33     Relative Function Convergence has been set to: le-008
34     Parameter Convergence has been set to: le-008
35
36
37
38                      Default Initial Parameter Values
39                         Background =            0
40                            Beta(l)  =      3.21631
41                            Beta(2)  =       5.7325
42
43
44               Asymptotic Correlation Matrix of Parameter Estimates
45
46               ( *** The model parameter(s)  -Background
47                     have been estimated at a boundary point, or have been
48    specified by the user,
49                     and do not appear in the correlation matrix )
50
51                    Beta(l)       Beta(2)
52
53       Beta(l)             1        -0.93
54
55       Beta(2)         -0.93            1
56
57
58
59                              Parameter Estimates
60
                                      D-109        DRAFT - DO NOT CITE OR QUOTE

-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
Variable
Background
Beta(l)
Beta (2)

Estimate Std. Err.
0 NA
3.01149 2.79594
6.44644 10.7674






NA - Indicates that this parameter has hit a bound
implied by some inequality constraint and thus
has no standard error.





Model
Full model
Fitted model
Reduced model

AIC:




Dose Est
i: 1
0.0000 0.
i: 2
0.0300 0.
i: 3
0.1000 0.
i: 4
0.3000 0.

Chi-square =


Benchmark Dose

Specified effect

Risk Type

Confidence level

BMD

BMDL




Analysis of Deviance Table






Log (likelihood) Deviance Test DF P-value
-50.8389
-50.8521 0.0264626 2
-84.6566 67.6355 3

105.704


Goodness of Fit

. Prob. Expected Observed Size

0000 0.000 0 35

0917 3.208 3 35

3062 10.718 11 35

7732 27.062 27 35

0.03 DF = 2 P-value = 0.9870


Computation

0.57

= Extra risk

0.95

0.197095

0.157781


0.9869
<.0001






ChiA2 Res.

0.000

-0.072

0.038

-0.010
















D-110
DRAFT - DO NOT CITE OR QUOTE

-------
                              Multistage Cancer Model with 0.95 Confidence Level
 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
      T3
      0)
      •5

      I
      o
      13
      ro
             0.5
             0.4
             0.3
        0.2
             0.1
                                  Multistage Cancer
                                 Linear extrapolation
                           BMDL
                                BMD
               0        0.5        1

   10:5912/282009

WENZ EL-HARTUNG19 9 OCH.OUT.txt
                                                1.5
                                               dose
                2.5
        Multistage  Cancer Model.  (Version:  1.7;   Date:  05/16/2008)
        Input  Data  File:
C:\USEPA\IRIS\PAH\lungimplant\Wenzell990\CH\msc_WenzelCH_MS_l_10.(d)
        Gnuplot  Plotting File:
C:\USEPA\IRIS\PAH\lungimplant\Wenzell990\CH\msc_WenzelCH_MS_l_10.plt
                                           Wed  Dec 23 11:48:10 2009
 HMDS Model  Run
   The form  of  the  probability function is:

   P[response]  =  background + (1-background)*[1-EXP(
                  -betal*doseAl)]

   The parameter  betas are restricted to be  positive
   Dependent variable = incidence
   Independent  variable = dose
 Total number  of  observations = 3
 Total number  of  records with missing values
 Total number  of  parameters in model = 2
= 0
                                       D-lll
                                                DRAFT - DO NOT CITE OR QUOTE

-------
 1     Total number of specified parameters = 0
 2     Degree of polynomial = 1
 3
 4
 5     Maximum number of iterations = 250
 6     Relative Function Convergence has been set to: 2.22045e-016
 7     Parameter Convergence has been set to: 1.49012e-008
 8
 9    ****   we are sorry but Relative Function and Parameter Convergence    ****
10    ****   are currently unavailable in this model.  Please keep checking  ****
H    ****   the wek sight for model updates which will eventually           ****
12    ****   incorporate these convergence criterion.  Default values used.  ****
13
14
15
16                      Default Initial Parameter Values
17                         Background =    0.0178361
18                            Beta(l) =     0.109158
19
20
21               Asymptotic Correlation Matrix of Parameter Estimates
22
23               ( *** The model parameter(s)  -Background
24                     have been estimated at a boundary point, or have been
25    specified by the user,
26                     and do not appear in the correlation matrix )
27
28                    Beta(l)
29
30       Beta(l)             1
31
32
33
34                                     Parameter Estimates
35
36                                                             95.0% Wald
37    Confidence Interval
38           Variable         Estimate        Std. Err.     Lower Conf. Limit
39    Upper Conf.  Limit
40         Background                0            *                *
41    *
42            Beta(l)          0.123432            *                *
43    *
44
45    * - Indicates that this value is not calculated.
46
47
48
49                            Analysis of Deviance Table
50
51           Model      Log(likelihood)  # Param's  Deviance  Test d.f.   P-value
52         Full model        -35.2935         3
53       Fitted model         -35.455         1      0.323044      2
54    0.8508
55      Reduced model        -43.0622         1       15.5374      2
56    0.0004228
57
58               AIC:          72.9101
59
60
                                      D-l 12        DRAFT - DO NOT CITE OR QUOTE

-------
1
2
3
A
4-
5
6
7
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
Dose Est
0.0000 0.
1.0000 0.
3.0000 0.
ChiA2 = 0.34


Benchmark Dose

Specified effect

Risk Type

Confidence level

BMD

BMDL

BMDU

Taken together, (
interval for the

Multistage Cancer


Goodness of Fit
. Prob. Expected Observed Size
0000 0.000 0.000 35
1161 4.064 5.000 35
3095 10.831 10.000 35
d.f. = 2 P-value = 0.8453


Computation

0.1

= Extra risk

0.95

0.853595

0.57298

1.36494

0.57298, 1.36494) is a 90 % two-sided
BMD

Slope Factor = 0.174526


Scaled
Residual
0.000
0.494
-0.304

















confidence





D-113       DRAFT - DO NOT CITE OR QUOTE

-------
 1
 2

 o
 J
 4
 5
 6
 7
D.4.  ORAL BIOASSAYS
Weyand et al.  2004  BcFE lung
      T3
      
-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
   The parameter betas are restricted  to be positive
   Dependent variable = incidence
   Independent variable = dose

 Total number of observations = 3
 Total number of records with missing values  =  0
 Total number of parameters in model =  2
 Total number of specified parameters = 0
 Degree of polynomial = 1
 Maximum number of iterations = 250
 Relative Function Convergence has been set to:  2.22045e-016
 Parameter Convergence has been set to: 1.49012e-008

****  We are sorry but Relative Function and  Parameter  Convergence
****  are currently unavailable in this model.   Please  keep checking
•k-k-k-k  the web sight for model updates which will  eventually
•k-k-k-k  incorporate these convergence criterion.   Default values  used.
                  Default Initial  Parameter Values
                     Background =             0
                        Beta(l) =  5.23754e+017
           Asymptotic Correlation Matrix  of  Parameter  Estimates

             Background      Beta(l)

Background            1        -0.45

   Beta(l)        -0.45             1



                                 Parameter Estimates
                                                                       •k-k-k-k

                                                                       •k-k-k-k
Confidence Interval
       Variable         Estimate         Std.  Err.
Upper Conf. Limit
     Background         0.233316             *
*
        Beta(l)        0.0289518             *
* - Indicates that this value is not  calculated.
                                                          95.0% Wald
                                                       Lower  Conf.  Limit
                                       D-115
                                               DRAFT - DO NOT CITE OR QUOTE

-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43

44

45

46
       Model
     Full model
   Fitted model
0.6567
  Reduced model

           AIC:
         Analysis of Deviance Table

   Log(likelihood)   # Param's  Deviance  Test d.f.
        -35.3639         3
        -35.4627         2      0.197606      1
        -58.7707
         74.9254
                   46.8136
                                  Goodness  of   Fit
     Dose
              Est. Prob.
             Expected
           Observed
Size
    0.0000
   13.6000
  197.0000

 ChiA2 =0.12
0.2333
0.4829
0.9974
    d.f.  =1
   Benchmark Dose Computation

Specified effect =            0.7

Risk Type        =      Extra risk

Confidence level =           0.95
             BMD =
            BMDL =
            BMDU =
                          41.5854
                          22.3673
                          81.9344
                                       P-value
                  <.0001
 Scaled
Residual
 6.766     7.000          29        0.103
13.520    13.000          28       -0.197
28.926    29.000          29        0.273

      P-value = 0.7253
Taken together,  (22.3673, 81.9344) is a  90
interval for the BMD
                                % two-sided confidence
Multistage Cancer Slope Factor =
                      0.0312958
                                      D-116
                                               DRAFT - DO NOT CITE OR QUOTE

-------
 1    D.5. BACTERIAL MUTAGENICITY
 2    Hass  1981 bact mut bap.out.txt
 3     ====================================================================
 4            Polynomial Model.  Revision:  2.2  Date:  9/12/2002
 5            Input Data File:  C:\BMDS\UNSAVED1.(d)
 6            Gnuplot Plotting  File:  C:\BMDS\UNSAVEDl.plt
 7                                              Wed Jul 06 11:29:07 2005
 8     ====================================================================
 9
10     HMDS  MODEL RUN
11    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
12
13       The form of the response  function is:
14
15       Y[dose]  = beta_0 + beta_l*dose + beta_2*doseA2 + ...
16
17
18       Dependent variable = MEAN
19       Independent variable = COLUMN1
20       rho is set to 0
21       Signs of the polynomial coefficients are not restricted
22       A  constant variance model is fit
23
24       Total number of dose groups = 4
25       Total number of records with missing values  = 0
26       Maximum number of iterations = 250
27       Relative Function Convergence has been set to: le-008
28       Parameter Convergence  has been set to: le-008
29
30
31
32                      Default Initial Parameter Values
33                              alpha =        194.5
34                                rho =            0    Specified
35                             beta_0 =        121.8
36                             beta_l =      297.029
37
38
39
40                                     Parameter Estimates
41
42                                                             95.0% Wald
43    Confidence Interval
44           Variable        Estimate        Std.  Err.     Lower Conf.  Limit
45    Upper  Conf.  Limit
46              alpha           132.71          54.1784             26.5217
47    238.897
48             beta_0            121.8          5.15188             111.702
49    131.898
50             beta_l          297.029          8.99387             279.401
51    314.656
52
53
54               Asymptotic Correlation Matrix of Parameter Estimates
55
56
57
58
59
                                      D-117        DRAFT - DO NOT CITE OR QUOTE

alpha
beta 0
beta 1
alpha
1
-1.4e-009
-1. le-008
beta 0
-1.4e-009
1
-0.76
beta 1
-1. le-008
-0.76
1

-------
1
2
3
4
5
6
7
/
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table of Data and Estimated Values of Interest
Dose N Obs Mean Obs Std Dev Est Mean Est Std Dev
Res .
_

0 3 124 8 122 11.5
0.25 3 194 16 196 11.5
0.5 3 269 13 270 11.5
1 3 420 17 419 11.5



Model Descriptions for likelihoods calculated


Model Al: Yij = Mu(i) + e(ij)
Var{e(ij) } = SigmaA2

Model A2: Yij = Mu(i) + e(ij)
Var{e(ij)} = Sigma (i)A2

Model R: Yi = Mu + e(i)
Var{e(i)} = SigmaA2


Likelihoods of Interest

Model Log (likelihood) DF AIC
Al -35.189802 5 80.379605
A2 -34.317788 8 84.635576
fitted -35.328976 2 74.657952
R -62.974684 2 129.949369

Test 1: Does response and/or variances differ among dose
levels
(A2 vs. R)
Test 2: Are Variances Homogeneous (Al vs A2 )
Test 3: Does the Model for the Mean Fit (Al vs. fitted)

Tests of Interest

Test -2*log (Likelihood Ratio) Test df p-value

Test 1 57.3138 6 <.0001
Test 2 1.74403 3 0.6272
Test 3 0.278348 2 0.8701

The p-value for Test 1 is less than .05. There appears
to be a
difference between response and/or variances among the
dose levels .
It seems appropriate to model the data

The p-value for Test 2 is greater than .05. A
homogeneous variance
model appears to be appropriate here
ChiA2


0.331
-0.309
-0.198
0.176















































D-118
DRAFT - DO NOT CITE OR QUOTE

-------
 1
 2
 3    The p-value for Test 3  is  greater than .05.   The  model
 4    chosen appears
 5    to adequately describe  the data
 6
 7
 8
 9     Benchmark Dose Computation
10    Specified effect =             1
11
12    Risk Type        =     Estimated  standard deviations  from the control  mean
13
14
15    Confidence level =          0.95
16
17                 BMD =      0.038784
18
19
20                BMDL =     0.0286028
21
22
                                      D-119        DRAFT - DO NOT CITE OR QUOTE

-------
 1    HASS_1981_BACT_MUT_BEP.OUT.txt
 2     ====================================================================
 3            Polynomial Model.  Revision: 2.2  Date: 9/12/2002
 4            Input Data File:  C:\DOCUMENTS AND SETTINGS\HCLYNCH\MY DOCUMENTS\PAH
 5    RPS\MODELING\HASS_1981_BACT_MUT_BEP.(d)
 6            Gnuplot Plotting File:  C:\DOCUMENTS AND SETTINGS\HCLYNCH\MY
 7    DOCUMENTS\PAH RPS\MODELING\HASS_1981_BACT_MUT_BEP.pit
 8                                              Wed Jul 06 13:42:38 2005
 9     ====================================================================
10
11     BMDS MODEL RUN
19    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
13
14       The form of the response function is:
15
16       Y[dose]  = beta_0 + beta_l*dose + beta_2*doseA2 + ...
17
18
19       Dependent variable = MEAN
20       Independent variable = COLUMN1
21       rho is set to 0
22       Signs of the polynomial coefficients are not restricted
23       A constant variance model is fit
24
25       Total number of dose groups = 4
26       Total number of records with missing values = 0
27       Maximum number of iterations = 250
28       Relative Function Convergence has been set to: le-008
29       Parameter Convergence has been set to: le-008
30
31
32
33                      Default Initial Parameter Values
34                              alpha =        117.5
35                                rho =            0   Specified
36                             beta_0 =       120.75
37                             beta_l =         77.5
38
39
40
41                                     Parameter Estimates
42
43                                                             95.0% Wald
44    Confidence Interval
45
46    U
47              alpha          98.6458           40.272             19.7142
48    1
49             beta0           120.75          4.19706             112.524
50    1
51             betal             77.5          7.66275             62.4813
52    9
53
54
55
56
57
58
59
60
                                      D-120        DRAFT - DO NOT CITE OR QUOTE
Variable
;r Conf. Limit
alpha
577
beta 0
976
beta 1
i!87
Estimate

98.6458
120.75

77.5
Asymptotic Correlation Matrix
alpha
alpha 1
beta 0 -8e-012
beta 1 l.le-011
beta 0
-8e-012
1
-0.73
Std. Err.

40.272
4.19706

7.66275
of Parameter
beta 1
l.le-011
-0.73
1
Lower Con

1
1

6
Estimates



-------
1
2
3
4
5
6
7
/
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table of Data and Estimated Values of Interest
Dose N Obs Mean Obs Std Dev Est Mean Est Std Dev
Res .
_

0 3 124 8 121 9.93
0.2 3 129 6 136 9.93
0.4 3 156 9 152 9.93
1 3 198 17 198 9.93



Model Descriptions for likelihoods calculated


Model Al: Yij = Mu(i) + e(ij)
Var{e(ij) } = SigmaA2

Model A2: Yij = Mu(i) + e(ij)
Var{e(ij)} = Sigma (i)A2

Model R: Yi = Mu + e(i)
Var{e(i)} = SigmaA2


Likelihoods of Interest

Model Log (likelihood) DF AIC
Al -32.165839 5 74.331679
A2 -30.272126 8 76.544252
fitted -33.549216 2 71.098432
R -47.594288 2 99.188576

Test 1: Does response and/or variances differ among dose
levels
(A2 vs. R)
Test 2: Are Variances Homogeneous (Al vs A2 )
Test 3: Does the Model for the Mean Fit (Al vs. fitted)

Tests of Interest

Test -2*log (Likelihood Ratio) Test df p-value

Test 1 34.6443 6 <.0001
Test 2 3.78743 3 0.2854
Test 3 2.76675 2 0.2507

The p-value for Test 1 is less than .05. There appears
to be a
difference between response and/or variances among the
dose levels .
It seems appropriate to model the data

The p-value for Test 2 is greater than .05. A
homogeneous variance
model appears to be appropriate here
ChiA2


0.567
-1.26
0.741
-0.0436















































D-121
DRAFT - DO NOT CITE OR QUOTE

-------
 1
 2
 3    The p-value for Test 3  is  greater than .05.   The model
 4    chosen appears
 5    to adequately describe  the data
 6
 7
 8
 9     Benchmark Dose Computation
10    Specified effect =             1
11
12    Risk Type        =     Estimated  standard deviations  from the control mean
13
14
15    Confidence level =          0.95
16
17                 BMD =      0.128156
18
19
20                BMDL =     0.0923937
21
22
                                      D-122        DRAFT - DO NOT CITE OR QUOTE

-------
 1    JOHNSEN_1997_BAC_MUT_BAP.OUT.txt
 2     ====================================================================
 3            Polynomial Model.  Revision: 2.2  Date: 9/12/2002
 4            Input Data File: C:\DOCUMENTS AND SETTINGS\HCLYNCH\MY DOCUMENTS\PAH
 5    RPS\MODELING\JOHNSEN_1997_BAC_MUT_BAP.(d)
 6            Gnuplot Plotting File:   C:\DOCUMENTS AND SETTINGS\HCLYNCH\MY
 7    DOCUMENTS\PAH RPS\MODELING\JOHNSEN_1997_BAC_MUT_BAP.pit
 8                                              Fri Jul 08 09:02:29 2005
 9     ====================================================================
10
11     BMDS MODEL RUN
19    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
13
14       The form of the response function is:
15
16       Y[dose]  = beta_0 + beta_l*dose + beta_2*doseA2 + ...
17
18
19       Dependent variable = MEAN
20       Independent variable = COLUMN1
21       rho is set to 0
22       Signs of the polynomial coefficients are not restricted
23       A constant variance model is fit
24
25       Total number of dose groups  = 3
26       Total number of records with missing values = 0
27       Maximum number of iterations = 250
28       Relative Function Convergence has been set to: le-008
29       Parameter Convergence has been set to: le-008
30
31
32
33                      Default Initial Parameter Values
34                              alpha =      70.2768
35                                rho =            0   Specified
36                             beta_0 =        115.5
37                             beta_l =         0.65
38
39
40
41                                     Parameter Estimates
42
43                                                             95.0% Wald
44    Confidence Interval
45           Variable         Estimate        Std. Err.     Lower Conf. Limit
46    Upper Conf. Limit
47              alpha          59.3512          27.9784             4.51449
48    114.188
49             beta_0            115.5          4.06035             107.542
50    123.458
51             beta_l             0.65         0.314513           0.0335651
52    1.26643
53
54
55               Asymptotic Correlation Matrix of Parameter Estimates
56
57
58
59
60
                                      D-123        DRAFT - DO NOT CITE OR QUOTE

alpha
beta 0
beta 1
alpha
1
-7.9e-010
-3.4e-012
beta 0
-7.9e-010
1
-0.77
beta 1
-3.4e-012
-0.77
1

-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
 Dose
Res .
     Table of Data and Estimated Values  of  Interest

            N    Obs Mean    Obs Std  Dev   Est  Mean    Est Std Dev   ChiA2
    0
   10
   20
        113
        127
        126
9.68
4.84
9.68
115
122
128
7.7
7.7
7.7
-0.562
1.12
-0.562
  Model Descriptions for likelihoods  calculated
 Model Al:        Yij
           Var{e(ij)}

 Model A2:        Yij
           Var{e(ij) }

 Model  R:         Yi
            Var{e(i) }
            Model
             Al
             A2
           fitted
              R
             Mu (i) + e (ij )
             SigmaA2

             Mu (i) + e (ij )
             Sigma(i)A2

             Mu + e (i)
             SigmaA2
            Likelihoods of Interest

            Log(likelihood)   DF
              -21.811395       4
              -21.026523       6
              -22.875626       2
              -24.653317       2
  AIC
51.622790
54.053045
49.751251
53.306634
 Test 1:  Does response and/or variances  differ  among dose
levels
          (A2 vs. R)
 Test 2:  Are Variances Homogeneous  (Al vs A2)
 Test 3:  Does the Model for the Mean  Fit (Al  vs.  fitted)
   Test

   Test 1
   Test 2
   Test 3
          Tests of Interest

-2*log(Likelihood Ratio)  Test df
            7.25359
            1.56974
            2.12846
p-value

  0.0266
  0.4562
  0.1446
The p-value for Test 1 is less than  .05.   There  appears
to be a
difference between response and/or variances  among  the
dose levels.
It seems appropriate to model the data

The p-value for Test 2 is greater than  .05.   A
homogeneous variance
model appears to be appropriate here
                                       D-124
                                               DRAFT - DO NOT CITE OR QUOTE

-------
 1
 2    The p-value for Test 3 is  greater than .05.   The model
 3    chosen appears
 4    to adequately describe the data
 5
 6
 7
 8     Benchmark Dose Computation
 9    Specified effect =             1
10
11    Risk Type        =     Estimated standard deviations from the control mean
12
13
14    Confidence level =          0.95
15
16                 BMD =       11.8523
17
18
19                BMDL =       6.27094
20
21
22
23
                                      D-125        DRAFT - DO NOT CITE OR QUOTE

-------
 1    D.6. MAMMALIAN MUTAGENICITY
 2    BARF_MUT_BAA.OUT.txt
 3     ====================================================================
 4            Multistage Model.  $Revision:  2.1 $ $Date: 2000/08/21 03:38:21 $
 5            Input Data File:  C:\DOCUMENTS AND SETTINGS\HCLYNCH\MY DOCUMENTS\PAH
 6    RPS\MODELING\BARF_MUT_BAA.(d)
 7            Gnuplot Plotting File:   C:\DOCUMENTS AND SETTINGS\HCLYNCH\MY
 8    DOCUMENTS\PAH RPS\MODELING\BARF_MUT_BAA.pit
 9                                              Thu Jun 30 12:46:38 2005
10     ====================================================================
11
12     BMDS  MODEL RUN
13    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
14
15       The form of the probability function is:
16
17       P[response]  = background + (1-background)*[1-EXP(
18    -betal*doseAl-beta2*doseA2-beta3*doseA3)]
19
20       The parameter betas are  restricted to be positive
21
22
23       Dependent variable = COLUMN2
24       Independent variable = COLUMN1
25
26     Total number of observations =  5
27     Total number of records with missing values = 0
28     Total number of parameters in model  = 4
29     Total number of specified  parameters = 0
30     Degree of polynomial = 3
31
32
33     Maximum number of iterations =  250
34     Relative Function Convergence has been set to:  le-008
35     Parameter Convergence has  been  set to: le-008
36
37
38
39                      Default Initial Parameter Values
40                         Background  = 3.89426e-006
41                            Beta(l)  = 3.46216e-007
42                            Beta(2)  =            0
43                            Beta(3)  = 1.93939e-012
44    **** WARNING:   Completion code = -2.   Optimum not found. Trying new starting
45    pont****
46
47
48
49               Asymptotic Correlation Matrix of Parameter Estimates
50
51               ( *** The model  parameter(s)  -Background    -Beta(2)    -Beta(3)
52                     have been  estimated  at a boundary point, or have been
53    specified by the user,
54                     and do not appear in the correlation matrix )
55
56                    Beta(l)
57
58       Beta(l)             1
59
                                      D-126        DRAFT - DO NOT CITE OR QUOTE

-------
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Parameter Estimates

Variable Estimate Std. Err.
Background 0 NA
Beta(l) 4.34385e-007 5.43792e-006
Beta (2) 0 NA
Beta (3) 0 NA

NA - Indicates that this parameter has hit a bound
implied by some inequality constraint and thus
has no standard error.



Analysis of Deviance Table

Model Log (likelihood) Deviance Test DF
Full model -1545.82
Fitted model -1548.6 5.57201 4
Reduced model -1597.17 102.713 4

AIC: 3099.21


Goodness of Fit

Dose Est. Prob. Expected Observed Size
i: 1
0.0000 0.0000 0.000 0 1000000
i: 2
20.0000 0.0000 8.688 12 1000000
i: 3
50.0000 0.0000 21.719 29 1000000
i: 4
100.0000 0.0000 43.438 34 1000000
i: 5
150.0000 0.0001 65.156 64 1000000

Chi-square = 5.77 DF = 4 P-value = 0.21


Benchmark Dose Computation

Specified effect = le-005

Risk Type = Extra risk

Confidence level = 0.95

BMD = 23. 0212

**** WARNING: Completion code = -2. Optimum not found.
point****

**** WARNING 0: Completion code = -2 trying new start***

**** WARNING 1: Completion code = -2 trying new start***
















P-value

0.2335
<.0001






ChiA2 Res.

0.000

0.381

0.335

-0.217

-0.018

66












Trying new starting


*

*
D-127
DRAFT - DO NOT CITE OR QUOTE

-------
 1
 2    ****  WARNING 2:   Completion code = -2 trying new start****
 3
 4    ****  WARNING 3:   Completion code = -2 trying new start****
 5
 6    ****  WARNING 4:   Completion code = -2 trying new start****
 7
 8    ****  WARNING 5:   Completion code = -2 trying new start****
 9
10    ****  WARNING 6:   Completion code = -2 trying new start****
11
12    ****  WARNING 7:   Completion code = -2 trying new start****
13
14    ****  WARNING 8:   Completion code = -2 trying new start****
15
16    ****  WARNING 9:   Completion code = -2 trying new start****
17
18    ****  WARNING:   Completion code = -2.   Optimum not found.  Trying new starting
19    point****
20
21    ****  WARNING 0:   Completion code = -2 trying new start****
22
23    ****  WARNING 1:   Completion code = -3 trying new start****
24
25    ****  WARNING 2:   Completion code = -3 trying new start****
26
27    ****  WARNING 3:   Completion code = -3 trying new start****
28
29    ****  WARNING 4:   Completion code = -3 trying new start****
30
31    ****  WARNING 5:   Completion code = -3 trying new start****
32
33    ****  WARNING 6:   Completion code = -2 trying new start****
34
35    ****  WARNING 7:   Completion code = -3 trying new start****
36
37    ****  WARNING 8:   Completion code = -3 trying new start****
38
39    ****  WARNING 9:   Completion code = -3 trying new start****
40
41
42    Warning:   completion code still negative
43    BMDL  did  not converge for BMR = 0.000010
44
45    Program execution is stopped
46
                                      D-128        DRAFT - DO NOT CITE OR QUOTE

-------
 1    BARF_MUT_BAP.OUT.txt
 2     ====================================================================
 3            Multistage Model.  $Revision: 2.1 $ $Date: 2000/08/21 03:38:21 $
 4            Input Data File:  C:\DOCUMENTS AND SETTINGS\HCLYNCH\MY DOCUMENTS\PAH
 5    RPS\MODELING\BARF_MUT_BAP.(d)
 6            Gnuplot Plotting File:   C:\DOCUMENTS AND SETTINGS\HCLYNCH\MY
 7    DOCUMENTS\PAH RPS\MODELING\BARF_MUT_BAP.pit
 8                                              Thu Jun 30 12:40:17 2005
 9     ====================================================================
10
11     BMDS MODEL RUN
19    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
13
14       The form of the probability function is:
15
16       P[response]  = background + (1-background)*[1-EXP(
17    -betal*doseAl-beta2*doseA2)]
18
19       The parameter betas are restricted to be positive
20
21
22       Dependent variable = COLUMN2
23       Independent variable = COLUMN1
24
25     Total number of observations =  4
26     Total number of records with missing values = 0
27     Total number of parameters in model = 3
28     Total number of specified parameters = 0
29     Degree of polynomial = 2
30
31
32     Maximum number of iterations =  250
33     Relative Function Convergence has been set to:  le-008
34     Parameter Convergence has been  set to: le-008
35
36
37
38                      Default Initial Parameter Values
39                         Background  = 1.39884e-006
40                            Beta(l)  = 5.34042e-006
41                            Beta(2)  =            0
42
43
44               Asymptotic Correlation Matrix of Parameter Estimates
45
46               ( *** The model parameter(s)  -Background    -Beta(2)
47                     have been estimated at a boundary point, or have been
48    specified by the user,
49                     and do not appear in the correlation matrix )
50
51                    Beta(l)
52
53       Beta(l)             1
54
55
56
57                              Parameter Estimates
58
59           Variable           Estimate             Std. Err.
60         Background                   0               NA


                                      D-129       DRAFT - DO NOT CITE OR QUOTE

-------
1
2
3
4
5
6
7
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Beta(l) 5.43367e-006
Beta(2) 0

NA - Indicates that this parameter
2.68102e-005
NA

has hit a bound




implied by some inequality constraint and thus
has no standard error.


Analysis of

Model Log (likelihood)
Full model -3273.08
Fitted model -3273.96
Reduced model -3395.25

AIC: 6549.92


Goodness of




Deviance Table

Deviance Test DF

1.75092 3
244.327 3




Fit

Dose Est. Prob. Expected Observed Size


i: 1
0.0000 0.0000 0.000
i: 2
10.0000 0.0001 54.335
i: 3
20.0000 0.0001 108.668
i: 4
30.0000 0.0002 162.997

Chi-square = 1.78 DF = 3


Benchmark Dose Computation

Specified effect = le-005

Risk Type = Extra risk

Confidence level = 0.95

BMD = 1. 84039

**** WARNING: Completion code = -3
point****

**** WARNING 0: Completion code =

**** WARNING 1: Completion code =

**** WARNING 2: Completion code =

**** WARNING 3: Completion code =

**** WARNING 4: Completion code =

**** WARNING 5: Completion code =




0 1000000

51 1000000

120 1000000

155 1000000






P-value

0.6257
<.0001






ChiA2 Res.



0.000

-0.061

0.104

-0.049

P-value = 0.6195












Optimum not found.


-3 trying new start***

-3 trying new start***

-3 trying new start***

-3 trying new start***

-3 trying new start***

-3 trying new start***













Trying new starting


*

*

*

*

*

*

D-130
DRAFT - DO NOT CITE OR QUOTE

-------
 1    **** WARNING  6:   Completion  code = -3 trying new start****
 2
 3    **** WARNING  7:   Completion  code = -3 trying new start****
 4
 5    **** WARNING  8:   Completion  code = -3 trying new start****
 6
 7    **** WARNING  9:   Completion  code = -3 trying new start****
 8
 9    **** WARNING:   Completion  code = -3.  Optimum not  found. Trying new starting
10    point****
11
12    **** WARNING  0:   Completion  code = -1 trying new start****
13
14    **** WARNING  1:   Completion  code = -1 trying new start****
15
16    **** WARNING  2:   Completion  code = -1 trying new start****
17
18               BMDL  =         1.68248
19
                                      D-131        DRAFT - DO NOT CITE OR QUOTE

-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
BARF MUT CH.OUT.txt
        Multistage Model. $Revision: 2.1 $ $Date: 2000/08/21  03:38:21  $
        Input Data File: C:\DOCUMENTS AND SETTINGS\HCLYNCH\MY DOCUMENTS\PAH
RPS\MODELING\BARF_MUT_CH.(d)
        Gnuplot Plotting File:  C:\DOCUMENTS AND SETTINGS\HCLYNCH\MY
DOCUMENTS\PAH RPS\MODELING\BARF_MUT_CH.pit
                                          Thu Jun 30  12:48:57 2005
 BMDS MODEL RUN


   The form of the probability function is:

   P[response] = background +  (1-background)*[1-EXP(
-betal*doseAl)]

   The parameter betas are restricted to be positive
   Dependent variable = COLUMN2
   Independent variable = COLUMN1

 Total number of observations = 3
 Total number of records with missing values =  0
 Total number of parameters in model = 2
 Total number of specified parameters = 0
 Degree of polynomial = 1
 Maximum number of iterations = 250
 Relative Function Convergence has been set to:  le-008
 Parameter Convergence has been set to: le-008
                  Default Initial Parameter Values
                     Background = 2.60526e-006
                        Beta(l) = 5.02638e-007
           Asymptotic Correlation Matrix of Parameter Estimates

           (  *** The model parameter(s)  -Background
                 have been estimated at a boundary point,  or  have  been
specified by the user,
                 and do not appear in the correlation matrix  )
   Beta (1)
Beta (1)

      1
                          Parameter Estimates
       Variable
     Background
        Beta(l)
          Estimate
                  0
       6.14293e-007
    Std.  Err.
       NA
1.93539e-005
                                      D-132
                                               DRAFT - DO NOT CITE OR QUOTE

-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
NA - Indicates that this parameter has hit a bound
     implied by some inequality constraint and thus
     has no standard error.
                        Analysis of Deviance Table
       Model
     Full model
   Fitted model
  Reduced model

           AIC:
    Log(likelihood)
         -504.191
          -505.38
         -522.575

          1012.76
      Deviance  Test DF
              P-value
         2.37752
         36.7681
                   0.3046
                  <.0001
     Dose
       Goodness  of  Fit

Est._Prob.    Expected    Observed     Size     ChiA2 Res,
i: 1
    0.0000
i: 2
   20.0000
i: 3
   50.0000

 Chi-square =
 0.0000

 0.0000

 0.0000

      2.53
 0.000

12.286

30.714

DF = 2
 0     1000000       0.000

17     1000000       0.384

26     1000000      -0.153

P-value = 0.2819
   Benchmark Dose Computation

Specified effect =         le-005

Risk Type        =      Extra risk

Confidence level =           0.95

             BMD =         16.279

**** WARNING:  Completion code = -1.  Optimum not found. Trying new starting
point****

**** WARNING 0:  Completion code = -1 trying new start****

**** WARNING 1:  Completion code = -1 trying new start****

**** WARNING 2:  Completion code = -1 trying new start****

**** WARNING 3:  Completion code = -1 trying new start****

**** WARNING 4:  Completion code = -1 trying new start****

**** WARNING 5:  Completion code = -1 trying new start****

**** WARNING 6:  Completion code = -1 trying new start****

**** WARNING 7:  Completion code = -1 trying new start****
                                      D-133
                                               DRAFT - DO NOT CITE OR QUOTE

-------
 1    ****  WARNING 8:   Completion code = -1 trying new start****
 2
 3    ****  WARNING 9:   Completion code = -1 trying new start****
 4
 5    ****  WARNING:   Completion code = -1.   Optimum not found.  Trying new starting
 6    point****
 7
 8    ****  WARNING 0:   Completion code = -3 trying new start****
 9
10    ****  WARNING 1:   Completion code = -3 trying new start****
11
12    ****  WARNING 2:   Completion code = -3 trying new start****
13
14    ****  WARNING 3:   Completion code = -3 trying new start****
15
16    ****  WARNING 4:   Completion code = -3 trying new start****
17
18    ****  WARNING 5:   Completion code = -3 trying new start****
19
20    ****  WARNING 6:   Completion code = -3 trying new start****
21
22    ****  WARNING 7:   Completion code = -3 trying new start****
23
24    ****  WARNING 8:   Completion code = -3 trying new start****
25
26    ****  WARNING 9:   Completion code = -3 trying new start****
27
28
29    Warning:   completion code still negative
30    BMDL  did not converge for BMR = 0.000010
31
32    Program execution is stopped
33
                                      D-134        DRAFT - DO NOT CITE OR QUOTE

-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
BARF MUT FA.OUT.txt
        Multistage Model. $Revision: 2.1 $ $Date: 2000/08/21  03:38:21  $
        Input Data File: C:\DOCUMENTS AND SETTINGS\HCLYNCH\MY DOCUMENTS\PAH
RPS\MODELING\BARF_MUT_FA.(d)
        Gnuplot Plotting File:  C:\DOCUMENTS AND SETTINGS\HCLYNCH\MY
DOCUMENTS\PAH RPS\MODELING\BARF_MUT_FA.pit
                                          Thu Jun 30  12:43:11 2005
 BMDS MODEL RUN


   The form of the probability function is:

   P[response] = background +  (1-background)*[1-EXP(
-betal*doseAl)]

   The parameter betas are restricted to be positive
   Dependent variable = COLUMN2
   Independent variable = COLUMN1

 Total number of observations = 3
 Total number of records with missing values = 0
 Total number of parameters in model = 2
 Total number of specified parameters = 0
 Degree of polynomial = 1
 Maximum number of iterations = 250
 Relative Function Convergence has been set to:  le-008
 Parameter Convergence has been set to: le-008
                  Default Initial Parameter Values
                     Background =  6.6658e-007
                        Beta(l) = 2.50006e-006
           Asymptotic Correlation Matrix of Parameter Estimates

           (  *** The model parameter(s)  -Background
                 have been estimated at a boundary point,  or have  been
specified by the user,
                 and do not appear in the correlation matrix )
   Beta (1)
Beta (1)

      1
                          Parameter Estimates
       Variable
     Background
        Beta(l)
          Estimate
                  0
       2.56672e-006
    Std. Err.
       NA
4.49565e-005
                                      D-135
                                               DRAFT - DO NOT CITE OR QUOTE

-------
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
NA - Indicates that this parameter has hit a bound
     implied by some inequality constraint and thus
     has no standard error.
                        Analysis of Deviance Table
       Model
     Full model
   Fitted model
  Reduced model

           AIC:
    Log(likelihood)
         -856.204
         -856.255
         -890.913

          1714.51
      Deviance  Test DF
              P-value
           0.103
          69.419
                   0.9498
                  <.0001
     Dose
       Goodness  of  Fit

Est._Prob.    Expected    Observed     Size     ChiA2 Res,
i: 1
    0.0000
i: 2
   10.0000
i: 3
   20.0000

 Chi-square =
 0.0000

 0.0000

 0.0001

      0.10
 0.000

25.667

51.333

DF = 2
 0     1000000       0.000

27     1000000       0.052

50     1000000      -0.026

P-value = 0.9494
   Benchmark Dose Computation

Specified effect =         le-005

Risk Type        =      Extra risk

Confidence level =           0.95

             BMD =        3. 89604

**** WARNING:  Completion code = -1.  Optimum not found. Trying new  starting
point****

**** WARNING 0:  Completion code = -1 trying new start****

**** WARNING 1:  Completion code = -5 trying new start****

            BMDL =              0
                                      D-136
                                               DRAFT - DO NOT CITE OR QUOTE

-------
 1    BARF_MUT_TPHEN.OUT.txt
 2     ====================================================================
 3            Multistage Model.  $Revision: 2.1 $ $Date: 2000/08/21 03:38:21 $
 4            Input Data File:  C:\DOCUMENTS AND SETTINGS\HCLYNCH\MY DOCUMENTS\PAH
 5    RPS\MODELING\BARF_MUT_TPHEN.(d)
 6            Gnuplot Plotting File:   C:\DOCUMENTS AND SETTINGS\HCLYNCH\MY
 7    DOCUMENTS\PAH RPS\MODELING\BARF_MUT_TPHEN.pit
 8                                              Thu Jun 30 12:52:56 2005
 9     ====================================================================
10
11     HMDS MODEL RUN
19    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
13
14       The form of the probability function is:
15
16       P[response]  = background + (1-background)*[1-EXP(
17    -betal*doseAl-beta2*doseA2)]
18
19       The parameter betas are restricted to be positive
20
21
22       Dependent variable = COLUMN2
23       Independent variable = COLUMN1
24
25     Total number of observations =  4
26     Total number of records with missing values = 0
27     Total number of parameters in model = 3
28     Total number of specified parameters = 0
29     Degree of polynomial = 2
30
31
32     Maximum number of iterations =  250
33     Relative Function Convergence has been set to:  le-008
34     Parameter Convergence has been  set to: le-008
35
36
37
38                      Default Initial Parameter Values
39                         Background  = 9.99937e-007
40                            Beta(l)  = 1.74289e-007
41                            Beta(2)  =            0
42
43
44               Asymptotic Correlation Matrix of Parameter Estimates
45
46               ( *** The model parameter(s)  -Background    -Beta(2)
47                     have been estimated at a boundary point, or have been
48    specified by the user,
49                     and do not appear in the correlation matrix )
50
51                    Beta(l)
52
53       Beta(l)             1
54
55
56
57                              Parameter Estimates
58
59           Variable           Estimate             Std. Err.
60         Background                   0               NA


                                      D-137       DRAFT - DO NOT CITE OR QUOTE

-------
1
2
3
4
5
6
7
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Beta(l) 1.85717e-007
Beta(2) 0

NA - Indicates that this parameter
4.42148e-006
NA

has hit a bound




implied by some inequality constraint and thus
has no standard error.


Analysis of

Model Log (likelihood)
Full model -755.63
Fitted model -755.773
Reduced model -781.782

AIC: 1513.55


Goodness of




Deviance Table

Deviance Test DF

0.2868 3
52.3039 3




Fit

Dose Est. Prob. Expected Observed Size


i: 1
0.0000 0.0000 0.000
i: 2
50.0000 0.0000 9.286
i: 3
100.0000 0.0000 18.572
i: 4
200.0000 0.0000 37.143

Chi-square = 0.29 DF = 3


Benchmark Dose Computation

Specified effect = le-005

Risk Type = Extra risk

Confidence level = 0.95

BMD = 53. 8457

**** WARNING: Completion code = -2
point****

**** WARNING 0: Completion code =

**** WARNING 1: Completion code =

**** WARNING 2: Completion code =

**** WARNING 3: Completion code =

**** WARNING 4: Completion code =

**** WARNING 5: Completion code =




0 1000000

10 1000000

20 1000000

35 1000000






P-value

0.9625
<.0001






ChiA2 Res.



0.000

0.077

0.077

-0.058

P-value = 0.9622












Optimum not found.


-2 trying new start***

-2 trying new start***

-2 trying new start***

-2 trying new start***

-2 trying new start***

-2 trying new start***













Trying new starting


*

*

*

*

*

*

D-138
DRAFT - DO NOT CITE OR QUOTE

-------
 1    ****  WARNING 6:   Completion code = -2 trying new start****
 2
 3    ****  WARNING 7:   Completion code = -2 trying new start****
 4
 5    ****  WARNING 8:   Completion code = -2 trying new start****
 6
 7    ****  WARNING 9:   Completion code = -2 trying new start****
 8
 9    ****  WARNING:   Completion code = -2.   Optimum not found.  Trying new starting
10    point****
11
12    ****  WARNING 0:   Completion code = -2 trying new start****
13
14    ****  WARNING 1:   Completion code = -5 trying new start****
15
16    ****  WARNING 2:   Completion code = -2 trying new start****
17
18    ****  WARNING 3:   Completion code = -2 trying new start****
19
20    ****  WARNING 4:   Completion code = -2 trying new start****
21
22    ****  WARNING 5:   Completion code = -2 trying new start****
23
24    ****  WARNING 6:   Completion code = -2 trying new start****
25
26    ****  WARNING 7:   Completion code = -5 trying new start****
27
28    ****  WARNING 8:   Completion code = -2 trying new start****
29
30    ****  WARNING 9:   Completion code = -5 trying new start****
31
32
33    Warning:   completion code still negative
34    BMDL  did  not converge for BMR = 0.000010
35
36    Program execution is stopped
37
                                      D-139        DRAFT - DO NOT CITE OR QUOTE

-------
 1    RAVEH_HUB_MUT_BAP.OUT.txt
 2     ====================================================================
 3            Multistage Model. $Revision:  2.1 $ $Date: 2000/08/21 03:38:21 $
 4            Input Data File:  C:\DOCUMENTS AND SETTINGS\HCLYNCH\MY DOCUMENTS\PAH
 5    RPS\MODELING\RAVEH_HUB_MUT_BAP.(d)
 6            Gnuplot Plotting File:   C:\DOCUMENTS AND SETTINGS\HCLYNCH\MY
 7    DOCUMENTS\PAH RPS\MODELING\RAVEH_HUB_MUT_BAP.pit
 8                                              Wed Jun 29 12:15:41 2005
 9     ====================================================================
10
11     BMDS MODEL RUN
19    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
13
14       The form of the probability function is:
15
16       P[response]  = background + (1-background)*[1-EXP(
17    -betal*doseAl)]
18
19       The parameter betas  are restricted to be positive
20
21
22       Dependent variable = COLUMN2
23       Independent variable = COLUMN1
24
25     Total number of observations =  3
26     Total number of records with missing values = 0
27     Total number of parameters in model  = 2
28     Total number of specified parameters = 0
29     Degree of polynomial = 1
30
31
32     Maximum number of iterations =  250
33     Relative Function Convergence has  been set to:  le-008
34     Parameter Convergence  has been  set to: le-008
35
36
37
38                      Default Initial Parameter Values
39                         Background  =            0
40                            Beta(l)  =   0.00102082
41    **** WARNING:  Completion code = -2.   Optimum not found. Trying new starting
42    pont****
43
44    **** WARNING 0:   Completion code =  -2 trying new start****
45
46    **** WARNING 1:   Completion code =  -2 trying new start****
47
48    **** WARNING 2:   Completion code =  -2 trying new start****
49
50    **** WARNING 3:   Completion code =  -2 trying new start****
51
52    **** WARNING 4:   Completion code =  -2 trying new start****
53
54    **** WARNING 5:   Completion code =  -2 trying new start****
55
56    **** WARNING 6:   Completion code =  -2 trying new start****
57
58    **** WARNING 7:   Completion code =  -2 trying new start****
59
60    **** WARNING 8:   Completion code =  -2 trying new start****


                                      D-140