-------
-------
-------
CONTENTS
Tables vi
Preface • vi
Abstract ix
Authors , Contributors , and Reviewers x
1. INTRODUCTION
1.1. BACKGROUND 1
1.2. METHODOLOGY 3
2 . LITERATURE SEARCH AND INFORMATION SOIRCES 5
2.1. SEARCH STRATEGY 6
2.2. SELECTION OF APPROPRIATE STUDIES 7
3. APPROACH TO DATA EVALUATION 9
3.1. QUALITATIVE PHASE: WEIGHT OF EVIDENCE OF
CARCINOGEN I CITY
3.1.1. Assessment of Evidence for Carcinogenicity from
Studies in Humans
3.1.2. Assessment of Evidence for Carcinogenicity from
Studies in Experimental Animals ........................ 1
3.1.3. Categorization of Overall Weight of Evidence for
Human Carcinogenicity .................................
t. j
3.2. QUANTITATIVE PHASE: ESTIMATION OF CARCINOGENIC POTENCY 16
3.2.1. Model Selection for Analysis of Dose Response
Data 16
3.2.2. Selection of Bioassay Response Data to be Used in
the Model to Calculate Potency 13
3.2.3. Adjustment (Transformation) of Dose Data 1Q
3.2.3.1. Calculation of Dose in mg/kg/day from
Doses Expressed as Dietary
Concentrations 20
111
-------
CONTENTS (continued)
3.2.3.2. Calculation of Dose in mg/kg/day from
Doses Expressed as Water
Concentrations 21
3.2.3.3. Calculation of Dose in mg/kg/day from
Doses Expressed as Air
Concentrations 21
3.2.3.4. Adjustment for Non-Continuous Exposure 25
3.2.3.5. Adjustment for Absorption, Distribution,
Metabolism, and Excretion 25
3.2.4. Calculation of the Human Potency Factor (F) 26
3.2.4.1. Animal Potency 26
3.2.4.2. Adjustment for Less Than Lifetime
Studies 26
3.2.4.3. Human Potency (F) 28
3.2.5. Grouping of Chemicals Based on Carcinogenic
Potency 2?
3.3. OVERALL HAZARD RANKING BASED ON COMBINED QUALITATIVE AND
QUANTITATIVE ASSESSMENTS 30
3.3.1. Use of Chemical and Environmental Fate and Trans-
formation Data in Hazard Ranking of Metals and
Their Salts 33
3.3.2. Special Problems in Hazard Ranking of Chemicals
Associated with Multimedia Exposure in Humans' 33
4. SUMMARY 34
5. REFERENCES 35
APPENDIX '-' HAZARD RANKING OF POTENTIAL CARCINOGENS 36
iv
-------
TABLES
2-1 Sources of bibliographic and numerical information for suspect
carcinogens 6
3-1 Tentative weight of evidence based on human and animal evidence 14
3-2 Sample table for the derivation of potency factor (F) 27
3-3 Hazard rankings scheme for reportable quantities under CERCLA 31
-------
PREFACE
This report describes the technical methodology the U.S. Environmental
Protection Agency (EPA or the Agency) has used in. developing a hazard ranking
for potential carcinogens in order to adjust reportable quantities (RQs) under
Section 102 of the Comprehensive Environmental Response, Compensation, and
Liability Act of 1980 (CERCLA). The RQ adjustment methodology is based, in
part, on the methodology used to establish RQs pursuant to Section 311 of the
Clean Water Act. Details of the methodology are given in the document
"Technical Background Document to Support Rulemaking Pursuant to CERCLA Section
102," Volume 1, March 1985, and in the final rule published on April i, 19S5
(U.S. EPA, 1985 a,b).
In deciding whether to assign primary criteria RQs for potential
carcinogens at all five RQ levels, the Agency examined the special properties
associated with these substances and evaluated them in light of the Ager.cv's
chronic toxicity methodology. The Agency decided not to use the two highest RC
levels, 1000 and 5000 pounds, for several reasons, all of which are explained
in the main body of the report "Technical Background Document to Support
Ruiemaking Pursuant to. CERCLA Section 102," Volume 3 (U.S. EPA, 1989.;.
As a consequence of the Agency's decision to adopt a 100-pound maximum ?.Z
for potential carcinogens, the Human Health Assessment Group was requested to
rank potential carcinogens on a three-tier scale (high, medium, and low) that
corresponds to RQ levels of 1, 10, and 100 pounds. This document describes the
methodology for using the weight of evidence and potency factor (U.S. EPA.
1986) to determine RQs based on the primary criterion of potential
carcinogenicity.
vi
-------
ABSTRACT
The Agency's Human Health Assessment Group (HHAG) has developed a
methodology for ranking CERCLA hazardous substances for the purpose of
establishing reportable quantities (RQs) based on the primary criteria of
potential carcinogenicity. The methodology combines the weight of evidence and
potency factor to determine a hazard ranking of high, medium, or low which
corresponds to an RQ of 1, 10, or 100 pounds, respectively.
An appendix is included which lists 194 compounds that were evaluated for
potential carcinogenicity along with their respective weight-of-evidence
categories, potency factors, and hazard rankings. Profiles.for each of these
19i chemicals are available as separate documents.
VI 1
-------
AUTHORS, CONTRIBUTORS, AND REVIEWERS
EPA's Office of Health and Environmental Assessment (OHEA) is responsible
for this methodology document. OHEA's Human Health Assessment Group developed
the methodology described in this document. The project manager, with overall
responsibility for coordinating and directing the development and production of
this document is:
Vincent James Cogliano, Ph.D.
U.S. Environmental Protection Agency
Office of Health and Environmental Assessment
Washington, D.C.
The first draft was prepared by Syracuse Research Corporation under
contract number 68-03-3112. Ms. Helen Ball and Mr. John Risher served as
project officers and Dr. Margaret M.L. Chu and Mr. Hugh L. Spitzer served as
technical project monitors. Work was completed by C-E Environmental, Inc.
under contract number 68-03-3452. Mr. Richard Field and Ms. Ivette 0. Vega
served as project officers and Dr. Vincent James Cogliano, Dr. Aparna M.
Koppikar, and Dr. K. Jack Kooyoomjian served as technical project monitors.
The authors of this methodology document are listed below.
Vincent James Cogliano. Ph.D.
Aparna M. Koppikar, M.D., D.P.H., D.I.H., Ph.D.
U.S. Environmental Protection Agency
Of f ic«-of Health and Environmental Assessment
Washington, O.C.
James M. Conis
Steven C. Gibson, M.S.
Jeffrey S. Gift, Ph.D.
vi 11
-------
Alan W. Messing, Ph.D.
Gregory R. Ricci
Bartholomew L. Tuffly
C-E Environmental, Inc.
Washington, D.C.
The following individuals provided peer review of this document or ear lie;
drafts of this document.
Donald Barnes
Office of Pesticides and Toxic Substances
U.S. Environmental Protection Agency
Washington, D.C.
Judith Bellin
Office of Solid Waste
U.S. Environmental Protection Agency
Washington, D.C.
Barbara Davis
Office of Waste Programs Enforcement
U.S. Environmental Protection Agency
Washington. D.C.
Thomas Gleason
Office of Health and Environmental Assessment
U.S. Environmental Protection Agency
Washington, D.C.
Barbara H. Hostage
Emergency Response Division
U.S. Environmental Protection Agency
Washington, D.C.
K. Jack Kooyoomjian
Science Advisory Board
U.S. Environmental Protection Agency
Washington, D.C.
Robert McGaughy
Office of Health and Environmental Assessment
U.S. Environmental Protection Agency
Washington, D.C.
John Riley
Emergency Response Division
U.S. Environmental Protection Agency
Washington, D.C.
-------
Patrick Tobin
Office of Water Regulations and Standards
U.S. Environmental Protection Agency
Washington, D.C.
Ivette Vega
Emergency Response Division
U.S. Environmental Protection Agency
Washington, D.C.
Dr. Herbert Cornish
School of Public Health
University of Michigan
Ann Arbor, Michigan
Dr. Rolf Hartung
School of Public Health
University of Michigan
Ann Arbor, Michigan
Dr. Benjamin Van Duuren
Institute of Environmental Medicine
New York University Medical Center
New York, New York
-------
1. INTRODUCTION
1.1. BACKGROUND
This report describes che technical methodology the Agency has used in
developing a hazard ranking for potential carcinogens in order to adjust
reportable quantities (RQs) under Section 102 of the Comprehensive
Environmental Response, Compensation, and Liability Act of 1980 (CERCLA).
Section 103 of CERCLA requires immediate notification to the National Response
Center by any person in charge of a vessel or facility who releases an a- :.-
of a hazardous substance equal to or greater than its RQ. Under CERCLA •; • ion
102(b), the RQ of any hazardous substance designated in Section 101(1^"! is 1
pound unless a different RQ has been established pursuant to Section 311(b)(i)
of the Federal Water Pollution Control Act. Under Section 102(a) these
statutory RQs may be adjusted by regulations establishing different quantities
to be reported upon release of a hazardous substance. Section 102(a also
gives the U.S. Environmental Protection Agency (EPA or the Agency,> authority re
establish a single RQ for each hazardous substance, regardless of the
environmental medium into which the substance is released.
The RQ adjustment methodology is one with which the regulated community is
familiar and is based, in part, on the methodology used to establish RQs
pursuant to Section 311 of the Clean Water Act (CVA). Details of the
methodology are given in the technical background document (U.S. EPA, 1933a..
and in the final rule (U.S. EPA, 1985b). The methodology begins with an
evaluation of the intrinsic physical, chemical, and toxicological properties
associated with each hazardous substance. The intrinsic properties evaluated,
called primary criteria, are: aquatic toxicity, mammalian toxicity (oral,
-1-
-------
dermal, and inhalation), ignitability, reactivity, chronic toxicity, and
potential carcinogenicity. The Agency ranks each intrinsic property (ocher
than potential carcinogenicity, which is discussed below) on a five-tier scale.
associating a specific range of values on each scale with a particular RQ
value. This five-tier scale uses the RQ levels of 1, 10, 100, 1000, and 5000
pounds, which were originally established pursuant to the CWA Section 311.
Each hazardous substance receives several tentative RQ values based on its
particular properties. The lowest of all of the tentative RQs becomes the
primary criteria RQ for that hazardous substance. The primary criteria RQ can
then be raised by one level using biodegradability, hydrolysis, and photolvsis
as secondary criteria. The Agency has determined that no potential carcinogen
shall be assigned a primary criteria RQ above 100 pounds. The Agency has
always regarded potential carcinogens with special concern and in its
regulatory actions has sought to minimize carcinogenic risks. This concern is
justified by scientific factors particular to cancer:
o It has not been demonstrated that there is a threshold level of
exposure below which potential carcinogens do not present sorr.e risk
of cancer. Therefore, a release of any amount of a potential
carcinogen represents an increased risk of cancer to the exposed
population. This is in contrast with most other toxic effects, for
which thresholds can be demonstrated (U.S. EPA, 1987 pp. 8140, 31-5).
o Cancer risks are considered to be cumulative. A number of small
releases can be as serious as a single large release (U.S. EPA, 1937
pp. 8140, 8146).
•2-
-------
o Cancer is not immediately manifested. There is a latent period
between exposure to a carcinogen and the manifestation of cancer that
makes it impossible to directly observe carcinogenic risks from
substances newly released into the environment (U.S. EPA, 1987 pp.
8140, 8146). This is in contrast to acute toxic effects, which are
more immediately manifested.
In deciding whether to assign primary criteria RQs for potential
carcinogens at all five RQ levels, the Agency examined the special properties
associated with these substances and evaluated them in light of the Agency's
chronic toxicity methodology. The Agency decided not to use the two highest RQ
levels, 1000 and 5000 pounds, for several reasons, all of which are explained
in the main body of the report "Technical Background Document to Support
Rulemaking Pursuant to CERCLA Section 102," Volume 3, (U.S. EPA, 1989).
As a consequence of the Agency's decision to adopt a 100-pound rr.aximum RQ
for potential carcinogens, the Human Health Assessment Group (HHAG) was
requested to rank potential carcinogens on a three-tier scale (high, medium.
and low) that corresponds to RQ levels of 1, 10, and 100 pounds.
1.2. METHODOLOGY
The HHAG developed a methodology for ranking potential carcinogens based
originally on a combination of the International Agency for Research on
Cancer's (IARC) weight-of-evidence scheme and the HHAG's potency factor. This
methodology had been reviewed and was described in a previous draft of this
report. On September 24, 1986, the Agency published its "Guidelines for
Carcinogen Risk Assessment" (U.S. EPA, 1986), which refined the IARC
weight-of-evidence criteria. The EPA Reportable Quantity Work Group determined
on June 4, 1985, that the ranking methodology should be revised to be
-3-
-------
consistent with the Agency's final guidelines, and should include che Agency's
new weight-of-evidence criteria. This has been done in the methodology
described in this report.
The following sections of this report present the objectives and
methodology applied in arriving at a carcinogenic hazard assessment for each of
the hazardous substances under study. The major findings are summarized in
tabular form in a separate, but attached, appendix (Hazard Ranking of Potential
Carcinogens). A more detailed discussion of the available studies, weight-of-
evidence determinations, potency factor assignments for each potential
carcinogen under study, and a bibliography of other pertinent references is
contained in a profile for each chemical. The profiles, entitled "Evaluation
of Potential Carcinogenicity of [substance name] in Support of Reportable
Quantity Adjustments Pursuant to CERCLA Section 102," collectively form an
additional appendix to, but are not attached as part of, this methodologv
document. The profiles can be accessed separately in the record that supports
CERCLA RQ adjustment rulemakings, which are available in Room M242" at the
U.S. Environmental Protection Agency. 401 M Street, S.W.. Washington. D.C.
20460 (Docket Number 102 RQ-273C). The docket is available for inspection
between the hours of 9:00 a.m. and 4:00 p.m., Monday through Friday, excluding
Federal holidays. To review docket materials, you may make an appointment by
calling 1-202/382-3046. The public may copy a maximum of 50 pages from any
regulatory docket at no cost. Additional copies cost $.20 per page.
-4-
-------
2. LITERATURE SEARCH AND INFORMATION SOURCES
2.1. SEARCH STRATEGY
The objective of Che information search was 'to identify all relevant
published reports concerning the potential carcinogenicity of the chemicals
under study. For the most part, only reports published prior to 1983 were
considered in this review. Epidemiologic studies and the published results of
controlled investigations with experimental laboratory animals were sought from
the worldwide biomedical literature. In order that the information search be
exhaustive, both on-line and hard-copy sources of bibliographic information
were consulted. A list of the data bases searched for this project is
presented in Table 2-1. Retrieval of old literature was accomplished through
searches of hard-copy sources and through researching bibliographies of
relevant publications. Every attempt has been made to rely upon primary
publications as opposed to data summaries or abstracts contained in secondary
sources such as monographs, surveys, review articles, criteria documents, etc.
Searches were conducted using specific Chemical Abstracts Service (CAS1'
names, CAS registry numbers, common synonyms, designated substructures, and key-
words related to carcinogenicity and mutagenicity when appropriate. All of the
chemicals included in the study were first searched in the CHEMLINE on-line
chemical, dictionary to identify proper CAS registry numbers and all available
synonyms^:. On-line bibliographic data bases were then searched by developing a
list of key words, translating them into data base specific index terms and
searching the file. If the citations retrieved were relevant, they were
printed. If these results indicated new key words, the data base was searched
again.
-5-
-------
TABLE 2-1. SOURCES OF BIBLIOGRAPHIC AND NUMERICAL INFORMATION
FOR SUSPECT CARCINOGENS
On-Line Sources
CHEMLINE (National Library of Medicine)
RTECS (National Library of Medicine)
Hazardous Substances Data Bank (National Library of Medicine)
TOXLINE (National Library of Medicine)
CANCERLINE (National Library of Medicine)
Chemical Abstracts (DIALOG Information Services)
Hard-Copy Sources
"Survey of Compounds Tested for Carcinogenicity," PHS-149 (all volumes)
"International Agency for Research on Cancer—Monographs on the Evaluation
of Carcinogenic Risks of Chemicals to Humans," Volumes 1-29
National Toxicology Program, Carcinogenesis Testing Program, "Chemicals or.
Standard Protocol" (as of October 7, 1982)
National Cancer Institute --Technical Report Series
"Genetox Carcinogen List" (as of July 9, 1982)
"TOX-TIPS" (National Library of Medicine)
"Chemical Carcinogens," C.E. Searle (ed.), ACS Monograph 173. 19"c
"Documentation of the Threshold Limit Values for Substances in Vorkroor.
Air." American Conference of Governmental Industrial Hygier.ists
"Ambient Water Quality Criteria Documents," U.S. EPA, ORD
National Institute for Occupational Safety and Health--Criteria Docurr.er.rs .
Technical Reports, Special Occupational Hazard Reviews, Current
Intelligence Bulletins, Information Profiles on Potential Occupational
Hazards
"Current Contents--Life Sciences," Institute for Scientific
Information
-6-
-------
In most cases, Che decision to obtain an article in hard copy was based
upon a review of information contained in abstracts which were obtained from
the on-line computer search. For the sake of completeness, many articles were
obtained in hard copy even if it was not clear from the title that useful data
for assessment of carcinogenic risk would be contained in the report. Thus, a
large body of literature relating to the chemicals under study was collected,
although no attempt was made to cite all of the articles retrieved.
Only published literature was reviewed for the assessments made in this
report. However, the HHAG is now in the process of reviewing unpublished
carcinogenicity study data submitted to the Agency. The summary table (Hazard
Ranking of Potential Carcinogens), which is attached as an appendix to this
report, may be revised upon the completion of that review.
2.2. SELECTION OF APPROPRIATE STUDIES
It is important to recognize that many published studies may be acceptable
tc provide qualitative evidence of carcinogenicity, yet still may be
inappropriate for quantitative estimation of carcinogenic potency.
It is also important to consider the relevance of the route of
administration of the test substance in experimental animals to the anticipated
route of exposure in humans. For example, the results of animal studies in
which test substances are administered by intravenous or intramuscular
injection may provide strong qualitative evidence of carcinogenic potential and
yet be inappropriate to predict the tumor incidence in humans resulting from
ingestion or inhalation of the substance. The problem is often compounded by a
lack of data concerning the extent of absorption from different routes of
exposure (i.e., oral, inhalation, and dermal contact), thereby leading to
inaccurate estimates of actual absorbed doses.
-7-
-------
Before a study can be used to support either a qualitative or quantitative
assessment of carcinogenicity, several criteria should be met. These general
criteria, Chat are applied by reviewers when evaluating the output of a
literature search, are listed as follows.
Factor: Experimental Design
Scoring Elements:
1. What are the objectives of the study?
2. Does the design address the issues?
3. Does the design represent the state-of-the-art?
4. Are there areas which might produce ambiguous results?
5. Were the sampling and handling procedures adequate?
Factor: Experimental Procedure
Scoring Elements:
1. Were standard protocols followed?
2. Were any variations from design noted, explained, and/or considered
in reporting results?
3. Were analytics! and quantitative parameters provided?
Fac tor: Results and Conclusions
Scoring Elements:
1. Were sufficient data presented to allow a credible case to be
established?
2. -.Were the results understandable?
3. "Were results statistically valid?
4. Were the investigators' conclusions supported by the results"
5. Does the paper allow for additional conclusions to be reached
concerning correlation of results with the findings of other
investigators?
A publication need not necessarily be rejected from consideration if all
of the above criteria are not met, although deficiencies in the study should be
indicated by the reviewer if the data are intended to be used for qualitative
and quantitative assessment purposes.
-8-
-------
3. APPROACH TO DATA EVALUATION
Each primary publication retrieved during the literature search phase of
the project is critically evaluated, both with respect to its relevance to an
assessment of carcinogenicity and to the quality of the reported data. In
developing a hazard ranking methodology, the Agency recognizes that a
distinction must be made between the evaluation of the qualitative strength of
the case that a substance causes cancer, and the quantitative estimate of the
strength of the substance to cause cancer. The qualitative assessments are
expressed as an overall weight of evidence of the likelihood that the substance
is a human carcinogen (see Section 3.1.). The quantitative assessment, on the
other hand, is a numerical estimate of the strength of the substance to cause
cancer (see Section 3.2.).
Because of the large number of substances to be evaluated, it is necessarv
to be as systematic as possible in conducting both the qualitative and
quantitative assessments. A two-phase approach was developed that would
facilitate the grouping of substances based on the overall weight of evidence
of carcinogenicity and on the calculated carcinogenic potency.
A subsequent quality control review, under the direction of the HHAG. has
been conducted to assure that all determinations and study interpretations
within each profile are consistent with qualitative and quantitative analysis
in other Agency risk assessment reports. The risk assessment reports reviewed
were Health Effects Assessments (HEAs), Health and Environmental Effects
Profiles (HEEPs), Health and Environmental Effects Documents (HEEDs), Health
Assessment Documents (HADs), and Drinking Water Criteria Documents (DWCDs).
• 9-
-------
3.1. QUALITATIVE PHASE: WEIGHT OF EVIDENCE OF CARCINOGENICITY
The first phase of the HHAG ranking procedure is a qualitative evaluation
of the strength of the available data. This evaluation is based on the
Agency's "Guidelines for Carcinogen Risk Assessment" (U.S. EPA, 1986). The
remainder of this section paraphrases the guidelines.
3.1.1. Assessment of Evidence for Carcinogenicitv from Studies in Humans
Evidence of carcinogenicity from human studies comes from three main
sources:
o Case reports of individual cancer patients who were exposed to the
agent(s):
o Descriptive epidemiologic studies in which .the incidence of cancer in
human populations was found to vary in space or time with exposure co
the agent(s); and
o Analytical epidemiologic (case-control and cohort) studies in which
individual exposure to the agent(s) was found to be associated with
an increased risk of cancer.
Three criteria must be reec before a causal association can be inferred
between exposure and cancer in humans:
o There is no identified bias that can explain the association:
o The possibility of confounding variables has been considered and
ruled out as explaining the association; and
o The association is unlikely to be due to chance.
In general, although a single study may be indicative of a cause-effect
relationship, confidence in inferring a causal association is increased when
several independent studies show the association, when the association is
•10-
-------
scrong, when there is a dose-response relationship, or when a reduction in
exposure is followed by a reduction in the incidence of cancer.
From studies in humans, the evidence for carcinogenicityl is classified
as :
o Sufficient evidence of carcinogenicity, which indicates that there
is a causal relationship between the agent and human cancer;
o Limited evidence of carcinogenicity, which indicates that a causal
interpretation is credible, but that alternative explanations, such
as chance, bias, or confounding, co^ld not adequately be excluded:
o Inadequate evidence, which indicates that one of two conditions
prevailed: (a) there were few pertinent data; or (b) the available
studies, while shoving evidence of association, did no: exclude
chance, bias, or confounding and therefore a causal interpretation
is not credible;
o No data, which indicates that data are not available: or
o No evidence, which indicates that no association was found between
exposure and an increased risk of cancer in well-designed and
well-conducted independent analytical epidemiologic studies.
3.1.2. Assessment of Evidence for Carcir.ogenicitv from Studies ir.
Experimental Animals
These assessments are classified into five groups:
purposes of public health protection, agents associated with life
threatening benign tumors in humans are included in the evaluation.
•11-
-------
o Sufficient evidence^ of carcinogenicity, which indicates that there
Is an increased incidence of malignant tumors or combined malignant
and benign tumors-': (a) in multiple species or strains; (b) in
multiple experiments (e.g., with different routes of exposure) or of
unusual degree in a single experiment with regard to high incidence,
unusual site or type of tumor, or early age at onset. Additional
evidence may be provided by data on dose-response effects, as well
as information from short-term tests or on chemical structure;
o Limited evidence of carcinogenicity, which means that the data
suggest a carcinogenic effect but are limited because: (a) the
studies involve a single species, strain, or experiment and do not
meet criteria for sufficient evidence (U.S. EPA, 1986); (b) the
experiments are restricted by adequate dosage levels, inadequate
duration of exposure to the agent, inadequate period of follow-up.
poor survival, too few animals, or inadequate reporting; or ^c ar.
increase in the incidence of benign tumors only;
o Inadequate evidence, which indicates that because of major
qualitative or quantitative limitations, the studies cannot be
interpreted as showing either the presence or absence of a
carcinogenic effect;
^An increased incidence of neoplasms that occurs even with high spontaneous
background- Incidence (e.g., mouse liver tumors and rat pituitary tumors in
certain, strains) generally constitutes "sufficient" evidence of carcinogeni-
city, but may be changed to "limited" when warranted by the specific
information available on the agent.
•^Benign and malignant tumors will be combined unless the benign tumors are not
considered to have the potential to progress to the associated malignancies of
the same histogenic origin.
-12-
-------
o No data, which indicates that data are not available; or
o No evidence, which indicates that there is no increased incidence of
neoplasms in at least two well-designed and well-conducted animal
studies in different species.
The classifications "sufficient evidence" and "limited evidence" refer
only to the weight of the experimental evidence that these agents are
carcinogenic, and not to the potency of their carcinogenic action.
3.1.3. Categorization of Overall Weight of Evidence for Human
Carcinogenicitv
The overall scheme for categorizing the weight of evidence of
carcinogenicity of a chemical for humans uses a three-step process: (1) the
evidence in human studies or animal studies is summarized; (2) these lines of
information are combined to yield a tentative assignment to a
weight-of-evidence category (Table 3-1); and (3) all relevant supportive
information is evaluated to see if the designation of the overall weight of
evidence needs to be modified. Relevant factors to be included along virr. the
turr.or information from human and animal studies include structure-act ivi ;y
relationships: short-term test findings; results of appropriate physiological.
biochemical, and toxicological observations; and comparative metabolism ar.d
kinetic studies. The nature of these findings may cause one to adjust the
overall categorization of the weight of evidence.
The-substances are categorized in the manner shown below.
Group A--H"mar| Carcinogen
An agent is placed in this group only when there is "sufficient" evidence
from epidemiologic studies to support a causal association between exposure to
the agent(s) and cancer.
-13-
-------
TABLE 3-1. TENTATIVE WEIGHT OF EVIDENCE BASED ON HUMAN AND
ANIMAL EVIDENCE3
Animal Evidence
Human
Evidence
Sufficient
Limited
Inadequate
No Data
No Evidence
Sufficient
A
Bl
B2
B2
B2
No No
Limited Inadequate Data Evidence
A A A A
Bl Bl Bl Bl
C D . D D
C DDE
C DDE
3-he above assignments are presented for illustrative purposes. There maybe
instances in the classification of both animal and human data indicating that
different categorizations than those given in the table should be assigned.
Furthermore, these assignments are tentative and may be modified by ancillary
evidence. In this regard all relevant information should be evaluated ~o
determine if the designation of the overall weight of evidence needs to be
modified. Relevant factors to be included along with the tumor data frorr.
human and animal studies include structure-activity relationships, short-terrr.
test findings, results of appropriate physiological, biochemical and
toxicoiogical observations, and comparative metabolism and pharmacokinecic
studies. The nature of these findings may cause an adjustment of the overa".;
categorization of the weight of evidence.
SOURCE: U.S. EPA, 1986.
-------
Group B--Probable Human Carcinogen
This group includes agents for which the weight of evidence of human
carcinogenicity based on epidemiologic studies is "limited" and also includes
agents for which the weight of evidence of carcinogenicity based on animal
studies is "sufficient." The group is divided into two groups. Group Bl is
usually reserved for agents for which there is "limited" evidence of
carcinogenicity from epidemiologic studies. It is reasonable, for practical
purposes, to regard an agent for which there is "sufficient" evidence of
carcinogenicity in animals as if it presented a carcinogenic risk to humans.
Therefore, agents for which -here is "sufficient" evidence from animal studies
and for which there is "inadequate" evidence or "no data" from epidemiologic
studies (human) would usually be categorized under Group B2.
Group C--Possible Human Carcinogen
This group is used for agents with "limited" evidence of carcinogenicity
in animals in the absence of human data. It includes a wide variety of
evidence: for example: (a) a malignant tumor response in a single,
well-conducted experiment that does not meet conditions for "sufficient"
evidence; (b) tumor responses of marginal, statistical significance ir. studies
having inadequate design or reporting; (c) benign but not malignant tumors wit:
an agent showing no response in a variety of short-term tests for mutagenicity;
and (d) responses of marginal statistical significance in a tissue known to
have a high or variable background rate of cancer.
Group D--Not Classifiable as to Human Carcinogenicitv
This group is generally used for agent(s) with "inadequate" human and
animal evidence of carcinogenicity or for which "no data" are available.
-15-
-------
Group E--Evidence of Non-Carcinogenic lev for Humans
This group is used for agenc(s) that show no evidence for carcinogenic icy
in at least two adequate animal tests in different species or in both adequate
epidemiologic and animal studies.
The designation of a Group E agent is based on the available evidence and
should not be interpreted as a definitive conclusion that the agent will no; be
a carcinogen under any circumstances.
3.2. QUANTITATIVE PHASE: ESTIMATION OF CARCINOGENIC POTENCY
After the qualitative determination that a substance is a potential
carcinogen, a quantitative assessment can usually be performed. Such
quantitative assessments are most useful for: (a) estimating the cancer risk
associated with a particular level of exposure; and (b) making comparisons
among potential carcinogens based on their relative potencies. This latter
application is the objective of this methodology. More specifically, the
objective is to group potential carcinogens according to potency.
3.2.1. Model Selection for Analysis of Dose-Response Data
Given the stated objective of grouping potential carcinogens according to
potency, -there is a need for consistency and comparability in the evaluation
process. In accordance with the Agency's guidelines and with Agency practice
in numerous risk assessments, the multistage model is used for estimating
carcinogenic potency (U.S. EPA, 1986). Under the multistage model, the
lifetime probability of developing cancer with a constant dose (d) is given by.
(3-1) P(d) - 1 - exp [-(q0 + qid + ... + qk dk)1
-16-
-------
In accordance with the Agency's "Guidelines for Carcinogen Risk
Assessment" (U.S. EPA, 1986), when study results are such that another model
would provide adequate estimates of carcinogenic potency, the HHAG may choose
to use the more appropriate model. For instance,- because of the extraordinary
rate of early deaths of animals with tumors, the model used to estimate risks
from exposure to ethylene dibromide (EDB) incorporates a time parameter. This
time parameter allows the differential risks of less than lifetime exposure to
EDB to different age groups to be factored into the risk estimated from a
National Cancer Institute . ;i) bioassay. Any variance from the standard
multistage model is described in the individual potential carcinogen profiles.
For reportable quantity adjustments pursuant to CERCLA Section 102, the
potency factor (?) is defined as the reciprocal of the estimated dose in
mg/kg/day, associated with a lifetime cancer risk of 10 percent (ED^o). ^e
reciprocal of the ED]_Q is used because it is the more direct measure of
potency. The ED]_g itself is inversely related to potency.
The potency factor (F) is used in place of the upper bound on the lir.ear
*
coefficient (q. ) (U.S. EPA, 1989), that HHAG normally uses to estimate porer.c;.
because:
o It can be estimated without the use of many assumptions required for
*
(q, ) . This is possible because there is no need for extrapolation
below the experimentally observable dose range^*.
^Extrapolation belgw the experimentally observable dose range is the whole
purpose of the (q..) estimation. It is done because "risk at low exposure
levels cannot be measured directly either by animal experiments or by
epidemiologic studies" (U.S. EPA, 1986).
-17-
-------
o It is relatively insensitive to the choice of the dose-response
extrapolation model. Therefore, the potency rankings are not
distorted by the selection of any particular dose-response model.
o The point estimate of ED^Q, which has some optimal statistical
properties, can be used to calculate F. Therefore, it is not
necessary to use statistical upper bounds, which are needed to ensure
*
stable estimates of (q.).
The potency factor (F) is used together with the qualitative weight of
evidence of carcinogenicity in ranking the potential carcinogens.
3.2.2. Selection of Bioassav Response Data to be Used in the Model to
Calculate Potency
This section covers the selection of animal study response data for use in
the multistage model and estimation of 1/ED]_Q. Human epidemiologic data must
be reviewed and used on a case-by-case basis. Hence, if epidemiologic studies
are selected as suitable for derivation of a potencv estimate, their use is
described separately in the individual potential carcinogen profile. In
general, the data selection criteria are those described in EPA's guidelines
(U.S. EPA, 1986).
The following app-roach to selecting the data sets for calculating a
potency factor is used where several studies on a particular substance might
involve different animal species, strains, and sexes, at several doses and by
different^ routes of exposure, and may result in different tumor sites and
types.
The tumor incidence data are separated according to organ site and tumor
type. All biologically and statistically acceptable data sets are presented in
the potential carcinogen profiles. Because it is possible that human
-18-
-------
sensitivity is as high as the most sensitive responding animal species, in the
absence of evidence to the contrary, the biologically acceptable data set from
long-terra animal studies showing the greatest sensitivity is generally used,
with due regard to biological and statistical considerations.
All assumptions are presented in the profile along with a discussion of
any uncertainties in the extrapolation. Where two or more significantly
elevated tumor sites or types are observed in the same study, extrapolations
may be conducted on selected sites or types. These selections are made on
biological grounds. To obtain a total estimate of carcinogenic risk, animals
with one or more tumor sites or types showing significantly elevated tumor
incidence are pooled and used for extrapolation. The pooled estimates are
generally used in preference to potency estimates based on single sites or
types. Quantitative risk extrapolations are generally not performed on the
basis of totals that include tumor sites without statistically significant
elevations.
Benign tumors are generally combined with malignant tumors for potency
estimates unless the benign tumors are not considered to have the potential to
progress to the associated malignancies of the same histogenic origin.
3.2.3. Adjustment (Transformation) of Dose Data
Before a potency factor can be calculated all dose information must be
transformed to standard units of milligram (mg) (substance) per kilogram (kg.i
(animal weight) per day, administered over the entire length of the study. If
doses are given in units other than mg/kg/day, or if animals are dosed in a
non-continuous manner, or if the reviewer has evidence that the absorbed or
metabolized dose is significantly less than the administered dose, then the
dose data must be converted to a "transformed dose."
-19-
-------
The next three subsections discuss how this is done for three exposure
routes: diet, water, and air. The assumptions and procedures used parallel
those described in the Agency's "Guidelines and Methodology for the Preparation
of Health Effects Assessment Chapters of the Ambient Water Quality Criteria
Documents" (U.S. EPA, 1980).
3.2.3.1. Calculation of Dose in mg/kg/dav from Doses Expressed as Dietary
Concentrations - - If the authors provide information on body weight and food
consumption, then the dietary dose (d) is calculated directly. If these data
are not provided, then the dose may be estimated by using standard food
consumption estimates based on the fraction of body weight that is consumed
each day as food (f) (U.S. EPA, 1980):
Species f
Mouse 0.13
Rat 0.05
Human 0.028
In order to obtain the dietary dose (d) from the data given in parts per
million ('pprr.) , the daily experimental dose in ppm is multiplied by f:
(3-2) d - ppm x f
Note chat ppm in food has units of mg toxicant per kg food, and the
fraction f has units of kg food per kg body weight per day. Thus, the produc
has units of mg of toxicant per kg body weight per day.
-20-
-------
3.2.3.2. Calculation of Dose in me/kg/dav from Doses Expressed as Water
Concentrations - - If the authors of the studies provide information on body
weight and water consumption, then the dietary dose d is calculated directly.
If these data are not available then d may be estimated by using standard water
consumption estimates based on the fraction of the body weight consumed as
water per day (fw) (U.S. EPA, 1980). The assumptions and the procedure for
making this estimation are the same as for dietary concentrations (Section
3.2.3.1.) but the following rates for fw apply:
Species fw
Mouse 0.17
Rat 0.078
Human 0.029
The dietary dose (d) in mg/kg/day is calculated by multiplying the dailv
cose in ppnr. by the appropriate f w:
(3-3) d - ppm x fw
Note that ppm in water has units of mg toxicant per liter water and that
fw has units of liters of water per kg body weight per day. Thus the product
has unics of mg toxicant per kg body weight per day.
3.2.3.3. Calculation of Dose in mg/kg/day from Doses Expressed as Air
Concentrations--When exposure is via inhalation, the calculation of dose can be
considered for two cases where: (1) the carcinogenic agent is either a
completely water-soluble gas or an aerosol and is absorbed in proportion to the
-21-
-------
amount of air breached in; and (2) where Che carcinogen is a poorly
water-soluble gas that reaches an equilibrium between the air breathed and the
body compartments. After equilibrium is reached, the rate of absorption of
these agents is expected to be proportional to the metabolic rate, which is
proportional to the rate of oxygen consumption, which in turn is a function of
surface area (U.S. EPA, 1980).
For Case 1, agents that are in the form of particulate matter or virtually
completely absorbed gases, such as sulfur dioxide, can reasonably be expected
to be absorbed proportional to the inhalation rate. The inhalation rate (I)
for various species can be calculated from the observations (FASEB, 1974) chat
25-g mice breathe 0.0345 cubic meters (m ) per day and 113-g rats breathe 0.105
mvday. For mice and rats of other weights in kilograms (W) , the surface-area
proportionality can be used to find breathing rates in m3/day as follows:
(3-4) for mice, I - 0.0345 (W/0.025)2/3 m3/day; and
(2-5) for rats. I - 0.105 (V/0.113)2/3 m3/day.
The weight ratio is raised to the two-thirds power because, to a close
approximation, the surface area is proportional to the two-thirds power of the
weight as would be the case for a perfect sphere.
For humans, the value of I - 20 m3/day is adopted as the standard
breathing rate. This is calculated from the observation (ICRP, 1977) that the
average breathing rate is 10 cubic centimeters (cm3) per 8-hour workday and
2 x 107 cm3 in 24 hours.
-22-
-------
The empirical factors for the air intake per kg per day. i - I/W,
calculated from the previously stated relationships for standard weight
animals, are tabulated as follows:
Species
Mouse
Rat
Human
W
0.03
0.35
70
i -
1.
0.
0.
I/W
3
64
29
The inhalation dose (d) in mg/kg/day is calculated by multiplying the
substance's air concentration (v) by the appropriate intake factor (i) and the
absorption fraction (r):
(3-6) d - v x i x r
Note that v has units of mg toxicant per m . i has units of m^ per kg bod'.
weight per day, and r is dimensionless. Thus the product has unics of mg
toxicant per kg body weight per day.
In the absence of experimental information or a sound theoretical argu.T.en:
to the contrary, r is assumed to be the same for all species and therefore
drops out of che calculations.
For Case 2, the dose in mg/day of partially soluble vapors is proportional
to 02 consumption, which in turn is proportional to W'/^. The dose is also
proportional to the solubility of the gas in body fluids, which can be
expressed as an absorption coefficient (r) for the gas.
-23-
-------
Therefore, by expressing the Q£ consumption as 02 - k x U2/3, where k is a
constant independent of species, it follows that the average dose per day in mg
during administration of the agent (n) can be expressed as:
(3-7) n = k x W2/3 x v x r
As with Case 1, in the absence of experimental information or a sound
theoretical argument to the contrary, the absorption fraction (r) is assumed to
be the same for all species. Therefore, for these substances a certain
concentration in ppm or mg/mj in experimental animals is equivalent to che same
concentration in humans. This is supported by the observation that the minimum
alveolar concentration necessary co produce a given "stage" of anesthesia is
similar in man and animals (Dripps et al., 1977). When the animals are exposed
via the oral route and human exposure is via inhalation or vice-versa, the
assumption is made, unless there is pharmacokinetic evidence to the cor.trary.
chat absorption, is equal by either exposure route.
Ir. this case, the dose (d) in mg/kg/day is:
(3-8) '. d - n/kg (animal)
For either inhalation case, exposures given in terms of ppm (by volume; in
air can be converted to units of mg/nr (%') by the following formula:
(3-9) v - 0.041 x molecular weight (gas) x ppm
.2U-
-------
(Note that 1 mL in 1 m-* is 1 ppm (by volume); therefore, 0.041 x molecular
weight (MW) is the weight in mg of 1 mL of a gas.)
3.2.3.4. Adjustment for Non-Continuous Exposure -- To this point the dose (d)
calculated reflects the daily dose given over the -experimental treatment
period. To derive the final "transformed dose," the dose must be multiplied by
the fraction of the study over which the animal was actively dosed. If the
animal was dosed continuously over an entire treatment period (e.g., not 3
times per week or, for inhalation studies, 6 hours per day) then the
transformed dose is:
(3-10) transformed dose = d x 1s- x l£
Le Le
where le - duration of the treatment and Le - duration of the study.
If the animal was dosed for a fraction of a week (e.g., 5/7) or a fraction
of a day (e.g., 6/24), then the transformed dose becomes, for example:
(3-11^ transformed dose - d x I6- x —2— x —£—
Le 1 2*
3.2.3.5. Adjustment for Absorrtion. Distribution. Metabolism, and Excretion- -
Whenever there is usable information on the absorption, distribution,
metabolism, or excretion of the substance, the potency factor is adjusted to
reflect chis information. For example, if the effective dose to the target
organ itv an.animal, due to any of these four factors, is known to be a fraction
of the administered dose, then the effective dose is used to estimate the
potency factor. However, in the absence of information or differences in
absorption, distribution, metabolism, or excretion between animals and humans.
no such adjustments are made.
-25-
-------
3.2.4. Calculation of the Human Potency Factor (F)
The information needed for calculating a human potency factor (F) can be
found in the profiles entitled "Evaluation of the Potential Carcinogenicity of
[substance name] in Support of Reportable Quantity Adjustments Pursuant to
CERCLA Section 102." A sample potency factor derivation table from the profile
for chloroform is presented in Table 3-2.
3.2.4.1. Animal Potency--The first step in the derivation of F is the
calculation of the animal potency from the transformed dose and response (tumor
incidence) data.
The animal potency is estimated by fitting a multistage dose - response
model to the transformed dose - response data, as described in EPA's Notice of
Availability of Water Quality Criteria Documents (U.S. EPA, 1980). The
mathematical assumptions made for the model used are described in a separate
appendix to the technical background document (U.S. EPA, 1989) for the
reportable quantity rule. • For chloroform, the dose causing an increased cancer
risk of 10 percent of the population is calculated to be ED]_Q = 9.62 mg.-kg/day .
The anirr.al potency is the reciprocal of this dose. 0.104 (mg/kg/day} " L .
3.2.4.2. Adjustment for Less Than Lifetime Studies --Under the current
procedures used by HHAG, the risk levels are derived only for full lifetime
experiments. In dealing with experimental data in which the observation period
(Le) is less than the lifespan (L) of the experimental animal, the potency
factor derived from the experimental data, which would give a risk estimate
over the fraction Le/L of the animal's lifespan, is increased by a faccor of
(L/Le)3 to obtain an estimated potency for full lifetime risk. As explained by
the EPA (1980):
•26-
-------
TABLE 3-2. SAMPLE TABLE FOR THE DERIVATION OF POTENCY
FACTOR (F)
Agent: Chloroform
Reference:
Exposure route:
Species:
Strain:
Sex:
Vehicle or physical state:
Body weigh;:
Duration of treatment (le):
Duration of study (Le):
Lifespan of animal:
Target organ:
Tu.r.or type :
Experimental doses/exposurec
(mg/kg):
Transformed doses
NCI (1976)
oral (gavage)
mice
B6C3F1
F
corn oil
0.03 kga
546 days
644 - 651 days
730 days5
Liver
hepatocellular carcinoma
477
238
irr.g/kg/day) :
Turr.or incidence:
Lifetime animal potency :
Human potency factor (F) :
288
39/41
0.104
1.97
143 0
36/45 0/20
aReportetf.
Assumed.
cExposures were 5 days/week. Duration of study was assumed to be 647 days.
To derive the transformed dose from the experimental dose data: experimental
dose (mg/kg/day) x (5 treatment days per week/7 days per week) x duration.
-27-
-------
We assume chat if the average dose (d) is continued, the age-specific race
of cancer will continue to increase as a constant function of the
background rate. The age-specific rates for humans increase at least by
the second power of the age and often by a considerably higher power, as
demonstrated by Doll (1971). Thus we would expect the cumulative tumor
rate to increase by at least the third power of age...
(3-12) Lifetime Animal Potency - Observed Animal Potency x
(L/Le)3
This adjustment is conceptually consistent with the proportional hazard
model considered by Crump and Watson (1982). For chloroform, the lifetime
animal potency is:
(3-13) Lifetime Animal Potency - 0.104 x (730/648)3 - 0.149
3.2.4.3. Human Potency (F)--Finally. the potency must be adjusted for humans
(if derived from animal data). The human potency adjustment is made using the
following surface-area correction:
(3-14) Human Potency (F) - Lifetime Animal Potency x
(70
-28-
-------
where Wa is the weight of the animal and 70 kg is the assumed average weighc of
humans. This is in accordance with the Agency's cancer guidelines (U.S. EPA,
1986)5.
For chloroform, the human potency (F) is:
(3-15) Human Potency (F) - 0.149 x (70/0.03)1/3 - 1.97.
3.2.5. Grouping of Chemicals Based on Carcinogenic Potency
After the potency factors are estimated, the substances are placed into
three potency groups. The most potent substances, those with potency factors
above 100, are placed in Potency Group 1; substances with potency factors
between 1 and 100 are placed in Potency Group 2; and substances with potency
factors below 1 are placed in Potency Group 3. The Potency Group will be used
along with the weight-of-evidence group in assigning tentative RQs for
potential carcinogenicity.
In the case where available data are inadequate for estimating a potency
factor, the HHAG has identified three possible alternatives:
Agency's guidelines reflect that animal potency is converted to human
potency by first multiplying the animal ED^o by c^e rati° °f cne Cne weight of
the aniaal. to the weight of man, and dividing the entire ED^Q by the racio °£
the surface area of the animal to man.
Human ED10 - animal ED10_x^WaZWmanl . ^.^ ^ (Ua/Wman)l/3
Then, the reciprocal of the Human ED^Q is the adjusted human potency.
-29-
-------
o If Che best available data suggest that the substance is a possible
strong carcinogen, but there is no basis for assigning a specific
ED^Q dose (e.g., all treated animals at every dose developed tumors),
then the substance is assigned to the highest potency group (i.e.,
Potency Group 1).
o If the best available data are inadequate for calculating a potency
factor and no quantitative inferences can be made (e.g., in animal
studies where control groups were not used or number of animals
treated were not specified), then the substance is assigned to the
mid-range potency group (i.e., Potency Group 2).
o If the best available data suggest that the substance is a possible
weak carcinogen, but there is no basis for assigning a specific ED^Q
dose (e.g., due to uncertainties associated with the pharmacokinetics
of the substance), then the substance is assigned to the lowest
potency group (i.e., Potency Group 3).
Hazard rankings can then be assigned by following the standard procedure
for combining the weight-of-evidence group and the assigned potency group.
2.2. OVERALL HAZARD RANKING BASED ON COMBINED QUALITATIVE AND QUANTITATIVE
ASSESSMENTS
The culmination of the hazard ranking process performed in this study is
accomplished by combining the qualitative weight of evidence for carcinogen!-
city (Section 3.1.) with the potency group (Section 3.2.) to arrive at a final
hazard ranking for each substance. Substances are ranked as "high," "medium."
or "low" hazard according to the scheme shown in Table 3-3.
•30-
-------
TABLE 3-3. HAZARD RANKING SCHEME FOR REPORTABLE QUANTITIES
UNDER CERCLA
Weight-of-Evidence
Group
Potency Group : 1 2 3
Potency Factor : F > 100' F - 1 to 100 F < 1
A - Carcinogenic
to humans
B - Probably carcino-
genic to humansa
C - Possibly carcino-
genic to humans
D - Not classifiable
as to human
carcinogenicity
E - Evidence of non-
carcinogenicity for
humans
High
High
Medium
High
Medium
Medium Low
Low
No hazard ranking
No hazard ranking
alr.cludes weight-of-evidence Groups Bl and B2.
•31-
-------
Hazard rankings are based jointly on two factors--weight of evidence and
potency--that the Agency believes are important in describing carcinogenic
hazards. Hazard rankings of high, medium, and low are assigned so that the
hazard ranking increases as either the weight of evidence or the potency
increases.
Depending on whether a substance falls into Potency Groups 1, 2, or 3, a
hazard ranking of high, medium, or low is assigned to Group B carcinogens.
Hazard rankings are one level higher (high, high, or medium) for Group A
carcinogens. This increased concern is justified because there is direct human
evidence establishing that Group A substances cause cancer. Hazard rankings
are one level lower (medium, low, or no hazard ranking is assigned at this
time) for Group C carcinogens. This reduced concern is justified because the
evidence implicating Group C substances either is unreplicated or is of
marginal biological or statistical significance. Before settling on these
hazard ranking assignments, alternative ranking schemes were considered.
Proposals that all Group A substances be ranked high or that all Group C
substances be ranked low were rejected because the Agency believes strongly
that potency, too, is important in describing a carcinogenic hazard.
Similarly, a proposal to base hazard rankings on potency alone was rejected
because the Agency believes that the weight of evidence must be considered as
well. It is the Agency's judgment that the hazard scheme finally selected
gives proper consideration to both weight of evidence and potency.
For substances placed in weight-of-evidence Groups D or E, primary
criteria other than potential carcinogenicity must be used to assign an
adjusted RQ.
•32-
-------
3.3.1. Use of Chemical and Environmental Fate and Transformation Data _in
Hazard Ranking of Mecals and Their Salts
The chemical and environmental specification, oxidation state,
solubility, chemical and environmental fate and half-life, and
disproportionation reactions are important determinants of the toxicity of
inorganic compounds. Furthermore, toxicity data relevant to potential
carcinogenicity and other chronic effects are lacking for many of the metals
and their salts. Therefore, in cases where toxicity data are not available on
a particular metal salt, and an evaluation of the above parameters indicates
its convertibility to the toxic (i.e., carcinogenic) species under realistic
human exposure conditions, then an appropriate hazard ranking assignment will
be performed based on this evidence.
3.3.2 Special Problems in Hazard Ranking of Chemicals Associated wich
Multimedia Exposure in Humans
Carcinogenic hazard ranking in the present study must take into accoun:
the potential for multimedia exposure. Thus, all routes of human exposure
(oral, inhalation, and dermal) must be considered in the hazard ranking scheme
The final hazard ranking is based on the route that gives the highest potency
factor.
-33-
-------
4. SUMMARY
The Agency's Human Health Assessment Group (HHAG) has developed a unique
method for ranking CERCLA hazardous substances fo'r potential carcinogenicity.
This methodology is not a risk assessment and it does not yield an absolute
measure of harm. Rather, the methodology simply represents a means of sorting
potentially carcinogenic substances into categories, which may then be equated
to RQ levels.
The methodology for ranking potential carcinogens begins by reviewing all
information available in the scientific literature on each substance identified
as a potential carcinogen. This information is then evaluated using a
two-stage process. The first stage is a qualitative assessment of the
likelihood that a particular hazardous substance is a human carcinogen. During
this stage, the available data is evaluated using EPA's weight-of-evidence
classification system, developed in the September 24, 1986. "Guidelines for
Carcinogen Risk Assessment" (U.S. EPA, 1986). The second stage is a
quantitative assessment designed to estimate the relative strength of a
hazardous substance to elicit a carcinogenic response (potency factor). The
quantitative stage allows the Agency to rank potential carcinogens on a
numerical scale. The results of the qualitative and quantitative assessments
are then eoabined to arrive at a hazard ranking for each hazardous substance
evaluated for potential carcinogenicity.
There are two separate appendices to this methodology document. The first
is a summary table of the hazard ranking results. The second consists of all
of the individual chemical profiles prepared to support these results.
.34.
-------
5. REFERENCES
Crump, K.S.; Watson, U.W. (1982). GLOBAL 82: a foreran program to extrapolate
dichotomous animal carcinogenicity data to low doses. National Institute
of Environmental Health Sciences, Contract No. l-ES-2123.
Doll, R. (1971). Weibull distribution of cancer: implications for models of
carcinogenesis. J. Roy. Stat. Soc. A. 13:133.
Dripps, R.D.; Eckenhoff, J.E.; Vanden, L.D. (1977). Introduction to anes-
thesia: the principles of safe practice. 5th Ed. Philadelphia, PA:
W.B. Saunders Company, pp. 121-123.
FASEB (Federation of American Societies for Experimental Biology) (1974).
Library of Congress No. 72-87738.
ICRP (International Commission on Radiological Protection) (1977).
Recommendation of the ICRP, Publication No. 26, adopted Jan. 17, 1977.
Oxford, United Kingdom, Pergamon Press.
U.S. Environmental Protection Agency (1980). Guidelines and methodology for
the preparation of health effects assessment chapters of the ambient water
quality criteria documents. Federal Register 45:79318.
U.S. Environmental Protection Agency (1985a). Technical background document
to support rulemaking pursuant to CERCLA Section 102, Volume 1.
U.S. Environmental Protection Agency (1985b). Notification requirements;
reportable quantity adjustments. Federal Register 50:13456-13522.
U.S. Environmental Protection Agency (1986). Guidelines for carcinogen risk.
assessment. Federal Register 51:33992-34003.
U.S. Environmental Protection Agency (1987). Reportable quantity adjustments.
Federal Register 52: 814C-817l'.
U.S. Environmental Protection Agency (1989). Technical background document to
support rulemaking pursuant to CERCLA Section 102, Volume 3.
-35-
-------
APPENDIX
HAZARD RANKING OF POTENTIAL CARCINOGENS
-------
HAZARD RANKING Of POIEN1IAI CARCINOGENS
Substance
CASRN
Weight -
Degree of Evidence of-Evidence
Humans Animals Gioup
Potency Potency Hazard
Factor Croup Ranking
1
2
3
4
5
6
7
B
9
10
11
12
13
U
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
Acetamide, H- f luoren-2-yl
Acryloni tri le
Aldrin
Aroitrole
Arsenic
Arsenic acid .
Arsenic disulfide
Arsenic pentoxide
Arsenic trichloride
Arsenic trioxide
Arsenic trisulfide
Cacodyl ic acid
Calcium arsenate
Calc iuro arseni te
Cupric acetoarseni te
D i ch I or opheny I ars i ne
Diethylarsine
Lead arsenate
Potassium arsenate
Potassium arsenite
Sodium arsenate
Sodium arsenite
Asbestos
Auramine
Azaserine
Aziridine
Benz(c)acridine
Benz(a)anthracene
Benzene
Beniidine and its salts
Benzo(b) f luoranthene
Benjo(k )* luoranthene
Benio(a)pyrene
00053963
00107131
00309002
00061825
07440382
01327522
07778394
01303328
01303282
07784341
01327533
01303339
00075605
07778441
52740166
12002038
00696286
00692422
07784409
07784410
10124502
07631892
07784465
01332214
00492808
00115026
00151564
00225514
00056553
00071432
00092875
00205992
00207089
000503?8
No Data
L i timed
Inadequate
Inadequate
Suf f ic ient
d
d
d
d
Suf f ic ient
d
No Data
d
d
d
No Data
No Data
d
d
Suf f icient
d
d
Suf f icient
1 nadequate
No Data
No Data
No Data
No Data
Suf f ic ient
Suf f ic i ent
No D.ita
No (>Jtn
Inadequate
Suf f ic ient
Suf f icient
Suf f ic ient
Suf f icient
Inadequate
No Data
No Data
No Data
No Data
Inadequate
No Data
Inadequate
No Data
No Data
Inadequate
No Data
No Data
Inadequate
No Data
Inadequate
No Data
No Data
Suf f ic ient
Suf f icient
Suf f icient
Suf f ic ient
Limi ted
Suf f ic ient
Suf f ic ient
Suf f icient
Suf f ic ient
Inadequate
Suf f ic ient
B2
B1
B2
B2
A
d
d
d
d
Ad
d
D'9
d
d
d
O'9
O'9
d
d
Ad
d
d
A
B2
B2
B2
C
B2
A
A
B2
D
B2
148
2.28*
239
3.30
142
d
d
d
d
d
d
None
d
d
d
None
None
d
d
1
d
d
P
0.48
169
336
a
21.1
0.27f
2220f
248
None
248
1
2
1
2
1
1
1
1
1
1
1
None
1
1
1
None
None
1
1
1
1
1
P
3
1
1
2
2
3
1
1
None
1
HIGH
MED
HIGH
NED
HIGH
HIGH
HIGH
HIGH
HIGH
HIGH
HIGH
Noneb
HIGH
HIGH
HIGH
Noneb
Noneb
HIGH
HIGH
HIGH
HIGH
HIGH
HIGH
LOU
HIGH
HIGH
LOU
MED
NED
HIGH
HIGH
Noneb
HIGH
37
-------
HA/ARD RANKING 01 POTfNIIAl CARCINOGENS (Continued)
34
35
36
37
18
39
40
41
42
43
44
4%
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
Substance
Benzotrichloride
Beniyl chloride
Beryl I turn
Beryl 1 iura chloride
Beryllium fluoride
Beryl 1 iura nitrate
alpha - BHC
beta - BHC
gamma - BHC (Lindane)
Bis(2-chloroethyl ) ether
Bis(chloromethyl ) ether
Bis(2-ethylhexyl ) phthalate
Cadmium
Cadmium acetate
Cadmium bromide
Cadmium chloride
Carbon let rachlor ide
Chlorambucil
Chlordane
Chlornaphaiine
Chloroform
Chloromethyl methyl ether (technical grade)
4-Chloro-o-toluidine, hydrochloride
Chromium'
Ammonium bichromate
Ammonium chromate
Calc ium chromate
Chromic acid
L i thium chromate
Potassium bichromate
Potassium chromate
Sodium bichromate
Sodium chromate
CASRN
00098077
00100447
07440417
07787475
07787497
13597994
00319846
00319857
00058899
00111444
00542881
00117817
07740439
00543908
07789426
10108642
00056235
00305033
00057749
00494031
00067663
00107302
03165933
07440473
07789095
07788989
13765190
11115745
14307358
07778509
07789006
105B8019
07/751 13
Degree of
Humans
I i mi ted
Inadequate
Inadequate
No Data
No Data
No Data
No Data
No Data
Inadequate
No Data
Suf f ic ient
No Data
L imi ted
k
k
k
Inadequate
L imi ted
Inadequate
Inadequate
Inadequate
Suf f icient
No Data
No Data
h
h
h
h
h
h
h
h
h
Evidence of
Anitn.il s
Suf f icient
Suf f ic ient
Suf f ic ient
No Data
Suf f ic ient
No Data
Suf f ic ient
L imi ted
Suf f ic ient/
L imi ted
Suf f ic ient
Suf f ic ient
Suf f ic ient
Suf f ic ient
No Data
No Data
Suf f ic ient
Suf f ic ient
Suf f icient
Suf f ic ient
Limited
Suf f icient
Inadequate
Suf f icient
No Data
No Data
No Data
Inadequate
Inadequate
No Data
Inadequate
Inadequate
Inadequate
1 nndequat e
Ueight-
-E vidence
Group
B1
82
B2U
u
u
u
B2
C
B2/C
B2
A
B2
B1
k
k
k
B2
B1
B2
C
B2
A
B2
D
Ah
Ah
Ah
Ah
Ah
Ah
Ah
Ah
Ah
Potency
factor
58.0
0.66
79.70U
ab
ab
ab
51.48
10.67
7.39
13.29
10377
0.194
57. 9k
k
k
k
59.9
a
15.13
a
1.97
n
0.40
None
h
h
h
h
h
h
h
h
h
Potency
Group
2
3
2
1
1
1
2
2
2
2
1
3
2
2
2
2
2
2
2
2
2
1
3
None
1
1
1
1
1
1
1
1
1
Hazard
Ranking
MED
LOU
MED
HIGH
HIGH
HIGH
MEO
LOU
MED
MEO
HIGH
LOU
MED
MED
MED
MED
MED
MED
MED
LOU
MED.
HIGH
LOU
None6
HIGH
HIGH
HIGH
HIGH
HIGH
HIGH
HIGH
HIGH
HIGH
JB-
-------
HAZARD RANKING Of P01EN11AL CARCINOGENS (Continued)
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
Substance
Strontium chromate
Chrysene
Coke Oven Emissions <
Creosote
Cyc t ophosphami de
Daunomycin
ODD
DOE
DDT
Dial late
Diaminotoluene (mixed)
Dibenz (a ,hj anthracene
1 ,2:7.8-Dibenzopyrene
1 , 2-Dibrorao- J-chloropropane
3,3' -Oichlorobenzidine
1 , 2-0 ich lor oe thane
1 . 1 -Dichloroethylene (Vinyl idene chloride)
Dieldrin
1 , 2:3,4- Diepoxybutane
1 ,2-Diethylhydrazine
Diethylsti tbestrol
Dthydrosaf role
3,3' -Oimethoxybenz idine
Dimethyl sulfate
D imethylaminoaiobenzene
7, 12-0imethylbenz(a)anthracene
3,3' -Dimethylbenzidine
Dimethylcarbamoyl chloride
1, 1-Dimethylhydrazine
1 ,2-Dimethylhydrazine
Dini trotoluene (mixed)
2,4 -Oini trotoluene
?.6-Oinitrotoluene
1 ,4 • 0 \ oxane
CASRN
07789062
00218019
N.A.
06001589
00050180
20830813
00072548
00072559
00050293
02303164
00095807
00053703
00189559
00096128
00091941
00107062
00075354
00060571
01464535
01615801
00056531
00094586
00119904
00077781
00060117
00057976
00119937
00079447
00057147
00540738
25321146
00121142
00606202
00125911
Degree of
Humans
h
No Data
Suf f ic ient
L imi ted
1 imi ted
No Data
Inadequate
Inadequate
Inadequate
No Data
No Data
No Data
No Data
No Data
No Data
No Data
No Data
Inadequate
No Data
No Data
Suf f icient
No Data
No Data
Inadequate
No Data
No Data
No Data
1 nadequat e
No Datn
No Data
No Data
No Data
No U.it a
1 fi.nlcqu.it e
Evidence
Animals
Suf f ic ient
I imi ted
Suf f icient
Suf f ic ient
Suf f icient
Suf f icient
Suf f icient
Suf f icient
Suf f ic ient
L imi ted
Suf f icient
Suf f icient
Sufficient
Sufficient
Sufficient
Suf f icient
L imi ted
Suf f ic ient
Suf f icient
Suf f icient
Suf f icient
Sufficient
Suf f ic ient
Sufficient
Suf f icient
Suf f icient
Suf f icient
Suf f icient
Suf f icient
Suf f ic ient
Suf f ic ient
Suf f ic ient
I imi ted
Suf f ic ient
Uc ight -
of - 1 v idence
Group
Ah
C
A
B1X
81
B2
B2
82
B2
C
82
B2
82
B2
82
B2
C
82
82
82
A
B2
B2
82
82
82
B2
B2
B2
82
B2'3
82
C
H2
Potency
factor
h
a
1.53f
ab
17.5
a
1.30
3.82
5.58
4.28
23.2
ab
a
1240
7.49
0.13
5.19
236
28.0
a
4740
1.08
3.07
a
a
540
27.4
505
82.5
4210
3.82'3
3.82
a
0.034
Potency
Group
1
2
2
1
2
2
2
2
2
2
2
1
2
1
2
3
2
1
2
2
1
2
2
2
2
1
2
1
2
1
2
2
2
3
Hazard
Rank ing
HIGH
LOU
HIGH
HIGH
MED
MED
MED
MED
MED
LOU
MED
HIGH
MED
HIGH
MED
LOU
LOU
HIGH
MED
MED
HIGH
MED
MED
MED
MED
HIGH
MED
HIGH
MED
HIGH
MED
MED
LOU
LOU
-------
HAZARD RANKING Of POTENTIAL CARCINOGENS (Continued)
101
102
103
104
105
106
107
108
109
110
111
112
113
1U
115
116
117
IIS
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
Substance
1 , 2-Diphenylhydrazine
Epichlorohydr in
Ethyl carbamate (ureth^nt)
Ethyl 4.4' -dichlorobenillate
Ethylene dibrowide
Ethylene oxide
E thy tenet hi ourea
Ethyl methanesul fonate
formaldehyde
Glycidylal dehyde
Heptachlor
Heptachlor epoxide
Hexach I orobenzene
Hexachlorobutadicne
Hexachloroethane
Hydrazine
Indenod ,2,3-cd)pyrene
Isosaf role
Kepone
Lasiocarpine
Lead
Lead acetate
lead phosphate
Lead stearate
Lead subacetate
Lead sulfide
Nelphalan
Methyl chloride
3-Nethylcholanthrene
4,4' -Hethylenebis(2-chloroani 1 ine)
Methyl iodide
N- Methyl -N'-nitro-N-ni t rosoquanidi ne
Methyl thiourac i 1
CASRM
00122667
00106898
00051796
00510156
00106934
00075218
00096457
00062500
00050000
00765344
00076448
01024573
00118741
00087683
00067721
00302012
00193395
00120581
00143500
00303344
07439921
00301042
07446277
07428480
01335326
01314870
00148823
00074873
00056495
00101144
00074884
00070257
0005604?
Degree of
Humans
No Data
Inadequate
No Data
Inadequate
Inadequate
L imi ted
No Data
No Data
L imi ted
No Data
1 nadequate
No Data
No Data
No Data
No Data
Inadequate
No Data
No Data
No Data
No Data
L imi ted
No Data
No Data
Inadequate
No Data
No Data
Inadequate
Evidence
Animals
Suf f icient
Suf ( icient
Suf f ic ient
Suf f icient
Suf f icient
Suf f icient
Suf f icient
Suf f icient
Suf f icient
Suf f ic ient
Suf f ic ient
Suf f ic ient
Suf f ic ient
L imi ted
L imi ted
Sufficient
L imi ted
Suf f icient
Suf f icient
Suf f ic ient
Suf f ic ient
L imi ted
Suf f ic ient
Suf f ic ient
L imi ted
Suf f ic ient
Suf f ic i ent
Weight-
of -Ev idence
Group
B1
B2
B2
B2
B2
B1
B2
B2
B1
B2
B?
B2
62
C
C
B2
C
B2
B2
62
B1z2
C
62
62
C
B?
82
Potency
Factor
4.31
0.37
0.64
1.79
390
1.34
1.30
295
2.96
2.90
117
290
39.0
0.59
0.077
107
a
0.54
48.0
48.9
810
0.050
25.5
1.52
a
54.7
a
Potency
Group
2
3
3
2
1
2
2
1
2
2
1
1
2
3
3
1
2
3
2
2
1
3
2
2
2
2
2
Hazard
Ranking
MED
LOU
LOU
MED
HIGH
MED
MED
HIGH
MED
MED
HIGH
HIGH
MED
•
*
HIGH
LOU
LOU
MED
MED
*«
*.»
*f
««
««
**
HIGH
LOU
MED
MED
LOU
MED
MED
-40-
-------
HAZARD RANKING OF POTENTIAL CARCINOGENS (Continued)
Weight•
134
135
136
137
138
139
UO
HI
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
Substance
M i t otnyc i n C
1 -Naphthylamine
2-Naphthy lamine '
Nickel
Nickel ammonium sulfate
Nickel carbonyl
Nickel chloride
Nickel cyanide
Nickel hydroxide
Nickel nitrate
Nickel sulfate
2-Ni t ropropane
N -N i trosodi-n- butyl am ine
N-Ni trosodiethanol amine
N-Ni trosodi ethyl ami ne
N-Ni trosodime thy 1 amine
N-Ni trosodi -n-propylamine
N-Nitroso-N- ethyl urea
N-Nitroso-N-methylurea
N-Ni troso-N-methylur ethane
N-NitrosoMthyl vinyl amine
N-Ni trosopi peri dine
N-Ni trosopyrrol idine
5-Ni tro-o- toluidine
Pentachloroethane
Pentachloroni trobenzene
Pentechlorophenol
Phenacet in
Polychlorinated biphenyls (PCBs)
Aroclor 1016
Aroclor 1221
Aroclor 1232
Aroclor 1242
CASRN
00050077
00134327
00091598
07440020
15699180
13463393
07718549
00557197
12054487
14216752
07786814
00079469
00924161
01116547
00055185
00062759
00621647
00759739
006B4935
00615532
04549400
00100754
00930552
00099558
00076017
00082688
00087865
00062442
01336363
12674112
11104282
11141165
53469219
Degree of
Mum.ins
No Data
Inadequate
Suf f ic ient
Inadequate
No Data
Inadequate
1 nadcquate
No Data
No Data
No Data
Inadequate
Inadequate
No Data
No Data
No Data
No Data
No Data
No Data
No Data
No Data
No Data
No Data
1 nadequate
No Data
No Data
No Data
Inadequate
Inadequate
Inadequate
Inadequate
1 nadequate
Inadequate
1 n.idequcite
E v idenre
Animals
Sufficient
L imi ted
Suf f icient
L imi ted
No Data
Sufficient
L imi ted
No Data
L imi ted
No Data
L imi ted
Suf f icient
Suf f ic icnt
Suf f ic ient
Suf f i c i ent
Suf f ic ient
Suf f icient
Suf f icient
Suf f ic ient
Suf f icient
Suf f ic ient
Suf f icient
Suf f icient
Limited
L imi ted
L imi ted
No Data
Sufficient
Suf f ic ient
No Data
No Data
No Data
No Data
of -Ev i donee
Group
B2
C
A
C
z7
B2
z7
z7
z7
z7
z7
B2
B2
B2
82
B2
B2
B2
B2
B2
B22
82
82
C
C
C
D
B2
B2
t
t
t
t
Potency
factor
a
None
4.77
z7
z7
z7
z7
z7
z7
z7
z7
a
43.70
ab
969
61.2
a
137
2100
ab
a
37.5
279
0.17
1.26
1.42Z1°
None
0.028
50.47*
t
t
t
t
Potency
Group
2
2
2
z7
z7
z7
z7
z7
z7
z7
z7
2
2
1
1
2
2
1
1
1
2
2
1
3
2
2
None
3
2
2
2
2
2
Hazard
Rank ing
MED
LOU
HIGH
LOU
LOU
MED
LOU
LOU
LOU
LOU
LOU
MED
MED
HIGH
HIGH
MED
MED
HIGH
HIGH
HIGH
MED
MED
HIGH
LOU
LOU
LOU
Noneb
LOU
MED
MED
MED
MED
MED
-------
HA/ARD RANKING OF POUN1IAL CARCINOGENS (Continued)
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
Substance
Aroclor 1248
Aroclor 1254
Aroclor 1260
1,3 -Propane suit one
1 , 2-Propylenimine
Saccharin
Safrole
Selenium sulfide
Streptozotocin
2,3, 7,8- letrachlorodibenio-p-dioxin (TCDD)
1,1, 1,2-Ietrachloroethane
1, 1 ,2,2-letrachloroethane
letrachloroethylene
Ihioacet amide
Thiourea
o- lotuidine
p- Toluidine
o-Toluidine hydrochlor ide
loxaphene
1 , 1 , 2-lrichloroethane
Tr ichloroethylene
Jrichlorophenol (mixed)
2 , 4 . 5 - 1 r i ch 1 oropheno I
2,4.6-lrichlorophenol
Iris(2.3-dibromopropyl ) phosphate
Irypan blue
Uracil mustard
Vinyl chloride
CASRN
12672296
11097691
11096825
01120714
00075558
00081072
00094597
07488564
18883664
01746016
00630206
00079345
00127184
00062555
00062566
00095534
00106490
00636215
08001352
00079005
00079016
25167822
00095954
00088062
00126727
00072571
00066751
00075014
Degree of
Humans
Inndequate
1 nadequate
Inadequate
No Data
No Data
Inadequate
No Data
1 nadequate
No Data
Inadequate
Inadequate
Inadequate
Inadequate
No Data
No Data
Inadequate
No Data
Inadequate
No Data
No Data
Inadequate
No Data
Inadequate
No Data
No Data
No Data
No Data
Suf f ic ient
Evidence
Animal s
No Data
Suf f ic ient
Suf f ic ient
Sufficient
Sufficient
L imi ted
Sufficient
Suf f icient
Suf f icient
Suf f icient
L imi ted
L imi ted
Suf f icient
Suf f icient
Suf f icient
Suf f icient
L imi ted
Sufficient
Sufficient
L imi ted
Suf f icient
Sufficient
Inadequate
Suf f icient
Sufficient
Suf f icient
Sufficient
Suf f ic ient
Weight -
of -Evidence
Group
t
B2
B2
82
B2
C
B2
B2
82
B2
C
C
82
82
82
82
C
B2
B2
C
82
B2'4
D
82
82
82
82
A
Potency
Factor
t
t
50.5
10.0
259
0.007
0.18
0.93
109
659,000
0.85
1.66
0.29
24.8
1.05
0.069
0.94
0.069
9.79
0.36
0.10
O.OB*4
Hone
0.08
9.76
a
a
18.1
Potency
Group
2
2
2
2
1
3
3
3
1
1
3
2
3
2
2
3
3
3
2
3
3
3
None
3
2
2
2
2
Hazard
Ranking
MED
MED
MEO
MED
HIGH
LOU
LOW
LOU
HIGH
HIGH
LOU
LOU
LOU
MED
MED
LOU
LOU
LOU
MEO
LOU
LOU
LOU
Noneb
LOU
MED
MED
MED
HIGH
-------
FOOTNOTES
GENERAL FOOTNOTES
a Data available are inadequate for calculating a potency factor and no
quantitative inferences can be made. Hence,'the substance is assigned to
Potency Group 2, the mid-range potency group.
ab The bioassay used to calculate the potency factor suggests that the
substance may be highly carcinogenic (i.e., all treated animals developed
tumors, and therefore there is no basis for calculating a specific ED^Q
dose). Hence, the substance is assigned to Potency Group 1.
abc The bioassay used to calculate the potency factor suggests that the
substance is a possible weak carcinogen, but due to uncertainties
associated with the pharmacokinetics of the substance, there is no basis
for calculating a specific ED^Q dose. Hence, the substance is assigned
the lowest potency group, Potency Group 3.
b No RQ can be assigned based on potential carcinogenicity. Other primary
criteria must be used for assigning RQs.
f Potency factor estimate is derived from human epidemiology data.
m When the weight of evidence is expressed as a range, for example, B2/C,
the hazard ranking is based on the higher weight-of-evidence group.
s This particular compound is not classified due to the inadequate nature or
nonexistence of data. However, there is a potential of its beir.e
converted into a carcinogenic form when released into the environ.mer.t.
For reportable quantity ranking purposes it should be considered as having
a weight of evidence similar to that of the known carcinogenic form.
== The basis for the Agency's determination that lead and lead compounds are
potential carcinogens is undergoing review by EPA's Science Advisory
Board. No hazard ranking will be assigned to these compounds at this
time.
-43-
-------
CHEMICAL-SPECIFIC FOOTNOTES
Arsenic--The weight of evidence for the carcinogenicity of inorganic
arsenic compounds is based on positive human studies in which exposure was
by either water or air. The weight of evidence is group A. The exact
species of inorganic arsenic that is directly carcinogenic to humans is
not known, but it is assumed that since arsenic is chemically convertible
among the chemical species both in vitro and in vivo, that all inorganic
species of arsenic are of equal concern. The potency factor is the same
as that given for "arsenic" (potency factor - 142.31). Arsenic trioxide
and potassium arsenite are classified as having sufficient human evidence
because human studies that specifically identify those compounds have beer.
conducted and show sufficient evidence of causal association.
Chromium metal--The latest Health Assessment Document on Chromium
(EPA-600/8-83-014F, August 1984) states that chromium metal is
biologically inert and has not been reported to produce toxic effects or
other harmful effects in man (p. 7-1).
Chromium compounds (hexavalent)--Grouping is based on weight of evidence
for chromate production, workers and animal data which indicate that the
inhalation of hexavalent chromium is carcinogenic. The potency estimate
is based on epidemiological data for the inhalation of hexavalent chromium
by chromate workers (potency factor - 388.99, weight-of-evidence group A).
Benzo(b)fluoranthene--Calculated, using the potency factor estimate for
benzo(a)pyrene as a reference.
Cadmium weight of evidence and potency are based on epidemiology daca for
cadmium workers exposed to cadmium oxide and/or cadmium fume. Although
human data for cadmium salts are lacking, due to the responsiveness of
animals to soluble cadmium compounds, especially cadmium chloride, the
weight of evidence for cadmium acetate, bromide and chloride are
considered to be the same as those cadmiuir. compounds to which workers are
exposed.
Chloromethyl methyl ether [CMMEj--Technical grade chloromethyl methyl
ether is contaminated with 1% - 8% bis(chloromethyl)ether which is a known
human carcinogen. Hence the human evidence for this compound and the
hazard ranking is based on the evidence for bis(chloromethyl)ether.
Asbestos--A potency factor estimate for asbestos is inappropriate here
because the carcinogenic potential of asbestos is related to specific
fiber shapes, sizes, and atmospheric concentrations. Air concentrators
are usually measured either as a number of fibers or mass. However, no
direct relationship exists between air, fiber/ml (75 microns) concentra-
tions (by the phase contrast light microscope method) and mass concen-
trations in mg/nr (determined by electron microscopy). The relationship
depends on the type of environmental sample, the type of asbestos in the
air, and the size of the fibers. As a delibrate policy choice, asbestos
is assigned a "HIGH" hazard ranking, as are most group "A" substances.
• 44-
-------
CHEMICAL-SPECIFIC FOOTNOTES (Continued)
c PCBs--The Aroclors are mixtures of polychlorinated biphenyls (PCBs). The
manufacturing process for commercial PCB products, such as the Aroclors,
yields products composed of a mixture of 20-6.0 different PCB compounds.
Individual lots of Aroclors of the same average chlorine content may
differ greatly in both their components and amount of each component.
Only Aroclors 1254 and 1260 have been tested for carcinogenic potential
and both produce a positive response. Therefore, for the purpose of RQ
hazard ranking, all Aroclors are considered to have carcinogenic potential
similar to Aroclor 1254, and Arocolor 1260.
u Beryllium--Every soluble beryllium compound that has been tested,
including beryllium sulfate, fluoride, oxide, phosphate, as well as beryl
ore, zinc beryllium silicate, and beryllium metal have been shown to be
carcinogenic. It is therefore considered highly likely that all soluble
forms of beryllium are carcinogenic in animals. The potency factor for
all soluble forms of beryllium is based on beryllium sulfate exposure in
rats. It is believed that these forms of beryllium would pose a similar
hazard to humans. The potency factor for beryllium metal is based on human
occupational exposure to much less soluble forms of beryllium, mostly
beryllium oxides.
v Benzo(a)pyrene--This compound is generally found in the environment as
part of a complex mixture of polycyclic aromatic hydrocarbons.
Benzo(a)pyrene has been found to be carcinogenic in animals and thus its
presence in the mixtures usually indicates the presence of a known animal
carcinogen. Many mixtures containing benzo(a)pyrene have also been
causally related to human cancer, e.g., soots, tars, coke oven emissions.
cigarecte smoke. Therefore, benzo(a)pyrene should be treated as a human
carcinogen.
x Creosote --There are no adequate studies of workers exposed to creosote
wood preservatives. It has been demonstrated that chimney sweeps exposed
to the creosote from the burning of wood or coal have an elevated risk of
cancer. In addition, creosote, including creosote wood preservative.
contains many of the compounds in other polycyclic aromatic hydrocarbon
mixtures such as roofing tar pitch and coke oven emissions that have been
found to be carcinogenic.
y Lead phosphate--The animal evidence is based in part on consistent
findings for other inorganic lead salts.
z N-Nitrosomethylvinylamine--No control data for animal studies exist.
However, based on more than one study and structure-activity relationship
with other nitrosamines, the weight of evidence is considered to be
sufficient.
• 45-
-------
CHEMICAL-SPECIFIC FOOTNOTES (Continued)
z2 Melphalan--The "limited" designation given to the human evidence for
carcinogenicity is based only upon three independent series of cases of
multiple myelomas that were treated with melphalan. The case studies
represent meager evidence of the carcinogenicity of melphalan and hence
the evidence is less than "limited" but greater than "inadequate."
z3 Dinitrotoluene--In the absence of analytical data, it should be considered
that the mixture contains 2,4-dinitrotoluene which is a potential human
carcinogen. Therefore, for hazard ranking purposes, the mixture should be
considered as hazardous as 2,4-dinitrotoluene.
z4 Trichlorophenol--In the absence of analytical data, it should be
considered that the mixture contains 2,4,6-trichlorophenol which is a
potential human carcinogen. Therefore, for hazard ranking purposes, the
mixture should be considered as hazardous as 2 ,<*, 6-trichlorophenol.
z7 Nickel--The latest Health Assessment Document on Nickel states that the
nickel ion (Ni+2) could be the ultimate carcinogenic form of nickel.
Although this is unproven, it is considered prudent to make this
assumption for covalent nickel (forms that generate Ni+2) and nickel
salts. Proven carcinogenic forms of nickel are nickel refinery dust and
nickel subsulfide (both in man) and nickel carbonyl (demonstrated in test
animals). The former two substances are in weight-of-evidence group A,
while the latter substance is in weight-of-evidence group B2. The sales of
nickel show some carcinogenic activity. The testing of these nickel salts
is inconclusive for assessment of cancer at this time due to limitations
of the data base on these nickel salts, but since there is some cancer
activity it is recommended by the HHAG that the hazard ranking under
CERCLA be reported as "LOW." This "LOW" hazard ranking reflects the
current data base on the nickel salts.
z9 Organic arsenic compounds are considered to be chemically different from
the inorganic arsenic compounds such that they are assessed for
carcinogenicity separately from the inorganic arsenic compounds. There
are no data (weight-of-evidence group D) implicating organic arsenic
compounds so that the carcinogenicity is indeterminate at this time.
z-10 Pentachloronitrobenzene - The potency estimate is for technical grade
pentachloronitrobenzene, which contains the probable human carcinogen
hexachlorobenzene.
NOTE: The gaps in the letters assigned to the footnotes in this table exist so
that consistency with the numbers assigned to the footnotes in the
summary table in OHEA-C-073 (Appendix A in "Technical Background
Document to Support Rulemaking Pursuant to CERCLA Section 102, (Proposed
Rulemaking)." Volume 3, March 1985), is maintained.
-U6-
------- |