-------

-------

-------
                                    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-

-------