Document XO8L6OEZzoMrM5d1z6rZ5jBvd

United States Environmental Protection Agency Research and Development IOiW-i73C.`f-5Ar4.> 0HEA-C-073 June 1988 Final?? -AS* METHODOLOGY FOR EVALUATING POTENTIAL CARCINOGENICITY IN SUPPORT OF REPORTABLE QUANTITY ADJUSTMENTS PURSUANT TO CERCLA SECTION 102 Prepared for OFFICE OF EMERGENCY AND REMEDIAL RESPONSE Prepared by HUMAN HEALTH ASSESSMENT GROUP Office of Health and Environmental Assessment Washington, D.C. 20460 CMA 121637 OHEA- C-073 June 1988 Final METHODOLOGY FOR EVALUATING POTENTIAL CARCINOGENITICY IN SUPPORT OF REPORTABLE QUANTITY ADJUSTMENTS PURSUANT TO CERCLA SECTION 102 Prepared for Office of Emergency and Remedial Response Office of Solid Haste and Emergency Response Prepared by Office of Health and Environmental Assessment Office of Research and Development U.S. Environmental Protection Agency Washington, d.C. CMA 121638 DISCLAIMER This document has been reviewed in accordance with U.S. Environmental Protection Agency policy and approved for publication. Mention of trade names or commercial products does not constitute Agency endorsement or recommendation for use. A-ii i I CMA 121639 CONTENTS PAGE Tables............................................................................. A-vl Preface.................................................................................................................. A-vil Abstract......................................................................... ....................................... A-lx Authors, Contributors, and Reviewers................................................ A-x 1. INTRODUCTION......................................................... '...................................... A-l 1.1. BACKGROUND.............................................. ....................................... A-l 1.2. METHODOLOGY........................................... ...................................... A-4 2. LITERATURE SEARCH AND INFORMATION SOURCES......................... A-6 2.1. SEARCH STRATEGY......................................................................... A-6 2.2. SELECTION OFAPPROPRIATE STUDIES..................................... A-8 3. APPROACH TO DATA EVALUATION........................................................ A-11 3.1. QUALITATIVE PHASE: WEIGHT OF EVIDENCE OF CARCINOGENICITY............................... ..... .................................... A-12 3.1.1. Assessment of Evidence for Carcinogenicity from Studies in Humans....................................... .................................... A-12 3.1.2. Assessment of Evidence for Carcinogenicity from studies in ExperimentalAnimals...;..................................... A-14 3.1.3. Categorization of Overall Weight of Evidence forHumanCarcinogenicity.............. A-16 3.2. QUANTITATIVE PHASE: ESTIMATION OF CARCINOGENIC POTENCY................................................. ......................................... A-19 3.2.1. Model Selection for Analysis of Dose Response Data.......................................................... A-20 3.2.2. Selection of Bioassay Response Data to be Used in the Model to Calculate Potency............................... ....................................... .. a-22 A-iii CMA 121640 CONTENTS (continued) PAGE 3.2.3. Adjustment (Transformation) of Dose Data..................................................................... ..... A-23 3.2.3.1. Calculation of Dose in mg/kg/day from Doses Expressed as Dietary Concentrations.......... A-24 3.2.3.2. Calculation of Dose in mg/kg/day from Doses Expressed as Water Concentrations.............. A-2 5 3.2.3.3. Calculation of Dose in mg/kg/day from Doses Expressed as Air Concentrations................... A-26 3.2.3.4. Adjustment for Non-Continuous Exposure......... ...................................... A-29 3.2.3.5. Adjustment fof Absorption, Distribution, Metabolism, and Excretion.... ^................................... A-30 3.2.4. Calculation of the Human Potency Factor (F)............................. A-30 3.2.4.1. Animal Potency................................... A-32 3.2.4.2. Adjustment for Less Than Lifetime Studies.............................. A-32 3.2.4.3. Human Potency (F)......................... A-33 3.2.5. Grouping of Chemicals B*sed on Carcinogenic Potency............................................ A-34 3.3. OVERALL HAZARD RANKING BASED ON COMBINED QUALITATIVE AND QUANTITATIVE ASSESSMENTS................. A-35 3.3.1. Use of Chemical and Environmental Fate and Transformation Data in Hazard Ranking of Metals and Tfyeir Salts.............. A-38 3.3.2. Special Problems in Hazard Ranking of Chemicals Associated with Multimedia Exposure in Humans................................................ A-38 A-iv CMA 121641 CONTENTS (continued) PAGE 4. SUMMARY....................................................................................................... A_39 5. REFERENCES........................................................... A_41 APPENDIX - HAZARD RANKING OF POTENTIAL CARCINOGENS................ A-42 A-v CMA 121642 TABLES 2- 1 Sources of bibliographic and numerical information for suspect carcinogens.............................................................. .. 3- 1 Tentative weight of evidence based on human and animal evidence............................................. 3-2 Sample table for the derivation of potency factor (F)........................................................ 3-3 Hazard rankings scheme for reportable quantities under CERCLA...................................................... PAGE A-7 A-18 A-31 A-36 A-vi CMA 121643 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 l," March 1985, and in the final rule published on April 4, 1985 (U.S. EPA, 1985). 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 maximum RQ for potential carcinogens, the Human Health Assessment Group was request'd to rank p tential c rcinogens on a A-vii CMA 121644 three-tier scale (high/ medium, and low) that corresponds to rq 1 vels of l, 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. A-viii CMA 121645 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 194 chemicals are available as separate documents. A-ix CMA 121646 AUTHORS, CONTRIBUTORS, AND REVIEWERS The Human Health Assessment Group within EPA's Office of Health and Environmental Assessment (OHEA) was responsible for the preparation of this document and provided overall direction and coordination during the production effort. The first draft of this document was prepared by Syracuse Research Corporation (John Risher, Project Officer). The OHEA Human Health Assessment Group was primarily responsible for methodology development and peer review of the hazard ranking of carcinogens. AUTHORS James Cogliano Human Health Assessment Group U.S. Environmental Protection Agency Washington, D.c. Apama Koppikar Human Health Assessment Group U.S. Environmental Protection Agency Washington, D.C. James M. Conis Steven C. Gibson Jeffrey S. Gift Alan w. Messing Gregory R. Ricci Bartholomew L. Tuffly C-E Environmental, Inc. Washington, D.C. CONTRIBUTORS Roy E. Albert, M.D. Steven Bayard, Ph.D. David L. Bayliss, M.S. Robert p. Bellies, Ph.D. Chao W. Chen, Ph.D. Arthur Chiu, M.D., Ph.D. Margaret M.L. Chu, Ph.D. Vincent Jam s Cogliano, Ph.D. A-x CMA 121647 William H. Farland, Ph.D. Herman J. Gibb, B.S., M.P.H. Bernard H. Haberman, D.V.M., H.S.R. (deceased) Charalingayya B. Hiremath, Ph.D. James w. Holder, Ph.D. Apamft M. Koppxlcat, M.D., D.P.H., D ^ TH Robert E. McGaughy, Ph.D. Jean C. Parker, Ph.D. William E. Pepelko, Ph.D. Charles H. Ris, M.S., P.E. Dharm V. Singh, D.V.M., Ph.D. Hugh L. Spitzer, B.A. Todd W. Thorslund, Sc.D. Human Health Assessment Group U.S. Environmental Protectin Agency REVIEWERS The following individuals provided peer review of this document and/or earlier drafts of this document (outside reviewers including other EPA offices). Donald Barnes Office of Pesticides and Toxic Substjances U.S. Environmental Protection Agency Washington, D.C. Judith Beilin Office of Solid Waste U.S. Environmental Protection Agency Washington, D.C. James Cogliano Human Health Assessment Group 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 Prot ction Agency Washington, D.C. A-Xi CMA 121648 K. Jack Kooyoomjian Science Advisory Board U.S. Environmental Protection Agency Washington, D.C. Apama Koppikar Human Health Assessment Group U.s. Environmental Protection Agency Washington, D.C., Robert McGaughy Office of Health and Environmental Ajssessment 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 Dr. Rolf Hartung School of Public Health University of Michigan Dr. Benjamin Van Duuren Institute of Environmental Medicine New York University Medical Center A-xii I CMA 121649 1. INTRODUCTION 1.1. BACKGROUND I This report describes the 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 amount of a hazardous substance equal to or greater than its RQ. Under CERCLA Section 102(b), the RQ of any hazardous substance designated in Section 101(14) is 1 pound unless a different RQ has been established pursuant to Section 311(b)(4) 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 to 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 (CWA). Details of the methodology are given in the technical background document (U.S. EPA, 1985a), and in the A-l 0MA 121650 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, dermal, and inhalation), ignitability, reactivity, chronic toxicity, and potential carcinogenicity. The Agency ranks each intrinsic property (other 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 photolysis 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: * It has not been demonstrated that there is a threshold level of exposure below which potential carcinogens do not present some risk of cancer. Therefore, a release A-2 CMA121651 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, 8145). * Cancer risks are considered to be cumulative. A number of small releases can be as serious as a single large release (U.S. EPA, 1987 pp. 81*0, 8146). Cancer is not immediately manifested. There is a latent period between exposure to a carcinogen and the manifestation of cancer that mhkes 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 maximum RQ for potential carcinogens, the Human Health A-3 CMA 121652 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) v ight-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 consistent with the Agency's final guidelines, and should include the Agency's new weight-of-evidence critetia. 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-ofvidence determinations, potency factor assignments for each potential carcinogen under study, and a bibliography of other pertinent references is contained in a profile for each A-4 CMA 121653 ch mical. 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 methodology document, lthe profiles can be accessed separately in the record that supports CERCLA RQ adjustment rulemakings, which are available in Room H2427 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. A-5 CMA 121654 2. LITERATURE SEARCH AND INFORMATION SOURCES 2.1. SEARCH STRATEGY The objective of the 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 1985 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 (CAS) names, CAS registry numbers, common synonyms, designated substructures, and key words related to carcinogenicity and mutagenicity when appropriate. All of the ch micals 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 A-6 CMM21655 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 on 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, 1976 "Documentation of the Threshold Limit Values for Substances in Workroom Air," American Conference of Governmental Industrial Hygienists "Ambient Water Quality Criteria Documents," U.s. EPA, ORD National Institute for Occupational Safety and HealthCriteria Documents, Technical Report?, Special Occupational Hazard Reviews, Current Intelligence Bulletins, Information Profiles on Potential Occupational Hazards "Current Contents--Life Sciences," Institute for Scientific Information A-7 CMA 121656 bibliographic data bases were then searched by developing a list f 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. In most cases, the 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 th 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 to provide qualitative evidence of carcinogenicity, yet still may be inappropriate for quantitative estimation of carcinogenic potency. 1> A-8 CMA 121657 It is also important to consider tha ralavanca of th 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. Before a study can be used to support either a qualitative or quantitative assessment of carcinogenicity, several criteria should be met. These general criteria, that 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? A-9 CMA 121658 2. Were any variations from design noted, explained, and/or considered in reporting results? 3. Were analytical and quantitative parameters provided? Factor: 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 pt 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 indicsted by the reviewer if the data are intended to be used for qualitative and quantitative assessment purposes. A-10 CMA 121659 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 recognizee 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 necessary 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 assessm nt reports reviewed were i. - A-ll CMA 121660 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 (DWCDsi). 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 Carcinogenicity from Studies In Hunans Evidence of carcinogenicity from human studies comes from three main sources: Case reports of individual cancer patients who were exposed to the agent(s); Descriptive epidemiologic studies in which the incidence of cancer in human populations was found to vary in space or time with exposure to the agent(s); and Analytical epidemiologic (case-control and cohort) studies in which individual exposure to the agent(s) was found to be associated vitp an increased risk of cancer. A-12 CMA121661 Three criteria must be met before a causal association can be inferred between exposure and cancer in humans: There is no identified bias that can explain the . association; The possibility of confounding variables has been considered and ruled out as explaining'the association; and The association is unlikely to be due to chance. In general/ although a single study bay 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 strong, 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 carcinogenicity1 is classified as: Sufficient evidence of carcinogenicity, which indicates that there is a causal relationship between the agent and human cancer; Limited evidence of carcinogenicity, which indicates that a causal interpretation is credible, but that alternative explanations, such ms chance, bias, or confounding, could not adequately be excluded; --- y ^-For purposes of public health protection, agents associated with life-threatening benign tumors in humans are included in the evaluation. i. A-13 CMA 121662  Inadequate evidence, which indicates that one of two conditions prevailed: (a) there were few pertinent data; or .(b) the available studies, while showing evidence of association, did ndt exclude chance, bias, or confounding and therefore a causal interpretation is not credible; No data, which indicates that data are not available; or 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 Cargihocenicitv from studies in Experimental Animals These assessments are classified intp five groups: Sufficient evidence2 of carcinogenicity, which indicates that there is an increased incidence of malignant tumors or combined malignant and benign tumors3: (a) in multiple species or strains; (b) in multiple experiments (e.g., with different routes of 2An 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 carcinogenicity, but may be changed to "limited" when warranted by the specific information available on the agent. 3Benign 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. A-14 CMA 121663 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; 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) an increase in the incidence of benign tumors only; 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; No data, which indicates that da[ta are not available; or No evidence, which indicates that there is no increased incidence of neoplasms in at leapt two well-designed and well-conducted animal studies in different species. ! A-15 CMA 121664 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 Carcinogenicity The overall scheme for categorizing the weight of evidence i of carcinogenicity of a chemical for humans uses a three-step process: (l) 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 with the tumor information from human and animal studies include structure-activity relationships; short-term test findings; results of appropriate physiological, biochemical, and toxicological observations; and comparative metabolism and kinetic studies. The nature of these findings may cause one to adjust the overall categorization of the Veight of evidence. The substances are categorized in thd manner shown below. Group A--Human 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. A-16 CMA 121665 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 B1 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 there is "sufficient" evidence from animal studies and for which there is "inadequate" evidence or "ho 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 in studies having inadequate design or reporting; (c) benign but not malignant tumors with an agent showing no response in a variety of short-term tests for mutagenicity; and (d) responses of A--17 CMA 121666 TABLE 3-1. TENTATIVE WEIGHT OF EVIDENCE BASED ON HUMAN AND ANIMAL EVIDENCE Human Evidence Sufficient Animal Evidence Limited Inadequate No Data No Evidence Sufficient A A A AA Limited B1 B1 B1 B1 B1 Inadequate B2 C D DD No Data B2 C D DE No Evidence B2 C D DE aThe 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 to determine if the designation of the overall weight of evidence needs to be modified. Relevant factors to be included along with the tumor data from human and animal studies include structure-activity relationships, short-term test findings, results of appropriate physiological, biochemical and toxicological observations, and comparative metabolism and pharmacokinetic studies. The nature of these findings may cause an adjustment of the overall categorization of the weight of evidence SOURCE: U.S. EPA, 1986. A-18 CMA 121667 marginal statistical significance in a tissue known to have a high or variable background rate of cancer. Group D--Not Classifiable as to HumanCarcinogenicity This group is generally used for agent(s) with "inadequate" human and animal evidence of carcinogenicity or for which "no data" are available. Group E--Evidence of Non-Carcinoaenicitv for Humans This group is used for agent(s) that show no evidence for carcinogenicity 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 i$ based on the available evidence and should not be interpreted as a definitive conclusion that the agent will not 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: (1) estimating the cancer risk associated with a particular level of exposure; and (2) making comparisons ai^ong 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. 1 A-19 CMA 121668 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 [-(qo + + ... + q* d*) ] In accordance with the Agency's "Guidelines for Carcinogen i 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 (NCI) bioassay. Any variance from the standard multistage model is describe4 in the individual potential carcinogen profiles. ), A-20 CMA121669 ft I For reportable quantity adjustments pursuant to CERCLA Section 102, the potency factor (F) is defined as the reciprocal of the estimated dose in mg/kg/day, associated with a lifetime cancer risk of 10 percent (ED^o) The reciprocal of the ED10 is us d because it is the more direct measure of potency. The ED10 itself is inversely related to potency. The potency factor (F) is used in place of the upper bound on the linear coefficient (q*i) (U.S. EPA, 1989), that HHAG normally uses to estimate potency because: It can be estimated without the use of many assumptions required for q*i> This is possible because there is no need for extrapolation below the experimentally observable dose range4. 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. The point estimate of ED^o, 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*i* Extrapolation below the experimentally observable dose range is the whole purpose of thE q*i 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) A-21 CMA 121670 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 bata 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 l/ED^o* Human epidemiologic data must be reviewed and used on a case-bycase basis. Hence, if epidemiologic studies are selected as suitable for derivation of a potency 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 approach 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 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-term animal studies showing the greatest is A-22 CMM21671 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 flose Data Before a potency factor can be calculated all dose information must be transformed to standard units of milligram (mg) (substance) per kilogram (kg) (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 abs rbed or m tabolized dose is significantly less than the A-23 CMA 121672 administered dose, then the dose data muOt be converted to a "transformed dose." The next three subsections discuss hov 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 mq/ko/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 d se may be estimated by using standard fpod consumption estimates based on the fraction of body weight that is consumed each day as food (f) (U.S. EPA, 1980): Species Mouse Rat Human ____ 0.13 0.05 0.028 In order to obtain the dietary dose (d) from the data given in parts per million (ppm), the daily experimental dose in ppm is multiplied by f: (3-2) d - ppm x f A-24 CMA 121673 Note that 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 product has units of mg of toxicant per kg body weight per day. 3.2.3.2. Calculation of Dose in mc/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 Mouse Rat Human _&__ 0.17 0.078 0.029 The dietary dose (d) in mg/kg/day is calculated by multiplying the daily dose in ppm by the ajppropriate fw: (3-3) d - ppm x fw A-25 CMA 121674 Note that ppm in water has units of big toxicant per liter water and that fw has units of liters of water per kg body weight per day. Thus the product has units of mg toxicant per kg body weight per day. 3.2.3.3. Calculation of Dose in ma/kc/day from Doses Expressed as Air Concentrations--When exposure is via inhalation, the calculation of dose can be considered for two cases where: (l) i th carcinogenic agent is either a completely water-soluble gas or an aerosol and is absorbed in proportion to the amount of air breathed in; and (2) where the 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 g^ses, 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 (FASE8, 1974) that 25-g mice breathe 0.0345 cubic meters (m3) per day and 113-g rats breathe 0.105 m3/day. 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: A-26 CMA 121675 * (3-4) (3-5) for mice, I 0.0345 (W/0.025J2/3 m3/day; and for rats, I - 0.105 (W/0.113)2/3 m3/day# The weight ratio is raised to the two-thirds power because, to a close approximation, the surface arata 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/diay is adopted as the standard breathing rate. This is calculated from the observation (ICRP, 1977) that the average breathing rate is 107 cubic centimeters (cm3) per 8-hour workday and 2 x 107 cm3 in 24 hours. 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.,. i m 1.3 0.64 0.29 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) dvx ixr A-27 CMA 121676 Note that v has units of mg toxicant per a3, i has units of a3 per kg body weight per day, and r is dimensionless. Thus the product has units of ag toxicant per kg body weight per day. In the absence of experiaental information or a sound theoretical argument to the contrary, r is assumed to be the sane for all species and therefore drops out of the calculations. For Case 2, the dose in ag/day of partially soluble vapors is proportional to O2 consumption, which in turn is proportional to W2/3. 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. Therefore, by expressing the O2 consumption as 02 * k x W2/3, where k is a constant independent of species, it follows that the average dose per day in ag 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/m3 in experimental animals is equivalent to the same concentration in huaans. This is supported by the observation that the minimum alveolar concentration necessary to produce a given NstageN of anesthesia is similar in man and animals (Dripps A-2B CMA121677 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 contrary, that absorption is equal by either exposure route. In this case, the dose (d) in mg/kq/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/m3 (v) by the following formula: (3-9) v - 0.041 x molecular weight (gas) x ppm (Note that 1 ml in 1 cm3 is 1 ppm (by volume); therefore, 0.041 x molecular weight (MW) is the weight in mg of l ml of a gas.) 3.2.3.4. Adjustment for Non-Contlnuous 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 5 times per week or, for inhalation studies* 6 hours per day) then the transformed dose is: (3-10) transformed dose d x le A-29 CMA 121678 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) r a fraction of a day (e.g., 6/24), then the transformed dose becomes, for example: (3-11) transformed dose d x ^ x ^ x.^ 3.2.3.5. Adjustment for Absorption. 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 this information. For example, if the effective dose to the target organ in 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. 3.2.4. Calculation of the Human Potency factor m The information needed for calculating a human potency factor (F) can be found in the profiles entitled "Evaluation of the Potential Carcinogenicity of tsubstance 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. A-30 CMA 121679 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 weight: Duration of treatment (le): Duration of study (Le): Lifespan of animal: Target organ: Tumor type: Experimental doses/exposurec (mg/kg): Transformed doses*1 (mg/kg/day): Tumor incidence: Lifetime animal potency : Human potency factor (f): NCI (1976) oral (gavage) mice B6C3F1 F com oil 0.03 ka 546 days 644 - 651 days 730 daysb Liver hepatocellular carcinoma 477 238 0 288 39/41 0.104 1.97 143 36/45 0 0/20 "Reported. "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/week)/ 7(days/week) x duration. A-31 CMA121680 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 ED10 " 9.62 mg/kg/day. The animal potency is the reciprocal of this dose, 0.104 (mg/kg/day)-1. 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 factor of (L/Le)3 to obtain an estimated potency for full lifetime risk. As explained by the EPA (1980): We assume that if the average dose (d) is continued, the age-specific rate 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... A-32 CMA 121681 is n (3-12) Lifetime Animal Potency = Observed Animal Potency x (VI*)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 Jt (730/648)3 - 0.149 3.2.4.3. Human Potency () -- Finally, tlie 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 Anialal Potency x (70 kg/Wa)V3; where Wa is the weight of the animal and 7io kg is the assumed average weight of humans. This is in accordance with the Agency's cancer guidelines (U.S. EPA, 1986)5. sThe Agency's guidelines reflect that animal potency is converted to human potency by first multiplying the animal ED^q by the ratio of the the weight of the animal to the weight of man, and dividing the entire ED^q by the ratio of the surface area of the animal to man. Hunan ED10 - MlMl EDio X (Wa/Branl - aninal EDi0 (W-/Wnan) 1/3 <a/man>2/3 Then, the reciprocal of the Human ED10 is the adjusted human potency. A-33 CMA121682 For chloroform, the human potency (F) is: (3-15) Human Potency (F) - 0.149 X (70/0.03) V3 - 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 l 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: If the best available data suggest that the substance is a possible strong carcinogen, but there is no basis for assigning a specific ED10 dbse (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). 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 A-34 CMA121683 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). If the best available data suggest that the substance is a possible weak carcinogen, but there is no basis for assigning a specific ED^g 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. i 3.3. 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 carcinogenicity (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. 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 \. A-35 CMA 121684 TABLE 3-3. HAZARD RANKING SCHEME FOR REPORTABLE QUANTITIES UNDER CERCLA w ight-of-Evidence Group Potency Group : 1 Potency Factor : F > 100 2 F - 1 to 100 3 F< 1 A - Carcinogenic to humans High High B - Probably carcino genic to humansa C - Possibly carcinog nic to humans High Medium Medium Low D - Not classifiable as to human carcinogenicity No hazard ranking E - Evidence of non carcinogenicity for humans No hazard ranking aIncludes weight-of-evidence Groups B1 and B2. Medium Low Low CMA12185 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 dirtect 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 0 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 pranked 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 adjust d RQ. A-37 CMA 121686 3.3.1. Use of Chemical and Environmental Fate and Transforma tion Data In Hazard Ranking of Metals 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 ar 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 abovp 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 with Multimedia Exposure in Humans Carcinogenic hazard ranking in the present study must take into account 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. A-38 CMA121687 4. SUMMARY The Agency's Human Health Assessment Group (HHAG) has developed a unique method for ranking CERCLA hazardous substances for 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 weightof-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 combined to arrive at a hazard ranking for each hazardous substance evaluated for potential carcinogenicity. A-39 CMA 121688 There are two separate appendices to this methodology document. The first is a summary table Of the hazard ranking r suits. The second consists of all of the individual chemical profiles prepared to support these results. The profiles are not attached but are available in Room M2427 at the U.S. Environ* mental 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 holidayjs. 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. A-40 / 121689 5. REFERENCES Crump, K.S.; Watson, W.W. (1982). GLOBAL 82: a fortran program to extrapolate dichotomous animal carcinqgenicity data to low doses. National Institute of Environmental Health Sciences, Contract No. l-ES-2123. Doll, R. (1971). Welbull 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 anesthesia: the principles of safe practice. 5th Ed. Philadelphia, PA: W.B. Saunders Company, pp. 121-123. FASEB (Federation of American Societies fbr 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, Phrgamon Press. U.S. Environmental Protection Agency (1980* Nov. 28). 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 (1985, March). Technical background document to support rulemaking pursuant to CERCLA Section 102, Volume 1. U.S. Environmental Protection Agency (1985, April 4). Notification requirements; reportable quantity adjustments. Federal Register 50: 13456-13522. U.S. Environmental Protection Agency (1986* Sept. 24). Guide lines for carcinogen risk assessment. Federal Register 51: 33992-34003. U.S. Environmental Protection Agency (1987, March 16). Repor table quantity adjustments. Federal Register 52: 8140-8171. u.s. Environmental Protection Agency (1989, July). Technical background document to support rulemaking pursuant to CERCLA Section 102, Volume 3. A-41 CMA 121690 APPENDIX HAZARD RANKING OF POTENTIAL CARCINOGENS A-42 CMA 121691 C M A121692 Substance 1 Acetamide, N-fluoren-2-yl 2 Acrylonitrile 3 Aldrin A Amitrole .5 Arsenic 6 Arsenic acid r Arsenic disulfide s Arsenic pentoside 9 Arsenic trichloride 10 Arsenic trioside 11 Arsenic trisulfide 12 Cacodylic acid 13 Catcium arsenate 14 Calciuai arsenite 15 Cupric acetoarsenite 16 Diehlorophenylarsine 12 Diethyl arsine 18 Lead arsenate 19 Potassium arsenate 20 Potassium arsenite 2L Sodium arsenate 22 Sodium arsenite 23 Asbestos 24 Auramine 25 Ataserine 26 Aiiridine 22 BenKOacridine 28 BenKalanthracene 29 Bentene 30 Beniidine and its salts 11 Bento(b)fluoranthene 32 Benio(k)f luoranthene 33 Bento(a)pyrene HAZARD RANKING OF POTEMtlAL CARCt CASRM Degree of E videne e Humans Animals Ueightof -Evidence Group Potency Factor Potency Group Hazard Ranking 00053963 00102131 00309002 00061825 02440382 01322522 02228394 01301328 01303262 02284341 01322533 01303339 00025605 02228441 52240166 - 12002038 00696286 00692422 02264409 02284410 101245Q2 02631892 02284465 01332214 00492806 00115026 00151564 00225514 00056553 00021432 00092825 00205992 00202089 00050328 Ho Data Limited Inadequate Inadequate Sufficient d Sufficient Sufficient Sufficient Sufficient Inadequate Mo Data d d d Sufficient d No Data d d d No Data No Oata d d Sufficient d d Sufficient Inadequate No Data Ho Data Ho Data No Data Sufficient Suf fic tent Mo Data Mo Data Inadequate No Data Ho Data Mo Data Inadequate Ho Data Inadequate Mo Data Mo Data Inadequate Ho Data No Oata Inadequate Mo Data Inadequate Ho Data No Oata Sufficient Sufficient Sufficient Suf ficient timited Sufficient Sufficient Suf ficient Suf ficient inadequate Sufficient A-43 B2 B1 B2 B2 A d d d d Ad d D*9 d d d Dl9 Dl9 d d Ad d d A B2 B2 B2 C 62 A A B2 0 B2 148 2.28* 239 3.30 142 d 1 2 1 2 1 1 d d d d d Hone d d d Hone None d d 1 a d P 0.46 169 336 a 21.1 0.22* 2220* 246 Mone 248 1 1 1 1 1 None 1 1 1 Hone Hone 1 1 1 t 1 P 3 1 1 2 2 3 1 1 Mone 1 HIGH HED HIGH HED HIGH HIGH HIGH HIGH HIGH HIGH HIGH Mone** HIGH HIGH HIGH Noneb Noneb HIGH HIGH HIGH HIGH HIGH HIGH LOU HIGH HIGH LOU HE0 HE0 HIGH HIGH Noneb HIGN Substance HAZARD RANKING OF POTENTIAL CARCINOGE ntInued) CASRN Degree of Evidence Humans Animats Weight of Evidence Group Potency Factor Potency Group Hazard Ranking 34 Beniotrichloride 35 Benzyl chloride 36 Beryltiun 37 Beryllium chloride 30 Berylliun fluoride 39 Beryltiun nitrate 40 alpha - BHC 41 beta - BHC 42 gamma - BHC (Lindane) 00098077 00100447 07440417 07787475 07787497 13597994 00319846 00319857 00058899 43 Bis(2-chloroethyl) ether 44 Bisfchloromethyl) ether 45 Bis(2-ethylhe*yl) phthalate 46 Cadmiuet 47 Cadmium acetate 48 Cadmium bromide 49 Cadmium chloride 50 Carbon tetrachloride 51 Chlorambucil 52 Chtordane 53 Chlornaphazine 54 Chloroform 55 Chloromethyt methyl ether (technical grade) 56 4-Chloro-o-totuidine, hydrochloride 57 Chromium9 58 Ammonium bichromate 59 Ammonium chromate 60 Calcium chromate 61 Chromic acid 62 lithiun chromate 63 Potassium bichromate 64 Potassiurn chromate 65 Sodium bichromate 66 Sodium chromate 00111444 00542881 00117817 07740439 00543908 07789426 10108642 00056235 00305033 00057749 00494031 00067663 00107302 03165933 07440473 07789095 07788989 13765190 11115745 14307358 07778509 07789006 10586019 07775113 Limited Inadequate Inadequate Ho Data Ho Data Ho Data Ho Data No Data Inadequate Ho Data Sufficient Ho Data Limited k k k Inadequate Limited 1nadequate Inadequate Inadequate Suf ficient Ho Data Ho Data h h h h h h h h h Sufficient Sufficient Sufficient Ho Data Sufficient Ho Oata Sufficient Limited Sufficient/ Limited Sufficient Sufficient Suf ficient Sufficient Ho Data Ho Data Sufficient Sufficient Sufficient Sufficient timited Sufficient 1nadequate Sufficient Ho Oata Ho Data Ho Data Inadequate Inadequate Ho Data Inadequate Inadequa t e !nadequate 1nadequa t e B1 B2 B2U u u u B2 c B2/CTM B2 A B2 Bl k k k B2 Bl B2 C B2 A B2 D Ah Ah Ah Ah Ah Ah Ah Ah Ah 58.0 0.66 79.70u ab ab ab 51.48 10.67 7.39 2 3 2 1 1 1 2 2 2 13.29 10377 0.194 57.9k k k k 59.9 a 15.13 a 1.97 n 0.40 Hone h h h h h h h h h 2 1 3 2 2 2 2 2 2 2 2 2 1 3 Hone 1 1 1 1 1 1 1 1 1 HED LOU HED HIGH HIGH HIGH HE0 LOU HED HED HIGH LOU HED HED HED HED HED HED HED LOU HED HIGH LOU Hone** HIGH HIGH HIGH HIGH HIGH HIGH HIGH HIGH HIGH C69V3VVWO mr\ A - 44 Substance HAZARD RANKING OF POtENtlAL CARCINOGEN *t fnued) CASRH Degree of Evidence Humans Animals Ueightof -Evidence Group Potency Factor Potency Group Haza Rank 67 Strontium chromate 68 Chrysene 69 Coke Oven Emissions 70 Creosote 71 Cyclophospham i de 72 DaunomycIn 73 ODD 7* DOE 7S DDT 76 Dial late 77 Diami notoluene (mixed) 78 Dibeni(a,hi anthracene 79 1,2:7,8-Dibenzopyrene 80 1,2-Dibromo-3-ch1oropropane 81 3,3'-Dichlorobenzfdine 82 1,2-Dich1oroethane 83 1,1-Oichloroethylerte (Vinylidene chloride) 86 Dteldrin 83 1,2:3,A-Diepoxybutane 86 1,2-Oiethylhydrazine 87 Diethylstilbestrol 88 Oihydrosafrole 89 3,3'-0imethox ybenz Tdine 07789062 00216019 N. A. 06001589 00050180 20830813 00072548 00072559 00050293 02303164 00095807 00053703 00189559 00096128 00091941 00107062 0007S354 00060571 01464535 01615801 00056531 00094586 00119904 h No Data Sufficient Lfmited Limited No Data Inadequate Inadequate Inadequate No Data No Data No Data No Data No Data No Data No Data No Data Inadequate No Data Ho Data Sufficient No Data No Oata Sufficient Limited Sufficient Sufficient Sufficient Sufficient Sufficient Sufficient Sufficient Limited Sufficient Sufficient Sufficient Sufficient Sufficient Sufficient Limited Sufficient Sufficient Sufficient Sufficient Sufficient Sufficient 90 Dimethyl sulfate 91 Dimethyl aminoazobenzene 92 7,12-0imethylbenz(a)anthracene 93 3,3'-Dimethyl benzidine 96 Dimethylcarbaatoyl chloride 95 1,1-Dimethyl hydrazine 96 1,2-Dimethythydrazine 97 Oini trotoluene (mixed) 98 2,4-Dinitrotoluene 99 2,6-Dinitrotoluene 1U0 1,4-Dioxane 00077781 00060117 00057976 00119937 00079447 00057147 00540738 25321146 00121142 00606202 00123911 Inadequate No Data No Data Ho Data 1nadequa t e No Oata No Data No Data No Data No Data Inadequate Suf ficient Sufficient Sufficient Sufficient Sufficient Sufficient Sufficient Sufficient Sufficient Limited Sufficient A-45 Ah C A 81* B1 B2 B2 B2 B2 C B2 B2 B2 82 B2 B2 C B2 B2 B2 A B2 B2 B2 B2 B2 B2 B2 B2 B2 B2*3 B2 C 82 h a 1.53* ab 17.5 a 1.30 3.62 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.B2*5 3.82 a 0.034 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 HIGH LOU HIGH HIGH NED NED NED NED KED LOU MED HIGH NED HIGH NED LOU LOU HIGH HE0 NED HIGH NED NED NED NED HIGH NED HIGH NED HIGH NED NED LOU LOU CMA 121694 Substance HAZARD RANKING OF PDTEMtlAL CARCINOGE ntinued) CASRN Degree of Evidence Humans Animals Weightof-Evidence Group Potency Factor Potency Group Hazard Ranking 101 1,2-Diphenyl hydratine 102 Epfchlorohydrin 103 Ethyl carbamate (urethane) 104 Ethyl 4,4`-dtchtorobenzilate 105 Ethylene dlbroatide 106 Ethylene oxide 107 Ethylenethiourea 10S Ethyl methanesulfonate 109 Formaldehyde 110 Glycldylaldehyde 111 Heptachlor 112 Heptachlor epoxide 113 Hexachtorobentene 114 Hexachlorobutadiene 115 Hexachloroethane 116 Hydrazine 117 1ndeno(1,2(3-cdIpyrene 110 Isosafrote \ 119 Kepone 120 Lasiocarpine 121 Lead 122 tead acetate 123 Lead phosphate 124 Lead stearate 125 Lead subacetate 126 tead sulfide 127 Melphalan' 120 Methyl chloride 129 3-Nethytcholenthrone 130 4,4--Methylenebfs(2-chloroaniline) 131 Methyl iodide O 132 N-Methyl-H*-nitro-N-nitrosoquanidine 5 133 Methylthiouracf1 00122667 00106690 00051796 00510156 00106934 00075218 00096457 00062500 00050000 00765344 00076446 01024573 00118741 00087663 00067721 00302012 00193395 00120581 00143500 00303344 07439921 00301042 07446277 07428460 01335326 01314870 00146623 00074673 00056495 00101144 00074884 00070257 00056042 No Data Inadequate Mo Data Inadequate Inadequate Limi ted/ Inadequate No Data No Data Limited No Data Inadequate Mo Data Mo Data No Data No Data Inadequate Mo Data Mo Data No Data Ho Data Sufficient Sufficient Sufficient Sufficient Sufficient Sufficient Sufficient Sufficient Sufficient Sufficient Sufficient Sufficient Sufficient Limited Limited Sufficient Limited Sufficient Sufficient Sufficient Limited No Data No Date Inadequate No Data No Data Inadequate Sufficient Limited Sufficient Sufficient Limited Suf ficient Sulficient B1 B2 82 82 62 B1/B2* S2 B2 61 B2 B2 B2 62 C C B2 C 62 62 62 B1lZ C 62 B2 C B2 62 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 2 3 3 2 1 2 2 1 2 2 1 1 2 3 3 t 2 3 2 2 810 0.050 25.5 1.52 a 54.7 a 1 3 2 2 2 2 2 MED LOU LOU MEO HIGH MED HE0 HIGH MED MED HIGH HIGH MEO * * HIGH LOU LOU MED MED 00 00 00 00 00 00 HIGH LOU MED MED LOW MEO MED 9691-21' VIAIO JI7 A - 46 HA2AR0 RANKING Of POTENTIAL CARCINOGENS tinued) Substance CASRN Oegree of Evidence Humans Animals Ueightof-Evidence Group Potency Factor Potency Group Hazard Ranking 134 Mitomycin C 135 1-Haphthylamine 136 2-Naphthyl amine 137 Nickel 138 Nickel ammonium aulfate 139 Nickel carbonyl 140 Nickel chloride 141 Nickel cyanide 142 Nickel hydroxide 143 Nickel nitrate 144 Nickel sulfate 145 2-Hitropropane 146 H-Nftrotodi-n-butyl amine 147 N-Nitrosod f ethanol amine 148 N-Ni trotodiethylamine 149 N-Nitrosodimethylamine 150 N-Nitrotodi-n-propylamine 151 N-Nitroao-N-ethylurea '-152 N-Nitroto-N-methylurea 153 N-H i troto-N-methylurethane 154 N-Hitrosomethylvinyl amine 155 N-NifrdaopTperTdrne 156 N-Nitroaopyrrolidine 157 5;Nitro-o-toluidine 158 Pentachloroethane 159 Pentachloroni trobemene 160 Pent achIoropheno1 161 Phenacetin 162 Polychlorinated biphenyls (PCBs) 163 A roc,lor 1016 164 Aroclor 1221 165 Aroclor 1232 166 Aroclor 1242 00050077 00134327 00091598 07440020 15699180 13463393 07718549 00557197 12054487 14216752 07786814 00079469 00924163 01116547 00055185 00062759 00621647 00759739 00684935 00615532 04549400 00100754 00930552 00099558 00076017 00082688 00087865 00062442 01336363 12674112 11104282 11141165 53469219 No Data Inadequate Sufficient Inadequate No Data Inadequate Inadequate No Data No Data No Data Inadequate Inadequate No Data Ho Data No Data No Data No Data Ho Data Ho Data No Oata No Oata No Data 1nadequa t e Ho Data No Oata No Data Inadequate Inadequate 1nadequate Inadequate 1nadequa t e Inadequate Inadequate Sufficient Limited Sufficient Limited Ho Data Sufficient Limited No Data Limited No Data Limited Sufficient Sufficient Sufficient Sufficient Sufficient Sufficient Sufficient Sufficient Sufficient Sufficient Sufficient Sufficient Limited Limited Limited Ho Data Sufficient Sufficient Ho Data Ho Oata Ho Data No Data S2 C A C z7 B2 z7 z7 z7 z7 t7 B2 B2 B2 B2 B2 B2 BZ B2 B2 821 B2 B2 C C C D B2 82 t r t t A-42 a2 None 2 4.77 z7 2 z7 z7 z7 z7 z7 z7 z7 z7 z7 z7 z7 z7 z7 z7 z7 a2 43.70 2 ab 1 969 1 61.2 2 a2 137 1 2100 1 ab 1 a2 37.5 2 279 1 0.17 3 1.26 1.42Z10 2 2 Hone None 0.028 50.47* 3 2 t2 t2 t2 t2 NED LOU HIGH LOU LOU NED LOU LOU LOU LOU LOU HE0 NED HIGH HIGH HE0 NED HIGH HIGH HIGH NED NED HIGH LOU LOU LOU None** LOU NED HE0 NED NED NED CMA 121696 Substance HAZARD RANKING OF POIENTIAL CARCINOGEN^BFont tnued) CASRN Degree of E videne e Humans Animals Ueightof-Evidence Group Potency Potency Factor Group Hazard Ranking 167 Aroclor 1248 168 Aroclor 1254 169 Aroclor 1260 170 1,3-Propane sultone 171 1,2-Propyteniialne 172 Saccharin 173 Safrole 174 Selenium sulfide 175 Streptozotocfn 176 2,3,7,8-Tetrachlorodfbenzo-p-dioxin (TCDD) 177 1,1,1,2-Tetrachloroethane 178 1,1,2,2-Tetraehforoethane 179 T etrach1oroethylene 180 Thioacetamide 181 Thiourea 182 o-Tolutdine 183 p-Toluldine 184 o-Toluidine hydrochloride 185 Toxaphene 186 1,1,2-Trichloroethane 187 Trichloroethylene 188 Trichlorophenol (nixed) 189 2,4,5-Trichlorophenol 190 2,4,6-Trichlorophenol 191 Tri*(2,3-dibronopropyl) phosphate 192 Trypan blue 193 Uracil aaistard 194 rVinyl chloride 12672296 11097691 11096825 01120714 00075 558 00081072 00094597 07488564 18883664 01746016 00630206 00079345 00127184 00062555 00062566 00095534 00106490 00636215 08001352 00079005 00079016 25167822 00095954 00088062 00126727 00072571 00066751 00075014 Inadequate Inadequate Inadequate No Data No Data inadequate No Data Inadequate No Data 1nadequate 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 Sufficient No Data Sufficient Sufficient Sufficient Sufficient limited Sufficient Sufficient Sufficient Sufficient limited Ifmi ted Sufficient Sufficient Sufficient Sufficient Limited Sufficient Sufficient timited Sufficient Sufficient Inadequate Sufficient Sufficient Sufficient Sufficient Sufficient t B2 B2 B2 B2 C B2 82 82 B2 C C B2/C" B2 82 B2 C B2 B2 C B2 B2*4 D B2 B2 B2 82 A t2 t2 50.5 2 10.0 2 259 1 0.007 3 0.18 3 0.93 3 109 1 659,000 1 0.85 3 1.66 2 0.29 3 24.8 2 t .05 2 0.069 3 0.94 3 0.069 3 9.79 2 0.36 3 0.10 0.08*4 3 3 None None 0.08 3 9.76 2 a2 a2 18.1 2 ME0 NED . NED NED HIGH IOU LOU LOU HIGH HIGH IOU LOW LOU NED NED LOU LOU LOU NED LOU LOU LOU Noneb LOW NED NED NED HIGH CMA 121697 A-48 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^o 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 EDio dose. Hence, the substance is assigned the lowest potency group. Potency Group 3. b No RQ can be assigned based on potantial 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 being converted into a carcinogenic form when released into the environment. 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. I. A-49 CMA 121698 CHEMICAL-SPECIFIC FOOTNOTES d Arsenic--The weight of evidence for the carcinogenicity of inorganic arsenic compounds is babed 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 been conducted and show sufficient evidence of causal association. g 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). h 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 tor the inhalation of hexavalent chromium by chromate workers (potency factor * 388.99, weight-of-evidence group A). i Benzo(b)fluoranthene--Calculated, using the potency factor estimate for benzo(a)pyrene as a reference. k Cadmium weight of evidence and potency are based on epidemiology data 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 fjor cadmium acetate, bromide and chloride are considereid to be the same as those cadmium compounds to which workers are exposed. n Chloromethyl methyl ether [CMME]--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 fjor bis(chloromethyl)ether. A-50 CMA 121699 p 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 concentrations are usually measured either as a number of fibers or mass. However, no direct relationship exists between air, fiber/ml (75 microns) concentrations (by the phase contrast light microscope method) and mass concentrations in mg/m3 (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. t 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 20-60 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 forbs 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, cigarette smoke. Therefore, benzo(a)pyrene should be treated as a human carcinogen. i. A-51 CMA 121700 X Creosote--There are no adequate studies of workers exposed to creosote wood preservatives. It has been desonstrated that chimney sweeps exposed to thb 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. Z2 Helphalan--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 2.4.6- 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. Althpugh 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 salts of nickel show some carcinogenic activity. The testing of these nickel salts is inc nclusive for A-52 CMA 121701 assessment of cancer at this time due to limitations of the data base on these nickel salts, but since there is n 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. Thetfe are no data (veight-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 hexaphlorobenzene. NOTE: The gaps in the letters assigned to the footnotes in this table exist so that consistency viih the numbers assigned to the footnotes in the summary table in OHEA-C-073 (Appendix A in NTechnical Background Document to Support Rulemaking Pursuant to CERCLA Section 102, (Proposed Rulemaking)," Volume 3, March 1985), is maintained. A-53 \ CMA 121702