National Toxicology Program

TABLE OF CONTENTS

1. Introduction and Background
2. Peer Review of Agents,Substances, Mixtures and Exposure Circumstances Nominated for Listing in or Delisting from the 9^th Report on Carcinogens:
1. Tetrafluoroethylene
2. Cadmium and Cadmium Compounds
3. Chloroprene
4. 1,3-Butadiene
5. UV Radiation
6. Tobacco Smoking
7. Smokeless Tobacco
8. Strong Inorganic Acid Mists Containing Sulfuric Acid
9. Tamoxifen
10. Phenolphthalein
11. Saccharin
12. 2,3,7,8-Tetrachlorodibenzo-p-dioxin(TCDD)
13. Trichloroethylene
14. Dyes Metabolized to Benzidine(Benzidine-Based Dyes as a Class)

The National Toxicology Program (NTP) Board of Scientific Counselors' Report on Carcinogens Subcommittee (the Subcommittee) held its third meeting on October 30 and 31, 1997, at the National Institute of Environmental Health Sciences (NIEHS), Research Triangle Park, North Carolina. (Attachment1: Federal Register meeting announcement; Attachment 2: Agenda and Roster of Members and Expert Consultants.) Members of the Subcommittee are Drs. Arnold Brown (Chairperson), John Bailer, Steven Belinsky, Eula Bingham, Clay Frederick, George Friedman-Jimenez, Carol Henry, Kim Hooper, and Franklin Mirer. Expert Consultant to the Subcommittee is Dr. Hiroshi Yamasaki. Dr. Bingham was present only on October 31. Dr. Henry was unable to attend; however, she was able to provide written reviews which were read into the record. Additionally, for this meeting the Subcommittee was supplemented by the participation of three ad hoc Expert Consultants: Drs. Stephen Hecht, University of Minnesota Cancer Centers; Karl Kelsey, Harvard School of Public Health and Medical School; and Shelia Ahem, National Cancer Institute.

1. Introduction and Background:

   Dr. George Lucifer, Director, Environmental Toxicology Program (ETP), noted that Congress had directed the Department of Health and Human Services to publish a Report on Carcinogens which would contain a list of substances (1) which either are known to be human carcinogens or may reasonably be anticipated to be human carcinogens, and (2) to which a significant number of persons in the United States are exposed. The National Toxicology Program was assigned responsibility for preparing the report. Seven complete reports have been published, and the eighth Report on Carcinogens (formerly the Annual Report on Carcinogens)will be submitted shortly to the Secretary. Dr. Lucifer said that the 14 agents, substances and mixtures to be reviewed by the Subcommittee at this meeting were intended for the ninth Report projected for publication in 1999. He commented on the revised criteria,used for the first time by the Subcommittee a year ago, for determining which agents should be listed in or considered for delisting from the Report. These criteria allow for use of mechanistic information along with epidemiological and animal cancer data, permitting use of all relevant information and allowing for scientific judgment be made. Dr. Lucifer reviewed the sources of information supporting the nominations for listing or delisting selected for review. He noted that complex mixtures and exposure circumstances are now considered for inclusion in the Re port.. He concluded by stating that the open review by the Subcommittee allowed the opportunity for public comment.

   Dr. Bill Jameson, NIEHS,commented on the preparation of draft background documents for each nomination which were provided to the reviewers as well as to the public. Dr. Jameson noted that written public comments had been received and also provided to the reviewers. Dr. J.Carl Barrett, Scientific Director, NIEHS, thanked the reviewers and the staff and commented on the national and international importance of the Reports in the protection of public health. Dr. Brown went over the review format to be used with each nomination.Each nomination will be presented by an NIEHS scientist who will discuss the nomination, data relating to human cancer, animal cancer, mechanistic information, summaries of the arguments for or against listing or delisting, and will provide the recommendations,including the votes of the two previous Federal scientific review groups, the NIEHS/NTP Review Group (RG1), and the NTP Executive Committee's Interagency Working Group (RG2). Then the primary reviewer from the Subcommittee will present his/her evaluations of the nominations, followed by the secondary reviewer who should emphasize differences or areas of agreement with the primary reviewer. There will be time for public comments followed by further discussion among the Subcommittee and expert reviewers concluding with motions and votes by Subcommittee members on recommendations to be forwarded to the NTP. Dr. Brown said there had been 16 requests to make formal public statements addressing seven of the 14 nominations,and as well, written comments had been received from a number of individuals and organizations and made available to the reviewers and the public.

2. Peer Review of Agents,Substances, Mixtures and Exposure Circumstances Nominated for Listing in or Delisting from the 9^th Report on Carcinogens:

1. Tetrafluoroethylene- Dr. Joseph Haseman, NIEHS, said that tetrafluoroethylene (TFE) was nominated for listing as reasonably anticipated to be a human carcinogen by RG1 because of an NTP 2-year inhalation rodent bioassay study in which TFE increased the incidence of malignant neoplasms at multiple sites in rats and mice. He said potential human exposure is primarily occupational from the production of polyfluoroethylenes and resultant leakage from closed-capture systems. The primary target sites in the NTP study were kidney,liver, and hematopoietic (mononuclear cell leukemia) in rats and liver and histiocytic sarcoma in mice. There were no human cancer data found for this chemical and available genetic toxicity results were negative. Dr. Haseman reported that both RG1 and RG2 recommended unanimously with 10 and eight votes, respectively, that TFE be listed as reasonably anticipated to be a human carcinogen.

   Dr. Belinsky, the primary reviewer, agreed with the proposed listing. He commented that the summary statement should be amended to accurately describe theras mutation studies since TFE was not negative for the induction of H-ras mutations in mouse liver; rathera 15% incidence was found which is significantly lower than seen in controls suggesting that tumors were induced via a ras independent pathway. He thought there should be a description of the cytotoxic effects of TFE in human renal proximal tubule cells.

   Dr. Hooper, the secondary reviewer, agreed with the proposed listing. He said it would be useful to compare the dose levels in rats and mice with those in the structurally-related tetrachloroethylene which in an NTP study was associated with a similar spectrum of tumors, and then compare dose levels of TFE that produced effects with what would be presumed occupational exposures. Because TFE is an aggressive animal carcinogen, occupationally exposed people should be made aware.

   Dr. Belinsky moved that the nomination of tetrafluoroethylene for listing in the Report asreasonably anticipated to be a human carcinogen be accepted. Dr. Hooper seconded the motion, which was accepted unanimously with five votes.

2. Cadmium and Cadmium Compounds-Dr. Michael Waalkes, NCI at NIEHS, said that cadmium and cadmium compounds(cadmium) were nominated for listing as known to be a human carcinogen(currently listed in the Report as reasonably anticipated to be a human carcinogen) based on findings of increased risk of cancers in exposed workers or populations and evidence of malignant tumor formation by multiple routes of exposure at various sites in multiple species of experimental animals. He said that cadmium was listed as a Category 1 human carcinogen by the International Agency for Research on Cancer (IARC) in 1993. There is occupational and environmental exposure with most recent estimates that more than half a million workers are exposed to cadmium in the U.S. Cohort studies provide consistent evidence of elevated lung cancer risk in cadmium exposed workers. Cadmium causes a variety of genetic damage including mutations, chromosomal damage, cell transformation, DNA strand breaks, disrupted DNA repair, effects on gene expression, and, in humans, chromosomal aberrations. Dr. Waalkes reported that RG1 voted 7 to 1, andRG2 voted unanimously with 8 votes in favor of the recommendation to list cadmium and cadmium compounds in the 9th Report on Carcinogens as known to be a human carcinogen.

   Dr. Kelsey, the primary reviewer, agreed with the proposed listing but thought the epidemiology studies showed a complex picture in that there is no way to completely determine that human exposure to cadmium can be separated from exposure to arsenic or other compounds. He noted also the conflicting data reported by NIOSH on a cohort from two plants, one where there clearly was a significant risk for lung cancer, the other where exposure to cadmium appeared to have a protective effect. The overall data in support of listing are quite strong including the data on prostate and bladder cancer in humans, the clear carcinogenic effects in animals, the genotoxic data, and mechanistic data.

   Dr. Frederick, the secondary reviewer, also commented on the complexity in the epistemology findings and said that he was influenced in his review by the1997 British paper that intensively reanalyzed the original NIOSH cohort study. They stated that three possible conclusions could be drawn, two of which were that cadmium oxide and arsenic trioxide together were human lung carcinogens, and the third was that arsenic trioxide was a human lung carcinogen and cadmium was not. Dr. Frederick asked if Dr. Waalkes could provide additional insight. Dr. Waalkes responded that the British study did not (1) adjust for mobility of workers moving into less contaminated areas of the plant, (2) look at a post-1940 employment group when the cadmium feed stock was free of arsenic contamination, whereas one of the authors of the NIOSH study had followed this up and found a significant positive correlation with cadmium exposure levels and lung carcinogenesis. Finally, The British study never mentioned the actual levels of arsenic, while the IARC determined how many of the tumors would be associated with the arsenic exposure, and concluded that of the 24 or so tumors only one or two would be accountable by the levels of arsenic exposure. Dr. Mirer commented on how difficult it is to measure quantitative exposure levels in the occupational setting leading to the likelihood of misclassification errors. There was further discussion around weighting chemical exposure appropriately relative to job function.

   Dr. Kelsey moved that the nomination of cadmium and cadmium compounds for listing in theReport as known to be a human carcinogen be accepted. Dr. Frederick seconded the motion, which was accepted unanimously with six votes.

3. Chloroprene-Dr. Jameson said that chloroprene was nominated for listing as reasonably anticipated to be a human carcinogen by RG1 based primarily on a NTP 2-year inhalation study which demonstrated that chloroprene was a potent, multiple organ, trans-species carcinogen. Further,chloroprene is structurally related to the known human carcinogen,vinyl chloride, and is the 2-chloro analogue of 1,3-butadiene,currently listed in the Report as reasonably anticipated to be a human carcinogen and nominated to be changed to known to be a human carcinogen. Dr. Jameson reported the environmental release and estimated occupational exposure, reviewed the target sites and levels of evidence for carcinogenicity from the NTP study, and described the limited evidence for carcinogenicity in humans. Chloroprene was negative in a number of genotoxicity assays However, lung and Harderian gland tumors in chloroprene exposed mice in the NTP study exhibited a high frequency of uniqueK-ras mutations. Dr. Jameson said that RG1 voted 7-0(with two abstentions) and the RG2 voted unanimously with 8 votes to recommend that chloroprene be listed as reasonably anticipated to be a human carcinogen.

   Dr. Henry, the primary reviewer, was unable to attend the meeting but had submitted her review, which Dr. Larry Hart, NIEHS, read into the record. Dr.Henry agreed with the proposed listing. She cited the convincinganimal data while noting that the concentration of chloroprenecyclic decomposition products comprised less than 0.1% of thechloroprene concentration in the exposure chambers. She thoughtdiscussion would have been useful on stability of chloroprenein test samples especially those used in genotoxicity assays. Dr. Henry said that more information about the unique K-rasmutations would have helped.

   Dr. Bailer, the secondaryreviewer, agreed with the proposed listing. He commented thatmore attention should have been given in the summary statementof the background document to the structural analogy to knowncarcinogens. He said the human studies were quite weak so characterizingthem under "limited evidence" was a fairly generousdescription. Dr. Frederick wondered as to why the NTP TechnicalReport on Chloroprene was still in draft form. Dr. John Bucher,NIEHS, responded that this study was one of those where therewas evidence for Helicobacter hepaticus infection in mice,an infection shown to increase the incidence of liver tumors inmale mice. After extensive analysis, the NTP concluded that theinfection did not impact on the findings in the chloroprene reportand the final report would be published in the near future.

   Public Comment: Mr. Michael Lynch, DuPont Dow Elastomers L.L.C., said that hiscompany was the only intentional producer of chloroprene in theUnited States. He pointed out the existence of a chronic inhalationbioassay of chloroprene sponsored by an industry group in the1970s and conducted on Syrian golden hamsters for 18 months andWistar rats for two years. The unpublished study showed thatchloroprene was not carcinogenic in either sex of either speciesat concentrations up to 50 ppm. In their opinion, the differencesin the findings could be attributed to differences in the vaporgeneration techniques. Dr. Lynch reported that his company andothers that produce chloroprene have initiated pharmacokineticand other studies to better assess human health hazards

   Dr. Bailer moved thatthe nomination of chloroprene for listing in the Reportas reasonably anticipated to be a human carcinogen be accepted. Dr. Frederick seconded the motion, which was accepted unanimouslywith six votes.

4. 1,3-Butadiene-Dr. Buchersaid that 1,3-butadiene was nominated by RG1 for listing as knownto be a human carcinogen (currently listed in the Report asreasonably anticipated to be a human carcinogen) basedon studies in humans which have consistently found excess mortalityfrom lymphatic and hematopoietic cancers associated with occupationalexposure to butadiene, studies in experimental animals which haveshown that 1,3-butadiene (butadiene) induces benign and malignantneoplasms at multiple tissue sites in multiple species, and supportingmechanistic data. Specifically, since IARC in 1992 categorizedbutadiene as probably carcinogenic to humans, new epidemiologydata have strengthened the evidence linking exposure with increasedhuman cancer risk, and recent research indicates that the metabolicbehavior of butadiene is qualitatively similar in humans and laboratoryanimals. Dr. Bucher reviewed the information indicating a potentiallylarge exposure of worker, primarily in butadiene monomer manufacturingand in the butadiene styrene synthetic rubber production industry,and emphasized that more recent epidemiology studies addressedmany of the limitations of earlier studies, including use of modelingefforts that quantitatively estimated exposure to butadiene andstyrene and performance of an interaction study that suggesteda negative interaction between styrene and butadiene exposure. Dr. Bucher reported that RG1 voted 9 yes with one abstentionand RG2 was unanimous with eight votes in support of listing butadieneas known to be a human carcinogen.

   Dr. Mirer, the primaryreviewer, agreed with the proposed listing. He stated that theepidemiology presents a strong picture if the hematopoietic tumorsare considered together and believed there was biological plausablityfor doing so. He noted especially the multiple studies findingof Delzell et al., which found an SMR excess and exposureresponse, and those of Devine and Hartman.

   Dr. Zahm, the secondaryreviewer, agreed with the proposed listing. She emphasized howthe epidemiologic data that have accumulated since the 1992 IARCreview support the proposed change in classification for butadiene. In particular, she stressed the strengths of the Delzell etal. study which had almost 16, 000 subjects, an excellentexposure assessment, demonstration of dose-response, and evaluationof any confounding by styrene. Dr. Zahm said there were two studiesomitted which should be cited, one of which was negative. Thisstudy does not detract significantly from the compelling evidencefor human carcinogenicity provided by the other studies.

   Public Comments. Dr. John Acquavella, Monsanto, representing the InternationalInstitute of Synthetic Rubber Producers, Inc., said that he wasthe project officer for most of the styrene-butadiene rubber (SBR)worker studies that were done. He stated that the NTP has characterizedthe butadiene epidemiologic literature as showing a consistentexcess of lymphatic and hematopoietic cancers (LHCs) associatedwith butadiene exposure. He said that, in fact, the epidemiologicliterature shows variable results for LHCs. The evidence forleukemia is not consistent across studies, though one large studyprovides credible, internally consistent evidence of a relationshipwith butadiene exposure. Dr. Acquavella commented that whiletwo studies show elevated mortality among short-term butadienemonomer workers, the larger study did not find excess mortalityfor long-term exposed workers and there was no exposure responserelationship. In addition, none of the SBR worker studies provideevidence to suggest a relationship bedtween butadiene and non-Hodgkin'slymphoma. He concluded that the butadiene epidemiologic literatureshould not be characterized as showing a consistent relationshipbetween butadiene exposure and the various LHCs.

   Dr. A. Philip Leber, Goodyear Tire and Rubber Company, said that Goodyear's economic interestin butadiene is related to its manufacture of butadiene copolymersused in numerous industrial, consumer, medical device, and foodadditive products. He contended that the NTP criteria call forsufficient evidence in humans indicating a causal relationshipbetween the agent and human cancer, and briefly cited such evidencefor industrial organic chemicals currently listed. He said thatfor butadiene there is significant question about its causality,as the three worker studies cited in support provide contradictorydata. Further, since many chemicals besides butadiene are usedwithin SBR operations, there is the possibility that other chemicalsare confounders and likely involved in leukemia etiology in theworkers.

   Dr. James Swenberg, University of North Carolina, representing the Olefins Panel of the ChemicalManufacturers Association, said he would share a few new findingsfrom his laboratory on the molecular dosimetry and molecular epidemiologyof butadiene. He said the metabolism section in the backgrounddocument should be expanded to reflect epoxy butenediol as themajor metabolite in mice, rats and humans. In looking at formationof monepoxide and diepoxide adducts in the liver, at low concentrationsrats and mice have identical molecular dosimetry but at 625 ppm,the mouse has over twofold greater numbers of adducts.

   Dr. Elizabeth Ward, NIOSH, an author on one of the epidemiologic studies cited was askedby Dr. Brown for any comments. Dr. Ward stated that the Delzellet al. study used the best possible epidemiologic methodologyand strongly established the association between butadiene exposureand leukemia.

   There ensued a discussion about metabolism of butadiene pertaining to differences betweenmice and rats in formation of reactive metabolites and with regardto which species humans more closely resembled in metabolism ofbutadiene and formation of reactive metabolites and detoxificationproducts.

   Dr. Mirer moved that thenomination of 1,3-butadiene for listing in the Report as knownto be a human carcinogen be accepted. Dr. Zahm seconded themotion, which was accepted by four yes votes to one no vote (Belinsky)with one abstention (Frederick). Dr. Frederick abstained forreasons of company affiliation.

5. UV Radiation-Dr. FrejaKamel, NIEHS, said that UV radiation (UVR) was nominated by RG1for listing as known to be a human carcinogen based onthe evidence that human studies have shown that exposure to solarradiation is causally related to skin cancer, and that use ofsunlamps or sunbeds is associated with skin and eye cancer. Shedefined UV radiation as the portion of the optical spectrum betweeen10 and 400 nanometers (nm) with UVA being 315-400 nm, UVB being280-315 nm, and UVC being from 100-280 nm. The nomination wasbased on the rather complex 1992 IARC classification, which classifiedsolar radiation as carcinogenic to humans, UVA, UVB, and UVC radiationas well as sunlamps and sunbeds as probably carcinogenic to humans,and fluorescent lighting as not classifiable. Dr. Kamel discussedthe various meteorological and personal behavior factors thatcause wide variability in dosimetry. She reviewed the varioussources of artificial radiation and the types of neoplasms seenwith solar and UV radiation, including carcinogenic effects inexperimental animals. UV radiation causes genetic damage in humanand animal cells with UVC somewhat more potent than UVB, and bothconsiderably more potent than UVA. RG1 voted unanimously with11 votes to support the proposed listing, while RG2 voted by sevenyes to one no votes to defer action until the background documentwas revised to address the full spectrum of UV radiation.

   Dr. Henry, the primary reviewer, was unable to attend the meeting but had submitted herreview, which Dr. Hart read into the record. Dr. Henry did notagree with the proposed listing. She noted that the body of evidencesupporting the proposed listing was from four positive human studiesdating from the mid-1980s or later. Dr. Henry pointed out thatthe most significant weakness in these studies was the lack ofdata regarding what the subjects were actually exposed to andhow much. In addition, no causal mechanism has been formulated. She concluded that with limited evidence of carcinogenicity inhumans, yet sufficient evidence in experimental animals, the mostappropriate listing would seem to be reasonably anticipatedto be a human carcinogen.

   Dr. Kelsey, the secondary reviewer, agreed with the proposed listing. He said the datasupporting UV radiation as a human carcinogen are clear, consistentand quite striking.

   Dr. Kelsey moved that the nomination of UV radiation for listing in the Report as knownto be a human carcinogen be accepted. Dr. Friedman-Jimenezseconded the motion, which was accepted unanimously with six votes.

6. Tobacco Smoking-Dr. Buchersaid that tobacco smoking was nominated by RG1 for listing inthe Report as known to be a human carcinogen based onstudies in humans which indicate a causal relationship betweenexposure to tobacco smoke and human cancer. The rationale fornomination included the fact that there have been numerous previousevaluations of human data on cigarette smoking or tobacco smokingand cancer including the cancer societies of various northernEuropean countries, the American Cancer Society, and the CanadianDepartment of Health and Welfare. In 1964 there was a reportissued by the Advisory Committee to the U.S. Surgeon General linkingsmoking to cancer of the lung, lip, and larynx, and in 1979, theReport of the Surgeon General added cancer of the esophagus tothis list. In 1986, the IARC reviewed tobacco smoking and thatthere was sufficient evidence that tobacco smoke was carcinogenicto humans, and new epidemiology data that have come out continueto confirm and expand the evidence linking tobacco smoking withknown and new tumor types and sites. The IARC review determinedthat there was also sufficient evidence in animals after inhalationexposure or topical application of tobacco smoke condensates. The review groups, RG1 and RG2, unanimously recommended listingof tobacco smoking as known to be a human carcinogen..

   Dr. Frederick, the primary reviewer, agreed with the proposed listing and wondered why tobaccosmoking had not been listed long ago. He noted that the nominationdoesn't address environmental tobacco smoke and assumed that wouldbe brought forward at a future meeting.

   Dr. Zahm, the secondary reviewer, agreed with the proposed listing. She said the reviewdocument cites numerous studies, utilizing every study designbut relying most heavily on cohort studies, which demonstratethat tobacco smoke is a human carcinogen.

   Dr. Mirer stated thatthe sentence in the summary statement that "between 80 to90 % of all human lung cancers and approximately 30 % of humancancers of all types are attributed to tobacco smoking" shouldbe deleted. He said that more and more occupational and otherfactors are known to affect lung and other cancer risks, includinga number of chemicals and chemical mixtures. Thus the methodsof estimation essentially ignore interactions and make no effortto apportion the risks of co-exposures which may be multiplicative. Dr.Friedman-Jimenez disagreed with deleting the statement that80-90 % of lung cancers are attributable to smoking. In asbestosworkers who smoke, you can also have 80 % of cancers attributableto the asbestos exposure; they don't have to add up to 100 %. Dr. Belinsky agreed saying we don't want to diminish the impactthat smoking has on cancer rates.

   Dr. Frederick moved thatthe nomination of tobacco smoking for listing in the Report asknown to be a human carcinogen be accepted. Dr. Zahmseconded the motion, which was accepted unanimously with six votes.

7. Smokeless Tobacco-Dr.Bucher said that smokeless tobacco was nominated by RG1 for listingin the Report as known to be a human carcinogen basedon studies in humans which indicate a causal relationship betweenexposure to smokeless tobacco and human cancer. In 1985, theIARC evaluated the human data and determined that there was sufficientevidence that the oral use of snuffs of the type commonly usedin North America and Europe is carcinogenic to humans. Therewas limited evidence that chewing tobacco of the types commonlyused in these areas was carcinogenic. But there were a numberof epidemiological studies that did not distinguish between chewingtobacco and snuff, and used as a whole provided sufficient evidenceof carcinogenicity for oral use of smokeless tobacco products. Since that time, there have been new epidemiology data that haveconfirmed the evidence linking exposure with increased human cancerrisk. Dr. Bucher said that the IARC found inadequate evidenceto evaluate the carcinogenicity of smokeless tobacco productsin experimental animals. More recent studies have provided someevidence of carcinogenicity in rats. Finally, there have beenstudies reporting positive relative risks for tumors at othersites in humans with the oral use of smokeless tobacco products,including rectum, kidney, and most strongly, the prostate. TheRG1, with nine yes votes and one abstention, and the RG2, unanimouslywith eight votes, recommended listing of smokeless tobacco asknown to be a human carcinogen.

   Dr. Hecht, the primaryreviewer, agreed with the proposed listing. He commented thatthere are large amounts of carcinogens in smokeless tobacco, particularlynitrosamines, and that exposure to these carcinogenic nitrosaminesis at least 10 times greater than through any other non-occupationalexposure. Dr. Hecht thought the summary document superficialin its coverage of the published literature since 1985, and provideda number of references.

   Dr. Bailer, the secondaryreviewer, agreed with the proposed listing. He stated that studieswhich do not adjust or control for smoking should be excludedunless evidence of similar patterns of smoking between cases andcontrols is available or unless the case for the specificity ofcancer site unique to smokeless tobacco can be made. Dr. Hooperasked whether more recent human studies reporting rectal, kidneyor prostate cancer could be associated with nitrosamines. Dr.Hecht responded that although nitrosamines are systemic carcinogens,neoplasia of the rectum or prostate have not been demonstratedin animal studies.

   Dr. Hecht moved that thenomination of smokeless tobacco for listing in the Report as knownto be a human carcinogen be accepted. Dr. Bailer secondedthe motion, which was accepted unanimously with six votes.

8. Strong Inorganic Acid Mists Containing Sulfuric Acid-Dr. Jameson said that strong inorganicacid mists containing sulfuric acid were nominated by the International Union, United Auto Workers (UAW), for listing in the Report as known to be a human carcinogen based on studies of occupational exposures that indicate a causal relationship between exposureto strong inorganic acid mists containing sulfuric acid and humancancer. Sulfuric acid, itself, is the largest volume chemicalproduced in the U.S., and yearly, over 770, 000 workers are exposed. Dr. Jameson reported five studies reviewed by the IARC that providedsufficient evidence of increased risk of laryngeal or lung cancerin workers exposed to strong inorganic acid mists containing sulfuricacid. A 10-year followup to one of the larger studies reportedlaryngeal cancer rates consistent with previous findings fromthis cohort. He said that there are no adequate experimentalanimal carcinogenicity studies reported in the literature. Themechanism for carcinogenicity may be related to genotoxicity ofthe low pH environment. Dr. Jameson said that the RG1 unanimouslywith eight votes and the RG2 by seven yes to one no votes supportedthe recommendation to list strong inorganic acid mists containingsulfuric acid as known to be a human carcinogen..

   Dr. Yamasaki, the primaryreviewer, agreed with the proposed listing. He thought the proposedmechanism of an enhanced depurination rate of DNA resulting fromthe low pH environment to be speculative in that studies fromexposed humans indicate increased chromosome aberrations or sisterchromatid exchanges, neither of which are linked to depurination.

   Dr. Bailer, the secondaryreviewer, agreed with the proposed listing. He thought it remarkableto have a case where there were human data to consider but notany adequate animal data.

   Public Comment. Dr. James Hathaway, Rhone-Poulenc, Inc., representing the Panelon Inorganic Acid Mists of the CMA, said that they were convincedthat the IARC decision to classify occupational exposure to stronginorganic acid mists as a known human carcinogen was flawed inthat the cohort and case control studies cited as the basis wereflawed. He noted that reanalyses of some of these studies werenegative for cancer of the larynx as were other similar studiesnot cited in the summary document. Dr. Hathaway said that allanimal studies were negative for tumors including studies doneat the NIEHS at or above the maximum tolerated dose. Thus, thebody of evidence is at best characterized as inadequate.

   Dr. Hooper asked for commentfrom staff on the NIEHS studies. Dr. Jameson reported that studieswere conducted under contract in the mid-1970s to address issuesrelating to catalytic converters, and were designed to exposeanimals to a combination of ozone and sulfuric acid. The resultswere negative and not published. Dr. Mirer said that an SMR studyshowing exposure response remains very strong evidence, and anumber of indirect assessments detailed in the IARC documentfurther support the conclusions. Dr. Hooper asked whether theprimary and secondary reviewers were familiar with the reanalysesmentioned by Dr. Hathaway. Drs. Yamasaki and Bailer said theywere and indicated that even though dose-response was lackingor hard to demonstrate the epidemiologic evidence was still quitecompelling.

   Dr. Yamasaki moved thatthe nomination of strong inorganic acid mists containing sulfuricacid for listing in the Report as known to be a human carcinogen be accepted. Dr. Bailer seconded the motion, which was acceptedunanimously with six votes.

9. Tamoxifen-Retha Newbold,NIEHS, said that tamoxifen was nominated for review by RG1 forlisting in the Report as known to be a human carcinogen basedon epidemiological studies that indicate a causal relationshipbetween exposure to tamoxifen and cancers of the uterine endometrium,and based on the IARC classification of the chemical as a Group1, "Known Human Carcinogen"in 1996. Tamoxifen as itscitrate salt is a pharmaceutical agent that is successfully usedas a primary therapy to inhibit metastatic breast cancer and asan adjuvant therapy to prevent the recurrence of breast cancerin both pre- and postmenopausal women. It is known for its nonsteroidalantiestrogenic activity. RG1 and RG2 both recommended that astatement be included with the listing that "there is alsoconclusive evidence that tamoxifen therapy reduces the risk ofcontralateral breast cancer in women with a previous diagnosisof breast cancer, and for these women, the benefits clearly outweighthe risks." Among its other pharmaceutical uses, tamoxifenis being used in chemoprevention trials which enlist disease freewomen; currently about 16,000, who are at high risk for developingbreast cancer are being recruited into these studies. The proposedlisting was derived from evaluation of the occurrence of primaryuterine cancers diagnosed following tamoxifen treatment for breastcancer. Findings from carcinogenicity studies in animals weresupportive of what was seen in humans. A likely mechanism of carcinogenicaction is that while tamoxifen acts as an antiestrogen in thebreast, accounting for its usefulness in preventing contralateralbreast cancer, it acts as an estrogen agonist in the uterus. Other mechanisms may involve DNA adducts, micronuclei formation,increased aneuploidy, and chromosomal aberrations. The RG1 unanimouslywith 10 votes and the RG2 with seven yes votes and one abstentionrecommended listing as known to be a human carcinogen.

   Since Dr. Brown was tobe a primary reviewer for this nomination, he temporarily turnedthe chair over to Dr. Lucier.

   Dr. Brown, the primaryreviewer, agreed with the proposed listing. He acknowledged concernsin a recent paper by MacMahon that sufficient attention had notbeen give to confounding factors such as prior hysterectomy and/orhormone replacement therapy. Dr. Brown stated that he did notbelieve relative rates of up to 7.5 for endometrial cancers couldbe disregarded on the basis of confounding factors.

   Dr. Kelsey, the secondaryreviewer, agreed with the proposed listing, noting that the epidemiologicdata present a strong and coherent picture of an association betweentamoxifen use and endometrial cancer. He commented on recentdata showing that tamoxifen induces DNA adducts at relevant sitesincluding uterine cells. Dr. Kelsey supported inclusion of arisk/benefit statement pertaining to its efficacy in women whohave had breast cancer.

   Public Comments. Dr. Mark Steinberg, Zeneca Pharmaceuticals, noted that Zenecawas the discoverer and developer of tamoxifen, the most widelyprescribed breast cancer treatment in the world. He thought therewere two points on which there could be agreement, with the firstbeing that the data do not support an association of tamoxifenwith primary hepatocellular carcinomas in humans, and, second,the endometrial cancers seen in women who have taken tamoxifenhave a similar stage, grade, and prognosis to those seen in thegeneral population. Dr. Steinberg then spoke of case controlstudies in breast cancer patients by Dr. Leslie Bernstein whichindicated that when data from women who had received unopposedestrogens were eliminated from her analysis, the association betweentamoxifen and endometrial cancer virtually disappeared. He saidnew and emerging data from the chemoprevention trials will supportthis conclusion. Dr. Steinberg stressed that listing tamoxifenas a human carcinogen will interfere with the patient-physicianrelationship.

   Ms. Joscelyn Silsby, CancerResearch Foundation of America, said the Foundation is a nonprofitorganization dedicated to cancer prevention through scientificresearch, education, and early detection. She wanted to sharetheir perspective on the value of tamoxifen and their belief thatlabeling it as a substance known to be carcinogenic to humanswithout further research would be premature at this time. Shesaid the research conducted to date has not established an absolutecausal relationship between tamoxifen and endometrial cancer. Ms. Silsby said patients and their families are already informedof the risk of secondary cancers and faced with risk/benefit decisionsthat would be immensely complicated by the proposed labeling inthe Report.

   In further discussionby the members, Dr. Brown commented that the charge to the reviewerssimply is to look at the data presented and decide whether ornot tamoxifen can cause human cancer, in this case, endometrialcancer. He added that there are a number of cancer chemotherapeuticagents which are known human carcinogens, and as with tamoxifen,it is the duty of the informed physician to clearly present thisrisk/benefit issue to the patient. Dr. Frederick agreed sayingthat the social overlay of how the information is used is beyondthe scope of this committee.

   Dr. Brown moved that thenomination of tamoxifen for listing in the Report as knownto be a human carcinogen be accepted with the addition ofthe statement endorsed by IARC, RG1 and RG2 that there is conclusiveevidence that tamoxifen reduces the risk of contralateral breastcancer, and that for women with breast cancer, the benefits oftamoxifen are clearly greater than the risks. Dr. Kelseyseconded the motion, which was accepted unanimously with six votes.

10. Phenolphthalein-Dr. JuneDunnick, NIEHS, said that phenolphthalein was nominated by RG1for listing in the Report as reasonably anticipated to be ahuman carcinogen based on the results of the NTP 2-year dosedfeed study in which there were treatment related neoplasms inthe kidney and adrenal medulla in male F344 rats, in the adrenalmedulla in female rats, histiocytic sarcomas and malignant lymphomasof thymic origin in male and female B6C3F₁ mice, andovary tumors in female mice. Dr. Dunnick reported that a subsetof control thymus and thymic lymphomas from mice were stainedwith an antibody to p53 tumor suppressor genes and p53 proteinaccumulation was found in thymic lymphomas from treated mice suggestingthat p53 gene alterations were involved. To learn more aboutmechanisms of phenophthalein carcinogenicity, the chemical wasgiven in the feed to p53 deficient mice at doses ranging from200 up to 12,000 ppm for up to six months. (The animal model wasdeveloped by Donehower et al.) There were large numbersof thymic lymphomas seen in phenolphthalein treated groups, especiallyin the two highest dose groups which were equivalent to the lowand high dose in the 2-year bioassay. Molecular biologic analysisshowed loss of the p53 wild allele, while thymuses from animalroom controls showed the normal p53 pattern of one null and onewild type allele. Increases in micronuclei formation in p53 transgenicmice were seen at dose levels within 10-20 times recommended humanexposure levels. Dr. Dunnick said there are no published epidemiologystudies that adequately examine the human cancer risk from exposureto phenolphthalein. The review groups, RG1 by 10 yes to 1 novotes, and RG2 by seven yes votes with one abstention recommendedthat phenolphthalein be listed in the Report as reasonablyanticipated to be a human carcinogen.

    Dr. Belinsky, the primaryreviewer, agreed with the proposed listing. He commented thatthe doses that induced tumors both in the 2-year bioassay andin the 26-week transgenic studies were significantly higher thanany anticipated human exposures and this should be noted in thesummary statement.

    Dr. Henry, the secondaryreviewer, was unable to attend the meeting but had submitted herreview, which Dr. Hart read into the record. Dr. Henry agreedwith the proposed listing. She commented that in September 1997,the FDA proposed reclassification of the use of phenolphthaleinin over-the-counter laxative products from "generally recognizedas safe and effective..." to "not generally recognizedas safe and effective..." Dr. Henry thought that a briefdiscussion about the incidence of colorectal cancer in Westernsocieties would help explain why phenolphthalein was a logicalsubstance for NTP to evaluate.

    Public Comments. Ms. Laura Zeoli, Novartis Consumer Health Inc., said that Novartisis the manufacturer of Ex-Lax, the largest selling phenolphthaleincontaining over-the-counter laxative. She stated that Novartisdisagrees with the conclusion that phenolphthalein was a potentialhuman carcinogen via a genotoxic mechanism, and requests thatany decision on listing in the Report be delayed. Further, whilethe p53 transgenic mouse may prove to be a valuable research tool,Ms. Zeoli said there is presently significant disagreement amongtoxicologists on the appropriateness of using these test resultsquantitatively for extrapolation to the human population and asthe basis for a regulatory decision.

    Dr. Hart read into therecord written statements from Dr. John French, NIEHS, and Dr.Raymond Tice, Integrated Laboratory Systems, both of whom couldnot be present. Dr. French stated that he disagreed with theNovartis contention that phenolphthalein was not a potential humancarcinogen via a genotoxic mechanism. He reported that phenolphthaleingiven in the diet to male and female B6C3F₁ mice fortwo years demonstrated the potential for genotoxicity in vivo by (1) increasing p53 protein overexpression in the nuclei ofthymic lymphomas but not in the thymuses of untreated controlmice, and (2) by loss of heterozygosity in mouse chromosome 11contiguous with the p53 locus. He said that normal p53 tumorsuppressor gene function is required for suppression of cancerand homeostatic maintenance of a nonmalignant phenotype in bothhumans and mice. Data collected to date on the function of p53and the mechanisms of inactivation of p53 function are consistentbetween species. Dr. Tice was principal investigator for theproject in which the phenolphthalein p53 study was conducted. He said that in a number of in vivo mouse studies, phenophthaleininduced a significant increase in the frequency of micronucleatederythrocytes. Based on these results, phenolphthalein is classifiableas an in vivo genotoxic agent.

    There was some discussionaround the loss of the wild type p53 allele in thymic lymphomasfrom phenophthalein exposed animals and whether these were comparablewith thymuses from control animals. Dr. Dunnick commented thatin looking at non-target tissues, the kidney and ear, in animalswith thymic tumors, these tissues had one wild type and one null p53 allele.

    Dr. Belinsky moved thatthe nomination of phenolphthalein for listing in the Report asreasonably anticipated to be a human carcinogen be accepted. Dr. Mirer seconded the motion, which was accepted unanimouslywith six votes.

11. Saccharin-Dr. Robert Maronpot,NIEHS, said that the Calorie Control Council had petitioned theNTP to consider delisting saccharin from the Report based on extensiveresearch supporting the safety of saccharin for human consumption. Saccharin is currently listed as reasonably anticipated tobe a human carcinogen. Dr. Maronpot listed the extensivefood and non-food uses of saccharin and reviewed the regulatoryactivity in the U. S. and elsewhere, as well as various pronouncementsby the IARC and the Joint Expert Committee on Food Additives (JECFA)of the World Health Organization from 1980-1993. In 1984, JECFAdetermined the no observed effect level (NOEL) for urinary bladdercancer in rats to be 1% in the diet, and they set an estimatedtemporary average daily intake (ADI) for humans at the upper limitof 2.5 mg/kg body weight. Dr. Maronpot posed the question asto what is different now than before? He said that more papershave been published pertaining to carcinogenicity and, also, theNTP has revised criteria allowing use of mechanistic informationin considering listing or delisting. He summarized the essentialelements: (1) urinary bladder carcinogen in male rats given highdietary levels for prolonged periods, (2) urinary bladder carcinogenin mice after bladder implants of sodium saccharin-containingcholesterol pellets, (3) thyroid carcinogen in mice given sodiumsaccharin in water by gavage in one study, and (4) numerous initiation/promotionstudies in bladders of rats with many positive. Among data gaps,saccharin has been less rigorously studied in female rats or mice,initiation/promotion studies for bladder cancer in mice are limited,and studies in hamsters and monkeys are inadequate for evaluation. Dr. Maronpot reported that there are numerous epidemiology studies,most of which focus on use of artificial sweeteners, rather thanon saccharin alone. Therefore accurate saccharin exposures aregenerally not very precise or are unknown. Dr. Maronpot describeda proposed mechanism of "male rat-specific" urinarybladder carcinogenesis predicated on high concentrations of urinaryprotein, mucopolysaccharides, and certain electrolytes associatedwith a calcium phosphate precipitate and a cascade of effectsrelating to urothelial toxicity, enhanced cell proliferation,and, ultimately, urothelial cancer.

    Dr. Maronpot said theRG1 by seven yes to three no votes, and the RG2 by six yes votesto two no votes recommended that saccharin be delisted from theReport. He said that the conclusory paragraph in the backgrounddocument states that under this proposed mechanism, "thefactors thought to contribute to tumor induction by sodium saccharinin rats would not be expected to occur in humans." Further,"although it is impossible to absolutely conclude that itposes no threat to human health, sodium saccharin is not reasonablyanticipated to be a human carcinogen under conditions of generalusage as an artificial sweetener." Dr. Maronpot then spokeof the assumptions pertaining to under conditions of generalusage. First, the NOEL in male rats for bladder cancer is1% equaling 500 mg/kg/day, while in 1993 JECFA set the ADI foradult humans at an upper limit of 5 mg/kg/day, so there is a 100-folddifference between rat and human consumption. However, lookingat consumption in children the U.S. Department of Agricultureindicates that children eat approximately twice as much saccharinas adults so therefore in children the difference is half he said.Further, if the surface area metric were used, and if the 90^thpercentile of human consumption were used instead of the average,then the safety factor drops to about 10 to 13-fold. Finally,he noted that the NOEL for cell proliferation in the rat urinarybladder was 1/10^th that for a tumor response, or 0.1%.

    Dr. Hooper, the primaryreviewer, said that he was not convinced that the available dataestablished the mechanism of cancer in rodents, nor was it sufficientlyclear that it would differ from humans. Better understandingof the mechanisms of bladder cancer in rodents and humans wouldbe helpful. He commented that in rats, the fetus binds four tofive times as much saccharin in the bladder as does the maternalanimal, while repeated dosing gives higher concentrations in thebody than does single large doses. Looking at Dr. Maronpot'sfigures, Dr. Hooper said that the rat doses may be only 2 to 5or 20 to 50 times the average human consumption. He noted thatin the two generation study in female rats there are positiveresults yet no indication of an amorphous precipitate, and, againin implantation studies in mice, tumor bearing animals were notreported to have an amorphous precipitate.

    Dr. Zahm, the secondaryreviewer, expressed concern about the lack of two-generationaland early life studies in mice, given that this is a substancelikely to be heavily consumed by young women, and lack of an explanationfor the positive findings in mice. Looking at the epidemiology,she focused on the largest study, the Hoover study, and said therewas an excess of bladder cancer risk in the whole group, men andwomen combined, and there were excesses in some of the subgroups,e.g., nonsmoking women, where there would be a low risk of bladdercancer normally making an increase easier to detect. Dr. Zahmsaid it was important to note that the studies in the 1970s and1980s were really artificial sweetener studies with probably heavyexposure to cyclamates. She thought the issue of whether sufficientlatency had elapsed since the widespread use of saccharin beganwas critical, since bladder cancer is believed to have an averagelatency of 30 years or more. Further, one runs into the competingrisk of cause of death as bladder cancer is something where theaverage age is over 70. Dr. Zahm concluded that there are sometroubling subgroup findings that would mean the epidemiology couldnot be used to say that saccharin does not play a role in bladdercancer.

    Public Comments. Dr. Robert Gelardi, President, Calorie Control Council, saidthat the Council had petitioned the NTP to have saccharin delistedfrom the Report on Carcinogens on the basis of NTP's newcriteria incorporating the use of mechanistic data. He statedthat the extensive data obtained during the past 20 years on saccharinclearly demonstrate that the bladder tumor findings in rats arenot relevant to humans, and added that studies conducted in mice,hamsters, and monkeys have not resulted in any saccharin-relatedtumor development. Dr. Gelardi reviewed the mechanistic studiesin male rats that indicate formation of an amorphous urinary precipitateat high doses lead to bladder tumors. He added that more than30 human studies indicate saccharin safety at human levels ofconsumption.

    Dr. Michael Jacobson,Executive Director, Center for Science in the Public Interest(CSPI), said that he spoke also on behalf of a number of scientistswho cosigned a statement received by the Subcommittee, and whobelieve that numerous animal and human studies provide sufficientevidence that saccharin poses a risk and the only prudent coursefor the NTP would be not to delist saccharin. Their first concernwas that saccharin causes bladder tumors in rats and at dose levelsas low as 0.5% with smaller increases at 0.1%, which is certainlyin the range of ingestion by children. Their second concern isthat saccharin may cause tumors at sites other than the bladderand there is evidence for this in several rodent studies. Theirthird concern was that half a dozen case control studies founda significantly higher incidence of bladder cancer in people exposedto artificial sweetener.

    Dr. Frederick describedstudies he did in female mice while at NCTR that were designedto see if mice were sensitive to saccharin and whether saccharincould serve as a promoter for a relatively low dose of a genotoxiccarcinogen. Because of the initiation/promotion component, thestudy was taken out to 33 months, and to give a complete dose-responsecurve, doses ranged from 0.1 to 5 %. No toxic effects or preneoplasticlesions were seen in the mice. He said the other key issue hereis whether human urine is significantly diffent from rat urine. Dr. Frederick emphasized that many of the epidemiology studieswere dealing with exposure to a mixture of materials, especiallycyclamate. Dr. Bailer complained that we are trying to draw conclusionsabout the potential carcinogenicity of saccharin when most ofthe human studies involved mixtures of sweeteners. He thoughtthere could be more discussion about the issue of dietary confounders. Dr. Zahm said there are not a lot of strong associations betweendietary factors and bladder cancer.

    Dr. Bailer asked whetherthere were any studies that identified possible changes in humanbladders analogous to those seen in the male rat. Dr. Brown askedDr. Samuel Cohen, University of Nebraska Medical Center, if hecould address this issue. Dr. Cohen said he would but wantedto first comment on epidemiology studies of which he said theonly one that gave a statistical overall risk of bladder cancerin humans was the Howe study, and that study had a major confoundingfactor, cigarette smoking, which was not taken into account. With regard to changes in the human bladder, Dr. Cohen reportedthere is only one study, that by Auerbach and Garfinkle publishedin Cancer in the late 1980s. Cigarette smokers were positivecontrols and showed increased cell proliferation in the bladder,while there was no increased level of proliferation in nonsmokersexposed to high doses of artificial sweeteners. Dr. Cohen gavea discourse on the composition of urine from rats and mice exposedto saccharin, noting that calcium and phosphate levels are 10-20times lower in mouse than in rat urine, and the form of proteinin male rat urine vs female rat urine may be key to formationof the precipitate. In response to a question from Dr. Yamasaki,Dr. Cohen opined that there is not just a quantitative differencebetween human and rat urine but also a qualitative one.

    Dr. Bingham was troubledby the supposition that if there is only a slight increased riskin the overall study population then higher risks among subgroupscould be dismissed. Dr. Zahm agreed that within an occupationalsetting, there are subgroups where the cancer risk may be muchhigher than the overall cohort. Dr. Mirer stated that all theissues around differences between rodents and humans in urinecomposition may enter into risk assessment but in terms of hazardidentification based on the criteria, the animal data supportreasonably anticipated to be a human carcinogen, and the epidemiologicdata are on the border. Dr. William Allaben, FDA/NCTR, commentedthat whether or not saccharin remains listed or is delisted therewill be no change in how FDA regulates at least until 2002 dueto extension of the Congressional moratorium until then. Dr.Friedman-Jimenez pointed out that the subgroups in the Hooverstudy were determined a priori based on other reasoningand not on the data. With regard to the Howe study, he said thatin the later reanalysis where smoking was controlled for therewas little change in the relative risks. Dr. Hooper returned tomechanistic possibilities that had not been explored with saccharin. He said it binds to proteins, it binds to epidermal growth factor,it is known to be a protease inhibitor, it stimulates increasesin P450 enzymes, and it induces ornithine decarboxylase. So untilthe mechanism of bladder cancer in humans is known he said hewould be reluctant to say that rodent findings are not relevantto humans.

    Dr. Hooper moved thatsaccharin not be delisted from the Report as reasonably anticipatedto be a human carcinogen. Dr. Zahm seconded the motion, whichwas accepted by four yes votes (Bingham, Friedman-Jimenez, Hooper,Mirer) to three no votes (Bailer, Belinsky, Frederick).

12. 2,3,7,8-Tetrachlorodibenzo-p-dioxin(TCDD) -- Dr. Arnold Schecter, NIEHS, said that 2,3,7.8-tetrachlorodibenzo-p--dioxin (TCDD) was nominated by RG1 for listing in the Reportas known to be a human carcinogen (currently listed asreasonably anticipated to be a human carcinogen) basedon several types of evidence. The nomination took into accountthe IARC classification of TCDD as a Group 1 "Known HumanCarcinogen" (IARC Monograph Vol. 69, 1997). Dr. Schecterdescribed the chemical structure relationships of the dibenzodioxinsand dibenzofurans and noted the dioxins are unwanted contaminantsin wastes, industrial processes, and various consumer productswith general population intake about 95% from food. With regardto epidemiology, he spoke of the combined international cohortand selected industrial cohort studies with high exposure levelsfocused on by IARC, which included the NIOSH (Fingerhut) studyof U.S. chemical workers, the first study to use blood measurementsof TCDD as an estimate of exposure. Dr. Schecter then reportedon two studies published or in press since the IARC meeting. The first of these was the 15-year cancer mortality followup ofan environmental exposure to TCDD in Seveso, Italy (Bertazzi etal.), which reported cancer mortality in three zones rangingfrom highest exposed (Zone A) through next highest exposed (ZoneB) to third highest exposed (Zone R). The second study dealtwith cancer mortality follow-up of an occupationally exposed Dutchcohort, a portion of whom were accidentally exposed to TCDD (Hooiveldet al.). Dr. Schecter summarized the many laboratory animalstudies, noting that TCDD causes cancer by multiple routes, inmultiple species, multiple strains, both sexes, and in multipleorgans and tissues. With regard to mechanism, the scientificconsensus is that binding to the aryl hydrocarbon (Ah) receptoris a necessary early step for all known TCDD effects. The Ahreceptor is found in the cytoplasm of vertebrate cells from fishto humans. He said that there are numerous Ah receptor responsesthat have been characterized in experimental systems, many ofwhich are relevant to plausible mechanisms of chemical carcinogenesis. The half-life of elimination for TCDD ranges from 2 to 4 weeksin laboratory rodents to between 5.8 and 11 years in humans. Dr. Schecter summarized the arguments supporting the proposedlisting: (1) studies in humans strongly point to a causal associationbetween exposure to TCDD and an increased incidence of cancersin highly exposed occupational cohorts; (2) all studies in rodentshave been positive; (3) mechanistic studies support a commonmode of action in humans and rodents and body burdens necessaryto produce dioxin mediated responses are similar in rodents andhuman. Arguments against listing TCDD are (1) humans exposedto dioxins are also exposed to mixtures of other carcinogens,and (2) human cancer data alone my not be sufficient to establishcausality between dioxin exposure and human cancer. Dr. Schecterreported that RG1 unanimously with 10 votes and RG2 unanimouslywith eight votes recommended listing TCDD as known to be ahuman carcinogen .

    Dr. Yamasaki, the primaryreviewer, stated that there is not sufficient evidence of carcinogenicityfrom studies in humans (epidemiological and mechanistic data combined)to support a causal relationship between TCDD exposure and humancancer. The lack of specific sites make the human data more difficultto interpret. He said that the mechanistic information availableto us indicates that TCDD binding to the Ah receptor is a necessarybut not sufficient event for induction of tumors. Since TCDDbinding to the Ah receptor induces the transcription of severalgenes, some of which are involved in cell growth and differentiation,it is biologically plausible that TCDD could induce tumors. Yet,more direct information in exposed humans is needed.

    Dr. Frederick, the secondaryreviewer, commented that TCDD is the 'chameleon carcinogen' asit seems to have different target sites for humans in differentgroups which have been studied. He demonstrated this by reviewingthe tumor data associated with the three zones in the Seveso study. There did not seem to be a common lesion or organ site amongthe male and female cancer deaths nor did there seem to be correlationswith tumors seen in occupational sites other than perhaps softtissue sarcomas. Dr. Frederick then turned to proposed mechanismof tumor formation. He said that the early steps of TCDD bindingto the Ah receptor and to DNA response elements have been elegantlyelucidated but are not enough to explain its carcinogenicity.

    Public Comments. Ms. Lisa Finaldi, Greenpeace International, said that Greenpeacestrongly endorsed the decision by the NTP to consider new scientificevidence on the human carcinogenicity of TCDD. She stated thatthe recent action by IARC, the dioxin reassessment by EPA, andthe conclusion by the Institute of Medicine about dioxin-contaminatedherbicides lend support for similar conclusions and action byNTP linking TCDD to cancer in humans. She said Greenpeace recommendsthat NTP proceed with urgency to acquire sufficient resourcesfor a broad but thorough assessment of all health effects of TCDDand like chemicals, with not only cancer as an outcome, but mosturgently the effects of TCDD exposure on the developing fetusand nursing infant as well as the young child and adolescent.

    Dr. Raymond Greenberg,Vice President and Provost, Medical University of South Carolina,representing the American Forestry and Paper Association, saidthe IARC working group concluded that there was limited evidencein humans for the carcinogenicity of TCDD because the associationsobserved were quite weak compared with those observed for knownhuman carcinogens, the pattern of outcomes was unusual for a humancarcinogen, the evidence for a dose-response relationship wasquite limited, and there was strong evidence that confoundingcould not be ruled out.

    Dr. Nathan Karch, Karchand Associates, Inc., representing the Chlorine Chemistry Councilof the CMA, said that in his view, the epidemiologic evidencedoes not support a reclassification of TCDD with each key studyacknowledging the potential for confounding or bias, a view consistentwith EPA's conclusion. This is marked by the lack of consistentelevations in specific tumor sites among various cohorts. Dr.Karch said the mechanistic data also provided little support inthat the mechanism by which TCDD induces cancer in animals isnot known, much less in humans.

    Mr. Steve Lester, CCHWCenter for Health, Environment, and Justice, supported the listingof TCDD as a known human carcinogen based on the strong evidencein human studies of an association, the compelling evidence inanimal studies, and the mechanistic data showing a basic similaritybetween animals and humans. He noted the IARC's overall conclusion,and the additional evidence from the continuing study of the Sevesoresidents. Mr. Lester said this was not a decision about riskassessment but rather about hazard identification and not everysingle aspect of dioxin's mechanism needs to be known to makethis decision.

    Mr. Jim Tozzi, MultinationalBusiness Services, Inc., providing staff support to the Centerfor Regulatory Effectiveness, stated that the proposal to listTCDD as a human carcinogen based on mechanistic data is a clearviolation of Health and Human Services regulations. He said thereason for this is that when the criteria were recently revised,the criteria for listing as a known human carcinogen was not changed,and this was stated publically in writing. Thus, he suggestedthat to use mechanistic data to upgrade human data is not authorizedby the law and is subject to judicial review.

    There ensued further discussionby the Subcommittee of the epidemiology data. Dr. Hooper saidhe was not surprised by the wide variety of tumors seen in theSeveso residents and other cohort studies in view of TCDD beinga multisite carcinogen in animals and also due to the nature ofthe purported mechanism. Some of the lack of target organ consistencymay relate to the quite different exposure scenarios between theSeveso residents and occupational cohorts. He said the lack ofgenotoxicity makes defining a carcinogenic mechanism more difficult. Dr. Belinsky said that some of the mechanistic data is not usefuland he found the epidemiology data confusing. Dr. Lucier spokeabout the pieces of information used by RG1 and RG2 to supportthe listing. This included the compelling data on animal cancer,the agreement that the Ah receptor was the necessary channel fortoxic events in both animals and humans, and the similar bodyburdens of TCDD associated with tumors in both species. Dr. Mirersaid the mechanistic information for humans was not what it couldbe. Dr. Hooper wondered if deferral of the proposed listing untilmore human data were available could be an option. Dr. Bailerthought the epidemiological evidence as presented was fairly compellingparticularly with all cancers and lung cancer. Dr. Yamasaki saidhis concern with the lung cancers was the confounding of smoking. Dr. Schecter noted that smoking considered in the NIOSH studyand the conclusion was that smoking would contribute a small effect,if any. Dr. Douglas Sharpnack, NIOSH, said that NIOSH does havea policy of classifying TCDD as a potential human carcinogen,and their epidemiologists he thought are not as concerned aboutsite specific epidemiology but place greater weight on the relativerisks overall. Dr. Elizabeth Ward, NIOSH, agreed and said thather concern was increased by seeing increased risks of a largenumber of cancers at a number of sites, and particularly sinceTCDD causes cancer at multiple sites in animals.

    Dr. Yamasaki moved that2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) continue tobe listed in the Report as reasonably anticipated to be a humancarcinogen. Dr. Frederick seconded the motion, which wasdefeated by three no votes (Bailer, Bingham, Friedman-Jimenez)to two yes votes (Belinsky, Frederick) with two abstentions (Hooper,Mirer). Dr. Mirer said he would like to see a deferral untilthe epidemiology information could be clarified. Dr. Binghamthen moved that TCDD be listed in the Report as known to bea human carcinogen . Dr. Bailer seconded the motion, whichresulted in a tie vote with three yes votes (Bailer, Bingham,Friedman-Jimenez) to three no votes (Belinsky, Frederick, Hooper)with one abstention (Mirer). The Chairman, Dr. Brown, then votedyes to break the tie. The final vote was then four yes votesto three no votes with one abstention.

13. Trichloroethylene -- Dr.Jameson said that trichloroethylene was nominated by RG1 for listingin the Report as reasonably anticipated to be a human carcinogen based on evidence by the NTP in 1990 and by Maltoni et al.in 1988 where, by inhalation and gavage exposure to laboratoryanimals, trichloroethylene was demonstrated to be a multiple organ,trans-species carcinogen. Trichloroethylene also is structurallyrelated to the known human carcinogen, vinyl chloride. Dr. Jamesonreported environmental exposure to be more than 25 million pounds,and occupationally, more than 400, 000 workers are potentiallyexposed. There is limited evidence of carcinogenicity to humansderived primarily from three cohort studies of workers. Thereis evidence for genotoxicity in some systems and not in others. Dr. Jameson said there is cancer site concordance between tumorsobserved in humans and animals, including liver, kidney, and lymphomas. Renal cell carcinomas from trichloroethylene workers exhibiteddifferent exon mutation patterns of the von Hippel-Landau (VHL)tumor suppressor gene, a gene associated with renal cell carcinoma,than did non-exposed renal cell carcinoma patients. He said thatthe glutathione S-transferase enzyme pathway that produces theultimate electrophile - chlorothioketene - in proximal tubularcells is more prevalent in humans than in rats. Dr. Jameson saidthat RG1 voted seven to two, and RG2 voted seven to one to recommendthat trichloroethylene be listed as reasonably anticipatedto be a human carcinogen.

    Dr. Mirer, the primaryreviewer, agreed with the proposed listing. He said the animaldata were straightforward and supported the listing. He commentedthat there were inconsistent tumor site data among the epidemiologystudies. Dr. Mirer said the concluding statement supporting thelisting implied that one needed to know the mechanism and suggestedchanging the wording to read "no compelling data indicatingthat the agent acts through mechanisms which do not operate inhumans."

    Dr.Kelsey, the secondaryreviewer, was not present so as requested, Dr. Hart read his reviewinto the record. Dr. Kelsey agreed with the proposed listing. He said the human data were indeed limited by small numbers,but thought the data on cancers of the kidney to be compelling. Dr. Kelsey thought the mechanistic studies on the VHL gene inexposed workers to be of great interest. The mutational spectrafor trichloroethylene workers appeared to be different from thatin non-exposed workers. He commented that the genotoxicity andmetabolism data indicate there are definite electrophilic metabolites,a fact consistent with mechanistic data indicative of carcinogenicrisk.

    Dr. Mirer moved that thenomination of trichloroethylene for listing in the Report as reasonablyanticipated to be a human carcinogen be accepted. Dr. Frederickseconded the motion, which was accepted unanimously with sevenvotes.

14. Dyes Metabolized to Benzidine(Benzidine-Based Dyes as a Class) -- Dr. H. B. Matthews, NIEHS,said that benzidine-based dyes were nominated by RG1 for listingin the Report as known to be a human carcinogen basedon the fact that these dyes are metabolized to a known human carcinogen,benzidine, and on the fact that all three of the benzidine-baseddyes tested induced evidence of liver cancer in rats in 13-weekstudies. Benzidine-based dyes are readily synthesized by reactionof benzidine with a variety of aromatic amines to form azo linkage,and they are readily metabolically reduced by intestinal microflorato benzidine and the respective aromatic amine(s). He said thatanimal studies have demonstrated almost complete metabolic conversionto free benzidine in rodents, thus exposure to the benzidine-baseddye is equivalent to exposure to an equimolar dose of benzidine.Epidemiological evidence for the carcinogenicity of benzidine-baseddyes has been difficult to demonstrate because exposure is almostalways associated with co-exposure to benzidine. However, anIARC evaluation concluded that "Although the epidemiologicaldata were inadequate to evaluate the carcinogenicity to man ofthe individual dyes, they together with the presence of benzidinein the urine of exposed workers provides sufficient evidence thatoccupational exposure to benzidine-based dyes represents a carcinogenicrisk to humans." He said that benzidine carcinogenicityis attributed to its metabolism to the proximate carcinogen, N-hydroxy-N-acetylbenzidine,in the urinary bladder of humans and liver of rodents. Dr. Matthewssaid that RG1 by seven yes to one no votes and RG2 by unanimousvote supported the proposed listing.

    Dr. Bingham, the primaryreviewer, agreed with the proposed listing. She noted that whilethe U.S. is not producing these dyes, we do buy clothing fromall over the world including places where these dyes are stillused.

    Dr. Hecht, the secondaryreviewer, agreed with the proposed listing.

    Dr. Bingham moved that the nomination of benzidine-based dyes for listing in the Reportas known to be a human carcinogen be accepted. Dr. Hecht seconded the motion, which was accepted unanimously with seven votes.