Tobacco Documents Online

Pulmonary Carcinogens in ETS (September 25, 1990) Page 1 Non-Epidemiologic Studies on Potential Pulmonary Carcinogens in Environmental Tobacco Smoke: A Critique of the Environmental Protection Agency's Designation of Environmental Tobacco Smoke as a Group A Carcinogen Domingo M. Aviado, M.D. Former Member, EPA Clean Air Scientific Advisory Committeel President, Atmospheric Health Sciences 152 Parsonage Hill Road, P.O. Box 307 Short Hills, New Jersey 07078 CONTENTS Introduction 2 Suspected Carcinogens in MSS and SSS 3 Human carcinogens with sufficient epidemiologic evidence 5 Pulmonary carcinogens 6 Reference standard for pulmonary carcinogens 6 Extrapulmonary carcinogens 7 Experimental animal carcinogens with limited epidemiologic studies 8 Polycyclic aromatic hydrocarbons and heterocylic 9 N-Nitroso compounds 11 Miscellaneous compounds 12 Biologic markers of exposure to ETS 15 Concluding Remarks 17 References (1) to (31) 20 Table I Sidestream smoke constituents, maximal emission and indoor concentrations. 23 Table II EPA Group A - Human carcinogens with sufficient epidemiologic animal inhalation studies to support a causal association. 24 Table III Carcinogens without sufficient evidence from epidemiologic studies to support a causal association. 25 Gb Table IV National Toxicology Program results of inhalation -g bioassay. 27 Table V Estimated number of cigarettes in 100 cubic meter IA enclosure for SSS emission to attain TLV levels. 28 W is

Pulmonary Carcinogens in ETS (September 25, 1990) Page 2 INTRODUCTION This submission is in response to a published notice2 that the Environmental Protection Agency (EPA) concludes that environmental tobacco smoke (ETS) is a "Grbup A or known human carcinogen." My comments apply to the first of six summary arguments contained in the EPA external Review Draft: Biological plausibility. ETS is taken up by the lungs and distributed throughout the body. The similarity of carcinogens identified in SS and MS along with the established causal relationship between lung cancer and smoking make it reasonable to suspect that ETS is also a lung carcinogen. (page 1-3, "Health Effects of Passive Smoking: Assessment of Lung Cancer in Adults and Respiratory Disorders in Children.") The remaining five summary arguments relate to the epidemiologic basis for the proposal that ETS is a "Group A or known human carcinogen." I have reviewed the available literature on non- epidemiologic studies of cancer associated with ETS exposure, specifically the chemical and toxicologic studies of ETS constituents. I conclude that there are several reasons for challenging the claim-regarding the "biological plausibility" that ETS is a pulmonary carcinogen.' The principal reason for rejecting the Review Draft's argument is that constituents identified as "suspected carcinogens" in mainstream or sidestream smoke have not been proven as site specific pulmonary carcinogens. My research interest in tobacco dates back to the 1950s when I investigated reactions elicited by nicotine and other chemical substances.3 During the 1960s, my interest in tobacco smoke

Pulmonary Carcinogens in ETS (September 25, 1990) Page 3 included investigation of cardiopulmonary effects.4,5 The possible health effects of ETS and other inhalants were my research objectives during the 1970s and 1980s.6 - 11 During the past three decades, inhalation toxicolbgy has evolved as a sub-specialty in occupational, environmental and regulatory toxicology. This research has permitted the distinction between pulmonary carcinogens and nonpulmonary carcinogens, based on primary lesions appearing as a result of inhalation of chemical vapors and particulates. SUSPECTED CARCINOGENS IN MSS AND SSS At the outset, it is important to note that although there are over 3200 constituents identified in cigarette mainstream smoke (MSS), approximately a hundred have been detected in sidestream smoke (SSS). Of the 100-plus SSS constituents emitted from lighted cigarettes in the laboratory, 20 have actually been detected in cigarette-filled rooms.12 - 14 The remaining 80-plus SSS constituents have not been detected indoors because of limitations in monitoring and analytic techniques for low level environmental constituents. The "biological plausibility" argument for ETS as a Group A "known human carcinogen" is restated in a second Support Document: Of the 99 compounds in tobacco smoke that have been studied in detail, at least 43 are complete carcinogens, each able on its own to cause the development of cancer in humans or animals. Other ETS

Pulmonary Carcinogens in ETS (September 25, 1990) Page 4 constituents are tumor initiators, capable of carrying out the first steps in cancer development. Still others are tumor promoters, able to accelerate the development of cancer. ETS also contains chemicals that are co-carcinogens, able to cause cancer when combined with other substances. It contains cancer precursors, compounds that pave the way for formation in the body of other carcinogenic chemicals. And it contains other compounds that damage the cilia, or cleansing hairs, of the lungs, making them less able to clear the lungs of deposited tars. This allows cancer-causing chemicals to remain. (pages 8, 9, "Environmental Tobacco Smoke: A Guide to Workplace Smoking Policies") The above quotation of EPA Review Draft was derived from the 1989 Surgeon General's Report.13 A table of 43 suspected carcinogens in MSS was in turn derived from an unpublished article by Hoffmann and Hecht,15 still In Press as a reference in the Review Draft (page R-11). Not all "43 suspected carcinogens" in MSS have been detected in SSS. Table I is my own list of 33 suspected carcinogens in SSS. The selection of constituents present in SSS is based on recent reviews.9 - 16 The selection of 33 SSS constituents includes 18 constituents derived from the source table of 43 MSS constituents.13, 15 The remaining 15 suspected carcinogens in SSS (Table I) are from the literature published during the past decade consisting of animal experiments.9-11, 17 From the original list of 43 MSS suspected carcinogens, the remaining 25 MSS constituents (43 less 18) are excluded from Table I because they have not been detected in SSS. The 25 MSS constituents are

Pulmonary Carcinogens in ETS (September 25, 1990) Page 5 listed in Tables II and III with evaluation of carcinogenicity based on animal experimentation. HUMAN CARCINOGENS WITH SUFFICIENT EPIdEMIOLOGIC EVIDENCE There are at least five scientific groups that continue to evaluate the scientific literature relating to carcinogenicity: the National Toxicology Program (NTP) with its Annual List of Carcinogens and Fiscal Year Reports;18 - 23 the National Institute of Occupational Safety and Health (NIOSH), with its Criteria Documents24 and its Registry of Toxic Effects of Chemical Substances (RTECS),17 also available from TOXLINE database; the American Conference of Governmental Industrial Hygienists (ACGIH) with its annual revision of Documentation of the Threshold Limit Values (TLV);25 and the International Agency for Research on Cancer (IARC) with monographs on the evaluation of the carcinogenic risks of chemicals to humans.26, 27 With the exception of NIOSH, which simply identifies carcinogens without further categorization, the four other groups have two major ratings for human carcinogenicity, namely: human carcinogen W= sufficient epidemiologic evidence of carcinogenicity and sufficient animal studies; and carcinogen without epidemiologic evidence and carcinogenicity based on sufficient or insufficient animal studies. The classification proposed by EPA for ETS is human carcinoaen with sufficient eoidemioloaic evidence, or GrouQ A in terminology used by EPA and

Pulmonary Carcinogens in ETS (September 25, 1990) Page 6 IARC, known carcinogen by NTP, and Al confirmed carcinogen by NIOSH. Pulmonarv Carcinocens. Table II lists chemicals or manufacturing processes that have been designated to cause human cancer in the lung and elsewhere, respectively sub-grouped into pulmonary and extra-pulmonary. There are seven entries listed as "pulmonary carcinogens" with acceptability ratings of sufficient eQidemioloaic evidence by two, three, or all four agencies, namely: NTP-known carcinogen; ACGIH-class Al; IARC-group 1; and NIOSH-carcinogen. The list of seven entries for pulmonary carcinogens includes inorganic substances (arsenicals, asbestos, chromium and nickel), and organic compounds such as mustard gas; bis(chloro- methyl) ether; and coal tar volatile, soot, tar, and coke oven emissions. All seven entries have been proven as pulmonary carcinogens by inhalation studies in experimental animals. Only one (nickel) is present in SSS but its concentration is less than that encountered in nickel refining or roasting industry. The other metals (arsenicals and chromium) have been detected in trace quantities in MSS. "Biological plausibility" is not supported by traces of nickel and undetected arsenic and chromium in SSS because their presence has not been demonstrated in ETS. The MSS levels are in nanogram quantities and depend upon metallic content of the soil nurturing tobacco. Reference Standard for Pulmonary Carcinoaens. It is necessary to briefly describe a selected reference pulmonary 0? '~i

pulmonary Carcinogens in ETS (September 25, 1990) Page 7 carcinogen such as bis(chloromethyl)ether. Exposure of workers to this substance is reputed to increase the risk of lung cancer. The exposure occurs in chemical plant workers, ion exchange resin makers, laboratory workers, and polymer woikers. Absorption of the chemical is through the lungs and skin. It is rated by NTP as a "known carcinogen" because there is sufficient evidence of carcinogenicity from studies in humans, which suggests a causal relationship between the agent and lung cancer.l8. 19 Bis(chloromethyl)ether produces tumors at the site of application in mice and rats, i.e., lungs after inhalation, subcutaneous tissues after injection, and skin after repeated topical a follows:1 pplication. 7 The features of inhalation studies are as rat inhalation - 100 ppb for 6 hours daily for 6 weeks rat inhalation - 100 ppb for 6 hours daily for 26 weeks rat inhalation - 75 ppb for 6 hours daily for 2 years mouse inhalation- 1 ppm daily for 82 days. The above results support the rating by IARC of "sufficient evidence" as an animal carcinogen and the rating by NTP of "known carcinogen." The threshold limit value (TLV) is 0.001 ppm, which provides an acceptable factor of safety based on additional short-term inhalation studies in humans and animals.25 Extra-Pulmonary Carcinoaens. There are five organic chemicals that cause cancer in the urinary bladder, liver or bone marrow but not in the lungs (Table II). These substances, referred to as "human extrapulmonary carcinogens," have been adequately investigated by epidemiologic studies of exposed

Pulmonary Carcinogens in ETS (September 25, 1990) Page 8 workers and animal inhalation experiments. The reported neoplastic diseases of workers are: leukemia by benzene; liver cancer by vinyl chloride; and urinary bladder cancer by 4- aminobiphenyl, benzidine and 2-naphthylamirie. Benzene, 4- aminobiphenyl and 2-naphthylamine are present in SSS but have not been detected in ETS. It is not "biologically plausible" to attribute lung cancer in nonsmokers to these three extrapulmonary carcinogens. EXPERIMENTAL ANIMAL CARCINOGENS WITH LIMITED EPIDEMIOLOGIC STUDIES With the exception of four human carcinogens listed in Table II, the remaining 27 SSS constituents are classified as without sufficient evidence from epidemiologic studies to support a causal associa tion in hu man cancer. The experimental animal carcinogens ar groups as foll e listed i ows: n Table III. Each one has been rated by NIOSH-ca (carcinogen) 3 NTP-ac (reasonably anticipated to be carcinogenic) 23 ACGIH-A2 (suspected human carcinogen) 8 IARC-g2A (probably carcinogenic in humans) 2 IARC-g2B (possibly carcinogenic to humans) 6 IARC-g3 (sufficient evidence of carcinogenicity in experimental animals) 37 Not rated by above 8 MSS entries ..... 53 SSS entries ..... 29 m For purposes of description of animal experiments, the `~ O! C1t constituents of cigarette smoke are grouped into three classes: ~ 00 (D

Pulmonary Carcinogens in ETS (September 25, 1990) Page 9 25 polycyclic aromatic hydrocarbons (PAHs) and heterocyclic compounds; 9 N-Nitroso compounds; and 19 miscellaneous compounds, including 3 inorganics. Polycyclic Aromatic Hydrocarbons and Heterocyclic Comnounds. Among the 25 constituents of MSS listed in Table III, 12 are present in SSS, and only 3 SSS constituents are recognized either by NTP, or ACGIH, or both, as having sufficient evidence from experimental animals, namely: benzo[a]pyrene, benz[a]anthracene, and dibenz(a,h]anthracene. All 25 MSS constituents have been tested by mouse skin painting with positive results. From the standpoint of pulmonary carcinogenesis, the following procedures have been applied and reviewed in RTECS:17 Oral administration in rat (lung neoplasm) dibenz[a,h] anthracene 4160 mg/kg for 126 weeks anthracene 20 g/kg for 79 weeks Subcutaneous injection in mice (lung neoplasm) dibenzo[a,i]pyrene 72 mg/kg for 9 weeks Subcutaneous injection in rat (neoplasm at site) anthracene 3300 mg/kg for 30 weeks Intrathoracic implant in rat (lung neoplasm) benzo(a]pyrene 150 mg/kg benzo(k]fluoranthene 5 mg/kg fndeno[1,2,3-cd]pyrene 20.75 mg/kg benzo[b]fluoranthene 5 mg/kg dibenz(a,h]acridine 1.5 mg/kg benzo(j]fluoranthene 5 mg/kg benzo(k]fluoranthene 5 mg/kg Intratracheal injection in hamster (lung neoplasm) benzo(a]pyrene 120 mg/kg for 17 weeks dibenzo[a,i]pyrene 33 g/kg for 8 weeks Intratracheal injection in mouse (lung neoplasm) benzo(a]pyrene 200 mg/kg for 10 weeks Intratracheal injection in rat (lung neoplasm) ~ benzo[a]pyrene 68 mg/kg for 15 weeks Inhalation of mouse (lung neoplasm) ~ benzo(a]pyrene 200 ng/cuM/6 hours daily for 13 week f~jts Inhalation of hamster (lung neoplasm) benzo(a]pyrene 1500 ug/cuM 4 hours for 96 weeks ~

Pulmonary Carcinogens in ETS (September 25, 1990) Page 10 In the above tests, RTECS has'rated lung neoplasm results as "equivocal" yielding "uncertain but seemingly positive" results.17 The positive results not rated as "equivocal" were derived by intrathoracic implantation or ihtratracheal injection. Only the inhalation study is relevant to the potential health effects of ETS, and these results are designated "equivocal". The test exposure concentrations of benzo(a]pyrene were reported as 9500 ug/cuM in the hamster, and 200 ng/cuM in the mouse. The peak benzo[a]pyrene concentration reported in the literature is 0.07 ug/cig or 70 ng/cig for sidestream smoke emitted by one cigarette. The mixture inhaled in one hamster study consisted of 9500 ug, which, when divided by 0.07 ug, is equivalent to 135,714 burning cigarettes contained in one cuM. In the mouse study, the mixture consisted of 200 ng (divided by 20 ng equals 10 burning cigarettes in one cuM of inspired air), and the results were "equivocal". Although the concentration in the mouse is lower than that used in the hamster (by a factor of 6785), the burning cigarettes per cuM of enclosure would be intolerable in terms of mucosal irritation. Benzo(a]pyrene is a research chemical and only researchers would be exposed to the pure substance. However, occupational exposure is widespread because benzo(a]pyrene is present in coal tar, coke oven emissions, and creosote, all of which have corresponding work exposure standards.18, 19 Benzo[a]pyrene andt other PAHs occur in the combustion products of coal, oil,