Tobacco Documents Online

. WN ` r? t.w w. © ~ ® 0 0 0 e ® ® © Cancer Letters, 14 (1981) 285-290 .285 Elsevier/North-Holland Scientific Publishers Ltd. EVIDENCE OF A MULTIPLICATIVE' EFFECT BETWEEN CIGARETTE SMOKING AND OCCUPATIONAL EXPOSURES IN THE AETIOLOGY OF BLADDER CANCER P. VINEISa, N. SEGNANb, G. COSTAb and B. TERRACINIa °Unit of Cancer Epidemiology, University of Torino and Torino Main Hospital and bHealth Adn:inistration, City of Torino„Torino (Italy) (Received 31 July 1981) (Revised version received 14 September 1981) (Accepted 17 September 1981) SUMIbIARY A case-control study on urinary tract cancer in men is ongoing in the Province of Torino. The analysis of tobacco smoke and occupational histories of the first 225 case-control pairs is described lThe overall relative risks (RR) for tobacco smoke and for occuoations known to entail a risk for bladder cancer were respectively 4.3' (a5%confidence limits (C.L.) 2.4--9.3) and 1.7 (1.0-2 i). There was a clearcut dose-response relationship between different categories of cigarette consuraotion and risk for urinary tract cancer. For each level of cigarette consumption, relative risks were compared between men reporting and' non-reporring occupations at risk. This analysis suggested a multiplicative effect between rhe 2 factors/ INTROD U C'i'ION Tobacco smoke is know:l to be carcinogenic for the human urinary tract. Several cay=-control an& ronorr studies have shown an overall RR in the order of 2-3, with a direcr dose-re-sponse relationship [25]. The well-known bladder carcinogens [12] 2-naphthylamine and 4-aminobiphenyl have been found in tobacco smoke [25; and the intake of a smoker of 20 unfiltered cigarettes per day has been estimated~ to be in the order of 3 mg 2-naphthyl- *This study was supported by the ltalian Ligue Against Cancer, the Italian Association for Cancer Researcn, the Health Administration of Torino and the Italian National Research Council ;contract no. 80:02312:96). Address all cor, espondence to: Dr. Paolo Vineis, Servizio di Epidemiologia dei Tumori„ Universita di Torino e Ospedale Maggiore, via Santena 7, 101,26 Torino, Italy. 0304-3835/81 /0000-0000/502.7 5 01981 Elsevier/North-Holland Scientific Publishers Ltd. t 9 e R .

amine in 2 years [27]. However, an exclusive role of aromatic amines in the bladder carcinogenicity of tobacco smoke is yet to be demonstrated. On the other hand, an excess risk for cancer of the urinary tract has been shown among workers engaged'in dyestuff manufacture [1,5,10,11,23], rubber and --cable production [5,6,9,15], leatherwork [5,7], petrol refining [11,19], gas plants [84, textile industry [1]. These associations are partly but not totally explained by the presence of and exposure to carcinogenic aromatic amines in the workplace. The risks associated with industrial exposures to specific aromatic amines are reviewed in Ref. 12. The risk has decreased where the use of carcinogenic aromatic amines was discontinued [2]; how- ever, a risk has been shown for some occupations where these chemicals are not used, such as men engaged in 'component building' in some English rubber industries [2]. . The possible interaction between tobacco smoke and occupational exposures in the aetiology of bladder cancer has been considered in 3 previous epidemiologic studies. One of them suggested an overall additive effect, but RRs were not reported separately for the different categories of cigarette smokers [4]. On the contrary, the study of Howe et al. [11], in which models of logistic regression were applied, has suggested a multiplica- tive effect between smoking and occupation. Finally, a small study indicating an increased risk among workers in the electrolysis department of an alumi- num reduction plant, also suggested a multiplicative effect with tobacco smoke: risks (relative to 1.0 for non-smokers never employed in such activity) were 1.8, 1.9 and 5.7, respectively, in non-occupationally exposed smokers, occupationally exposed :ron-smokers and occupationally exposed smokers [22]. New findings from an on-going case-control study seem to confirm thee possibility of a more than additive effect between occupation and tobacco smoke in the aetiology of bladder cancer. personnel, c occupationa codes for in( Ratios (notE [14]. In adc' occupationa RRs were c RR• RR = I RESULTS AN As a preli separately fc (non-smoker did not diffe analyses wer Einploym of the follov as 'occupatic of tyres and TABLE 1 DISTRIBUTIC CONSUIMPTIC Age-classes Exposed to occ <50 Ca. Co 50-59 Ca: Co 60+ Ca5 Co: Non-exposed't( <50 Caf Co: 50-59 Ca' Q Co: ~' 0 C~` O F+ 6 + Co. I•i 'As defined in

, TABLE 1 RESULTS AtiD DISCUSSION 60+ 50-59 {$ Cases 2 Controls 11 Cases 6 Controls 21 8As defined in the text. Non-exposed to oceupa!ional'risirs. <50 Cases 1! 7 Controls 5 8 Age-classes Average cigarette consumption <I DISTRIBUTION OF CASES AtiD CON-TROLS BY AGE AND AVERAGE DAILY CONSUMPTION OF'CIGARETTES (AFTER AGE 20) Exposed :o occupational risksa Cases 0 3 1 5 0. 9 <50 Controls 2 ? 1 0 0 5 -59 C~es 1 3 13 2 0 19 50 Controls 2 2 6 1 0 11 + V~s 0 4 11 1 2 18 60 Controls 0 10 2 1 1 14 8' 4 2 22 10 5 1 29 12 37 8 4 63 21 23 8 3 66 17 43 16 12 94 33 34 7 5 100 personnel, using a standard questionnaire covering lifelong smoking and occupational histories. The latter were subsequently coded using pre-existing codes for industrial'branches [13] and occupational titles [3]. Exposure-Odds Ratios (noted as RRs) were age-adjusted with the Mantel-Haenszel method [14]. In addition, RRs based on discordant pairs have been estimated when occupationa.i' exposures alone were analysed. Ninety-five percent C.L. of RRs were calc•slated with the formula suggested by Miettinen [181: RR' R,. = RR(i`)(a/Xp); Xp for pairs was McNemar's test [16]. As a preliminary analysis, RRs for tobacco smoke were determined separately for incident and prevalent bladder cancers: within each category (non~smokers, smokers of 1-9, 10-19, 20--29 and 30+ cigarettes/day) RRs did not differ more than 15% between the two groups. Therefore, further analyses were carried out on the whole series. Employment for att least 6 months at any time during life in one or more of the following industrial branches and/or occupational titles was considered as 'occupational exposure': chemical industry; dye production; production of tyres and' other rubber goods; petrol refining; gas plants; typographers; 0 z-J 10-19 20-29 30+ Totals 287 © 0 © ® e he :he je ea- ing :li- Lzy) rs, ~rs ain 70 ne was 1a 's. ~ i- :Ons dd had 9'.0

TABLE 2 AGE-ADJUSTED RELATIVE RISKS ASSOCIATED WITH CIGARETTE SMOKE (AVERAGE DAILY CONSUMPTION AFTER AGE OF 20) AMONG NONLOCCLTPA- TIONALLY EXPOSED PEOPLE, OCCUPATIONALLY EXPOSED PEOPLE'AND IN THE WHOLE SERIES' Average Non-occupa- Occupation- Whole series (95% C.L.)b cigarette tionally ally exposedO consumption exposeda Non-smokers 1.0 1.2 1.0 1-9 2.4 2.8 2.5 (1.1- 5.3) 10-19 5.4 11.1 5.9 (2.9-12.0). 20-29 ( ( 7.0 (3.0-16.3). 30+ ` 16.6 11.2 7.8 (2.9-21.2) ' Reference category: non-smokers non-occupationally exposed'~ b Reference category: non-smokers. furnacemen; leather; shoe repair; textile. At least one of these activities was reported by 46 cases and 30 controls, which corresponds to an RR of 1.7 (95% C.L. 1,0-2.7 ). Analyses restricted' to ages < 50, 50-59 and 60+ yielded, respectively, RRs of 2.4 (1.7 8',0), 1.8 (0.8-4.1) and 1.4 (0.6-2.9). Nine- teen cases and 10 controls (all in discordant pairs) had worked in the rubber indusuy, thus entailing a R R of 1.9 (0.9-4.0). Table 1 reports absolute numbers of cases and controls broken down by age, tobacco smoke and occupational exposure. Table 2 shows the age- adjusted RRs for cigarette smoke estimated: (1) from'non-occupationally exposed people; (2) from occupationally exposed people; (3) in the whole series. Age adjustment under (1) and (2) was made on only 2 age groups (<50 and 50+ ) due to smail' absolute numbers in some cells of Table 1. The overall RR for smokers compared to non-smokers was 4.8 (2.4-9.3): this value is higher than those found in studies carried out elsewhere, possibly bec--:a;e of the age limit of ': 0 years in the present study. A1 further RRs were- ...mated assuming non-occupationally exposed non-smokers as reference category. The relative risk for smokers (regardless of the amount) was 4: z(2 ~2-3'.1) in non-occupationally exposed men vs. 6.8 ('2.9--15.6) in occsaaTionally exposed men. A dose-response relationship jvith ciQa..rett.e smoke was present in both occupationally exposed and non- exposed men (Table 2). The observed RRs given in Table 2 fit better with a multiplicative statistical interpretation of the combined effect of smoking and o=spational exposures. This is reminiscent of the interaction between asbestos and' cigarette smoke in the aetiology of lung cancer [21] and between smoke and alcohol in the aetiology of cancer of the larynx and oesophagus [24,26]. A major problem of this kind of analysis is the difficulty in finding people in extreme c exposed nor rate ratio [] difference b• (1.2 as a Ma on the crudt Little is k stones, so tl- However, a small group has given rel those obtair In spite o effect betwc bladder carc in all workp' whether or i ACKNOWLEI The techr edged. REFERENCE I Anthony, of the occ 895. 2 Baxter, P.~. 76. Her M; 3 Bureau In- fessions, E 4 Cole, P. ar Br. J. Prev 5 Cole, P., H urinary trz 6 Davies, J.14 146. 7 Decoufle, leather prc 8 Doll, R., V report of a ..19~_ 9 Fox, A.J. : O cablemakii . P.-\. "L4` -"L64. tv vuiuwamr C .3 _.,. _i __a .a.. r ..,, . F 11 Howe, G.F Fm various nu M %J

Anthony, H.M. and Thomas, G.M. (1'.970) Tumors of the urinary bladder: an analysis of the occupation of 1030 patien:.s in Leeds, England:. J. Natl. Cancer Inst., 45, 879- 895. Baxter, P.J. and Werner, J.B. (1980) Mortality in the British rubber industries 1967- 76. Her Majesty's Stationery Office, London. Bureau International du Travail (1969) Classification Internationale Type des Pro- fessions, Edition revisee. Geneoe: Cole, P. and Maci4Iahon, B. (3971) Attributable risk percent in case-control studies. Br. J. aev. Soc. Med., 25, 24v"-244. Cole, P., Hoover, R. and Fr:eaefi, G.H. (1972) Occupation and cancer of the lower urinary tract. Cancer, 29, 12a0-1^60. Davies, .;'2M. (1965) iBladder tumou.-s in the electric-cable industry. Lancet, 2, 143- 146. Decoufle,, P. (1979 ) Cancer risks associated with employment' in the leather and leather products industry. Arch. E: viron. Health, 34, 33-37. Doll, R., Vessey, M.P., Beasley, R.W.R. et al. (1972) Mortality of gasworkers - final report of a prospective study. Br. J. Ind. Med'., 29, 394-406. Fox, A.JJ and Collier, P.F. (1976) A survey of occupational cancer in the rubber and cablemaking industries: analysis of deaths occurring in 1972-74. Br. J. Ind. Med., 33, 249-264. Goldwater, L.J., Rosso, A.J. and Kleinfeld, M. (1965) Bladder tumors in a coal tar dye plant. Arch. Environ. Health, 11, 814-817. Howe, G.R., Burch, J.D., Miller, A.B. et al. (1980) Tobacco use, occupation, coffee, various nutrients and bladder cancer. J. Natl. Cancer Inst., 64, 701-713.

Cancer L, Elsevier/i~ 18 Miettinen, O.S. (1976) Estnnability and estimation in case-referent studies. Am. J. Epidemiol., 103, 226-235. 19 Milharn, S: (1976) Occupational mortality in Washington State 1950-1971. NIOSH Research Report, Vol. 1. U.S. Department of Health Education and Welfare, Washington DC. 20 Monson, R.R. and Fine, L.J., (1978) Cancer mortality and morbidity among rubber workers. J. Natl. Cancer Inst., 61, 1047-1053. 21 Selikoff, I.J., Hammond, E.C. and Churg, J. (1968) Asbestos exposure, smoking and neoplasia. J. Am. Med. Assoc., 204, 104-112. 22 Theriault, G., De Guire, L. and Cordier, S. (1981') Reducing aluminum: an occupation possibly associated with bladder cancer. Can. Med. Ass. J., 124, 419-423. 23 Tsuchiya, K., Okubo, T. and Ishizu, S. (1975) An epidemiological study of occupa- tional bladder tumours in the dye industry of Japan. Br. J. Ind. Med., 32, 203-209. 24 Tuyns, A.J., Pequignot, Gy. and Jensen, O.M. (1977) Le cancer de l°oesophage en Ille-et-Vilaine en fonction ces niveaux de consommation d'alcool et de tabac: des risques qui se multiplient. Bull. Cancer, 64, 45-60. 25 U.S. DHEW (1979) Smoking and health: a report, of the Surgeon General, P.H.S. Publication 79-50066. i7epartment of Health, Education and Welfare, Washington DC. 26 Wynder, E.L., Covey, L.S.. Ivlabuchi, K. and Mushinski, M. (1976) Environmental factors'in cancer of the larynx. A second look. Cancer, 38, 1591-1601. 27 Wynder„E.L. an&Golasr.iiLz. R, (1977)The epid'emiolbgyof bladder cancer. A second look. Cancer, 40, 1246-_268. i.TPTAK- PHOTO: POTEN' JOHAN h Departme; lfion tebell (Received' (Accepted Four tt potential agent hen Cell size r single cell, types had the untrar Hpdwasf membranE malignant to the pre: . The pre tumors, aw literature, tissue. INTRODUC An animals ; of the mox more than tumor tissu whether th properties c inadequate A numb,