Tobacco Documents Online

EXPOSURE TO ENVIRONMENTAL TOBACCO SMOKE AND OTHER INDOOR AIR POLLUTANTS ON THE INCIDENCE OF LUNG CANCER - A REVIEW Du Ying-xiu Abstra ct I. Introduction 1. Effects of environmental tobacco smoke (ETS) on health - though in certain areas, such as asthma, clear > conclusions have been reached, there is still a great deal of debate on the association of ETS and lung cancer; 2. Reasons for such debates - A. Complexity of the problem itself; B. Difficulties in research methodology - difficulties in both epidemiologic research methodology and laboratory research methodology. II. Necessary Focus of Research

1. Since lung cancer is closely related to cell types, any analyses of association of ETS and lung cancer must be correlated with consideration of cell types; 2. Attention to composition of mainstream and sidestream smoke; 3. Search for an ETS biomarker; 4. Mutual confirmation by epidemiological and laboratory methods. III. Available literature and inadequacies 1. Majority of available literature is on results of epidemiological research, with little correlation with laboratory results; 2. Some are not related to cell types; 3. In available literature, some contradictory results are reached; 4. Individual literature has pointed out several biomarkers, but their values still need to be ascertained. - 2 -

IV. Conclusions and suggestions of direction of research 1. Emphasis on the correlation of epidemiological research and laboratory research is needed; 2. Focus on adenocarcinoma when investigating female lung cancer; 3. Etiology of female lung cancer: ETS? Hormone? Air pollution? 4. Urgent research for biomarker that can reflect the relationship of ETS and lung cancer. introduction It has generally been established that smoking is an important risk factor of lung cancer in both men and women, but the relationship between environmental tobacco smoke (ETS) and lung cancer is still being debated. While many studies have not discovered a close relationship between ETS and lung cancer, others have reported that ETS can induce lung cancer in non-smoking women. However, the latter group could not agree among themselves on the lung cancer cell type: some are of the opinion that ETS is - 3 -

associated with only squamous cell carcinoma which is totally unrelated to other types of lung cancer; others think ETS is only related to adenocarcinoma. Between ETS and lung cancer there exists a very complex relationship. This is because lung cancer is a disease of not only multiple risk factors but long latency. To determine the relationship between ETS and lung cancer, other lung cancer risk factors or confounders must be controlled. Moreover, it is very difficult to design a study protocol due to the length of lapse of time and the complex ETS-exposure conditions of the study subjects. In addition, a biomarker that can accurately reflect the effects of ETS on health is lacking at present. Perhaps this is the main reason why a clear conclusion of the ETS-lung cancer relationship has not been reached. For purposes of providing a reliable basis for ETS-lung cancer relationship research, this paper presents a survey of world literature to date on this subject; makes some suggestions on questions the research should answer and potential problems the research may encounter as well as future directions such research might take. Focus`of ETS-Lung Cancer Research - 4 -

It is generally known that in investigating the etiology of a disease several steps must be taken: First, epidemiological , approach is used to form a hypothesis of a cause-effect relationship. Then, laboratory experiments are used to confirm the hypothesis developed by epidemiology. Finally, epidemiological method is again used to reconfirm laboratory results. In other words, any investigation into disease etiology must rely on the correlation of epidemiology and laboratory research, research on the relationship of ETS and lung cancer is, of course, no exception. That is to say, the clues provided by epidemiology should be explained by the mechanism of the disease, and the laboratory results should confirmed by intervention study of epidemiology. In correlating epidemiologic and laboratory research, a very important issue is a biomarker to accurately reflect the effects of ETS on health, and on which to base dose-response evaluations. Theoretically, this biomarker should be a direct marker for ETS-induced mutation of lung cells or at least an indirect marker for that biochemical change and which change is believed to be related to lung cancer. However, to date, neither a direct nor an indirect biomarker has been found that can reflect the ETS-lung cancer relationship, not even a biomarker that can reflect the effects of ETS on health in general. This is where the greatest difficulty of this research lies. - 5 -

It is commonly known that there are four cell types in lung cancer: squamous cell carcinoma, small cell carcinoma, adenocarcinoma and large cell carcinoma. The four cell types are differentiable by their etiology, morphology, structure, function and site within the lung. More importantly, many studies show that different causes may induce different cell types of lung cancer. Therefore, any study of ETS and lung cancer must be correlated with cytology of the lung cancer. It has been reported that smoking induces central type squamous cell carcinoma. Since research has found little difference in the constituents of mainstream and sidestream smoke, there is reason to believe if ETS is capable of lung cancer induction, it must also induce primarily squamous cell carcinoma. Any epidemiological analyses without correlative study of cancer cell types must be considered deficient. In summary, when researching the relationship between ETS and lung cancer, one, must consider the cell types, use precise biomarkers, adopt a methodology that correlates epidemiologic and laboratory results. These are the focal issues to be considered when designing a study protocol. Difficulties in the Study of ETS and Lung Cancer Relationship As stated above, the study of disease origin should seek mutual verification from both epidemiologic and laboratory - 6 -

research. At present, however, the great majority of research results are obtained from epidemiologic investigations. In epidemiologic research there are many difficulties to overcome, for example: 1. Regardless whether induced by active smoking, occupational exposure or air pollution, lung cancer is characterized by a long latency period. In other words, lung cancer is the result of cumulative effects of carcinogens. If ETS can induce lung cancer, the latency period is undoubtedly longer than in the case of active smoking. During such a prolonged period of time, the effects of ETS are likely to be variable and complex; and as a result, it is difficult to obtain true, quantitative dose- response relationship. 2. Lung cancer may have multiple causes. To ascertain the effects of ETS on lung cancer, in addition to working with only non-smoking study subjects, possible effects due to occupational exposure and air pollution must also be excluded - a task that may be difficult to accomplish. 3. In this investigation process, if the survey is conducted retrospectively or by correspondence, the data obtained may be difficult to control for accuracy. If the survey is based on hospital cases, in addition to selecting study subjects, - 7 -

attention must be given to the accuracy of the smoking habits of the active smokers to whose cigarette smoke the study subjects are exposed. 4. The smoking status of active smokers to whose smoke the passive smokers are exposed is not constant during a long period of time. It is also difficult to exclude the element of self-imposed avoidance of exposure by the nonsmoker due to aversion to the spouse's cigarette smoke. 5. It is not only difficult to estimate exposure to ETS in public places, but also in the home environment. If the living space is large, for example, then the effect of ETS is difficult to estimate. 6. Even though B(a)P, DMNA and other carcinogens can be found in sidestream smoke and even in amounts higher than those found in mainstream smoke, what these chemicals normally induce is central type of squamous cell carcinoma. It is apparent that the sidestream smoke absorbed by the body is greatly diluted, and less likely to penetrate deeply into the lungs, thus with less opportunity to cause peripheral type lung adenocarcinoma. Therefore, any study of ETS-lung cancer association must be correlated with analyses of lung cancer cell types. Any - 8 -

epidemiological studies without related cytological analyses will have questionable reliability. 7. At present, a reliable biomarker for ETS exposure has not been found; neither is a personal exposure measuring device available. Thus, exposure data to ETS are generally descriptive. The acceptability of descriptive epidemiology data is subject to certain limitations. 8. Due to the long latency period in the development of lung cancer, with added complications in the status of exposure to ETS, it is very difficult to design an animal model to study the)~ association of ETS and lung cancer. Please consult references: 1. Katzenstein, A.W.: Environ. Int. 18: 341-345, 1992. 2. Gori, G.B.: Regl. Toxicol. Pharmacol. 14: 88-105, 1991. 3. Vberta, K.: Z. Hautkr. 626 (Suppl. 2): 26-29, 1991. t - 9 -

These are some of the important yet difficult to control factors associated with epidemiologic research. Perhaps this is the reason why it is difficult to arrive at a clear conclusion on the association of ETS and lung cancer. Review of Literature We have collected worldwide publications since 1980, on the association of ETS and never-smoker lung cancer cases. Among the 35 studies, 12 did not find any ETS-lung cancer association, 24 did find an association. Among the 24 [23) positive findings, only 11 have correlated with cell type studies. Among these, Garfinkel, Dalager and Pershagen think that, like active smoking, ETS induces primarily squamous cell carcinoma and small cell carcinoma, while Wynder, Goodman, Lam, Elizabeth, Fortham think that adenocarcinoma is primarily induced. Thus, to those who do consider the close relationship between disease cause and cell type, the latter two groups actually hold opposite views on this matter. It is well known that squamous cell carcinoma is the result of squamatization of the mucous epidermal cells of the larger bronchus near the hilus of the lung, and mostly of the central type; whereas adenocarcinoma is formed by the mutation of mucous epidermal and glandular cells of the smaller bronchi, mostly of the peripheral type. - 10 -