Tobacco Documents Online

I I I I I I I RECENT PROGRESS IN THE EPIDEMIOLOGY OF LUNG CANCER IN HUMANS Du Ying-xiu Guangzhou Medical College, Guangzhou, China Abstract Lung cancer has been on the rapid rise worldwide during the last three or four decades. In China, the death rate from lung cancer is the highest among all types of malignant tumors in the urban population. Smoking, indoor and outdoor air pollution, and certain occupational exposures have been recognized as the main risks of lung cancer. This has been confirmed by many epidemiological research and laboratory studies. However, the significance of such risk factors may vary between different sexes or for different areas. Smoking is an important risk factor of lung cancer in both men and women; however, a large number of female lung cancer patients are never-smokers, indicating potential important risk factors other than smoking. There is a great debate over the association of passive smoking and lung cancer. Currently, available information cannot sufficiently confirm that passive smoking is capable of lung cancer induction. The relationship of atmospheric air pollution and lung cancer has long been noted; however the relationship is complex and the investigation requires data from long-term studies. Indoor air pollution is an important risk factor for lung cancer in women in China, but this is rarely reported outside China. At present, li carcinogens and 5 industrial processes have been confirmed as causes of occupational lung cancer. With further etiology research, more lung cancer causes may be discovered. To date, several observed phenomena are still without explanation. For example, why is smoking not an important factor in lung cancer in farmers? Or, what is the reason for the high incidence of adenocarcinoma in women? The answers may require research in the pathogenic mechanism of lung ancer. Introduction I According to a World Health Organization (WHO) report, for the past few years, stomach cancer and cardiovascular disease have decreased, while lung cancer is on a rapid rise, globally(1). There are already 35 nations where lung cancer is the number one malignant tumor in men; other nations may see lung cancer also becoming the number one malignant tumor in women. According to forecasts based on available information, AIDS and lung cancer will be the two most frequent health threats to mankind in early 21st century(l). In 1980, the number of new lung cancer cases in the world were estimated to be 600,500 (including 66,300 in China). If effective measures of prevention are not adopted, this number could reach 2 million in the year 2000 and 5 million in 2025. In China, according to annual nationwide health statistics, during the seven years 1982-1988,the average annual death rate of China's 16 largest cities was 565/100,000 with little change in the last seven years (regression coefficient b = 0.011, P > 0.05). The total cancer death rate, on the other hand, was on the rise; the average of 100/100,000 in 1982 was increased to 125/100,000 in 1988 (b = 0.0117, P < 0.05), including the lung cancer death rate which not only constitutes 25°l0 of all cancers but also increased most rapidly, 25/100,000 in 1982, 32/100,000 in 1988 (b = 0.0151, P< 0.01). The swiftness with which the lung cancer death rate has risen is not often seen in other diseases; this inevitably causes great concern. I

Regarding the cause of lung cancer, according to Brunner(2) there is reason to believe that the global prevalence of lung cancer is caused by conditions of the modern society and the unhealthy lifestyles of the people. Smoking, indoor and outdoor air pollution, and certain occupational exposures are considered the three most important factors in the etiology of lung cancer. Selawry and Hansen(3) suggest that at least 80% of lung cancer can be attributed to chemical carcinogens. Thus, research into the mechanism of carcinogenesis will be the basis to prevent lung cancer. Active Smoldng and Lung Cancer That smoking can cause lung cancer has been confirmed. Doll and Peto's 20-year retrospective study of 34,440 British male doctors found the adjusted death rate for nonsmokers was 10/100,000; for noncigarette smokers 48/100,000; for 15-24 cigarettes per day smokers 127/100,000; for over 25 cigarettes per day smokers 251/100,000(4). The occurrence rate for lung cancer decreased by 11 % of the estimated occurrence within 15 years of smoking cessation, while no change occurred for the incidence of other tumors. These data strongly suggest the close relationship between smoking and lung cancer. Many case-control studies on the relationship between smoking and lung cancer have been conducted in many areas of China. For example, the Wuhan Medical College study reported a relative risk (RR) value of 5.33 for smoking and lung cancer; Liaoning Health Investigation Bureau reported a RR of 8.45; Nanjing Health and Antiepidemic Station reported a RR of 6.51 for smokers of fewer than 20 cigarettes per day, and RR as high as 17.95 for smokers of more than 21 cigarettes per day. In 1985, our study of 849 cases and controls of lung cancer in Guangzhou showed smoking had an important significance for both men and women(5). The RR for men, at a 95% confidence level, was 3.53 (2.44- 5.11, P < 0.01) and 1.93 (1.30-2.27, P < 0.01) for women. The reason for the lower relative risk for women was that many female lung cancer patients were nonsmokers, which in turn indicated that potential risk factors other than smoking existed for female lung cancer. Eatough reviewed world literature on the chemical composition of mainstream and sidestream tobacco smoke and found that among the 108 traceable chemicals, in addition to the six that had already been designated as carcinogens by IARC (2-naphthyalmine, 4-aminobiphenyl, benzo(a)pyrene, N- nitrosodimethylamine, formaldehyde, and acetamide), others may be potential carcinogens also(6). Zhan et al.(7) applied a metabolite of B(a)P, anti-BTBE, to bronchial epithelia] cells of human fetus and observed not only mutation at H-ras gene 12 but the damages also resembled the mutation phenomenon observed in human lung cancer specimens. Kapitulinik et al.(8) induced lung tumors in mice by anti- BTBE. Chen et aL(9) applied smoke aerosol to cells of human fetus and found that the cells underwent morphological transformation suggesting that smoking is associated with lung cancer. Additionally, Wu et al.(10) applied extracts from snuff tobacco and chewing tobacco to BALB/3T3 cells and found the cells underwent mutations and the cell growth from the transformed colony exhibited characteristics of neoplastic transformation. These findings indicate a possible carcinogenic effect of smokeless tobacco. In summary, whether based on epidemiological or laboratory research, evidence exists for the association of smoking and lung cancer. Smoking cessation is apparently an important measure to prevent lung cancer. - 2 - I I I I I I I I I I i I I I

I I 1 I I I I I I I I I I I Passive Smoking and Lung Cancer The relationship of environmental tobacco smoke (ETS) exposure and lung cancer is still being debated. In China, studies from Harbin, Shanghai, Guangzhou and Xuanwei all reported no association between smoking and female lung cancer. Studies from other countries have produced mixed results. Some found no relationship between the two; others, while finding ETS an important risk in female lung cancer, disagree on cell type. Some find an association of ETS and adenocarcinoma only, which is unrelated to lung cancer of any other cell type, others find ETS a risk for squamous cell carcinoma only. Since the etiology of a disease is closely related to cell types, the latter two groups actually hold opposing views on the effect of ETS. Lung cancer is characterized not only by having multiple risk factors but also by its long latency. The conditions of human exposure to ETS can also be complex. For these reasons, to ascertain the relationship between ETS and lung cancer, the research must include good controls for a number of factors, such as: 1. The study subjects must experience "true" exposure to ETS, i.e. other than being nonsmokers themselves, the study subjects' exposure to air pollutants and occupational exposure must be controlled; 2. Both the extent of the exposure to ETS and the active smoker's smoking status should be accurately measured; 3. Objective reference biological markers exist that can precisely reflect the exposure to ETS. Since these conditions cannot be simultaneously achieved, it is not surprising that the association of ETS and female lung cancer cannot be confirmed. We conducted a case-control study of nonsmoking lung cancer patients that included effects of the husbands' smoking on lung cancer of nonsmoking wives, which also analyzed the relationship of active and passive smoking with lung cancer cell tvpes(l1). We found no association between the two. Pershagen et al.(12) conducted a case-control study which surveyed 27,409 Swedish female nonsmokers by questionnaire. They found that when nonsmoking women were married to smoking husbands the RR (3.30) of squamous carcinoma for these women increased significantly (P < 0.05). It is noteworthy that in their 20-year follow-up, only a small number of lung cancer cases (67 total) were found, only 20 of which were squamous and small cell carcinoma. In order to explain why ETS only induces adenocarcinoma of the peripheral type and not the central type squamous carcinoma which is primarily induced by active smoking, Wynder et aL(l3) proposed the following hypothesis: when ETS passes through the nasal cavity, the vibrissae are able to block certain particulates, with the result that gaseous phase carcinogens in the sidestream smoke are able to penetrate deep into the lung, even deeper than active smoking, and thereby inducing peripheral type adenocarcinoma. This hypothesis invites discussion. Carcinogens in tobacco smoke have high vaporization temperatures: for example, 2-naphthylamine vaporizes at 3060C, 4-aminobiphenyl at 302oC, benzo(a)pyrene at 311OC and N-nitrosodimethylamine at 152oC. It is extremely unlikely that these chemicals will maintain their gaseous state in the ambient environment or in the body without coalescing into particulates, which will make their deep penetration into the lung less probable. Recently, Trichopoulos et al.(14) evaluated the effects of active and passive smoking by using the increase in squamous metaplasia and abnormalities of the bronchial and alveolar basal cells as evidence of EPPL (epithelial possibly precancerous lesions). By using an EPPL value of 60 as a baseline value, they found that nonsmoking women married to smokers had a higher EPPL value than those married to nonsmokers. This result was interpreted as evidence supporting the view that ETS can induce lung cancer. A number of findings in the same paper, however, are at variance with such a view. For example, the EPPL value for heavy smokers could be as low as 29; the OR of 6.0 for - 3 - I

I nonsmoking women married to smokers (when compared to those married to nonsmokers) is actually higher than the OR of 4.4 for active smokers (compared to nonsmokers). That passive smokers can have a higher risk of lung cancer than smokers is biologically implausible. In summary, it is reasonable to conclude that current data have not strongly proven an association existing between ETS and lung cancer. However, this does not mean that ETS is harmless to humans. The sidestream smoke of cigarettes contains many harmful substances which are apparently hazardous to health. Atmospheric Air Pollution and Lung Cancer Statistical data show that the 1988 death rate in China for large cities is 32.14 per 100,000 which is higher than for medium size and small cities, 17.00; which is, again, higher than for rural areas, 12.53. Two questions arise: 1. Is air pollution one of the causes of a higher death rate for urban populations than for rural populations? 2. There is little difference in the smoking rates of urban and rural population? Why is it that smoking does not appear as important to farmers' lung cancer? The atmospheric air pollution and lung cancer relationship has long been noted. Stocks studied lung cancer rates in various parts of Great Britain and reported that they are closely related to the local atmospheric deposit index, smoke index, population density, and atmospheric concentrations of benzo(a)pyrene (B(a)P), beryllium, molybdenum, vanadium and arsenic(16, 17). Blot analyzed regional lung cancer death rates in the United States and found higher male lung cancer death rates in locations where paper, chemical, petroleum, and locomotive manufacturing industries are located and that the death rate is also related to atmospheric air pollution(18). Blot and Xu conducted a case-control study in Shenyang and found more male and female lung cancer patients among those living near smeltering plants for many years(19). Wang et al.(20) studied the relationship of lung cancer regional distribution and industrial pollution in Shanxi Province. They found that lung cancer at a particular location in Shanxi is inversely correlated with its distance from an industrialized center but is directly associated with its degree of industrialization. The lung cancer death rate is also positively correlated with atmospheric air pollution. The direction of the spread of lung cancer from high-incidence areas is also related to prevailing wind directions, being higher in "down-wind" areas than "up-wind" areas. For example, the spread from Taiyuan as the focal center, the locations of high-incidence areas in Quingxu, Jiaocheng, Wenshui, Taigu and Yuci coincides with the prevailing wind direction of Taiyuan area. Since lung cancer is a slow developing disease, the atmospheric air pollution is under constant change, and the conditions of human exposure unstable, the determination of air pollution-lung cancer relationship must be based on information from long-term studies. In 1991 we compiled atmospheric monitoring data collected by Guangzhou authorities during a 17-year period (1972-1981) and calculated the air pollution index(21). We subjected the data and Guangzhou's lung cancer death rates during 1976- 1989 to further analysis. We found higher lung cancer death rates in districts with more serious air pollution. - 4 - 1 I I I I I II

I I I There are many types of carcinogens in the atmosphere, including benzo(a)pyrene and benzo(a)anthracene. Among them the most important is B(a)P. There are many sources of B(a)P, mainly from industrial and home coal-burning, automobile and airplane exhaust. It has been reported that burning of 1 kg of coal can produce 2.1 mg of B(a)P, and 100g of coal smoke contains as much as 6.4 mg of B(a)P. In locations where traffic flow rate is 540-1,050 car/hr, 0.79-3.25 µg/100M3 of B(a)P can be detected. Yu et al.(22) conducted research on the mutagenicity of particulates according to size (diameter) contained in the atmospheres of Beijing, Taiyuan, Wuhan, Shengyang and Xuanwei. They found all samples to be mutagenic, and in inverse relationship with size of the particulates. Particulates with diameter < 1.0 µm have the highest mutagenicity, and at the same time the < 1.0 µm size particulate are the easiest to be retained in the lungs. In summary, it can be confirmed that lung cancer is related to atmospheric air pollution. However, since the occurrence of lung cancer is affected by many factors, and since the conditions of atmospheric air pollution are subject to constant ongoing changes, the search for a quantitative relationship may not be realistic. Indoor Air Pollution and Lung Cancer 1 I I I In China, a high incidence of female lung cancer has been reported from Harbin, Shanghai, and Guangzhou. All have pointed to indoor air pollution, caused by coal-burning, as an important cause of lung cancer. Gao et al.(23) in Shanghai found indoor air pollution and the use of rapeseed oil for cooking to have significant effects on the occurrence of female lung cancer. Sun et al.(24) in Harbin found, after adjusting for smoking, that indoor coal stoves and fire pits for heating can increase the risk of female lung cancer. Wang et al.(25) in Nanjing found kitchen cooking fumes to be a cancer risk factor in both squatnous cell carcinoma and adenocarcinoma in the lung. Ye(26) in Tainjin found that, after excluding cigarette smoking as a factor, women who live in run-down one-story houses in close proximity to low boiler chimneys and fumes from workshops have higher risks of lung cancer. Ou et aL(27) found that coal-burning households not only have higher levels of suspended dust, suspended dust-B(a)P, sedimentary dust, sedimentary dust-B(a)P in the air than propane-burning households, but housewives in the former have higher urine B(a)P content. This is direct evidence that carcinogens in indoor air are capable of entering the body. Liang et al.(28) conducted an on-site study in Xuanwei County of Yuennan Province by letting mice and rats breathe air containing coal-burning smoke, wood burning smoke, or "unpolluted air" (control). After 15-19 months, the lung cancer rate of the coal smoke group was higher than that from the wood smoke group, and the wood smoke group was higher than that of the control group. The B(a)P concentration of kitchen air is related to methods of cooking, with frying and stir- frying meats producing the highest B(a)P concentration. It has also been reported that carcinogens that entered the body carried by food have been found in the lungs(29). The question of whether food-bome substances can cause lung cancer is worth noting. There are great differences in indoor air problems in China and in highly industrialized nations. The main fuel for cooking and heating is coal in China, electricity and gas in industrialized nations. Of course, there are indoor air pollution problems abroad, such as wall board and ceiling tile installations, harmful emissions from carpets and other allergenic particles. But these materials are not significantly related to lung cancer. In China, with increasing use of gas, the nature of indoor air pollution problem will change correspondingly. - 5 - I

It is understandable that indoor air pollution should be more important to female lung cancer than male lung cancer. A remaining unexplained phenomenon relates to the fact that whereas chemical carcinogens such as B(a)P in general induce squamous cell carcinoma, primarily adenocarcinomas have been observed in female lung cancer. Occupational Exposure and Lung Cancer The confirmation of the etiology of disease must be based on verification from epidemiologic studies and laboratory research. Based on this priticiple, Saracci(30) of IARC classified lung cancer causing industrial carcinogens and industrial processes into the following. 1. Lung cancer carcinogens with sufficient evidence: arsenic and arsenic compounds, asbestos, dichtoromethyl ether, 6-valence chromium, tar, mustard gas, coal smoke, talcum powder contaminated by asbestos fibers, vinylchloride, nickel and nickel compounds; 2. Industrial processes with sufficient evidence: aluminum production, coal gasification, tar production, charcoal production, hematite-refining and radon radiation, steel casting; 3. Lung cancer carcinogens with insufficient epidemiological information: vinyl cyanide, beryllium and beryllium compounds, cadmium and cadmium compounds, crystal silicone; 4. Unconfirmed potential lung carcinogens: dimethylsulfate, aluminum, mineral oil, formaldehyde and phenobarbital. In China, lung cancer among tin miners was the first to be noticed as an occupational lung cancer. Wu et al. (31) in their study of Yuennan tin mines found the workers' lung cancer death rate had increased yearly since the 50s, reaching 0.3% in the 70s. In order to control the incidence of occupational tumors, China established National Cooperative Occupational Tumor Working Groups, to conduct overall investigations into the relationship of certain occupations and lung tumor. For instance, the National Arsenic Workers Lung Cancer Working Group(32) surveyed ten refining plants in Shenyang, Shanghai, Yuennan and mines in Hunan, found arsenic workers had both higher lung cancer standard mortality rate (SMR) and higher relative risks (RR) than the control group. The crude lung cancer incidence rate was as high as 248/100,000. The Asbestos Working Group(33) conducted occupational tumor surveys in nine asbestos plants and found malignant tumors to be the number one cause of death (SMR = 2.19, P< 0.01), and lung cancer the number one among malignant tumors (SMR = 6.33, P< 0.01). The two death rates are higher than for the control group. The Chloromethyl Ether Working Group's(34) survey of 11 chloromethyl ether manufacturing or user plants found for all tumors an SMR of 336, P < 0.01, and lung cancer SMR of 1546, P< 0.0001. In addition, Na et al.(35) conducted a retrospective cohort study of four nickel plant and mining operations and found lung cancer risks to be statistically significantly elevated in nickel refining and finishing workers. Further surveys by Chen et al.(36) among hematite workers and by Zhang et al.(37) among asphalt workers, all support the view that occupational risk factors have an important and significant contribution in the incidence of lung cancer. In general, it is easier to establish the relationship of occupational exposure and incidence of lung cancer, because there is clear and accurate employment history to verify exposure. On the other hand, - 6 - 2081783292 I I I I I I

a 1 I I I I I 1 I i the use of chemical substances by man is on the rise, both in quantity and in type. It has been reported the number of registered chemicals has reached over 5 million, with 60,000 of the chemicals in constant use, and with 200-1,000 new chemicals per year being added to the "use" list. Clearly, research of chemical carcinogenicity will be an important subject facing mankind, and the prevention of occupational lung cancer will be an important task. Etiology and Lung Cancer Cel1 Types Lung cancer is classified into 13 types 2nd several sub-types, by WHO, according to cytopathological appearance. The most frequently seen are squamous cell carcinoma, adenocarcinoma, small cell carcinoma and large cell carcinoma, with squamous cell carcinoma and adenocarcinoma making up 80% of all lung cancers. The reason why lung cancer by different causes are of different cell types must be explained by the mechanism of pathogenesis of the disease. Squamous cell carcinoma is formed by mutation of squamous metaplasia of epithelial cells lining the larger bronchial membrane near the hilus of the lung; it is the central type. Adenocarcinoma is formed by mutation of alveolar cells or the epithelial and glandular cells of the smaller bronchi; it is the peripheral type. According to epidemiological research(5): 1. The most frequent lung cancer in men is squamous cell carcinoma, followed by adenocarcinoma as the next most frequent. The reverse is true for women, in whom adenocarcinoma is the most frequent, followed by squamous cell carcinoma; 2. Smokers of either sex have a higher rate of squamous cell carcinoma than nonsmokers, and the cancer rate is further affected by smoking index (number of cigarettes smoked per day times years of smoking), with a higher smoking index being correlated with the higher rate of squamous cell carcinoma; 3. Among smokers, squamous cell carcinoma is higher than adenocarcinoma in males , while in females squamous cell carcinoma and adenocarcinoma are about equal; 4. When smokers are excluded (when comparison is made only among nonsmokers) squamous cell carcinoma and adenocarcinoma rates are similar in males, while in females the rate for adenocarcinoma is much higher than for squamous cell carcinoma. The above results show that smoking mainly induces squamous cell carcinoma and that the higher ratio of squamous cell carcinoma in males may be associated with smoking (the majority of male lung cancer patients are smokers) while in females, there may exist unknown factors for adenocarcinoma. Selawry and Hansen(3) in their analysis of the relationship of cell type and lung cancer etiology, noted that cancers associated with smoking, air pollution, occupational exposure and other environmental factors are chiefly squamous and small cell lung cancers. Such an association probably means that upon entering the lung, it is easier for carcinogens to settle in the larger bronchus and ultimately cause central type squamous cell carcinoma. That being the case, it is hard to conceive how the peripheral type adenocarcinoma and the central type squamous cell carcinoma would have the same underlying biologic mechanism(s). Kobayashi et aL(38) found that adenocarcinoma cells show a higher positive estrogen receptor than other lung cancer cell types, thus raising the possibility that estrogen uptake and/or function may be linked to development of adenocarcinoma. When human fetal bronchial epithelial cells were treated with DMNA and estrogen, cells were found to survive for 45 weeks, which was longer than cells not treated with estrogen or androgen(38). Moreover, the cells also showed the appearance of gland-like - 7 - I

structures in early passages which could be indicative of adenocarcinoma phenotypes. The estrogen-lung cancer relation needs to be further investigated. Prevention There cannot be real prevention of a disease without studying its etiology. Research has shown smoking, indoor and outdoor air pollution and occupational exposure as the three most important factors for lung cancer. Therefore, promoting smoking cessation, eliminating indoor and outdoor air pollution and controlling occupational hazards are the most important measures in the prevention of lung cancer. Smoking is not only an important risk factor for lung cancer, but it is also one of the risk factors in tumors of the oral cavity, larynx, esophagus, cervix and kidney. China's smoking rate ranks tenth in the world and its total tobacco consumption ranks first. Ten percent of the world's cancer deaths occur in China. In Guangzhou, for the over-15 population, the smoking rate is 43% for males and 4% for females, with a trend towards a smoking increase among young people. Smoking cessation should, therefore, be an urgent matter at hand. The U.S. National Cancer Institute (NCI) reported a lung cancer rate in American males at 71/100,000 and for females at 20/100,000 in 1982. If smoking is not stopped, in 2025 male lung cancer deaths will reach 311/100,000 and females, 56.9/100,000. If the NCI smoking plan is adopted (decreasing smoking rate to 15°k, use of low-tar cigarettes), in 2025 the male death rate will be 31.2/100,000 and the female rate 13.4/100,000. In 1986 the European Action Against Tobacco Committee was established(40). Funds of over $20,000,000 per year were raised for various activities, such as control of the tar content of cigarettes, health warning labels on tobacco products, ban of direct or indirect advertising of cigarettes, setting cigarette pricing policy, antismoking education in schools, etc. China should also establish a stop-smoking policy according to its own circumstances. Home coal-burning may cause severe indoor air pollution. It has been proven that the carcinogens released as a result of coal-burning are able to enter the body. Therefore, elimination of indoor air pollutants will have important meaning for lung cancer in females. It is believed that with the wider use of gas, the lung cancer death rate among females will decrease. Regarding the relationship between atmospheric air pollution and lung cancer, long-term, wide- scale investigation and research is needed. Atmospheric pollution is mainly due to industrial and mining operations; the control and elimination of industrial pollution are important measures in the prevention of occupational lung cancer. Improvement of general health may be important in lung cancer prevention. Trials are in progress in the U.S. in which high-risk population take daily doses of 50mg of vitamin E and 20mg of (3-carotene(41). At the present, these are several questions with no satisfactory answers. For example, why is I I I smoking seemingly not as important a factor in the incidence of lung cancer for farmers? Or, what is the cause (or causes) of adenocarcinoma, especially its high incidence in women? All these await further research. -1 - 8 - 2081783294 1 I

z I References I I , I I I 1. Stjersward, J. and Stanley, K. "Lung Cancer -- A Worldwide Health Problem," Lung Cancer 4(Supplement): 11-14, 1988. 2. Brunner, K.W. "Foreword of Fifth World Conference on Lung Cancer," Lung Cancer 4(Supplement), 1988. 3. Selawry, O.S. and Hansen H.H. "Lung Cancer, Cancer Medicine," pp. 1709-1744, 1982. 4. Doll, R. and Peto R. "Mortality in Relation to Smoking: 20 Years Observation in British Doctors," Brit. Med. J. 2: 1525, 1976. 5. Du, Y, X., et aI. "An Epidemiological Study of Lung Cancer Risk Factors in Guangzhou," Proceedings of Third Lung Cancer Conference in Guangzhou, 1-23, 1992. (In Chinese) 6. Eatough, D.J. "The Chemical Characterization of Environmental Tobacco Smoke," Proceeding of the International Symposium at MeGill University, 3-39. 7. Zhan, D.J., et al. "Study of Mutagenicity of H-ras Gene of the Bronchial Epithelial Cell of Human Fetus Induced By Anti-BPDE," J. of Health Toxicology 6(4), 1992. (In Chinese) 8. Kapitulinik, J., et al. Nature (London) 266: 378, 1977. 9. Chen, J.K., et al. "Study of the Effect of Cigarette Smoke Aerosol on Kidney Epithelial Cells of Human Fetus," J. of Health Toxicology 5(3): 143-146, 1991. (In Chinese) 10. Wu, Z.L., et al. "Study of the Effect of Smokeless Tobacco on the Mutagenicity of BALB73T3 Cells," J. of Health Toxicology 5(3): 167-169, 1991. (In Chinese) 11. Du, Y.X. "Exposure to Environmental Tobacco Smoke (ETS) and Lung Cancer in Women," Proceedings of Third Lung Cancer Conference, Guangzhou, 1992, p. 24-35. (In Chinese) 12. Pershagen, G., et al. "Passive Smoking and Lung Cancer in Swedish Woman," Am. J. Eoidemiol. 125: 17-24, 1987. 13. Wynder, E.L., et al. "Smoking and Lung Cancer: Some Unresolved Issues," Epidemiology Review 5: 177-207, 1983. 14. Trichopoulos, D., et al. "Active and Passive Smoking and Pathological Indicators of Lung Cancer Risk in an Autopsy Study," JAMA 268(13): 1697-1701, 1992. N C 15. Ministry of Health, People's Republic of China, Annual Report on National Health Statistics, ~ 1955. (In Chinese) V tb W IV - 9 - <O N I

16. Stock, P. "Cancer and Bronchitis Mortality in Relation Atmospheric Deposit and Smoke," Brit. Med. J. 1: 74-79, 1959. 17. Stock, P. "The Relation Between Atmospheric Pollution in Urban and Rural Localities and Mortality from Cancer Bronchitis and Pneumonia," Brit. J. of Cancer 14(3): 397-418, 1960. 18. Blot, W.J. "Geographic Pattern of Lung Cancer, Industrial Correlation," American J. of Epidemiology 14(3): 397-418, 1976. 19. Xu, Z.Y. and Blot, W.J. "Smoking, Air Pollution and the High Rates of Lung Cancer in Shangyang, China," J. of the National Cancer Institute 81(23): 1,800-1806, 1989. 20. Wang, J.S., et al. "Study of the Association of Industrial Pollution and Characteristics of Lung Cancer Distribution in Shanxi Province." (from exchange of information at conference) (In Chinese) 21. Du, Y.X. "Atmospheric Pollution and Human Lung Cancer," Lung Cancer 7(Supplement): 2, 1991. 22. Yu, S.Y., et al. "Study on the Mutagenicity of Atmospheric Particulates of Various Size," Chinese J. of Preventive Medicine 25(2): 70-74, 1991. (In Chinese) 23. Gao, Y.T. "Case-Control Study of Lung Cancer in Women," Proceedings of Second Lung Cancer Conference, Guangzhou, p. 17, 1987. (In Chinese) 24. Sun, X.W. "Risk Factors of Female Lung Cancer from Heating Fuel and Respiratory Disease," Chinese Tumor J. 13(6): 413-415, 1991. (In Chinese) 25. Wang, G.X., et al. "Multi-Variant Analysis of Cooking Fume and Other Risks in A Case- Control Lung Cancer Study," Chinese J. of Prevention Medicine 26(2): 89-91, 1992. (In Chinese) 26. Ye, Z. "Study of Environmental Factors of Lung Cancer in Housewives of Tianjin," Clinical Tumor in China 17(4): 195-198, 1990. (In Chinese) 27. Ou, X.L., et al. "Relationship of Home Use of Coal and Lung Cancer in Housewives," Proceeding of Second Lung Cancer Conference, Guangzhou, 1987, p. 76-81. (In Chinese) 28. Liang, C.K., et al. "Lung Cancer in Animals Induced by Breathing Coal and Wood Burning Smoke - An On-Site Study," Health Research 14(2): 16-22, 1985. (In Chinese) 29. Zhang, P.C., et al. O. "The Determination of Methylated Purines (06 M Gua and M7-Gua) in Liver and Lung of Mice," Chinese J. Preventive Medicine 24, 136-138, 1990. (In Chinese) l ' , I I I