Tobacco Documents Online

RISK FACTORS FOR LUNG CANCER AMONG NON-SMOKERS WITH EMPHASIS ON LIFESTYLE FACTORS Gao, Yu Tang Shanghai Cancer Institute, Shanghai, China Abstract Exploration of risk factors for lung cancer among non- smokers, in particular among females, has attracted the attention of cancer researchers in China for a considerable period of time. Female lung cancer in some large Chinese cities is characterized by a relatively high incidence (although there is a relatively low smoking prevalence among females in the general population) and also by the high percentage of adenocarcinoma in lung cancer cases [1,2]. Based on results of two population-based, case-control studies, it has been estimated that the population attributable risks (PARs) due to smoking for female lung cancer in Shanghai and Shenyang were 0.24 (95% CI: 0.19-0.29) and 0.37 (95% CI: 0.29- 0.44), respectively [1,2], suggesting that the majority of female lung cancer cases (about 75% in Shanghai and 60% in Shenyang) can not be attributed to smoking. The causes of lung cancer in nonsmoking females remain to be explained. This review summarizes results of studies performed in China, focusing on risk factors for lung cancer among nonsmokers.

I. Indoor,air pollution 1. Coal burning The effect of coal fumes from heating or cooking in poorly ventilated houses on lung cancer risk has drawn the attention of numerous Chinese environmental scientists. A study in Xuan Wei County, China (where extraordinarily severe indoor air pollution due to burning smoky coal is known to exist) showed a close relationship between indoor air pollution (as measured by BaP concentration) and high lung cancer mortality rates (r=0.778, P 0.01). In the same study, no relationship was found between female lung cancer risk and tobacco smoking or exposure to environmental tobacco smoke [3,4]. Polycyclic aromatic hydrocarbon (PAH), a well-known human carcinogen, was found in the pollutant emitting from combustion of smoky coal. It is worthwhile noting, however, that indoor environmental conditions in high-risk areas, such as Xuan Wei County, are exceptional, since the average BaP concentration in houses without chimneys located in Xuan Wei was found to be extraordinarily elevated and has been reported to be as high as 626.9 µg/100 m3. A study in Shenyang and Harbin involving household conditions typically found in Northeast China [2,5], showed that the risk for lung cancer was 30-50% higher among women who spent most of their lives in homes heated by coal and who used coal as the primary cooking fuel. The effects of indoor air pollution by - 2 -

burning coal were stronger for squamous and oat cell cancer than for adenocarcinoma of the lung, for which the effects were similar to cigarette smoking. The concentrated use of coal-burning stoves in Shenyang was estimated to contribute to 10-20% of the lung cancer cases [2]. 2. Volatiles of some vegetable oils from wok cooking at high temperature. Another suspected risk factor for lung cancer is the volatiles generated from cooking oils heated at high temperatures. Such an idea is supported by both epidemiologic and laboratory studies. A large-scale population-based, case-control study conducted in urban Shanghai [6] showed that lung cancer risk was increased with the use of rapeseed oil. Specifically, different levels of reported eye irritation experienced during cooking (used as a subjective variable to represent the severity of exposure to cooking vapors) were associated with an excess lung cancer risk. Controls were women who used soybean oil but never or 'rarely reported eye irritation. The few women who never cooked were excluded. The overall increase in lung cancer risk associated with rapeseed oil use, compared with soybean oil use, was 1.4 (95% CI: 1.1-1.8). The patterns of risk were similar for squamous/oat cell cancer and adenocarcinoma of the lung. After adjusting for eye irritation, a 60% higher risk for lung cancer was observed among - 3 -

women who reported considerable or somewhat smoky conditions in their homes when cooking. This was considered as another rough measure of exposure to cooking vapors as well as the efficiency of household ventilation. In addition, the odds ratios (ORs) increased with the number of different dishes prepared by stir frying, deep frying, or boiling. No significant case/control differences were found in regard to the type of fuel used for cooking in the Shanghai study. A large-scale case-control study of female lung cancer carried out in Shenyang and Harbin [5] showed that, in addition to tobacco smoking, the following variables had a significant effect on risk for lung cancer. (P 0.05) These appeared in the regression model in the order shown: deep frying, eye irritation, pneumonia, household tuberculosis, burning kang, self-reported occupational exposure to burning fuel, passive smoking from any household member, and heated brick wall/floor. It is interesting to note that the two variables related to cooking (deep frying and eye irritation) appeared in the model as the first and second most significant variables. In multivariate analysis of a case-control study of lung cancer in Nanjing [7), both squamous cell carcinoma and adenocarcinoma of the lung were significantly associated with cooking vapors; similar ORs were obtained for both types of lung - 4 -

cancer. In addition, coal stoves used for heating in the winter and non-gaseous fuel were also associated with an increase in lung cancer risk, although only for the squamous cell type. Researchers at the Shanghai Cancer Institute did a number of laboratory studies on the genotoxicity of heated cooking oil vapors [8,9]. They repeatedly observed that condensates of volatile emissions from rapeseed and soybean cooking oils were genotoxic in short-term tests, including the Salmonella mutation assay, the SV40 forward-mutation assay, the sister chromatid exchange assay, and the mouse bone marrow micronucleus assay. Mutagenicity of volatile emissions from rapeseed oil was much stronger than those from soybean oil in the Salmonella mutation assay. In another study [10], volatile emissions from soybean oil (collected in a cold trap) also increased mouse bone marrow micronuclei, which was consistent with the results of similar studies done by the Shanghai Cancer Institute. In the same study, peanut oil and lard condensates were not mutagenic, irrespective of the assay used. Two other laboratory studies also provided evidence on genotoxicity of rapeseed oil vapors [11,12]. A dose-dependent induction of rat tracheal epithelial cell transformation was shown by infusing condensates of rapeseed oil (with doses up to 1.5 mg/kg) into rat tracheas [11]. Formation of DNA adducts resulting - 5 -

from cold-trapped condensates of rapeseed oil was studied with 32P- post-labeling techniques using a butanol enrichment procedure and conditions that amplify adduct detection. These studies showed that the condensate could react with calf thymus DNA to form DNA adducts without S9 [enzyme fraction] activation. A total of six spots were identified on thin layer plates. No spots were evident when the unheated rapeseed oil or the solvent were used as controls. These results suggest that rapeseed oil condensates contazn some electrophilic compounds that could react with DNA directly to form adducts [12]. Collectively, these laboratory find3.ngFs give support to the epi.demi.ologically-based hypothesis that exposure to volatile emissions from some types of cooking oil partially contributes to an elevated risk for lung cancer in females. Since the genotoxicity of rapeseed oil condensates disappeared (or decreased significantly) with the addition of butylated hydroxyanisole (BHA) or with hydrogenation, it may be suggested that oxidation and pyrolysis of unsaturated fatty acids in cooking oils contribute to the observed genotoxicity [8,9]. Consistent with such an idea, rapeseed and soybean oil are known to contain linolenic acid, which has 3 double bonds and, hence, is easily oxidized at high temperature to produce pyrolysates. Condensates of linolenic acid have shown high mutagenicity in the Ames test. Moveover, condensates from peanut oil, which were - 6 -

initially non-mutagenic, became mutagenic when the peanut oil was first supplemented with linolenic acid, implying that linolenic acid plays an important role in the mutagenicity of condensates derived from cooking oil (8,13]. 3. Environmental Tobacco Smoke (ETS) Despite the trendy suggestion that passive smoking contributes in some unexplained fashion to a slightly elevated risk of lung cancer [14], results of investigations in China on the relationship between ETS exposure and lung cancer risk were inconsistent and equivocal. odds ratios for lung cancer in non- smoking wives in relation to exposure to ETS from husbands were 2.16, 1.19 and 0.79 in Tianjing, Shanghai, and Shenyang and Harbin combined, respectively [15]. The effects of exposure to ETS on lung cancer risk are difficult to evaluate due to uncertainties in the methodology of investigation. II. General air pollution The contribution of air pollution in general as a possible contributing risk factor for lung cancer has been proposed for a number of decades by Chinese environmental scientists. Evidence supporting such a proposal, however, has been lacking. To evaluate the effect of general air pollution, and at the same time give consideration to other relevant important risk factors such as smoking, a prospective cohort study was carried out in three - 7 -

residential areas of Shanghai with substantially different levels of general air pollution [16]. About 220,000 male and female adult residents in these areas were involved in the study. Information on smoking from each subject was obtained. Results of a five-year follow-up showed that there was no discernible effect of general air pollution on risk for lung cancer among male and female non- smokers, but the risk for lung cancer was higher in urban smokers than in smokers residing in suburban areas and on the coast. Such a difference might reflect either an interactive effect of smoking with pollution or a delayed effect of smoking due to differences in smoking histories between residents of these three areas. III. Other relevant risk factors 1. Diet and nutrients A review paper pointed out that consumption of vegetables, in particular those rich in j3-carotene, may reduce risk for lung cancer [17]. In Hong Kong, an association between vegetable intake and a reduced risk for lung cancer was observed among non-smoking women [18]. Few studies in China have addressed the relationship between diet and lung cancer. The relationship between diet and lung cancer was studied in male residents in a Yunnan Province mining community [19]. Cases consumed less protein-rich foods and vegetables than did controls. The relative risk for lung cancer across increasing quartiles of meat consumption were 1.00, 0.67, 0.72, and 0.46 (P for trend 0.01). - 8 -

The relative risks for lung cancer across increasing quartiles of consumption of dark-green, leafy vegetables were 1.00, 0.62, 0.52, and 0.41 (P for trend 0.01). A similar trend was observed in a population-based, case-control study in Shanghai [6], in which the risks for lung cancer were found to be lower among those with reduced consumption of carotene-rich foods. No effect on risk was found for consumption of retinol-rich foods. In Shenyang, a more frequer~t intake of retinol and carotene-containing foods did not protect against lung cancer in smokers or nonsmokers [2]. 2. History of lung diseases In most case-control studies of lung cancer in China, it was found that a history of lung diseases such as tuberculosis, pneumonia, and emphysema were associated, to varying degrees, with an increase in risk for lung cancer. Smoking, which is usually associated with both chronic lung disease and lung cancer, was adjusted for in data analysis of these studies. After adjusting for smoking, the excess risk for lung cancer in association with history of lung diseases persisted [5-7, 20]. A retrospective cohort study of tuberculosis patients registered in the Shanghai TB registry since 1972 was carried out during 1987-89 to test the hypothesis that an association exists between lung cancer and pulmonary tuberculosis [21]. A total of 30,373 cases of pulmonary tuberculosis (born before January 1, 1957 - 9 -

and residing in urban Shanghai) were followed until 1986. The SMRs for lung cancer (calculated to be 1.38 and 2.73 in males and females, respectively) were statistically significant. When the risk was adjusted for smoking, the adjusted SMRs for lung cancer were 1.72 (95% CI: 1.11-2.53) in males and 2.79 (95% CI: 1.79-4.14) in females. Thus, the elevated risk for lung cancer among tuberculosis patients was independent of smoking. Neither INH treatment nor exposure to X-rays explained the higher risk. 4' Considering that chronic diseases of the respiratory tract are prevalent among the Chinese population, the elevated risk for lung cancer associated with these diseases and their contribution to total risk for lung cancer should not be neglected. 3. Menstruation In a population-based, case-control study of lung cancer in Shanghai [6,221, an unexpected observation was that the risk for lung cancer was higher among women with shorter menstrual cycles. The association existed primarily for adenocarcinoma and showed a strong dose-response relationship. Additionally, among age 55 and older females with natural menopause, the risk for adenocarcinoma. showed an increase with the total number of lifetime menstrual cycles. A study in Shenyang and Harbin suggests that the risk for lung cancer was positively associated with the age at which natural menopause occurs [5]. Additional studies are needed to clarify the - 10 -