Role of Household Exposure , Dietary Habits and Glutathione S-Transferases M 1 , T 1 Polymorphisms in Susceptibility to Lung Cancer among Women in Mizoram India

Lung cancer is the leading cause of cancer death worldwide with an annual mortality of 18.2% cancer death (Ferlay et al., 2010). In India, it continues to be important (D’Souza et al., 2013) and the incidence of lung cancer is highest in Aizawl district of Mizoram among females with an age adjusted rate of 38.7 per 105 population (NCRP, 2010). Mizoram situated in between 92.15 ́ and 93.29 ́E longitude and 21.58 ́ to 24.35 ́N latitude; surrounded by Myanmar in east and Bangladesh in west. Mizoram have their ancestral origin in China; later which shares a similar mortality trend with United States according to China Health Statistical Yearbook, published by Ministry of health in 2010 (China’s Health Statistics Yearbook, 2010). A unique pattern of tobacco consumption and dietary habits with high incidence of lung cancer was observed in Mizoram (Chaturvedi et al., 1998; Phukan et al., 2006; Malakar et al., 2014). Smoking of tobacco is undoubtly one of predominant risk factor for lung cancer (Tredaniel Abstract


Introduction
Lung cancer is the leading cause of cancer death worldwide with an annual mortality of 18.2% cancer death (Ferlay et al., 2010).In India, it continues to be important (D 'Souza et al., 2013) and the incidence of lung cancer is highest in Aizawl district of Mizoram among females with an age adjusted rate of 38.7 per 10 5 population (NCRP, 2010).Mizoram situated in between 92.15´ and 93.29´E longitude and 21.58´ to 24.35´N latitude; surrounded by Myanmar in east and Bangladesh in west.Mizoram have their ancestral origin in China; later which shares a similar mortality trend with United States according to China Health Statistical Yearbook, published by Ministry of health in 2010 (China's Health Statistics Yearbook, 2010).
A unique pattern of tobacco consumption and dietary habits with high incidence of lung cancer was observed in Mizoram (Chaturvedi et al., 1998;Phukan et al., 2006;Malakar et al., 2014).Smoking of tobacco is undoubtly one of predominant risk factor for lung cancer (Tredaniel et al., 1994;Pisani et al., 2006;Sellappa et al., 2009).Apart from smoking exposure, indoor smoke from household activities (Zhong et al., 1999;Reid et al., 2012;Yin et al., 2013) and dietary habits (Marchand et al., 2002;Shen et al., 2008) were also suspected to have casual relationship.Exposure to cooking oil fumes (COF) and wood smoke are known to contain considerable amount of carcinogens such as benzo[a]pyrene, 2, 4-decadienal, carbon monoxide, carbon dioxide and other polycyclic aromatic hydrocarbons (PAH) (Yang et al., 2000;Lissowska et al., 2005).Exposure to COF may potentially play a role in occurrence of lung cancer.Mizo cooking often involves stir-frying and deep-frying to a high temperature that produces ample amount of COF containing several harmful compounds (Straif et al., 2006).Incense burning, a traditional practice in Mizo household also produces a considerable amount of PAH, carbonyls and benzene (Koo et al., 1995).
Glutathione S-transferases (GST) are a family of phase II enzymes, metabolize carcinogenic compounds that are found in tobacco smoke and environmental carcinogens (Jourenkova-Mironova et al., 1998;Wenzlaff et al., 2005;Nosheen et al., 2010;Senthilkumar and Thirumurugan, 2012).Functional genetic polymorphisms of Phase II metabolic pathways genes GST also may account for differences in metabolic inactivation of dietary or other environmental carcinogens (Ye et al., 2006).Among these, the best known are deletion of GST mu 1 (GSTM1) or GST theta 1 (GSTT1) genes (null variant), which result in no functional activity (Kelsey et al., 1997).
The present study was a population based case control study design to investigate whether exposure from COF, house hold combustion and other indoor environmental sources plays a significant role in enhancing risk of lung cancer among women in Mizoram.We also wished to explore the join effect of these indoor environmental sources and whether the impact of these compounds and other dietary carcinogens is modified by GST polymorphisms.

Materials and Methods
The basic design of present study was an age and sex matched population based case-control study.The work was carried out at Regional Medical Research Centre (RMRC), Indian Council of Medical Research (ICMR), North East Region; India in collaboration with Civil Hospital, Aizawl, Mizoram.The study duration was from October 2009 to December 2012.All incident cases and controls subject willing to participate in the study were ethnic tribes of Mizo women.Histopathologically confirmed cases of no evidence of pulmonary inflammation or benign lung tumors and only lung as their primary site of cancer were included in the study.Cases too aged to be interviewed elaborately and who refused to be interviewed, were excluded from this study.Sources of data collection for the cases include Radiotherapy and Pathology Department of Civil hospital Aizawl, other private laboratory and nursing homes of Mizoram.Cancer free controls with age and sex matched were selected from healthy population of the state.
In order to reduce potential bias in our study none of the controls had consanguinity with the cases or had any non-communicable diseases.Information of dietary habits, socio-demographic characteristics and tobacco consumption were recorded in a structured pre designed questionnaire.Subjects who reported that they were regularly smoking during index year were defined as current smokers or current users, those who reported that they had stopped smoking the year before index year or before were defined as ex-smokers or ex-users, and those who reported that they never had smoked before or during index year were defined as never-smokers or never users (Phukan et al., 2005).A composite index, cooking disk years was constructed to quantify regular cooking habits based on frequency and duration of cooking, as previously described (Yu et al., 2006).One stir-frying dish year means cooking one stir fried dish with cooking oil daily for a year.The total cooking dish year was used to express amount of exposure to cooking oil emissions, which was also categorized into five groups with interval of 50 dish year.The time period for exposure of COF, heating sources for cooking, types of ventilation, position of kitchen in house, use of fume extractor and roasting of soda in kitchen was set for since last 25 years and all participants were asked to recall exposures for the above mention practices, 25 years before age of diagnosis for lung cancer cases as well as for controls group.In the dietary section, we calculated following items: smoked fish, dried fish, fresh fish, smoked meat, dried meat, fresh meat, intake of tea, soda (an alkali preparation), bamboo shoots, soybean, pickle, egg and fruits.Based on dietary style of Mizo population we divided consumption of these items into following groups: (i) Nil (ii) Occasionally (iii) Twice or more in a week.A total of 230 female lung cancer cases and 460 controls matched for age sex were enrolled in the study.Written informed consent was taken from all subjects for participation in the study in a protocol approved by institutional ethical committee of Regional Medical Research Centre, North East Region (Indian Council of Medical Research).

Statistical analysis
Difference in consumption of dietary and cooking habits, distribution of socio-demographic characteristics and genotype frequencies between cases and controls were evaluated using the Chi Square (χ2) test.Estimates of lung cancer risk, imparted by genotypes, other covariates as tobacco smoking and other dietary factors were determined by deriving odds ratio (OR) and its corresponding 95% confidence interval (95% CI) using multivariable logistic regression after adjusting for age, education and occupation.For all the tests a two sided Asian Pacific Journal of Cancer Prevention, Vol 15, 2014 3255 DOI:http://dx.doi.org/10.7314/APJCP.2014.15.7.3253 Household Exposure, Diet and GST M1, T1 Polymorphisms Relate to Lung Cancer among Women in Mizoram, India.p≤0.05 was considered statistically significant.All the statistical analysis were done using SPSS version 17.0.

Results
The distributions of socio-demographic characteristics of cases and controls subject were analyzed in Table 1.There was no statistically significant difference between cases and controls in terms of mean age (p=0.711) of study subjects.However, we observed a significant difference in terms of occupation (Housewife, p= 0.047).Furthermore, in terms of indoor air environmental sources; significant association was observed: exposure of cooking oil fumes (OR=2.56,CI=1.39-4.74),wood as heating source for cooking (OR=1.50,CI=1.01-2.22),kitchen inside living room (OR=1.79,CI=1.25-2.55),improper ventilated house (OR=2.10,CI=1.36-3.22)and roasting of soda in kitchen (OR=2.30,CI=1.38-3.80).Analysis of effects of cooking dish-years, both cases and controls were grouped by five dish year levels (with an interval of 50) (Table 2).A dose response gradient with total cooking dish years was displayed, with an OR >3 in second highest dish years and >4 in highest dish years.
We also observed that, with increasing level of sources for smoke inside the house, risk for lung cancer tend to  increase (OR= 4.45,).An enhanced risk was observed for stir frying after smoke emission with these sources (OR=5.30,CI=1.40-20.12).Although indoor smoking was not significantly associated with increase of lung cancer, risk was even more with the collegial effect of indoor smoking with these sources (OR=6.42,CI=1.15-

Discussion
We examined the effect of cooking exposure and dietary habits on lung cancer risk among women in Mizoram, and their modification by indoor tobacco smoking and GST polymorphisms.In Mizo tradition, males always occupy a superior position in family, and their behavior is not restricted; while females are mostly housewife.Because of traditional culture of Mizo, responsibility of cooking lies mostly within women.Meanwhile, we observed a strong association between exposure to cooking oil fumes from frying and lung cancer risk.This was indicated by the result of total cooking dish years which depicts a significant contribution and there was a clear exposure response trend observed in the study subject.Mizo cooking often involves stir frying at high temperature, which produces ample amount of COF containing several carcinogens (Straif et al., 2006).Our study was consistent with the studies from China for significant association of COF and risk of lung cancer among women (Metayer et al., 2002).An experimental approach performed by Hung et al. (2007) reported that COF was capable of causing cellular destruction.Another studies from China also reported that women were at higher risk for lung cancer if they were exposed to COF emitted at high temperature, and risk were higher when fumes were not reduced by an extractor (Ko et al., 2000).Most household in Mizoram depends on combustion of biofuels, primarily wood for heating with no fume extractor, poor ventilation with roasting of soda inside houses.Soda, an alkaline preparation frequently used as food additives was significantly associated with increased risk of stomach cancer in Mizoram (Phukan et al., 2006).Perhaps roasting of soda inside houses, may enhance smokier environment forming complicated reaction with wood smoke and COF.These sources combined with foods being cooked with cooking oil and indoor tobacco smoke, often make quite smoky environment of several harmful carcinogens.Combustion of wood also contains many hazardous pollutant including benzo(a)pyrene, formaldehyde, benzene and other PAH (Lissowska et al., 2005).The association of wood use as burning fuel and lung cancer risk as observed in our study was important, because IARC (International Agency for Research on Cancer) classified biomass use (primarily wood) as a Group 2A carcinogen due to limited epidemiological evidence (IARC, 2010).Significantly, fewer studies of lung cancer in association with combustion of wood were reported.Association of combustion of wood and lung cancer in our study will support conclusion that indoor air pollution from wood combustion was a possible carcinogens (Group 2A).
We also observed a significant association for consumption of smoked fish and smoked meat with lung cancer among women in Mizoram.Increase risk of lung cancer with consumption of smoked meat and fish in present study may be due to PAH and other nitrosamines present in smoked fish (Stolyhwo and Sikorski, 2005;Viksna et al., 2008).Precise type of cooking also exert an influence, since heterocyclic amines are formed when meat was cooked at high temperatures, particularly when it is fried at high temperatures; PAH are formed during grilling leading to PAH -DNA adducts, which are known to cause mutation, ultimately cancer (Rithman et al., 1990).However lack of association of fresh and dried fish as well as meat with lung cancer in the present study warrants further studies.Present study also indicates that consumption of both egg and bamboo shoot are related to significant risk reduction for lung cancer among women in Mizoram.It is known fact that eggs are rich source of n-6 PUFA (Lewis et al., 2000) that contributes to increase risk of breast cancer (Murff et al., 2011).Thus, the same role of association of egg intake and protective effect against lung cancer has to be studied in depth to make a conclusive statement.The protective effect of bamboo shoot on lung cancer risk may be due to presence of physterol and dietary polyphenolic compounds in bamboo shoot (Gong et al., 2010;Chongtham et al., 2011).Polyphenolic compounds inhibit promotion of carcinogenesis by inhibiting oxygen radical-forming enzymes that contributes to DNA synthesis, inhibiting DNA topoisomerase II activity which further contributes to proliferative signal transduction (Galati and Brien, 2004).
As expected, our results confirmed well-established association between smoking and lung cancer.Although, result was not statistically significant for indoor smoking; risk tends to increase with collegial effect of indoor smoking with the sources of indoor environment.A distinctive pattern of tobacco consumption was observed in Mizoram.Unlike other smokeless tobacco product, a unique tobacco-infused water (locally known as Tuibur) and locally made cigarette (Meiziol) is predominant in Mizoram.Both Tuibur and Meiziol demonstrated a high risk for lung cancer development among women in Mizoram.
Present study is the first study to describe significant association of GSTM1 null genotype with lung cancer risk among women in Mizoram (Table 6).For studying the role of gene-diet interaction that might modify susceptibility of lung cancers, we also made an attempt at evaluating potential interactions of GST polymorphisms with known dietary risk factors (Table 7).Significant risk estimates were observed for interaction of GSTM1 null genotype with soda, smoked fish and smoked meat.Risk estimates were also observed for interaction of GSTT1 null genotype with soda, smoked and dried meat.However an inverse correlation was observed when GSTT1 homozygous genotype interacts with consumption of fresh fish.Present study indicates carcinogen-specific modulation of cancer risk by GST genotype and was contrary to the classical roles assumed by GST genes.Understanding the mechanism of GST polymorphisms and dietary fish meat interaction will require further studies; though we can speculate that individual with GSTM1 null genotype would suffer an increase level of toxic metabolites while an increase level of PAH were found in smoked fish and smoked meat (Stolyhwo and Sikorski, 2005;Miculis et al., 2011).To our knowledge, this is the first study that has analyzed interaction between dietary fish and meat consumption with GST polymorphisms in lung cancer risk among women.Our current knowledge of diet related carcinogenesis is still limited, so individual variability in potential relationship between dietary constituents and cancer risk markers merits further investigation.
Our study has several strengths and findings.It is a population based case control study with a high participation rates and was carried only among the ethnic Asian Pacific Journal of Cancer Prevention, Vol 15, 2014 3259 DOI:http://dx.doi.org/10.7314/APJCP.2014.15.7.3253 Household Exposure, Diet and GST M1, T1 Polymorphisms Relate to Lung Cancer among Women in Mizoram, India.tribes of Mizo women.Newly diagnosed cases were included in the present study.Newly diagnosed cases were less likely to change their habits and exposure.Intense care has been taken while collecting epidemiological data to minimize recall bias and observer bias.Major strength of our studies was that great efforts were made to quantify cooking emission exposure on the frequency and duration of cooking exposure as expressed in total cooking dish year.In addition to these, dietary and GST gene polymorphisms were also taken into account to find out for possible association with lung cancer.Our studies were the first to find out protective effect of bamboo shoot with lung cancer.Exposure to Mizo cooking style of cooking emission may pose serious health impact.Our results underscore the necessity and priority of formulating an effective strategy to modify and improve cooking practice in Mizo populations.

Table 1 . Socio-Demographic Information and Risk of Lung Cancer
*Significant; **Adjusted OR were estimated by adjusting age in multiple logistic regression model

Table 2 . Total Cooking Dish Years and Risk of Lung Cancer
b Case (230) n (%) b Crude OR (95% CI) p value Adjusted OR (95% CI) a p value *Significant; a Adjusted OR were estimated by adjusting age, education and occupational status in multiple logistic regression model; b Calculated only for those cases that use these peculiar cooking methods (stir frying and deep frying); hence column-wise percentage does not add up to 100%

Table 4 . Risk Factors According to Tobacco Consumption Habits
Significant; **Adjusted OR were estimated by adjusting age education and occupational status in multiple logistic regression model *
*Significant; **Adjusted OR were estimated by adjusting age, education and occupational status in multiple logistic regression model Asian Pacific Journal of Cancer Prevention, Vol 15, 2014 3257 DOI