Risk Factors for Lung Cancer in the Pakistani Population

Lung cancer is one of the most prevalent malignancies in the world (Wahab et al., 2007; Tas et al., 2008). Incidence and mortality rates due to lung cancer are continuing to increase in developing countries like Pakistan (Lam, 2005), which is contrary to the situation in many developed countries, where it is showing declining trend (Kwong et al., 2005; Wahab at al., 2007). Genetic, non-genetic determinants and late detection are important reasons for high mortality rates due to lung cancer (Ahmed at al., 2013). In Pakistan, it is most frequent malignancy in males and incident rate is also increasing in females (Hussain et al., 2009). Lung cancer has been well co-related with smoking but high frequency of lung cancers cases (25%) are nonsmokers (Sun et al., 2007). Development of malignancy of lung is a multifactorial interacting process. Many suspect


Introduction
Lung cancer is one of the most prevalent malignancies in the world (Wahab et al., 2007;Tas et al., 2008).Incidence and mortality rates due to lung cancer are continuing to increase in developing countries like Pakistan (Lam, 2005), which is contrary to the situation in many developed countries, where it is showing declining trend (Kwong et al., 2005;Wahab at al., 2007).Genetic, non-genetic determinants and late detection are important reasons for high mortality rates due to lung cancer (Ahmed at al., 2013).
In Pakistan, it is most frequent malignancy in males and incident rate is also increasing in females (Hussain et al., 2009).Lung cancer has been well co-related with smoking but high frequency of lung cancers cases (25%) are nonsmokers (Sun et al., 2007).Development of malignancy of lung is a multifactorial interacting process.Many suspect

Risk Factors for Lung Cancer in the Pakistani Population
Muhammad Luqman 1,2 *, Muhammad Mohsin Javed 3 , Shakeela Daud 4 , Nafeesa Raheem 5 , Jamil Ahmad 6 , Amin-Ul-Haq Khan 1 factors including life style, social and also filial factors are contributory causes for development of lung cancer in nonsmokers.Humans are exposed to organic and inorganic pollutant through environmental and occupational sources.Epidemiological evidences shows link between exposure to occupational and non-occupational pollutants and risk of cancer (Bruske et al., 2000).The main occupational exposures occur in workers who are engaged in smelting and refining of metals, production of pesticides, pigments, dyes, glass, semiconductors, wood/cotton products and various pharmaceutical substances (Ferreccio et al., 2000).Non-occupational exposures mostly occur due to outdoor air pollution including residence near major industrial emission sources, asbestos, in-door air pollution, arsenic in drinking water, chlorinated by-products in drinking water, dioxins and electromagnetic fields (IARC, 1994;Ferreccio et al., 2000).
Many studies has have shown that environmental and genetic factors interactively play their role in development of lung cancer (Bailey, 2004).Lung cancer susceptibility genes have also been identified (Bailey, 2004), making certain individuals more susceptible to lung cancer over exposure to environmental factors than others.Smoking trend is increasing in Pakistan.Tobacco is being used in the form of high tar cigarettes, hukka, cigar, shisha, biri and pan in Pakistan (Alam, 1998).Growing incidence of lung cancer in non-smokers suggests that population is bare to several types of other environmental risk factors like lifestyle, dietary and occupational exposures.No epidemiological study to evaluate these common risk factors for lung cancer in Pakistani population has been reported.The current study is preliminary effort to evaluate these determinants for lung cancer in this country.

Materials and Methods
In this case-control study, 400 cases (histologically confirmed lung cancer patients) from all over the Pakistan were enrolled from various public and private sector hospitals and health centers during 2010 to 2013.Subjects were briefed about the study and their written consent was taken on consent form.Agreed subjects were interviewed and data was recorded in the pre-designed questionnaire.The questionnaire body was composed of questions regarding age, sex, education, occupational exposure, tobacco smoking/chewing habits, alcoholism, physical activity, geographical location, family history, chronic respiratory diseases and dietary habits.Total 800 Hospital controls were also enrolled in the study after their written consent.Controls were free of all types of cancers and any chronic respiratory ailments.Odds ratio, its 95% confidence interval, X 2 value and p values were calculated by using EpiTools Epidemiological Calculator (Sergeant 2014).Doze response analysis was performed for suitable risk factors to find out chi-square value for linear tren d and p value for one degree of freedom by using

Socio-demographic features of the participants (Table 1)
Out of 400 cases, 282 were male and 118 were female.Among controls 589 were male and 211 were female.Study participants were belonging to all five provinces of Pakistan and Azad Jammu and Kashmir (AJK).Maximum numbers of cases were collected from Sindh (132) and then Punjab (119), KPK (71), Balochistan (55) and AJK (22).It was our effort to collect controls from all these provinces and AJK with similar proportion to cases.Religious affiliations of the cases were very much comparable to their proportion in the population.Majority of the cases and controls were literate, but overall controls were having more literacy rate (81.25) than cases (76.25).High frequency of lung cancer cases (89.50%) was belonging to the age group of 50-70 years.9.25% of the cases were belonging to age group of >70 years and only 1.25% of the cases were less than 50 years of age.Most of the participants both in cases and controls were married (almost 98%).Similar pattern of gender, race and  positive association with lung cancer with varying degree of strength (Table 2).However some variables have shown preventive association and some others have shown very weak or no association.

Causative factors
Tobacco smoking has shown most strong association with lung cancer (OR=9.4,p≤0.0001) in Pakistani population, which means that smokers in Pakistan are almost 9-10 times at higher risk lung cancer.This outcome is analogous to the previous results of global and regional research on lung cancer (IARC, 1994;Stellman et al., 2001;Hussain et al., 2009;Ganesh et al., 2011., Matteis, 2013).But the current value of OR is among the highest values observed in any segment of population in Asia.High values of association with smoking is depicting factual picture of increasing lung cancer in Pakistan, where high percentage of population is smokers and smoking habit is on the rise (Alam, 1998).It is worth mentioning that Pakistani cigarettes have high tar contents than other countries in the world (Alam, 1998).
Other variable which have shown strong association with lung cancer was pesticide exposure (OR=5.0,p≤0.0001).Status of pesticide exposure as causative factor is not easy to conclude from the past studies (Pesatori et al., 1994;Menvielle et al., 2003;Michael et al., 2004;Michael et al., 2013).In the studies carried out on Caucasian and Asian populations, some of the results presented no association while others have shown positive association.Current high value for this variable strengthens the results which have documented this factor as causative one.Being an agricultural country 40% of rural population is directly related with agriculture profession in Pakistan.Pesticide use without protective measures is common practice.Similarly individuals involved in manufacturing and packing are 5 times at higher risk to get lung cancer.
Exposure to diesel exhaust has also shown a three times higher risk of lung cancer than unexposed (OR=3.1,p≤0.0001).Several studies have found that long-term, heavy exposure to diesel exhaust can cause lung cancer in lab animals (Mauderly et al., 1987) and it has also been documented as an established human carcinogen (IARC, 2012).Though many studies have found a modest link between diesel exhaust exposure and lung cancer, some have reported no link between them (Garshick et al., 1987;Steenland et al., 1990;1998;Attfield et al., 2012).The current study has shown a solid link between lung cancer and exposure to diesel exhaust at work place in Pakistani population.
Red meat consumers are three times at higher risk of lung cancer (OR=2.9,p≤0.0001).Some of previous studies have also reported this a causative variable (Alavanja et al., 2001;Lam et al., 2009;Yang et al., 2012;Gnagnarella et al., 2013), but OR values was much lower as compared to present study.Per capita Red meat consumption in Pakistan is higher than its neighboring countries like India and Afghanistan.Similarly significant association of chicken consumption with lung cancer has been observed (OR=2.8,p≤0.0001).Although some of the earlier studies had reported this variable as preventive factor for lung cancer (Lim et al., 2011), currently it is reported as causative factor in Asian populations.
Status of tea and coffee consumption as a risk factor of lung cancer is not clear.In many parts of the world tea and coffee consumption has been reported as causative factors (Baker et al., 2005;Tang et al., 2010), some others have reported them as preventive factors (Wu et al., 2006;Tand et al., 2009) and still some of the reports has shown no link between coffee and tea consumption with lung cancer (Baker et al., 2005).Current study has documented it as an important risk factor in Pakistani population, where per capita tea consumption is 1kg/year, which is among the highest values of the world.

Preventive factors
Use of vegetable (OR=0.3,95%CI=0.2-0.4),juices (OR=0.3,95%CI=0.3-0.4),fruits (OR=0.7,95%CI=0.5-0.9) and milk (OR=0.6,95%CI=0.5-0.8) has shown reduction in risk of lung cancer in Pakistani population.These results for vegetables and juices as preventive factors are in line with finding of many regional and global research (Wright et al., 2003;Mannisto et al., 2004) .Status of milk consumption as a preventive factor for lung cancer is debatable.Some earlier epidemiological studies have documented it as a causative factor and some other studies have recognized it as a preventive factor (Axelsson et al., 2002;Ahn et al., 2007;Davoodi et al., 2013).Current study strengthens the role of this suspect determinant as a preventive variable.

Dose-response relationships
Strong dose-response relationship has been observed for smoking (Χ 2 =333.8,p≤0.0000001), pesticide exposure (Χ 2 =50.9, p≤0.0000001) and exposure to diesel exhaust (Χ 2 =51.8, p≤0.0000001) in Pakistani population (Table 3).Strongly positive dose-response relationship has been observed for smoking with multifold increase in lung cancer risk with increase in exposure period.Current findings are analogous to the results reported from Indian population (Ganesh et al., 2011).In the present study, odds ratio for 1-8 year smoking period was observed to be 3.4, which increased to 5.1 and 12.1 when exposure period was doubled and tripled up to 16 and 24 years respectively.For the exposure history of more than 24 years, OR was recorded as 26.6.OR for pesticide exposure was recorded as 2.5, 4.1 and 5.6 for exposure periods of up to 8, 16 and 24 years respectively.Then for exposure history of more than 24 years, OR has increased to 10.7.Similarly increase in lung cancer risk with increase in exposure period has been observed for diesel exhaust.Risk increased from 0.5 to 13.5 times for diesel exhaust when exposure period increased from up to 8 years to more than 24 years with linear trend for intervening doses.Uniquely, risk of lung cancer has increases dramatically in Pakistani population for smoking, pesticides and diesel exhaust when exposure period increased from 24 years.
In conclusion, tobacco smoking, pesticide exposure, exposure to diesel exhaust, red meat and chicken consumption are leading lung cancer determinants in the Pakistani population.Other important variables were welding fumes, sedentary living, family history, wood dust, tea/coffee consumption, alcoholism and asbestos exposure.Lung cancer risk increases considerably as the period of exposure to certain variables like smoking, pesticide exposure, and diesel exhaust increases.Use of vegetable, juices, fruits and milk are preventive factors in Pakistani population.Healthy life style, active living with no smoking and alcoholism, using safety measures against pollutants at work place and consuming vegetables, fruits, milk and juices can reduce the risk of lung cancer risks in Pakistani population.

Table 3 . Exposure-Response Association Chi-Square for Linear Trend
*Extended Mantel-Haenszel chi square for linear trend