Abstract
Purpose: We examined the association between BMI and lung cancer while overcoming bias due to preclinical weight loss and insufficient adjustment on tobacco-smoking. We used data from the ICARE study, a large French multicenter population-based case-control study. BMI was available at age 30 and two years before diagnosis. Complete lifetime smoking habits were combined in an innovative Comprehensive Smoking Index (CSI) that specifically models the effect of tobacco-smoking on the risk of lung cancer.
Methods: A total of 2683 incident lung cancer cases and 3385 controls were included. Unconditional logistic regression controlled for age, centre, CSI, occupational exposure to a carcinogenic agent, antecedents of chronic bronchitis, family history of lung cancer and level of education was used to estimate adjusted odds ratios (aORs) and 95% confidence intervals (CIs). BMI was categorized according to the OMS international classification of underweight (<18.5 kg/m2), overweight (25-29.9 kg/m2) and obesity (>30 kg/m2) in adults, with normal range (18.5-24.9 kg/m2) used as the referent group. Influence of change in weight between age 30 and two years before diagnosis was examined among participants whose BMI was normal at age 30. Analyses were stratified on smoking status (current, former and never-smokers) and sex.
Results: When compared with normal BMI two years before diagnosis, lung cancer aORs (95% CI) among male current smokers with a BMI of <18.5 kg/m2, 25-29.9 kg/m2, 30-32.4 kg/m2 and ≥32.5 kg/m2, were 3.7 (95%CI: 1.0 – 14.3), 0.6 (95%CI: 0.5 – 0.8), 0.7 (95%CI: 0.4 – 1.1) and 0.5 (95%CI: 0.3 – 0.8), respectively (p trend <0.001). This association was weaker when considering BMI at age 30. When comparing with male current smokers who gained more than 3 kg during adulthood, those with minimal change in weight (<3 kg) as well as those who lost more than 3 kg had a higher risk of lung cancer (OR=3.1 (95%CI: 2.2 – 4.3) and OR=5.5 (95%CI: 2.3 – 12.8) respectively). No association was found among former smokers and never-smokers. Results were similar among men and women and will be presented for both genders.
Conclusions: The inverse dose-dependent association between BMI and lung cancer risk in current smokers was more pronounced when considering BMI two years before diagnosis as opposed to BMI at age 30, which is consistent with a larger “exposure-frame” to a low BMI. BMI may modulate the carcinogenic effect of smoking because low BMI has been associated with increased levels of markers of oxidative stress and DNA adducts among smokers.
Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 102nd Annual Meeting of the American Association for Cancer Research; 2011 Apr 2-6; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2011;71(8 Suppl):Abstract nr 1920. doi:10.1158/1538-7445.AM2011-1920