BODY COMPOSITION, FAT DISTRIBUTION, AND LUNG CANCER INCIDENCE: ATHEROSCLEROSIS RISK IN COMMUNITIES STUDY

In previous studies, the association between body figure and lung cancer incidence was analyzed using limited anthropometric indicators or among cohorts consisting of single race/ethnicity. In the Atherosclerosis Risk in Communities (ARIC) study, we estimated the predicted fat mass and lean mass and prospectively evaluated the association between predicted fat mass, predicted lean mass, body mass index (BMI), waist circumference and lung cancer incidence among 14,653 White and Black participants who were cancer-free at baseline visit (1989-1991) and followed for lung cancer incidence (N=689) through 2015. In our final Cox proportional hazard models, exposure measurements were categorized into sex-specific quartiles, multiple covariates (age, race, smoking history, alcohol consumption, and physical activity) were adjusted, predicted fat and lean mass were mutually adjusted, and BMI and waist circumference were mutually adjusted. After mutual adjustment, predicted fat mass was not significantly associated lung cancer incidence (HR: 0.88, 95% CI: 0.61, 1.26; p-trend = 0.25); predicted lean mass and BMI were both inversely associated with lung cancer incidence (HR: 0.59, 95% CI: 0.40, 0.87; p- trend = 0.02 and HR: 0.37, 95% CI: 0.26, 0.54; p-trend