ABSTRACT
PURPOSE: General and central adiposity are associated with the risk of developing prostate cancer (PCa), but the role of these exposures on PCa survival among men of African ancestry are less studied. This study aimed to investigate the association of anthropometry at diagnosis with all-cause and PCa-specific mortality and evaluate whether androgen deprivation therapy (ADT) modulated this risk. METHODS: Associations between body mass index (BMI), waist circumference (WC), and waist-to-hip ratio (WHR) at diagnosis and mortality were examined in 242 men with newly diagnosed PCa enrolled between 2005 and 2007 and re-evaluated 10.9 years later. Multi-variable Cox proportional hazard models were used to examine associations of body size variables (using standard WHO cut-points and as continuous variables) with mortality, adjusted for sociodemographic characteristics, Gleason score, smoking, diabetes, primary treatment, and ADT therapy. RESULTS: A total of 139 deaths (all-cause mortality 6.98/100 person-years) occurred (PCa-specific deaths, 56; other causes, 66; causes unknown, 17). In multi-variable analysis BMI, WC and WHR categories at diagnosis were not associated with all-cause mortality even after adjusting for ADT. While WHR (but not BMI or WC) when included as a continuous variable predicted lower PCa-specific mortality (multi-variable adjusted WHR per 0.1 difference: HR, 0.50; 95%CI 0.28, 0.93), the effect disappeared with ADT covariance and excluding deaths within the first 2 years. CONCLUSION: Our study suggests that central adiposity as measured by WHR may improve long-term survival among men of African ancestry. Metabolic studies to understand the mechanism for this association are needed.