We investigated the association of bone mineral density (BMD) measures with prostate cancer (PCa) risk in older men enrolled in the Osteoporotic Fractures in Men Study. We hypothesized that men with higher BMD, a marker of exposure to endogenous sex hormones, would have an increased incidence of PCa. The cohort included 4,597 men (89% White, 65 years or older) with no prior history of PCa. Baseline total body, total hip, and spine BMD were assessed using dual energy X-ray absorptiometry. Prostate cancer was confirmed by review of medical records. Cox regression was used to assess the association of BMD quartiles with incident PCa, adjusting for age, body mass index, and other covariates. During an average follow-up of 5.2years, 5.6% (n = 255) of men developed PCa. Total body BMD was inversely associated with incident PCa, with a significant trend for decreasing PCa risk with increasing BMD quartiles (Ptrend = 0.007). Men in the highest total body BMD quartile had a 41% reduced risk for PCa (hazard ratio, 0.59; 95% confidence interval, 0.40-0.86), compared with men in the lowest quartile. Total hip and spine BMD did not exhibit significant relationships with PCa. Associations of BMDmeasures differed for low-grade (Gleason sum, 2-6) versus high-grade tumors (Gleason sum, ≥7). Significant inverse relationships with high-grade disease were noted at the total body and total hip sites. However, no associations were observed with low-grade disease. Our results provide support for an inverse association between BMD and PCa risk. Possible pathophyisological mechanisms linking BMD and PCa should be elucidated.