History of diabetes mellitus and subsequent prostate cancer risk in the NIH-AARP Diet and Health Study.

OBJECTIVE: A history of diabetes has been hypothesized to decrease prostate cancer risk, but studies have not always considered confounding or effect modification by dietary or lifestyle factors. METHODS: We examined the association between diabetes history and subsequent prostate cancer risk in 328,316 men enrolled in the NIH-AARP Diet and Health Study. Participants were ages 50-71 years and without a prostate cancer diagnosis at baseline in 1995. A prior history of physician-diagnosed diabetes was assessed using a self-administered mailed questionnaire. Cases of prostate cancer were ascertained by matching the cohort to state cancer registries. Multivariable relative risks (RR) and 95% confidence intervals (CI) of prostate cancer were estimated using Cox regression. RESULTS: During 5 years and 1,432,676 person-years of follow-up, 11,193 prostate cancer cases were ascertained. The age-adjusted and multivariable RRs of prostate cancer comparing men with diabetes to those without diabetes were 0.69 (95% CI=0.64, 0.74) and 0.71 (95% CI=0.66, 0.76), respectively, indicating no important confounding. The inverse association between diabetes and prostate cancer was particularly strong among men in the highest category of routine physical activity at work or home (RR=0.41; 95% CI=0.23, 0.74; p value for test of interaction = 0.03). Findings were similar for organ-confined and advanced prostate cancer. CONCLUSION: Results from this large prospective study suggest that a history of diabetes is associated with a decreased risk of prostate cancer. The relationship strengthened with high levels of routine physical activity. Because increased physical activity is associated with lower circulating levels of insulin and testosterone, our findings support a role of hypoinsulinemia and low androgenicity linking diabetes to decreased prostate cancer risk.