Effect of diabetes mellitus on high-grade prostate cancer detection among Japanese obese patients with prostate-specific antigen less than 10 ng/mL.

OBJECTIVE: To investigate the association of diabetes mellitus (DM) with prostate cancer (PCa) risk and grade among Japanese patients undergoing extended biopsy and to investigate how obesity modifies these relationships.
METHODS: We retrospectively evaluated the data from 2038 patients with a prostate-specific antigen (PSA) level <10 ng/mL undergoing initial extended biopsy at our institutions. The DM history was determined by self-report and medication use. Multivariate analyses of DM for PCa risk and grade were done using logistic regression. Moreover, we examined whether these associations were modified by the body mass index using subgroup analyses (nonobese <25 kg/m(2) or obese ≥25 kg/m(2)) and interaction tests. Cancer grade was classified according to the Gleason score (GS): low-grade (GS ≤6), intermediate-grade (GS 7), and high-grade (GS 8-10).
RESULTS: Of 2038 patients, obesity and DM was observed in 606 (30%) and 213 (11%), respectively. Also, 836 patients (41%) had positive biopsy findings. On multivariate analysis, we found no significant association of DM with the risk of overall PCa (P = .106) or the risk of low-grade (P = .735), intermediate-grade (P = .119), or high-grade (P = .110) disease. When stratified by obesity, the relative risk (RR) of PCa detection for diabetic men apparently increased with higher cancer grade (low grade, RR = 1.19, P = .71; intermediate grade, RR = 2.01, P = .099; high-grade, RR = 4.03, P = .025). However, in the nonobese men, no association was noted between DM and PCa risk, irrespective of grade. Obesity modified the effect of DM on high-grade disease risk with a trend (P for interaction = .087).
CONCLUSION: DM was associated with more aggressive PCa detection among Japanese obese patients with gray-zone PSA levels undergoing extended biopsy.