Type 2 diabetes is associated with reduced risk of prostate cancer and low prostate-specific antigen levels for uncertain reasons. Recently, two studies demonstrated the time course of diabetes with prostate-specific antigen levels, which would be explained by the chronic renal complication. Therefore, we conducted a retrospective study to determine whether diabetic nephropathy is associated with prostate-specific antigen levels. Eligible patients were men aged a parts per thousand yen40 years, with type 2 diabetes and a recorded prostate-specific antigen level. Patients with a prior history of prostate cancer or prostatectomy or prostate-specific antigen level a parts per thousand yen10 ng/mL and patients with end-organ damage were excluded. Of the 247 patients included in the adjusted analysis, 51 (20.8 %) were diagnosed with diabetic nephropathy. A significant association of diabetic nephropathy with log-transformed prostate-specific antigen level was detected (Spearman's correlation coefficient -0.201, p = 0.003; adjusted for all other correlated variables) and fit into a linear regression model (B-coefficient -0.331, p = 0.003). Comparing diabetic nephropathy group with non-diabetic nephropathy group, the difference in log-transformed prostate-specific antigen levels was also significant (p = 0.002). The results indicated that patients with diabetic nephropathy have low prostate-specific antigen levels, suggesting fewer prostate cancers being detected or fewer incidences among this group.