Background: Epidemiologic studies have reported inconsistent associations of vitamin D and prostate cancer risk; however, few have adequately controlled for detection bias related to prostate-specific antigen (PSA) screening, and the results of many studies may be affected by occult prostate cancers among controls.

Methods: Data for this nested case–control analysis (n = 1,695 cases/1,682 controls) are from the Prostate Cancer Prevention Trial. Baseline serum was analyzed for 25-hydroxyvitamin D [25(OH)D]. The presence or absence of cancer was subsequently determined by prostate biopsy. Polytomous logistic regression models were used to estimate associations of 25(OH)D with risk of total, Gleason 2–6, Gleason 7, and Gleason 8–10 prostate cancer. Results are presented for placebo and finasteride arms separately and combined.

Results: There were no associations of serum 25(OH)D with total prostate cancer risk. For Gleason 2–6 cancers, results were inconsistent across treatment arms with a suggestion of increased risk in the placebo arm only; however, there was no dose–response relationship. For Gleason 8–10 prostate cancers, 25(OH)D concentrations were associated with a linear decrease in risk among combined treatment arms [quartile 4 vs. 1: OR, 0.55; 95% confidence interval (CI), 0.32–0.94; P_trend = 0.04]. These findings were somewhat stronger among men ≥65 versus 55–64 years at baseline (quartile 4 vs. 1: OR, 0.40; 95% CI, 0.18–0.88 vs. OR, 0.73; 95% CI, 0.35–1.52, respectively; P_interaction = 0.52).

Conclusions: Higher serum 25(OH)D may modestly increase risk of Gleason 2–6 disease and more substantially reduce risk of Gleason 8–10 prostate cancer.

Impact: Vitamin D may have different effects for different stages of prostate cancers. Cancer Epidemiol Biomarkers Prev; 23(8); 1484–93. ©2014 AACR.