Background: A 23% relative risk reduction (RRR) in prostate cancer (PCa) was shown in men receiving dutasteride in the 4-yr Reduction by Dutasteride of Prostate Cancer Events study, in whom biopsies were protocol dependent. Objective: Our aim was to explore PCa risk reduction in men with benign prostatic hyperplasia (BPH) from the Combination of Avodart and Tamsulosin (CombAT) study, in which biopsies were undertaken for cause. Design, Setting, and Participants: CombAT was a 4-yr randomized double-blind parallel group study in 4844 men ≥ 50 yr of age with clinically diagnosed moderate to severe BPH, International Prostate Symptom Score ≥ 12, prostate volume ≥ 30mL, and serum prostate-specific antigen (PSA) 1.5-10 ng/mL. Men underwent annual PSA measurement and digital rectal examination (DRE), and prostate biopsies were performed for cause. Intervention: All patients took tamsulosin 0.4mg/d, dutasteride 0.5mg/d, or a combination of both. Measurements: The primary end point was incidence of PCa. Secondary end points included postbaseline prostate biopsy rates and Gleason score of cancers. Results and Limitations: Dutasteride (alone or in combination with tamsulosin) was associated with a 40% RRR of PCa diagnosis compared with tamsulosin monotherapy (95% confidence interval, 16-57%; p=0.002) and a 40% reduction in the likelihood of biopsy. There were similar reductions in low- and high-grade Gleason score cancers. The biopsy rate in the groups receiving dutasteride trended toward a higher diagnostic yield (combination: 29%, dutasteride: 28%, tamsulosin: 24%). One limitation was the lack of a standardized approach to PCa diagnosis and grading. Conclusions: Dutasteride, alone or in combination with tamsulosin, significantly reduced the relative risk of PCa diagnosis in men with BPH undergoing annual DRE and PSA screening. Consistent with the increased usefulness of PSA for PCa detection, men receiving dutasteride had a numerically lower biopsy rate and higher yield of PCa on biopsy. Trial Registration: Clinicaltrials.gov identifier: NCT00090103 (http://www.clinicaltrials.gov/ct2/show/NCT00090103).