Résumé : PURPOSE. We evaluated pathological features of prostate cancer detected on repeat prostate biopsy in men with a serum total prostate-specific antigen (PSA) level between 4 and 10 ng/ml who were diagnosed with benign prostatic tissue after an initial biopsy and compared them to those cancers detected on initial prostate biopsy. MATERIAL AND METHODS. In this prospective European prostate cancer detection study, 1,051 men with a total PSA level between 4 and 10 ng/ml underwent transrectal ultrasound (TRUS)-guided sextant biopsy and two additional transition zone biopsies. All subjects whose biopsy samples were negative for prostate cancer (CAP) underwent a repeat biopsy after 6 weeks. Those with clinically localized cancers underwent radical prostatectomy. Pathological and clinical features of patients diagnosed with cancer on either initial or repeat biopsy and clinically organ confined disease who agreed to undergo radical prostatectomy were compared. RESULTS. Initial biopsy was positive (CAP) in 231 of 1,051 enrolled subjects and negative (benign histology) in 820 subjects. Of these 820 subjects, CaP was detected in 10% (83/820) upon repeat biopsy. Of cancers detected on initial and repeat biopsy, 148/231 (64%) and 56/83 (67.5%) had clinically localized disease, respectively, and were offered radical prostatectomy. 10/148 (6.7%) and 3/56 (5.3%), respectively, opted for radiation therapy and thus, 138/148 (93.3%) and 53/56 (94.7%), respectively, underwent radical retropubic prostatectomy. There were statistically significant differences with respect to multifocality (P = 0.009) and cancer location (P < 0.001) with cancers on repeat biopsy showing a lower rate of multifocality and a more apico-dorsal location. In contrast, there were no differences with respect to stage (P=0.2), Gleason score (P=0.36), percentage Gleason grade 4/5 (P=0.1), serum PSA (P = 0.62), and patient age (P = 0.517). CONCLUSIONS. At least 10% of patients with negative prostatic biopsy results will be diagnosed with CaP on repeat biopsy. Despite differences in location and multifocality, pathological and biochemical features of cancers detected on initial and repeat biopsy are similar, suggesting similar biological behavior and thus advocating for a repeat prostate biopsy in case of a negative finding on initial biopsy. Cancers missed on initial biopsy and subsequently detected on repeat biopsy are located in a more apico-dorsal location. Repeat biopsies should thus be directed to this rather spared area in order to improve cancer detection rates. © 2001 Wiley-Liss, Inc.