INTRODUCTION: The therapeutic approach of prostate cancer depends mainly on pathological criteria obtained through prostate biopsy. The low accuracy of prostate biopsy for Gleason grade determination is well known but its accuracy for bilateral or multifocal tumor has not been evaluated. The goal of this study was to assess the concordance between prostate biopsy and whole prostate specimen obtained after radical prostatectomy especially for bilateral and/or multifocal tumor. METHODS: We retrospectively compared the pathological results of prostate biopsy cores to the prostate specimen in patients who underwent radical prostatectomy in our department between the 01/01/1999 and the 31/12/2008. The criteria analyzed were the Gleason score, tumor bilaterality or multifocality. The impact of the number of prostate biopsy cores was also analyzed. RESULTS: Two hundred and five complete histological records were studied. Regarding the Gleason score overall concordance was 55%. In 38%, prostate biopsies downgraded the Gleason score. This concordance decreased with tumor differentiation (90.6% for Gleason 6 vs. 31% for Gleason greater than 7). For the tumor bilaterality, 78% of cancers affected both lobes at the definitive specimen analysis while only 49% were bilateral at prostate biopsies, achieving a concordance of 61%. Multifocal disease was observed in 36% at definitive pathology analysis with low concordance with prostate biopsies (36%). The number of biopsies increased the concordance for the Gleason score (60 to 81% for Gleason 7 and from 28 to 50% for Gleason greater than 7) and tumor location (44 to 70%). CONCLUSION: Pathological criteria and tumor mapping obtained from prostate biopsies were not very reliable especially when the tumor was poorly differentiated. An increased number of prostate biopsy core improved the sensitivity and specificity for the Gleason score diagnostic and of the tumor mapping.