<b><i>Objectives:</i></b> The Gleason score (GS) is the primary classification of clinical risk in prostate cancer (PCa). Here, we estimated the factors predictive of accordance of local and central pathologist-dependent GS and clinical risk classification in an increased number of cases. <b><i>Methods:</i></b> Between January 2009 and December 2013, 388 patients were diagnosed with PCa by 80 independent pathologists from local communities and were referred to our hospital. Validation of the GS with needle-core biopsy specimens was carried out by a single central pathologist, and clinical risk, according to the D'Amico risk classification, was determined. Discrepancies between the GS and risk classification, based on the GS estimated by the local or central pathologist, were reviewed, and predictive factors for accordance of clinical risk classification were estimated. <b><i>Results:</i></b> When pathological results were compared, 59.5% of cases were given concordant GSs by local and central pathologists. A significant discrepancy existed in the classification of intermediate risk (p < 0.0001). Multivariate analysis indicated that local pathologist-dependent GS7, lower prostate-specific antigen (≤10 ng/ml), and lower T stage (T1 or T2a) were significant predictive factors for discordance with the central pathologist-dependent risk classification. <b><i>Conclusion:</i></b> Review of bioptic GSs by central pathologists affected discrepancies in risk classification in patients with PCa.