Gleason scores provide more accurate prognostic information than grade groups

Prostate cancer grade is currently often reported both by Gleason scores and by grouping of the scores into five so-called International Society of Urological Pathology (ISUP) grades (also known as grade groups). Using population-based registry data from 172,112 men diagnosed with prostate cancer on needle biopsy, we recently investigated the outcome of Gleason score 8-10 prostate cancer with death due to prostate cancer and death from any cause as endpoints. There was a prognostic heterogeneity between Gleason scores 3+5, 4+4 and 5+3 (ISUP grade 4) and between Gleason scores 4+5, 5+4 and 5+5 (ISUP grade 5). This heterogeneity was lost when the grades collapsed into ISUP grades 4 and 5, respectively. On the other hand, there was also a prognostic overlap between these ISUP grades. The outcome of Gleason score 5+3 and 4+5 cancers was very similar. The prostate-specific mortality of Gleason scores 5+3 and 4+5 was 0.32 (95% confidence interval 0.27-0.36) and 0.30 (0.29-0.31), respectively, after 5 years and 0.44 (0.39-0.49) and 0.45 (0.44-0.46), respectively, after 10 years. The findings emphasise the importance of reporting the Gleason grades and scores for more accurate prognostic information of highly heterogeneous high-grade prostate cancers. It also questions the clinical value of the current recommendations of grouping of Gleason scores into ISUP grades or grade groups.