Validation of the novel 5-group Gleason grading system: 3+5 disease risk may be overestimated.

Abstract

Introduction/ background: The international Society of Urological Pathology, suggested to change the classical Gleason grading system to 5 prognostic grade groups (1: 3+3, 2: 3+4, 3: 4+3, 4: Gleason score 8, and 5: Gleason score 9-10). We compared the predictive value for biochemical recurrence (BCR) after surgery of the classical grading system to the new risk group system. Method: The radical prostatectomy database from a single centre was used. BCR was defined as a post-operative PSA of >=0.2 ng / ml and rising. the classical 9-group system was compared to the novel 5-group system. Kaplan-Meier curves with Log-Rank test for comparisons between curves were used. P-values of <0.05 were considered statistically significant. Results: For biopsy, the classical grading showed overlap between different subgroups of Gleason 7 and 8 between 8 and 9 ( all p>0.05). The new risk group 2 overlapped with 3 (p=0.282), and 3 with 4 (p=0.161), other group comparisons were significantly different. For prostatectomy, the classical grading showed overlap between different Gleason 9 subgroups and 10 ( all p>0.05). The new risk group 3 overlapped with 4 (p=0.893), other group comparisons were significantly different. Within risk group 4, 3+5 patients (3-year BCR-free 65%) had more favourable (p<0.01) BCR than 4+4 patients (18%) and were even similar (p=0.524) to 3+4 (73% [Figure]. No corrections for margin or lymph node status were made. Conclusion: Condensing the classical 9-group Gleason system to the suggested 5-group system resulted in less overlap between BCR rates, suggesting increased prognostic stratification value between some higher risk disease patients may be lost. Most markedly, the risk of prostatectomy Gleason 3+5 patients (new risk group 4) may be overestimated, as the BCR resembles that of Gleason 3+4 (new risk group) patients.