Predictive value of the 2014 International Society of Urological Pathology grading system for prostate cancer in patients undergoing radical prostatectomy with long-term follow-up.

Abstract

To assess the relationship between the International Society of Urological Pathology (ISUP) 2014 grading system, biochemical recurrence (BCR) and clinical recurrence (CLR) after radical prostatectomy (RP), to determine whether the 2014 ISUP grading system is a better predictor of survival compared with the previous Gleason scoring systems, and to investigate whether incorporation of the tertiary pattern/grade into the ISUP scoring system significantly improves its efficacy. A total of 635 RP cases (1991-1999) were identified from a database at a single institution. A histopathology review was performed to re-grade the cases as per the ISUP 2014 grading system. All relevant clinicopathological data and clinical follow-up (median [range] 15.25 [0.3-26] years) were obtained. Log-rank, Kaplan-Meier, Cox regression and Harrell's concordance c-indices analyses were performed. At a median follow-up of 15years, 276 patients (44%) had BCR and 41 (7%) had CLR. Grade Groups 1, 2, 3, 4 and 5 were seen in 112 (18%), 307 (48%), 129 (20%), 33 (5%) and 54 patients (9%), respectively: 337 (53%) were upgraded, while 70 (11%) were downgraded compared with the 1992 Gleason system. Grade Group (hazard ratio [HR] 4.9; P < 0.001) and preoperative prostate-specific antigen (PSA) level (HR 1.4; P < 0.001) were independent predictors of BCR. Only Grade Group 5 (HR 12.3; P = 0.02), preoperative PSA (HR 1.6; P < 0.001), stage pT3b (HR 3.1; P = 0.03) and pT4 (HR 12.4; P < 0.001) independently predicted CLR. Harrell's c-indices showed that the 2014 ISUP grading system was a significantly better predictor of BCR and CLR as well as prostate cancer-specific death, compared with the 2005 ISUP modified Gleason system. The replacement of the secondary pattern by the tertiary pattern did not alter the prognostic efficacy of the ISUP 2014 grading system. The ISUP 2014 grading system is a significant independent predictor of both BCR and CLR, outperforming the 2005 ISUP modified Gleason system. This classification system has the potential to influence clinical decision-making after RP.