Purpose: We investigated the value of percent Gleason grade 4/5 as a predictor of long-term outcome in men with prostate cancer diagnosed at transurethral resection who received deferred treatment.
Materials and methods: A series of 305 men with prostate cancer diagnosed at transurethral resection from 1975 to 1990 who had subsequent expectancy was analyzed. Mean patient age at diagnosis was 74 years (range 52 to 95). Slides were reviewed, and the Gleason score, percent Gleason grade 4/5 and modified Gleason score (the sum of the dominant and worst grades) were assessed.
Results: At followup 271 men (89%) had died, including 110 (36%) of prostate cancer. Gleason score, percent Gleason grade 4/5 and modified Gleason score were significant predictors of disease specific survival (p <0.001). Of all men 34% had tumors without any grade 4/5 pattern, of whom only 8% died of prostate cancer compared with 52% with any grade 4/5 pattern (p <0.001). Gleason score 6 tumors with focal grade 4 (less than 5%) had a worse prognosis than pure Gleason score 3 + 3 = 6 tumors (p = 0.008). There was nonsignificantly shorter survival for Gleason score 4 + 3 = 7 than for Gleason score 3 + 4 = 7 disease (p = 0.19). In Cox models including all possible pairs of Gleason score, percent Gleason grade 4/5 and modified Gleason score the percent Gleason grade 4/5 and modified Gleason score were stronger than Gleason score, although all 3 were independently significant prognosticators.
Conclusions: Percent Gleason grade 4/5, modified Gleason score and Gleason score are predictors of disease specific survival in patients with prostate cancer on deferred treatment. Our study indicates that any grade 4/5 pattern impairs the prognosis significantly.