Predicting radical prostatectomy outcome: Cell cycle progression (CCP) score compared with primary Gleason grade among men with clinical Gleason less than 7 who are upgraded to Gleason 7.

Abstract

13 Background: Improved prognostic markers for prostate cancer are an important part of addressing the issue of over- and under-treatment in prostate cancer. Gleason score (GS), prostate-specific antigen and clinical stage work well for population risk assessment but lack precision for individual patients. Concern over undergrading on biopsy is felt to be a major barrier to increasing use of active surveillance. Molecular analysis can further refine risk assessment as demonstrated by the cell cycle progression (CCP) score (Prolaris) which has been shown to predict prostate cancer aggressiveness in eight separate cohorts. In these studies, CCP scores typically ranged from −2 to +3 with each one−unit increase in CCP score corresponding to approximately a doubling of the risk of the studied event (recurrence or death from prostate cancer). In the present investigation, we sought to assess how upgrading compared with CCP in predicting failure after radical prostatectomy (RP). Methods: In this study, we examined men from three cohorts (UCSF, Martini Clinic, and Durham VAMC) with clinical GS (cGS) less than 7 who had pathologic GS (pGS) 3+4 or 4+3. The CCP assay was performed using RP specimen from UCSF and biopsy material from the other cohorts. We compared the rates of biochemical recurrence (BCR) as predicted by pGS, CAPRA, or CCP score alone versus CAPRA combined with CCP score using a Cox proportional hazards model stratified by cohort. Results: Among the 230 men with cGS less than 7 and pGS equal to 7 included in this analysis, 207 had pGS 3+4 and 57 had BCR; 23 had pGS 4+3; and eight had BCR. There was no difference in BCR based on GS (p=0.8) or CAPRA (p=0.4). In contrast, CCP score alone and a pre-defined score combining the CCP score and CAPRA were prognostic of BCR (HR=1.82 [95 % CI 1.32-2.52; p<0.001]; HR=1.84 [95 % CI 1.10-3.05; p=0.021] respectively). Conclusions: These data indicate that the risk of BCR is indistinguishable in men with cGS less than 7 who have pGS 3+4 or 4+3. However, the CCP assay does provide stratification of this risk. Tests that can predict BCR on biopsy will aid in initial therapeutic decision making.