Improved decision making in intermediate-risk prostate cancer: a multicenter study on pathologic and oncologic outcomes after radical prostatectomy.

BACKGROUND: Prognoses for intermediate-risk prostate cancer (PCa) remain heterogeneous. Improved substratification could optimize treatment and monitoring strategies. The objective was to validate this subclassification in a radical prostatectomy (RP) series. METHODS: Between 2000 and 2011, 4038 patients who underwent RP for intermediate-risk PCa in seven French academic centers were included. Unfavorable intermediate-risk (UIR) PCa was defined as having a primary Gleason score of 4, ≥50% positive biopsy cores (PPBC), or more than one D'Amico intermediate-risk factor (i.e., cT2b, PSA 10-20, or Gleason score 7). Remaining PCa cases were classified as favorable. Main endpoints were pathologic results (pT stage, final Gleason score, surgical margin status), and oncologic outcomes were assessed according to PSA recurrence-free survival (PSA-RFS). Univariate and multivariate analyses were performed using the log-rank test and the Cox proportional hazards model. RESULTS: Median follow-up was 48 months (95% CI = [45-49]). Patients with UIR had worse PSA-RFS (68.17 vs. 81.98% at 4 years, HR = 1.97, 95% CI = [1.71; 2.27], p < 0.0001) compared to those with a favorable disease. The need for adjuvant therapy was significantly greater for UIR patients (43.5 vs. 29.2%, p < 0.0001). In multivariate analysis, primary Gleason score of 4 (HR = 1.81, 95% CI = [1.55; 2.12], p < 0.0001) and PPBC ≥ 50% (HR = 1.26, 95% CI = [1.02; 1.56], p = 0.0286) were significant preoperative predictors for worse PSA-RFS. CONCLUSIONS: This study highlights the heterogeneity of NCCN intermediate-risk patients and validates (in a large RP cohort) the previously proposed subclassification for this group. This classification can significantly predict both pathologic and oncologic outcomes. This easy-to-use stratification could help physicians' decision making. Prospective study and new tools as genomic tests and novel molecular-based approaches can improve this stratification in the future for patient counseling.