Heterogeneous oncologic outcomes according to surgical pathology in high-risk prostate cancer: implications for better risk stratification and preoperative prediction of oncologic outcomes.

ABSTRACT

PURPOSE: To evaluate the better risk stratification based on surgical pathology, and to predict oncologic outcomes after radical prostatectomy (RP) with a better scoring system in high-risk prostate cancer (PCa) patients. METHODS: We evaluated high-risk PCa patients (PSA >20 ng/ml, ≥cT3a, or Gleason score 8-10) who underwent RP between 2007 and 2013 at our institute. We classified patients into three groups according to their pathologic outcomes: favorable (pT2, Gleason score ≤7, and node negative), intermediate (specimen-confined disease (pT2-3a, node negative PCa with negative surgical margins) but not in the favorable group), and unfavorable (the remaining patients). We developed a risk stratification scoring system to predict prognostic outcomes after RP and validated our scoring system to estimate its predictive accuracy. RESULTS: Among a total of 356 patients, 95 (26.7%), 115 (32.3%), and 146 (41%) were in the favorable, intermediate, and unfavorable prognostic groups, respectively. The 5-year biochemical recurrence-free survival rates of the patients in each group were 87.8, 64.6, and 41.4%, respectively. We developed a scoring system based on preoperative PSA, clinical stage, percentage of tumor positive core, and percentage of cores with a Gleason score 8-10. This demonstrated internally and externally validated concordance indices of 0.733 and 0.772, respectively. CONCLUSIONS: Using our scoring system, we can predict which patients with high-risk PCa would benefit more from RP. Thus, this system can be used in patient counseling to determine an optimal treatment strategy for high-risk PCa.