Performance of biopsy factors in predicting unfavorable disease in patients eligible for active surveillance according to the PRIAS criteria

Abstract

To assess the added value of biopsy factors, like maximum cancer length in a core (MCL), cumulative cancer length (CCL), cumulative length of positive cores (CLPC), percentage of cancer involvement in positive cores (CIPC) and the Prostate Cancer Research International: Active Surveillance (PRIAS) criteria in patients who underwent radical prostatectomy (RP) but eligible for active surveillance (AS). From January 2002 to December 2007, 750 consecutive subjects underwent RP. We identified 147 (19.07%) patients who were eligible for AS based on PRIAS criteria: clinical stage T1c or T2, PSA level of ⩽ 10 ng ml(-1), Gleason score ⩽ 6, PSA-D of <0.2 ng ml(-2) and one or two positive biopsy cores. We calculated the diagnostic accuracy of biopsy factors in determining pathological confirmed unfavorable disease. Decision curve analysis (DCA) were performed. Of all subjects, 95 patients (66.43%) had favorable whereas 48 had (33.57%) unfavorable disease. On multivariate analyses, the inclusion of MCL, CCL, CLPC and CIPC significantly increased the accuracy of the base multivariate model in predicting unfavorable disease. The gain in predictive accuracy for MCL in a core, CCL, CLPC and CIPC ranged from 13 to 27%. The DCA shows that adding MCL, CCL, CLPC and CIPC resulted in a greater net benefit when the probability of ranges between 15 and 50%. The models can be applied at the cost of missing not more than 16.83% of unfavorable disease. Our findings suggested that the addition of these biopsy factors to PRIAS criteria has the potential to significantly increase the ability to detect unfavorable disease.