PD30-04 COMPARING ONCOLOGIC OUTCOMES OF DIFFERENT DEFINITIVE TREATMENTS FOR PROSTATE CANCER AFTER A PERIOD OF ACTIVE SURVEILLANCE

Abstract

INTRODUCTION AND OBJECTIVES: Most current guidelines recommend active surveillance (AS) as a valid treatment option for clinically localized, low-risk prostate cancer (PCa) to reduce overtreatment. However, a considerable proportion of AS candidates harbors unfavorable PCa at final pathology when surgically treated. The aim of this study was to identify predictors of unfavorable PCa in AS candidates that were treated surgically. METHODS: We relied on the 2010e2011 Surveillance Epidemiology and End Results (SEER) database. We identified 3,300 patients treated with radical prostatectomy who could have been selected for AS according to the UCSF criteria: prostate-specific antigen (PSA) <10 ng/ml, biopsy Gleason score 6 with no pattern 4 or 5, clinical stage T1/T2a, and percentage of positive cores <33%. All patients had complete demographic, clinical, and pathologic data. The main outcome was unfavorable PCa, defined as pathologic Gleason score 4þ3 and/ or pathologic stage pT3b. We performed a sensitivity analysis defining unfavorable PCa as pathologic Gleason score 7 and/or pathologic stage pT3a. Multivariable logistic regression analysis tested the relationship between unfavorable PCa and age, race (White vs African American vs Hispanic vs Other), marital status (not married vs married), annual family income, insurance status (insured vs Medicaid covered vs uninsured), total PSA, clinical stage (T1 vs T2a), and percentage of positive cores. RESULTS: Overall, 195 patients (5.9%) harbored unfavorable PCa at final pathology. On the other hand, 1,309 patients (40%) showed unfavorable PCa when pathologic Gleason score 7 and/or pathologic stage pT3a were considered. At multivariable analysis, patient age (odds ratio [OR]: 1.03; 95% confidence interval [CI]: 1.01, 1.05; p1⁄40.006), African American patients (OR: 1.82; 95% CI: 1.19, 2.78; p1⁄40.016), Hispanic patients (OR: 1.63; 95% CI: 1.04, 2.57; p1⁄40.033), and total PSA (OR: 1.15; 95% CI: 1.07, 1.25; p<0.0004) were significantly associated with higher probability of harboring unfavorable PCa. These results were confirmed when unfavorable PCa was defined as pathologic Gleason score 7 and/or pathologic stage pT3a. CONCLUSIONS: A significant proportion of AS candidates harbors unfavorable PCa at final pathology, even considering stringent criteria as adverse cancer features (namely, pathologic Gleason 4þ3 or pathologic stage pT3b). Patient age, race, and total PSA are significant predictors of unfavorable PCa. These results should be taken into account when counseling AS candidates.