422 Predictors of early failure after timely administered salvage radiotherapy for biochemical recurrence after radical prostatectomy

Abstract

INTRODUCTION AND OBJECTIVES: The oncological safety of robot-assisted (RARP) and open radical prostatectomy (ORP) in patients with more aggressive disease is still debated. We aimed at comparing the biochemical recurrence (BCR) rates of RARP and ORP in patients with intermediate or high-risk prostate cancer (PCa). Additionally, since individuals experiencing early BCR are at increased risk of clinical progression and cancer-specific mortality, we assessed the impact of the surgical approach also on the risk of early BCR. METHODS: Overall, 2,181 patientswith intermediate or high-risk PCa (clinical stage T2b, biopsy Gleason score (GS) 7, and/or preoperative PSA 10ng/ml) treated at a single tertiary referral center between January 2006 and August 2013 were identified. No patient received neo-adjuvant or adjuvant therapies. BCR was defined as the detection of PSA 0.2ng/ml after surgery. Early BCR was defined as the occurrence of BCR within 2 years from surgery. Uniand multivariable (MVA) Cox regression analyses evaluated the association between the surgical approachand the risk of BCRandearly BCR, after accounting for confounders. Covariates consisted of clinical stage, biopsy GS, and preoperative PSA. Subsequently, we evaluated the impact of RARP on the risk of BCR and early BCR after accounting for pathological GS, pathological stage, surgicalmargin status (SM), and adjuvant treatments. RESULTS: Mean patient age was 64.9 years (median: 65). Overall, 1,522 (69.8%) and 659 (30.2%) patients were treated with ORP and RARP, respectively. Overall, the 5-year BCR-free survival rates were 79.8%. The 5-year BCR-free survival rates were 79.8% and 81.9% in patients treated with ORP and RARP, respectively (P1⁄40.7). At univariable Cox regression analysis, RARP was not associated with increased risk of BCR (Odds ratio [OR]: 1.27; 95% confidence interval [CI]: 0.72-1.58; P1⁄40.3) and early BCR (OR: 1.07; 95% CI: 0.78-2.06; P1⁄40.7). This was confirmed at MVA, where RARP was not associated with BCR and early BCR, after accounting both for preoperative (PSA, clinical stage, and biopsy GS) and postoperative (pathological stage, pathological GS, SM, lymph node invasion, and adjuvant treatments) confounders (all P>0.1). CONCLUSIONS: RARP and ORP have comparable 5-year BCR-free survival rates in patients with intermediateand high-risk PCa. Minimally invasive approach leads to comparable oncological outcomes in terms of BCR and early BCR. These results support the oncological safety of RARP in patients with intermediateand high-risk PCa.