75 Background: High-risk prostate cancer (PCa) is associated with greater risk of biochemical recurrence and cancer specific lethality. A multi-modal treatment is required for this group of patients, comprising surgery as part of it. However, the role of surgery as monotherapy is still under investigation. The purpose of this study is to analyze the influence of surgical margins on biochemical recurrence (BCR) among patients with high-risk prostate cancer (PCa) treated with robot assisted radical prostatectomy (RARP) since the start of our robotic program. Methods: We retrospectively analyzed our prospectively collected database of 5695 minimally invasive prostatectomies performed between 2000 and 2015. Clinical, pathological and oncological outcomes were evaluated in patients fulfilling Damico´s high risk characteristics. Primary endpoint was BCR, defined as post-operative PSA ≥ 0,2. Patients with neoadjuvant or adjuvant therapy were excluded. BCR was estimated with Kaplan-Meier curves. Cox proportional hazards regression was used to estimate variables associated with BCR. Results: We identified 199 high-risk PCa patients treated with RARP during the study period. Gleason score ≥ 8, PSA ≥ 20 and clinical stage ≥ T2c were present in 44%, 35% and 11% of the patients, respectively. The rate of positive surgical margins was 25%. With a median follow-up of 23 months (interquartile 12 – 34 months), 31% of the patients had BCR. Five-year BCR-free survival was 34,5%. Gleason score ≥ 8, PSA ≥ 20 and positive surgical margins were not predictors of BCR. A positive correlation of pathological stage ≥ T3 and BCR was found with (HR = 2.9; 95% CI = 1.2-6.9). Conclusions: The 5-years BCR-free survival was poor despite a low rate of positive surgical margins, when compared to historical series. We found that pathological stage ≥ T3 has a significant correlation with the BCR and that negative surgical margins do not assure good prognosis for high-risk patients.