First Postprostatectomy Ultrasensitive Prostate-specific Antigen Predicts Survival in Patients with High-risk Prostate Cancer Pathology

Background: Ultrasensitive prostate-specific antigen (uPSA) has untapped potential for optimizing management following radical prostatectomy (RP) in terms of facilitating early salvage, minimizing overtreatment, and identifying those at risk of occult systemic disease. Objective: To test first postoperative uPSA for prediction of outcome in patients with adverse pathology after RP. Design, setting, and participants: Patients with extraprostatic extension and/or a positive margin who did not receive immediate adjuvant therapy. Outcome measurements and statistical analysis: First uPSA was measured at 3 mo after RP. The study endpoints were biochemical relapse (BCR), defined as PSA >= 0.2 ng/ml, bone metastasis-free survival (BMFS), prostate cancer-specific survival (PCSS), overall survival (OS), and salvage radiation therapy (SRT) success. Outcome results were compared using the Kaplan-Meier method and multivariate analysis (MVA). Results and limitations: The cohort consisted of 269 RP patients from 1991-2015 with median follow-up of 77 mo. Sensitivity analysis identified first postoperative uPSA of >= 0.03 ng/ml as the optimal threshold for predicting BCR. First postoperative uPSA >= 0.03 versus <0.03 ng/ml was associated with worse 5-yr BCR (86%, 95% confidence interval [CI] 71-93% vs 39%, 95% CI 25-51%; p < 0.00001), 10-yr BMFS (75%, 95% CI 62-92% vs 95%, 95% CI 88-100%; p = 0.0001), 10-yr PCSS (84%, 95% CI 73-96% vs 100%, 95% CI 100-100%; p = 0.005), and 10-yr OS (81%, 95% CI 70-93% vs 98%, 95% CI 94-100%; p = 0.009). On MVA, first postoperative uPSA >= 0.03 ng/ml was an independent predictor of BCR (hazard ratio [HR] 9.4, 95% CI 5.8-15.4; p < 0.00001) and the only predictor for BMFS (HR 9.7, 95% CI 2.1-44.6; p = 0.0034), PCSS (HR 13.5, 95% CI 1.7-107.9; p = 0.014), and OS (HR 5.0, 95% CI 1.4-18.3; p = 0.014). Following SRT, first postoperative uPSA >= 0.03 ng/ml independently predicted worse BMFS (HR 5.9, 95% CI 1.3-26.9; p = 0.021), PCSS (HR 6.9, 95% CI 0.9-55.8; p = 0.07), and OS (4.5, 95% CI 1.0-20.1; p = 0.057). Limitations include the retrospective design and potential selection bias. Conclusions: First postoperative uPSA >= 0.03 ng/ml independently predicts BCR, BMFS, PCSS, and OS better than traditional risk factors. SRT alone may be insufficient for patients with high-risk disease when first postoperative uPSA is >= 0.03 ng/ml. Patient summary: When the first postprostatectomy ultrasensitive prostate-specific antigen level is >= 0.03 ng/ml, patients are at higher risk of recurrent and occult prostate cancer. They should be considered for early salvage radiotherapy, possibly with hormone therapy. (C) 2018 European Association of Urology. Published by Elsevier B.V. All rights reserved.