Patterns of Clinical Recurrence of Node-positive Prostate Cancer and Impact on Long-term Survival

Background: The patterns of recurrence of patients with node-positive prostate cancer (PCa) at radical prostatectomy (RP) are still unknown. Objective: To describe recurrence patterns, to identify predictors of progression, and to test the impact of the site of clinical recurrence (CR) on cancer-specific mortality (CSM). Design, setting, and participants: We included 1003 patients with node-positive PCa treated with RP and extended pelvic lymph node dissection. Patients who experienced biochemical recurrence (BCR; n = 370) and CR (n = 183) were identified. CR was defined as positive imaging after BCR. Patients were stratified according to the first site of CR: local and/or nodal (recurrence in the prostatic bed and/or pelvic nodes), retroperitoneal, bony, or visceral. Outcome measurements and statistical analysis: Multivariable competing-risks regression analyses identified predictors of systemic recurrence (ie, retroperitoneal, bony, and/or visceral) and tested the association between the site of recurrence and CSM after accounting for the risk of other-cause mortality. Results and limitations: When considering patients experiencing BCR, pathologic Gleason score, time to BCR, and the administration of a positron emission tomography/computed tomography scan represented predictors of systemic recurrence (all p <= 0.002). Among patients who experienced CR, recurrence was local and/or nodal in 56 (30.5%), retroperitoneal in 25 (13.7%), skeletal in 77 (42.1%), and visceral in 25 (13.7%). Among patients experiencing local recurrence, 27 (48.2%) had positive margins, 29 (51.8%) had adjuvant radiotherapy, and 22 (39.5%) had salvage radiotherapy. Patients experiencing local and/or nodal recurrence had higher 5-yr CSM-free survival rates compared with those with retroperitoneal, skeletal, and visceral recurrence (79.3%, 76.3%, 50.8%, and 35.3%, respectively; p < 0.001). The site of recurrence represented an independent predictor of CSM (p <= 0.04). Conclusions: In approximately one-third of patients who are pN+ and experience CR, the prostatic bed and pelvic lymph nodes represent the first sites of recurrence. These patients have a more favorable prognosis compared with those with skeletal and visceral metastases. These data have important implications for the selection of the optimal postoperative management of pN+ patients who experience CR. Although patients with local and/or pelvic nodal recurrence might benefit from nonsystemic salvage therapies, men with visceral and skeletal recurrence might represent ideal candidates for systemic approaches. Patient summary: Not all patients with pN+ prostate cancer who experience clinical recurrence harbor distant metastatic disease. Local and/or nodal recurrence occurs in one-third of these cases. These patients share a more favorable prognosis than their counterparts with systemic recurrence. These results are important for tailoring the optimal postoperative management for each node-positive patient with recurrent disease after surgery. (C) 2015 European Association of Urology. Published by Elsevier B.V. All rights reserved.