Nomograms for Metastasis-Free and Overall Survival for Pathologically Node Positive Prostate Cancer Patients Treated With or Without Radiation Therapy Plus Short-Term ADT.

Abstract

We aimed to develop nomograms to predict the risk reduction for metastasis and death in pathologically node-positive (pN +) prostate cancer patients treated with or without radiation therapy (RT). From a prospectively gathered institutional database, we identified patients with pN+M0 prostate cancer after surgery. We evaluated several regression models of known or suspected clinical-pathologic covariates and selected the model with the highest Harrell's concordance-index (c-index) and clinical utility to prognosticate metastasis for inclusion in a nomogram. Covariates in the final, competing-risk adjusted, metastasis model included PSA nadir after surgery, pathologic T-stage, margin status, Gleason score (GS), number of positive lymph nodes, and use of postoperative radiotherapy combined with androgen deprivation therapy (RT+ADT). The overall survival model also included Charlson comorbidity score and age. 336 pN+men with a mean age of 64.9 years and a median follow-up of 4.1 years who had a radical prostatectomy were included in the analysis. 83 men were recommended RT+ADT, of whom 4% refused the ADT and received RT alone. C-index was 0.85 and 0.71 for the MFS and OS models, respectively. On multivariable analysis (MVA) adjusted for competing risks, RT+ADT significantly improved MFS (HR=0.70 P=< .01) with number of nodes positive, GS 8-10, PSA nadir > 1 ng/mL, and pT3b prognostic for metastasis. MVA for OS demonstrates RT+ADT improves survival (HR=0.40, P=.02), with GS8-10 and PSA nadir > 1.0 prognostic for death. We developed predictive nomograms for patients with pN+ prostate cancer following radical prostatectomy. These models can discretely quantify an individual's risk of metastasis or death with and without post-prostatectomy radiotherapy.