ABSTRACT
PURPOSE: The comparative effectiveness of surgery and radiation therapy for high-grade, clinically localized prostate cancer remains a seminal, open question in urologic oncology, with no randomized controlled trials to inform management. We therefore emulated a hypothetical target clinical trial of radical prostatectomy (RP) versus external beam radiotherapy (EBRT) for high-grade, clinically localized prostate cancer. MATERIALS AND METHODS: We conducted observational analyses using the National Cancer Database from 2006-2015 to emulate a target clinical trial in men 55-69 years with cT1-3cN0cM0, PSA<20 ng/mL, Gleason 8 to 10 prostate adenocarcinoma treated with RP or 75 to 81 Gy EBRT with androgen deprivation therapy (EBRT+ADT). The associations of treatment type with overall survival (OS) were estimated using Cox regression with stabilized inverse probability weights (IPW). RESULTS: A total of 26,806 men formed the study cohort (RP: 23,990; EBRT+ADT: 2,816). Baseline characteristics were well-balanced after IPW-adjustment. Median follow-up was 48.4 (IQR 25.5-76.2) months. After IPW-reweighting, RP was associated with improved OS compared to EBRT+ADT (HR 0.54;95% CI 0.48-0.62; P<0.001), with 5- and 10-year OS of 93% vs 87%, and 76% vs 60%, respectively. RP was associated with improved OS across all categories of Gleason score, PSA, cT stage, age, and Charlson comorbidity index examined. In sensitivity analyses adjusting for biopsy tumor volume and a biopsy-specific Gleason score, RP remained associated with improved OS compared to EBRT+ADT (HR 0.62;95% CI 0.49-0.78; P<0.001). CONCLUSIONS: In observational analyses designed to emulate a target clinical trial of men with high-grade, clinically localized prostate cancer, RP was associated with improved OS compared with EBRT+ADT.
