Comparative effectiveness of treatments for high-risk prostate cancer patients

Abstract

BackgroundTo determine the comparative effectiveness of primary radical prostatectomy (RP) compared to external bean radiation therapy (EBRT) with androgen deprivation therapy (ADT), or EBRT plus brachytherapy (BT) with or without ADT among Medicare fee-for-service beneficiaries with high-risk prostate cancer, for 10-year, mortality (overall and prostate cancer-specific), complications, health service use, and cost. MethodsThis population-based cohort study used Surveillance, Epidemiology, and End Results – Medicare data. Eligible patients were men aged 66 or older and diagnosed with high-risk prostate cancer between 1996 and 2003. Outcomes evaluated were 10-year overall mortality and prostate cancer-specific mortality, complications, health service use, and cost. We used Cox regression, Poisson regression, and Generalized Linear Model (GLM) log-link models to assess the outcomes. Main findingsThe 10-year overall mortality of EBRT + ADT was comparable to that of the RP group (hazard ratio [HR] = 1.09, confidence interval [CI] = 0.72–1.66). The EBRT + BT ± ADT group had overall survival advantage compared to RP (HR = 0.47, CI = 0.31–0.73). Compared to the RP group, EBRT + ADT group had higher 10-year prostate cancer-specific mortality (HR = 2.19, CI = 1.92–5.21). Both EBRT + ADT and EBRT + BT ± ADT were associated with higher 10-year cost (odds ratio = 1.72, CI = 1.35–2.20; and odds ratio = 1.63, CI = 1.29–2.04), compared to RP group. Complications and health service use varied across 3 treatment groups and across phases of care. Principal conclusionsOur results also demonstrate long-term overall survival benefits for EBRT + BT ± ADT, and greater bowel and bladder side effects over a decade, compared to RP. The RP group had advantage for long-term prostate-cancer specific mortality, compared to EBRT + ADT group. Thus, RP can provide superior cancer control with clear cost advantage for older men with high-risk disease. In terms of value proposition, our results support RP as preferred treatment option, compared to EBRT + ADT and EBRT + BT ± ADT for high-risk prostate cancer patients.