A competing-risks analysis of survival after alternative treatment modalities for prostate cancer patients: 1988-2006

Background: The efficacy of prostate cancer (PCa) treatment modalities is a subject of continuous debate. Objective: We tested the hypothesis that significant differences in survival rates may exist among PCa patients treated with radical prostatectomy (RP), radiation therapy (RT), and observation. Design, setting, and participants: We focused on 404 604 patients with clinically localized PCa within 17 Surveillance, Epidemiology and End Results registries. Measurements: Competing-risks survival analyses were used to estimate cancer-specific mortality (CSM) and other-cause mortality (OCM) rates. Patients were stratified according to treatment type, age group, and PCa risk group (high risk: T2c and/or Gleason score 8-10; low to intermediate risk: all others). Results and limitations: The 10-yr CSM and OCM rates were 6.1% and 29.2%, respectively. In RP, RT, and observation patients, CSM rates were 3.6%, 6.5%, and 10.8% (p <0.001), respectively; OCM rates were 17.1%, 32.4%, and 48.9% (p <0.001), respectively. In low- to intermediate-risk patients, the lowest CSM (1.3-3.7%) and OCM (6.9-31.6%) rates within all age categories except octogenarians (8.9% and 62.8%, respectively) were recorded in RP. In high-risk patients, the lowest CSM (5.8-7.2%) and OCM (8.7-16.1%) rates in patients aged ≤69 yr were also recorded in RP. RT was equally favorable to RP in the 70-79 age category and appeared ideal in all octogenarian patients. Conclusions: Our results showed that RP provides the most favorable survival rates in most patients. The exception is octogenarian men, in whom RT provides the best results. Finally, the least-favorable outcomes were recorded after observation. However, these findings must be interpreted within the context of the limitations of observational data.