Competing risk analysis of mortality in prostate cancer treated with radical prostatectomy.

OBJECTIVE To determine the risk of cancer-specific mortality (CSM) versus the competing risk of mortality by other causes (MOC) in patients with localised prostate cancer (LPC) treated with radical prostatectomy (RP). MATERIAL AND METHOD An observational cohort study of 982 patients with LPC treated with RP selected from our department's PC registry database. A competing risk analysis was performed, calculating the probability of CSM in the presence of the competing risk of MOC. Cumulative incidence curves were constructed, and point estimates were performed at 5, 10 and 15 years. The analysis was stratified by age (≤65 vs. >65 years) and risk group: low (Gleason score ≤6 and pT2abc); intermediate (Gleason score of 7 and pT2abc) and high (Gleason score of 8-10 or pT3ab). RESULTS With a median follow-up of 60 months, the overall probability of dying from PC was 3.5%, and the probability of dying from other causes was 9%. A competing effect for MOC was observed. The risk of MOC was almost 3 times greater than that of CSM. This effect remained for all risk groups, although its magnitude decreased progressively according to the risk group level. At 10 years, CSM was only 0%, 1% and 2% for the low, intermediate and high-risk groups, respectively, while the likelihood of MOC was 4%, 4% and 10%, respectively. The mortality risk was shown after 10years of follow-up and was higher for other causes not attributable to PC and for patients older than 65years. CONCLUSIONS The benefit of RP might be overestimated, given that the risk of MOC is greater than that of CSM, regardless of the age group and risk group, especially after 10years of follow-up. The only parameter that varied was the magnitude of the CSM/MOC ratio. This information could help in choosing the active treatment for patients with LPC and short life expectancies.