Disease specific survival as endpoint of outcome for bladder cancer patients following radical cystectomy.

OBJECTIVE: To directly compare disease specific and overall survival as endpoints in the outcome analysis of a large number of cystectomy patients and to define predictors for survival.
MATERIALS AND METHODS: We retrospectively analyzed the records of 686 patients who underwent bilateral pelvic lymph node dissection (PLND) and radical cystectomy from 1980 to 1990 at Memorial Sloan-Kettering Cancer Center.
RESULTS: Disease specific survival characterized a clearly more favorable patient outcome than overall survival in the entire patient population as well as patients with organ confined (OC) and non-organ-confined disease (NOC): 10-year disease specific and overall survival rates for patients with OC disease (<or=P3a), negative nodes or NOC (>or=P3b) were 72.9% versus 49.1%, 61.7% versus 40.8% and 33.3% versus 22.8%, respectively. In node positive (N+) patients 10-year disease specific and overall survival rates were 27.7% and 20.9%, respectively. In a multivariate analysis organ confinement and nodal status were the strongest independent predictors of disease specific survival in all patient categories. However, stratification according to organ confinement and nodal status revealed additional prognostic parameters.
CONCLUSION: Organ-confined bladder cancer translates into high disease specific survival rates following radical cystectomy. Outcome is best characterized by disease-specific survival versus overall survival, which underestimates the impact of treatment in patients with favorable tumor and nodal stage. Subgroup analysis of patients with organ confinement and nodal status identified additional prognostic variables within the more favorable patient categories not apparent in the entire population. The poor prognosis of patients with NOC and/or N+ tumors emphasized the importance of future randomized trials in which such stratification variables may be of value.