Upper-tract tumors after an initial diagnosis of bladder cancer: argument for long-term surveillance.

Abstract

To determine the relative risk (RR) of upper-tract tumors (UTT) after bladder cancer, stratified by bladder tumor characteristics, demographic factors, and follow-up duration, in order to develop an improved risk-based surveillance strategy. The 1973 to 1996 Surveillance, Epidemiology, and End Results (SEER) database was used to determine the observed and expected number of UTT after bladder cancer. The RR with 95% confidence intervals (CI) were calculated, stratifying by race, sex, stage, grade, histology, and follow-up duration. The tumor characteristics and clinical outcome were compared in patients with UTT after bladder cancer and those with de novo UTT. A total of 94,591 patients had a first diagnosis of bladder cancer, of whom 91,245 had follow-up (median, 4.1 years), with no antecedent or synchronous UTT. UTT developed subsequently in 657 of 91,245 (0.7%), with 12.80 expected cases (RR = 51.3; 95% CI, 47.5 to 55.4). The respective RRs for UTT for white men and women were 64.2 (95% CI, 55.1 to 74.3) and 75.4 (95% CI, 57.7 to 96.9) at less than 2 years, 44.3 (95% CI, 36.7 to 53.0) and 40.5 (95% CI, 27.9 to 56.8) at 2 to 5 years, 50.8 (95% CI, 42.2 to 60.7) and 42.1 (95% CI, 28.8 to 59.4) at 5 to 10 years, and 43.2 (95% CI, 32.6 to 56.1) and 22.2 (95% CI, 10.1 to 42.2) at >or= 10 years. Similar RRs were seen among different strata of race, stage, grade, and histology. Patients with UTT after bladder cancer had lower stage and improved disease-specific survival compared with those with de novo UTT. The incidence of UTT is stable on long-term follow-up, with no significant risk factors identified. These findings suggest that upper-tract surveillance remain rigorous on extended follow-up of bladder cancer patients.