RESUMO
OBJECTIVE: To characterize Bacillus Calmette-Guérin (BCG) treatment patterns and associated outcomes in a large cohort of patients with non-muscle-invasive bladder cancer (NMIBC). METHODS: Our retrospective analysis of patients aged ≥66 years with stage 0-1 urothelial bladder carcinoma diagnosed between 2000 and 2012 in the United States Surveillance, Epidemiology, and End Results-Medicare database estimated proxies for recurrence and secondary events and both all-cause and bladder cancer-specific mortality. Proportional hazards models were used in conditional landmark analyses to compare adequate (≥5 induction instillations and ≥2 maintenance instillations) and inadequate BCG, stratified by National Comprehensive Cancer Network risk group. RESULTS: Of 39,532 patients who met the selection criteria, 16,225 (41.0%) received BCG; of them, 4602 (28.4%; 11.6% overall) received adequate treatment. Adequately treated patients were slightly younger and healthier than inadequately treated patients. Half of patients with intermediate- and high-risk NMIBC did not receive BCG; few received adequate treatment. At the 12-month landmark, adequate BCG treatment was associated with decreased risks of recurrence and of cancer-specific and all-cause mortality in patients with intermediate- and high-risk disease. CONCLUSION: We observed lower than expected use of adequate BCG treatment in patients with intermediate- to high-risk NMIBC despite evidence of improved outcomes, which suggested that practice patterns may not be in line with management recommendations in this population.