Abstract
BACKGROUND: Studies have suggested a positive association between bladder cancer (BC) outcome and comedication use, including nonsteroidal anti-inflammatory drugs (NSAID), metformin, and prednisone use. To validate these associations, we evaluated whether these medications were associated with clinical outcome in a Canadian cohort of BC patients. METHODS: This is a retrospective cohort study on BC patients undergoing radical cystectomy (RC) in Québec province in 2000-2015, as registered in the provincial health administration databases. Medication use was considered chronic when prescribed for ≥ 1 year. Overall (OS), disease-specific (DSS) and recurrence-free (RFS) survival were compared using multivariable Cox proportional hazards models. Covariates included age, Charlson's comorbidity index, region of residence, year of RC, distance to hospital, hospital type, hospital and surgeon annual RC volume, neoadjuvant chemotherapy use, and type of bladder diversion, as well as mutual adjustment for concomitant comedication use (statins, NSAIDs, metformin, and prednisone). RESULTS: Of 3742 patients included, 293, 420, and 1503 patients chronically used prednisone, metformin, and NSAIDs before surgery, respectively. In multivariable analyses, preoperative prednisone use was associated with improved OS (HR 0.67, 95%CI 0.55-0.82), DSS (HR 0.58, 95%CI 0.45-0.76), and RFS (HR 0.61, 95%CI 0.47-0.78). Patients who chronically used metformin preoperatively had a worse OS (HR 1.29, 95%CI 1.07-1.55), DSS (HR 1.38, 95%CI 1.10-1.72), and RFS (HR 1.41, 95%CI 1.13-1.74). Preoperative, chronic NSAID use was not significantly associated with all clinical outcomes, with adjusted HRs for OS, DSS, and RFS of 1.10 (95%CI 0.95-1.27), 1.24 (95%CI 1.03-1.48), and 1.22 (95%CI 1.03-1.45), respectively. Directionality of findings was similar when stratifying by comedication use in the year following surgery. Results were similar after propensity-score matching too. CONCLUSIONS: In our Canadian cohort of BC undergoing RC, chronic prednisone use was associated with improved clinical outcomes, while metformin and NSAID were not.