Pioglitazone and risk of bladder cancer among diabetic patients in France: a population-based cohort study.

Abstract Aims/hypothesis Previous studies have suggested an increased risk of bladder cancer with pioglitazone exposure. We aimed to investigate the association between pioglitazone exposure and bladder cancer in France. Methods This cohort study involved use of data from the French national health insurance information system (Système National d'Information Inter-régimes de l'Assurance Maladie; SNIIRAM) linked with the French hospital discharge database (Programme de Médicalisation des Systèmes d'Information; PMSI). The cohort included patients aged 40 to 79 years who filled a prescription for a glucose-lowering drug in 2006. The cohort was followed for up to 42 months. Pioglitazone exposure was modelled as a time-dependent variable and defined by having filled at least two prescriptions over a 6-month period. Incident cases of bladder cancer were identified by a discharge diagnosis of bladder cancer combined with specific aggressive treatment. Statistical analyses involved a multivariate Cox model adjusted for age, sex and exposure to other glucose-lowering drugs. Results The cohort included 1,491,060 diabetic patients, 155,535 of whom were exposed to pioglitazone. We found 175 cases of bladder cancer among exposed patients and 1,841 among non-exposed patients. Incidence rates were 49.4 and 42.8 per 100,000 person-years, respectively. Pioglitazone exposure was significantly associated with bladder cancer incidence (adjusted HR 1.22 [95% CI 1.05, 1.43