Pioglitazone use and bladder cancer - Hospital-based results from a nested case-control study in japan

Bachground Although many studies on the pioglitazone and bladder cancer have been published in recent years, the findings seemed to be inconsistent among the studies. Most studies have been arisen from Western countries except a single Japanese study. Objective We tried to investigate the association between the use of pioglitazone and bladder cancer since it is still controversial. Methods A nested case-control study was executed using the Toyama University Hospital Database between 2005 and 2011. Cases were defined as pathologically diagnosed bladder cancer and retrieved from patients with type 2 diabetes mellitus. Controls were selected by matching with gender, age within 6 years, and visit date within 60 days. Conditional logistic regression provided an adjusted odds ratio of pioglitazone to incident bladder cancer for age, hemoglobin Alc (HbA1c), and other antidiabetic medications. Results We identified a total of 58 patients with bladder cancer. Since some patients contained missing data in the index date of incident bladder cancer and/or did not match with any controls, there were 95 patients in matched analysis. Mean age was 69 years and 26% were women. Mean HbAlc value was 7.0%. Antidiabetic drug utilization was 6% for pioglitazone, 46% for sulfonylureas, 27% for alpha-glucosidase inhibitors, and 38% for insulin. Use of pioglitazone was not associated with the risk of bladder cancer (adjusted odds ratio 0.90 [95% CI 0.09-8.89]; P=0.93). Insulin, sulfonylureas, alpha-glucosidase inhibitors revealed an adjusted odds ratio 1.41, 1.16, and 0.70, respectively. Conclusion Our finding has suggested that pioglitazone might not increase the risk of bladder cancer.