Large analysis finds no link between pioglitazone and bladder cancer in type 2 diabetes patients

Pioglitazone wasn't associated with higher incidence of bladder cancer in people with type 2 diabetes, according to a recent, multi-country analysis.

Researchers obtained data on prescriptions, cancer, and mortality from 1.01 million people with type 2 diabetes in 6 places: Finland; Manchester, England; Rotterdam, Netherlands; Scotland; British Columbia; and the U.K. Clinical Practice Research Datalink. In each center, the researchers considered for study inclusion any patient who had type 2 diabetes since the introduction of thiazolidinediones (TZDs)—i.e., around 1999. Patients who were known to have had bladder cancer prior to their diabetes diagnosis, or whose cumulative exposure to TZDs couldn't be evaluated, were excluded. The researchers used prescription data to define all periods of exposure to TZD, metformin, insulin, and sulfonylurea for each patient, and follow-up lasted until the latest date for which patient data were available. Results were published in the December Diabetologia.

Results comprised 5.9 million person-years, in which there were 3,248 cases of bladder cancer and 5,262 bladder neoplasms during a median follow-up of 4 to 7.4 years. Overall, there were 117 bladder cancers and 204 bladder neoplasms in people ever exposed to pioglitazone by the end of follow-up. There was no significant association between cumulative exposure to pioglitazone and bladder cancer in men (rate ratio [RR] per 100 days cumulative exposure, 1.01; 95% CI, 0.97 to 1.06) or women (RR, 1.04; 95% CI, 0.98 to 1.03). These results were adjusted for age, year, duration of diabetes, and smoking. There was also no association between rosiglitazone and bladder cancer in either men or women.

Past observational studies on the risk of pioglitazone and bladder cancer have yielded contradictory results. All of these studies were “prone to inappropriate causal inference and potential confounding,” however, the authors noted. Still, based on results from those studies, the FDA recommended in 2010 that pioglitazone be avoided in patients with bladder cancer and prescribed with caution in those with previous bladder cancer, they noted. The current analysis is the only one to include a large number of patients with diabetes from several international centers that used identical methodology. The results do not support the existence of a causal effect of pioglitazone on bladder cancer, but more analyses with longer follow-up times are needed, the researchers concluded.