Does Pioglitazone Up Bladder Cancer Risk?

A study published in BMJ found no association between pioglitazone ever use and risk of bladder cancer vs. pioglitazone never use.

A study published in BMJ found no association between pioglitazone ever use and risk of bladder cancer vs. pioglitazone never use. 

Researchers conducted a retrospective cohort study using propensity score matched cohorts to examine the association between pioglitazone use and bladder cancer risk in patients with type 2 diabetes. They extracted information from country specific datasets of linked records on cancer, general practitioners, hospitals, prescriptions, and deaths. 

A total of 56,337 patients with type 2 diabetes who initiated pioglitazone were matched with 317,019 patients with type 2 diabetes in the same country exposed to a non-pioglitazone diabetes therapy. Patients were matched on treatment history and propensity score accounting for some factors tied to initiating pioglitazone. Researchers also performed sensitivity and stratified analyses to assess the robustness of the evidence. 

Related Articles

Among the pioglitazone cohort, 130 occurrences of bladder cancer were reported over a mean 2.9 year follow-up time. For the nearest match and multiple match cohorts with no pioglitazone exposure, 153 and 970 occurrences of bladder cancer were reported, with a mean 2.8 years and 2.9 years, respectively. 

The adjusted hazard ratio (aHR) for patients ever exposed vs. never exposed to pioglitazone related to bladder cancer risk was 0.99 (95% CI: 0.75–1.30) and 1.00 (95% CI: 0.83–1.21) in the nearest and multiple match cohorts, respectively. The study authors also found that increasing duration of pioglitazone therapy (>48 months, aHR 0.86, 95% CI: 0.44–1.66) and increasing cumulative dose (>40,000mg dose, aHR 0.65, 95% CI: 0.33–1.26) were not associated with the risk of bladder cancer in the nearest match cohort. 

Overall, the study did not find evidence of an association between pioglitazone ever use and the risk of bladder cancer when compared to never use. These findings remain consistent with data from previous studies that also included a long follow-up period. 

For more information visit