Glitazones Once Again Take a New Hit for Bladder Cancer Risk

The research team, led by Ronac Mamtani, MD, of the University of Pennsylvania in Philadelphia, concluded that the increased risk of bladder cancer associated with glitazones -- reached a relative increase of 72% in patients who started on the agents more than 5 years earlier -- "appears to be a class effect." However, the analysis of some 60,000 patients indicated that pioglitazone (Actos) posed no more risk than rosiglitazone (Avandia), according to an abstract of their study.

A large, retrospective study conducted in France and released last year found a 22% increase in rates of bladder cancer among patients taking pioglitazone compared with other diabetes drugs. That study prompted France and Germany to pull the drug from their markets in June 2011. The drug still remains available elsewhere in Europe and in the U.S.

The FDA issued a warning about the bladder cancer risk with pioglitazone but is awaiting results of a 10-year prospective study that began in 2002 before taking further action. An interim analysis based on data through 2007 showed a 40% increase in bladder cancer in patients taking the drug for two years (HR 1.40, 95% CI 1.03 to 2.0).

Rosiglitazone is the only other thiazolidinedione on the U.S. market and access to it is now severely restricted because of concerns about cardiovascular side effects. In the new study, Mamtani and colleagues examined records from the U.K.'s general practitioner database called the Health Improvement Network from 2000 to 2010.

Unlike the earlier studies, which compared diabetic patients taking glitazones with those on all other treatments, Mamtani and colleagues focused on patients taking sulfonylurea drugs such as glipizide (Glucotrol) as the reference group.

There were 18,459 patients in the database taking glitazones and 41,396 on sulfonylurea drugs. New cases of bladder cancer were diagnosed in 60 of the glitazones and 137 patients on sulfonylureas. The overall case frequency was similar in the two groups (HR 0.93, 95% CI 0.68 to 1.29), but Mamtani and colleagues indicated that most patients had only taken the drugs for a short time.

Hazard ratios indicating increased risk with glitazone drugs grew larger with increasing time since treatment was started (irrespective of duration of treatment):

• 3-4 years: HR 1.15
• 4-5 years: HR 1.40
• 5-plus years: HR 1.72 (P=0.033 for trend)

The rising risk with duration of treatment applied both to pioglitazone (P<0.001 for trend) and to rosiglitazone (P=0.006), the researchers indicated.

There was no significant difference in bladder cancer risk when the two glitazone drugs were compared directly, when examined either for time since initiation of treatment or for duration of on-drug therapy.

If pioglitazone is eventually removed from the market because of the bladder cancer risk, it would leave rosiglitazone as the only drug in the class still available. Currently it may be prescribed to new patients only when their doctors warrant that glycemic control cannot be achieved with other medications.

• Note that this study was published as an abstract and presented at a conference. These Patients followed after starting on glitazone drugs for type 2 diabetes were significantly more likely to develop bladder cancer than those taking sulfonylurea agents.
• Note that the analysis indicated that pioglitazone (Actos) posed no more risk than rosiglitazone (Avandia) with respect to risk of bladder cancer.
• Data and conclusions should be considered to be preliminary until published in a peer-reviewed journal.
  

Mamtani R, et al "Long-term therapy with thiazolidinediones and the risk of bladder cancer: A cohort study" ASCO 2012;Abstract 1503.