Canadian study affirms pioglitazone-bladder cancer link
Monday, 4 June 2012- The use of the anti-diabetic agent pioglitazone (Actos) is associated with an increased risk of bladder cancer, a nested case-control study confirmed.
Patients with type 2 diabetes who were ever treated with pioglitazone had an 83% higher risk for bladder cancer (adjusted rate ratio 1.83, 95% CI 1.10 to 3.05) than those who had never used the thiazolidinedione, according to Laurent Azoulay, PhD, of McGill University in Montreal, Canada, and colleagues.
This was a drug-specific effect, because patients taking rosiglitazone (Avandia) did not have an elevated risk (RR 1.14, 95% CI 0.78 to 1.68), the researchers reported online in BMJ. The results of this analysis differed from those of a study presented at the annual meeting of the American Society of Clinical Oncology, which suggested that the bladder cancer risk was a class effect for glitazones.
Other previous efforts to ascertain if pioglitazone is associated with bladder cancer have had conflicting results, but the FDA and the Canadian health authorities have added a warning of that risk to the drug's labeling, and some European countries have restricted its use.
To clarify the potential hazard with the drug, Azoulay and colleagues analyzed data from a large UK primary care database, identifying 115,727 patients who had been given prescriptions for anti-diabetic drugs between 1988 and 2009. Mean age of participants was 64, and mean time of follow-up was 4.6 years.
Patients who had been given a prescription for a thiazolidinedione tended to be more obese, to have smoked, and to have poorly controlled diabetes. A total of 0.5% were taking pioglitazone or rosiglitazone as monotherapy, with a mean duration of use of about 2.2 years.
During 526,559 patient-years of follow-up, there were 470 new cases of bladder cancer, for a rate of 89.4 per 100,000 patient-years.
For the primary analysis, in which the researchers attempted to account for disease latency, the researchers included only the 376 cases with a year of follow-up in the cohort before the cancer diagnosis. The cases were matched with 6,699 controls. The rate of bladder cancer increased with longer use of pioglitazone, with those taking the drug for longer than 2 years having a twofold increase in risk (RR 1.99, 95% CI 1.14 to 3.45, P for trend 0.050).
This was a "relatively short duration of use" for development of the cancers, the researchers noted. Moreover, a dose response was seen, with patients whose cumulative dosage exceeded 28,000 mg having an adjusted rate ratio of 2.54 (95% CI 1.05 to 6.14, P for trend 0.030). In contrast, duration of use and cumulative dosage did not influence risk with rosiglitazone.
Although the researchers found that the rate ratios were "relatively high," they cautioned that the risks "are in absolute terms low," with long duration of use increasing the rate by 88 cases per 100,000 patient-years and high dosage increasing the rate by 137 cases per 100,000.
In a sensitivity analysis that included all 490 cases, the overall findings did not change.
A possible explanatory mechanism for the bladder cancer risk is chronic irritation of the bladder from crystal formation, although further research will be needed to establish this as causative, according to the researchers.
Limitations of the study included its restriction to prescriptions written in primary care and a lack of information on patient adherence to treatment or other risk factors for cancer. Also, patients taking this class of drugs usually have worse diabetes and may already be at risk for bladder cancer.
Azoulay and colleagues concluded that there indeed was an increased likelihood of bladder cancer among patients with type 2 diabetes treated with pioglitazone. "Doctors, patients, and regulatory agencies should be aware of this association when assessing the overall risks and benefits of this therapy," they advised.
Source: L. Azoulay, et al. "The use of pioglitazone and the risk of bladder cancer in people with type 2 diabetes: nested case-control study‘," BMJ 2012; DOI: 101136/bmj.e3645.