5 April 2016  

But absolute risk remains low overall, finds large British study.

by Parker Brown
Staff Writer, MedPage Today
Pioglitazone (Actos) was associated with a higher risk of bladder cancer than other diabetes drugs in a large population-based cohort study.
Researchers looked at about 146,000 patients in the U.K. who had recently started taking an anti-diabetic drug; those on pioglitazone faced an increased risk of bladder cancer of 121.0 versus 88.9 per 100,000 person-years among those on other diabetes drugs (hazard ratio 1.63, 95% CI 1.22-2.19). Fifty-four patients receiving pioglitazone were diagnosed with cancer after a mean follow-up of 4.7 years.
But those on rosiglitazone (Avandia) didn’t see an increased risk of bladder cancer (HR 1.10, 95% CI 0.83-1.47), according to lead author Marco Tuccori, PhD, at the Jewish General Hospital in Montreal, Canada, suggesting that the increased risk was not a class effect. Tuccori and colleagues published their findings on Wednesday in The BMJ.
In 2015, Actos manufacturer Takeda agreed to pay $2.4 billion to settle lawsuits alleging that it hid bladder cancer risks associated with the drug from patients.
Where the Evidence Fits In
The new study is the latest in a long-running seesaw drama, with one study identifying an increased bladder cancer risk with the drug, only to be followed by another showing no link.
The latest results support findings from a 2005 trial that found a heightened risk of bladder cancer for those on pioglitazone versus placebo. But last year, a study published in the Journal of the American Medical Association found no association between the two, for example.
Silvio Inzucchi, MD, at the Yale School of Medicine, said he still considers that JAMA study to be the definitive one in this area.
“It’s interesting to note that the sum total of bladder cancer events in the pioglitazone-treated patients was only 54” in the BMJ study, he told MedPage Today in an email. “Given such small numbers any modest changes in the underlying risk for bladder cancer could have a big impact on measured incidence rates.”
The JAMA study, on the other hand, analyzed a larger number of cases (n=1,261) and had a longer duration of follow-up (with a median of 7.3 years), he pointed out.
Inzucchi noted that in the BMJ study, pioglitazone patients “tended to be older, had worse diabetes control, had a greater likelihood of previous bladder conditions, and had more urine testing.” He added that those factors, rather than the drug itself, could have been determining factors in bladder cancer diagnosis. (Tuccori and colleagues adjusted for age, HbA1c, and presence of bladder conditions, as well as certain other factors, in their statistical analyses.)
But there’s still a “substantial and fairly consistent body of evidence” that supports the link between bladder cancer and pioglitazone, according to Victor Montori, MD, at the Mayo Clinic in Rochester, Minn. “By some accounts the association between pioglitazone and bladder cancer should be added to the long list of suppressed information about harm that would have affected the informed use of a drug during its patent protected life,” he wrote in an accompanying editorial.
And David Nathan, MD, at Massachusetts General Hospital, told MedPage Today that the findings were not surprising. “This study is simply better (and larger) than many of the older observational studies,” he wrote in an email. “What needs to be emphasized is that the absolute risk for bladder cancer, which is quite small, adds to the potentially deleterious effects of the thiazolidinediones, such as volume retention and bone loss, which are more common.”
He added that he agreed with the editors that some of the contradictory earlier findings may have been due to methodological shortcomings of those trials.
Study Details
Data were taken from the U.K. Clinical Practice Research Datalink. Overall, 622 patients received a diagnosis of bladder cancer, with a crude incidence rate of 90.2 per 100,000 person-years (95% CI 83.2-97.6). The median time between starting a drug and getting diagnosed with bladder cancer was 4.4 years.
Patients who received a prescription for any non-insulin anti-diabetic drug — including metformin, sulfonylureas, prandial glucose regulators, thiazolidinediones, acarbose, dipeptidyl peptidase-4 (DPP-4) inhibitors, glucagon-like peptide (GLP-1) agonists, and sodium-glucose cotransporter-2 (SGLT-2) inhibitors — between 1998 and 2013 were included. All patients were at least 40 years old.
The authors adjusted for several variables that were recorded at cohort entry, including age, sex, year of entry, body mass index, smoking status, alcohol use status, and the presence of some disorders. They also controlled for duration of treated diabetes.
Compared with those on other anti-diabetic medications, those on pioglitazone were less likely to be obese but more likely to have higher HbA1c levels, have undergone urine protein testing before cohort entry, have a longer duration of treated diabetes, and have previous bladder conditions.
Limitations of the study include the possibility of residual confounding from unmeasured variables like diet, physical activity, family history of cancer, or race. There’s also the possibility of misclassification of drug use or of bladder cancer status.
Treating Patients
The absolute risk of bladder cancer for patients on pioglitazone is about three extra cases per 10,000 patients exposed for 1 year, according to Inzucchi. “This risk would need to be considered in the context of the known benefits of the drug, which is now generically available and extremely inexpensive compared to many other diabetes drugs, at less than $10/month in the U.S.,” he wrote.
“The problem now is whether the increased likelihood and undesirability of bladder cancer (which remains rare) justify withholding pioglitazone from adults with type 2 diabetes, given the benefits (such as stroke prevention) and potential harms (such as weight gain, heart failure, bone fractures) associated with its use, and the relative efficacy and safety of alternatives,” added Montori. “This is the kind of consideration that complicates the treatment of diabetes and demands reliable evidence of comparative effectiveness and safety.”
Nathan said doctors need to carefully consider the advantages and disadvantages of different classes of diabetes treatments. Empagliflozin (Jardiance) was found to lower risk of cardiovascular disease last year, raising questions about where other treatments still fit in.
But Inzucchi, who was an investigator on the trial that discovered the CV benefit, said he still thinks there’s room for pioglitazone — particularly when it’s paired with empagliflozin. “I actually believe that the pioglitazone-empagliflozin tandem could be quite powerful combination – pioglitazone to prevent or slow the development of atherosclerosis and then empagliflozin to reduce complications from coronary artery disease once it occurs,” he wrote. In addition, empagliflozin could reduce fluid retention that is sometimes associated with thiazolidinediones.
But he added that we need more studies to test that and develop evidence-based guidelines. “We’ve certainly come a long way in less than a year in understanding the CV effects of diabetes therapies and even more information will be forthcoming over the next 1-2 years,” he wrote.
The study by Tuccori et al was funded by the Canadian Institutes of Health Research. The authors disclosed no relationships with industry.
Montori disclosed no relationships with industry.
Inzucchi disclosed serving as an expert witness for Takeda, which markets Actos, in a patent suit and receiving clinical study drugs from the company. He also disclosed ties with Merck, Janssen, Sanofi/Regeneron, Poxel, Boehringer Ingelheim, Eli Lilly, Novo Nordisk, Intarcia, and Abbott.
Reviewed by F. Perry Wilson, MD, MSCEAssistant Professor, Section of Nephrology, Yale School of Medicine and Dorothy Caputo, MA, BSN, RN, Nurse Planner