It's far from the first study to address Avandia safety, but it's by far the largest to date, says FDA researcher and study leader David J. Graham, MD, MPH.
The study analyzes Medicare records for 227,571 patients whostarted treatment with Avandia or Actos between July 2006 and June 2009.The average age of patients in the study was 74.4.
"Our study shows very clearly that Avandia is much less safe than Actos in things that really matter -- things that will put you in the hospital or land you in the cemetery," Graham tells WebMD. "If you are a doctor, there is no earthly reason why you should continue to prescribe Avandia. There are safer alternatives."
Defending Avandia's Safety
GlaxoSmithKline, which makes Avandia, points to its analysis of six clinical trials of Avandia.
That's a case of missing the forest for the trees, says David N. Juurlink, MD, PhD, head of the division of clinical pharmacology at Sunnybrook Health Sciences Center and a scientist at the Institute for Clinical Evaluative Sciences in Toronto, Canada.
"The issue is you have two drugs on the market with identical indications for diabetes, and this increasing body of evidence that one is safer than the other," Juurlink tells WebMD. "Why would a patient want to go on the drug that is less safe and has no advantage?"
In an editorial accompanying the Graham study, Juurlink notes that the American Diabetes Association and its European counterpart have each advised against the use of Avandia.
Switch to Actos?
The Graham study of older people with diabetes finds that, compared to patients taking Actos, patients taking Avandia had:
- 27% higher risk of stroke
- 25% higher risk of heart failure
- 14% higher risk of death
How big a risk is this really? For every 60 older patients who take Avandia instead of Actos for one year, there would be one extra heart attack, heart failure, stroke, or death.
"If you are a patient currently being treated with Avandia, based on the data from our study it would be wise for you to contact your doctor and ask to be switched to a safer alternative," Graham says. "Even if you have been on Avandia for a while and have not had anything bad happen to you, why take those risks? They are completely unnecessary."
Graham is an outspoken safety advocate within the FDA. He stresses that his opinions are his own and not those of the FDA. He and his colleagues researched, wrote, and published the current study themselves. Despite partial funding by the FDA, the federal agency had no role in conducting, designing, or publishing the study.
Heart Attack Risk in Question
Earlier studies have suggested that Avandia may increase patients' risk of heart attacks -- findings that GSK categorically disputes.
The Graham study found no increased risk of heart attack in elderly patients. That, he says, is likely because too few elderly patients survive heart attacks long enough to make it to the hospital.
"We find patients taking Avandia have a higher risk of death than those taking Actos, but there is no way Avandia is causing death from all causes," Graham says. "The leading causes of death in older people are heart attack, stroke, lung disease, dementia, and cancer. It's far more likely that Avandia is causing older people to die more from heart attacks than from cancer or dementia or something else."
But Avandia does indeed increase overall heart attack risk, asserts another paper by Cleveland Clinic researchers Steven E. Nissen, MD, and Kathy Wolski, MPH. Their new analysis of 56 Avandia clinical trials essentially updates their controversial 2007 paper, which ignited a firestorm over Avandia safety.
Now Nissen and Wolski again find Avandia increases the risk of heart attacks. Again, GSK disputes the finding.
What's Next for Avandia?
GSK says it's looking forward to defending Avandia safety at the July 13-14 meeting of an FDA outside-expert advisory panel. The panel's charge: To evaluate Avandia and Actos safety and to make recommendations to the FDA on possible actions.
The Graham study, and the Juurlink editorial, were published online ahead of print on June 28 by the Journal of the American Medical Association. The Nissen paper appears in Archives of Internal Medicine, published online ahead of print on June 28.
Neither Graham, his colleagues, nor Juurlink report funding from or financial interest in GSK or Takeda, which makes Actos. Nissen consults for a number of pharmaceutical companies, including Takeda, through the Cleveland Clinic Center for Clinical Research. He has all honoraria or other fees contributed directly to charity so he derives no tax benefit.