Avandia, made by GlaxoSmithKline, is an oral drug that makes the body more sensitive to insulin.
But concerns that Avandia causes heart problems has led the American Diabetes Association's treatment guidelines committee to advise against prescribing Avandia in favor of Actos, another drug in the same class with fewer heart-safety concerns -- although both drugs increase a patient's risk of heart failure.
GlaxoSmithKline's RECORD study was supposed to answer these concerns. And according to study leader Philip D. Home, DPhil, of the U.K.'s Newcastle University, it did. Home presented the study findings at this week's meeting of the American Diabetes Association in New Orleans.
"The findings are essentially that in overall cardiovascular terms the drug is safe," Home said at a news conference. "There is no increased or decreased risk of death from heart disease, and that includes the heart failure data."
David Nathan, MD, chairman of the American Diabetes Association guideline committee, said the group would reconsider its recommendations in light of the study findings.
However, the study was unable to determine whether Avandia increases a patient's risk of heart attack. That concern was raised by several experts, including Steven Nissen, MD, chairman of cardiovascular medicine at the Cleveland Clinic.
Nissen remains unconvinced by the final report from Home and colleagues.
"The RECORD trial is seriously flawed," Nissen tells WebMD. "The authors don't reveal the number of patients who were still taking Avandia by the end of the study, but I would estimate this number to approach 50%. Obviously, it is impossible to assess the safety of a drug when patients are not actually taking it."
Home says patients assigned to Avandia treatment took the drug for 88% of the study time. But Nissen says Home's own previously published interim findings do not support this calculation.
Indeed, Home agrees that the study does not answer the question of whether patients taking Avandia have an increased risk of heart attack.
"But what we do know is that this is not associated with cardiovascular death," he said. "There were actually fewer deaths in the [Avandia] group."
In the study, all patients received standard treatment with metformin and/or a sulfonylurea. Half added Avandia to this treatment. The study was not blinded, meaning that study investigators and patients knew which treatment they were getting.
Nissen doubts that this unblinded study will convince experts to change their minds about Avandia. That, he says, will happen only if a new study -- the just-started TIDE study -- shows Avandia is truly safe. The TIDE study is a double-blind trial. And even though it is sponsored by GlaxoSmithKline, there will be a direct comparison of Avandia to Actos, made by Takeda Pharmaceuticals.
Home disagrees, and expects the American Diabetes Association committee to give serious consideration to the new findings, which appear in the June 5 early online edition of The Lancet.
In an editorial accompanying the study, Ravi Retnakaran and Bernard Zinman of Toronto's Mt. Sinai Hospital agree with Nissen that the study's open-label design -- and it's much lower-than-expected rate of cardiovascular deaths -- are problematic.
"Definitive conclusions about the relation between [Avandia] and cardiovascular risk remain elusive, owing to study limitations," Retnakaran and Zinman write. "Furthermore, the findings are inconclusive for [heart attack], for which a non-statistically-significant increased risk was noted in the [Avandia] group."
Unfortunately, a definitive answer isn't soon forthcoming. The TIDE trial isn't scheduled to end until October 2015.
Meanwhile, Retnakaran and Zinman suggest that doctors consider prescribing half doses of Avandia, noting that a half dose offers more than half the benefit of a full dose -- and fewer risks.