Whether Avandia Causes Heart Attacks Remains An Open Issue, Despite Recent RECORD Study Results

(Posted by Tom Lamb at DrugInjuryWatch.com)

On June 5, 2009, at the annual meeting of the American Diabetes Association (ADA) in New Orleans, GlaxoSmithKline announced the results of a much-anticipated study regarding its diabetes drug Avandia (rosiglitazone).

As background, from a June 5, 2009 news article "ADA: Rosiglitazone Increases Risk of Heart Failure, but Not Mortality", published online by MedPage Today:

The results of the Rosiglitazone Evaluated for Cardiac Outcomes and Regulation of Glycemia in Diabetes (RECORD) trial came from an assessment of 4,447 patients with type 2 diabetes at 338 centers in 23 countries in Europe, Australia, and New Zealand who had five to seven years of follow-up.

The study was warranted in 1999 after the thiazolidinedione class [e.g., Avandia and Actos] had been thought to cause fluid retention and possibly heart failure in diabetes patients.

According to that same article, Philip D. Home, M.D., Ph.D., of Newcastle University in England and chairman of the RECORD steering committee, told a press briefing at the June 2009 ADA meeting:

"The heart failure answer hasn't changed," Dr. Home said, referencing interim results from the trial that were released two years ago. "We confirmed that there is an increase in heart failure — a double increase with rosiglitazone compared to the control [group]."

"But in terms of overall cardiovascular risk, the drug is safe," Dr. Home said, noting a risk ratio of one between the two study groups. "There's no increased risk, no decreased risk. And that includes the heart failure element."

Soon after the RECORD study results were presented at the ADA meeting — and published online by The Lancet— Steven E. Nissen, M.D., of the Cleveland Clinic, weighed in with his comments and observations. 

First, you may recall a June 2007 New England Journal of Medicine article, "Effect of rosiglitazone on the risk of myocardial infarction and death from cardiovascular causes", in which Dr. Nissen and Kathi Wolski, M.P.H. reported a meta-analysis of 42 clinical trials involving 27,847 patients which essentially started the ongoing controversy about the safety of Avandia.  They concluded that treatment with Avandia was associated with an approximately 43% greater risk for myocardial infarction and an approximately 64% greater risk for cardiovascular death than placebo or other antidiabetic regimens.

Returning to the June 2009 MedPage Today article and what Nissen had to say about the RECORD study:

"At the time of publication of the interim analysis in 2007, approximately 30% of patients in the trial originally assigned to receive [rosiglitazone] were no longer taking the drug" [Nissen said]….

"In The Lancet manuscript, the authors don't reveal the number of patients who were still taking Avandia by the end of the study," he said. "Obviously, it is impossible to assess the safety of a drug when patients are not actually taking it."

Due to this missing key data, for Dr. Nissen and others the issue of whether Avandia is associated with an increased risk of heart attacks, or myocardial infarction, is still unresolved, today.

Matthew Herper, of Forbes, focuses on the point Dr. Nissen makes about this missing data in his June 5, 2009 report, "Glaxo Fails To Learn Lesson of Avandia":

Nissen argues that in a drug safety study it is also necessary to look at just the patients who were taking the drug the whole time. But the dropout rates were not included in the Lancet paper.

"That's a problem," says Bernard Zinman, a researcher at the Samuel Lunenfeld Research Institute of Mount Sinai Hospital in Toronto who wrote an editorial on the study in The Lancet. The study "doesn't tell us anything" about Avandia's propensity for causing heart attacks, Zinman says….

[W]hen Glaxo published an earlier analysis of RECORD in 2007, before the trial was completed, that interim paper in the New England Journal of Medicine took great care to mention how many patients had stopped taking Avandia and other drugs.

… 27% of the patients had stopped taking an assigned drug in the Avandia arm and 33% had stopped taking an assigned drug in the control group.

That's key information because it means side effects could be washed out by the number of patients who were included in the final tally but weren't on the drug.

But these numbers, which were present in the New England Journal, are missing from the final publication in The Lancet. The Lancet did not return an e-mailed request for comment.

GlaxoSmithKline provided the figures at Forbes' request. At the end of the study, 1,344 patients 61% of the Avandia group were still on the drug, while 1,131 patients, or 51% of the control group, remained on their medicines.

Those numbers are not surprising or unusual for a study where patients were being followed for more than five years, says Mary Anne Rhyne, a Glaxo spokeswoman.

Rhyne said she could not explain why those numbers were not in The Lancet paper because the paper was written by the outside researchers running the trial, not Glaxo.

The RECORD study was funded by GlaxoSmithKline.

I am sure we will be hearing more from Dr. Home, Dr. Nissen, and others about what the RECORD study shows about the safety of Avandia, and what it does not.