FDA Advisory Panel Votes Would Allow Avandia Sales

En español | An advisory panel to the U.S. Food and Drug Administration (FDA) today cast a contentious vote to allow blockbuster diabetes drug Avandia to remain on the market. But the vote was split, and overwhelmingly negative in its judgment about Avandia’s risks, with 12 voting to withdraw the drug from the market altogether, 10 voting for it to be sold only with additional warnings and use restrictions, seven calling for some added warnings, and only three voting for no changes. One panel member abstained.

The FDA is not legally bound to adopt the recommendation, but typically does.

The recommendation comes after 11 years of brisk global sales of Avandia—including millions of U.S. prescriptions—earning well into the billions for its maker, GlaxoSmithKline (GSK). The panel decision also comes a month after a study in the Journal of the American Medical Association associated the drug with a significant number of cases of heart failure, heart attack, stroke, and death among diabetics taking it. Several previous studies have reached similar conclusions.

The 33-member panel of experts, which combines the FDA’s Endocrinologic and Metabolic Drugs Advisory Committee and its Drug Safety and Risk Management Advisory Committee, met all day Tuesday and Wednesday to hear presentations from FDA researchers and other doctors and scientists, as well as from GSK, on the cardiovascular safety of Avandia.

Mounting furor

Avandia is used to lower blood sugar in people with type 2 diabetes. It’s usually prescribed as an add-on to traditional diabetes drugs, such as metformin. Indeed, a committee of diabetes experts in 2008 issued guidelines excluding Avandia, known generically as rosiglitazone, from any role in treating the disease. If called for, the committee said, doctors should prescribe Actos, another drug in the same class that has not been linked to cardiovascular disease.

The three years leading up to today’s climactic decision have seen mounting furor over Avandia’s risk profile. GSK has vigorously defended its safety. Some experts have insisted the data on heart risks are inconclusive and the drug should remain available. Others have cried foul at both the drugmaker and the FDA for failing to make patient safety a priority.

When it all started

The first salvo in this debate came in 2007, when Cleveland Clinic cardiologist Steven E. Nissen, M.D., published a meta-analysis—a statistical look at 42 drug trials taken together—that found a sharply increased rate of heart attack among people taking Avandia.

Largely in response to those findings, a joint FDA advisory committee later that year recommended adding heart risks to Avandia’s label, but voted almost unanimously—22 to 1—to keep the drug on the market.

Meanwhile the drug’s maker mounted a publicity campaign to counter concerns about Avandia, whose sales had plunged. Among other things, the company quickly published interim findings from its own ongoing trial comparing Avandia with other standard regimens, results that largely exonerated its drug of causing heart problems and strokes.

Critical studies

But 2010 has seen a steady drumbeat of official reports and medical journal articles casting further doubt on the drug and criticizing its defenders.

In late July, Nissen published a second meta-analysis examining results from 56 trials—and again, it showed Avandia users faced an increased heart attack risk. In the same week, outspoken FDA scientist David J. Graham, M.D., published a review of medical records from 227,571 Medicare beneficiaries who took Avandia or Actos from July 2006 to June 2009.

His analysis demonstrated a higher incidence of stroke, heart failure and death among Avandia patients.

A couple of recently published studies, however, would seem to provide reassurance on this score. One was an after-the-fact analysis of results of a study, presented last month at the annual meeting of the American Diabetes Association, which actually found a lower incidence of heart attack and stroke among Avandia patients. However, the original study was designed to look at a different set of questions.