WebMD Newsroom

“Malfeasance” was the last word in yesterday’s report from the FDA expert advisory committee’s two-day meeting to decide what to do about the GlaxoSmithKline diabetes drug Avandia.

That word took on special resonance with the New York Times report that GSK’s U.S. forebear, SmithKlineBeecham, knew of Avandia’s extra heart risks in 1999 — the year the drug was approved — and hid that information from patients and from the FDA. That allegation will not be discussed at today’s meeting, but it casts a shadow over the proceedings.

Tuesday’s meeting featured detailed arguments from FDA safety officers — particularly David Graham, MD — who pointed to observational data, particularly a large study of Medicare patients taking Avandia or Actos (a similar diabetes drug apparently without the heart issues that plague Avandia). Graham’s study suggested that Avandia is a much more dangerous choice.

But as a counterpoint, FDA drug evaluation officers went over clinical trial data in sometimes excruciating detail, arguing that their earlier recommendation to approve Avandia was and still is the right choice. Prominent among them was Karen Mahoney, MD, who offered a vigorous defense of data supporting GSK’s position that Avandia is safe. Mahoney also criticized the Graham study.

And so it went, back and forth, as FDA researchers demonstrated the vivid internal split not only over Avandia, but over how the agency should evaluate drug safety. Given that drug safety is the core of FDA’s public mission, this panel meeting is about far more than Avandia. It’s about the FDA’s soul.

The panel won’t just decide whether to ban Avandia. They’ll also decide whether the TIDE trial comparing Avandia to Actos should be allowed to continue.

TIDE study leader Hertzel Gerstein, MD, of Canada’s McMaster University, passionately defended the study. It needed defending, because  a string of experts, particularly Graham and  Steven Nissen, MD, savaged the trial as one seeking to give half its patients a drug suspected of being harmful. But Gerstein argued that the drugs possible benefit or harm never will be known if the study is stopped.

Researchers from the respected Institute of Medicine, who will address the panel today, have indicated that the trial may indeed not be ethical.

Gerstein admitted that TIDE results will not be available until 2016  or maybe 2015, well after GSK’s Avandia patent expires.  Is is worth waiting that long?

Dean Follman, PhD, of the NIH, told the committee that what’s now known about Avandia, and even about Actos, is incomplete. He suggested that only a clinical trial like TIDE could fully answer the questions the panel is today being asked to decide.

But panelist Ruth Day, PhD, of Duke University, sharply questioned whether the highly selected patients in a clinical trial really represent real-world patients better than the somewhat messy observational studies Follman discounted.

And that is the question. Should the FDA panel act on incomplete evidence? Or should it wait until more information is available years from now? And how should the FDA act in the future when considering new drug approval?

These are not academic questions. The lives of real patients, with real illnesses today as well as tomorrow, hang in the balance.