CardioExchange Archive

After two days of deliberation an FDA advisory panel has recommended that the severe restrictions (REMS) placed on rosiglitazone (Avandia, GlaxoSmithKline [GSK]) be modified. The vote constitutes a modest revival in fortune for the embattled drug and its maker after many years of controversy and bad news.

Only 5 panel members voted to continue the current severe restrictions. One member (the consumer representative) voted for the withdrawal of rosiglitazone. By contrast, 20 members voted to either modify (13 votes) or remove (7 votes) the REMS. But the vote should not be interpreted as a broad statement by the panel that rosiglitazone is safe. Panel members made clear that they still had concerns about the safety of the drug. But their concerns were less pressing than in the past.

The key to restoring at least some of rosiglitazone’s tarnished reputations was the re-adjudication of the RECORD trial by the Duke Clinical Research Institute (DCRI). Despite an attempt by Tom Marciniak, an FDA reviewer acting on his own initiative, to question the reliability of the re-adjudication, the rest of the FDA and the panel members all expressed praise for DCRI’s work. The panel endorsed the FDA’s perspective that the re-adjudication of the trial had been well performed, thus confirming the integrity of the original analysis.

But it would be a mistake to conclude that assurance of the integrity of the RECORD data leads to the conclusion that rosiglitazone is safe. Over the course of the two-day meeting, panel members continually stated their belief that the underlying flaws of RECORD — especially its open-label, noninferiority design and a large amount of missing data — ensured that it could never definitively dispel the cloud of uncertainty over rosiglitazone’s safety. DCRI’s Ken Mahaffey, who was in charge of the re-adjudication, conceded under questioning that although no significant misconduct was observed in the trial, RECORD did not provide strong evidence demonstrating the safety of rosiglitazone. But the DCRI analysis did allay the 2010 fears that RECORD may have been even more severely compromised due to misconduct, incompetence, or fraud.

On the second day of the meeting, panel members spent a significant amount of time considering the impact of the current rosiglitazone REMS, which resulted in a greater than 99% reduction in the use of the drug in the U.S., and asked the FDA about possible modifications of the severe REMS. It was clear that most members wanted to see rosiglitazone become more easily available though they seemed wary about removing all restrictions. The general sentiment of the committee was that the restrictions on rosiglitazone should be loosened but not entirely removed. But the committee could reach no agreement about how to achieve this goal.

Conspicuous by his absence at the panel meeting was Steve Nissen, whose original meta-analysis helped ignite the Avandia controversy and who was a vocal  presence at the 2010 advisory panel meeting. Nissen had earlier announced that the FDA had not invited him to present his results to the panel (except as a speaker during the public access portion of the meeting). Nissen provided the following statement:

Little practical impact. Despite maximal efforts by FDA to convince the panel to remove restrictions, the majority voted to continue restricted access. If FDA loosens the requirements, a few additional patients may receive the drug, but effectively, Avandia is gone, which is a good thing.

I disagree with the findings regarding RECORD. Duke used the dossiers supplied by GSK to re-adjudicate. It was a foregone conclusion that they would get a similar answer.

We have three sources of data regarding Avandia. The meta-analyses, include FDA’s, show an increased risk of heart attack ranging from 40 to 80%. The observational data virtually all show an increased hazard for MI. The outlier is RECORD and it was a fatally flawed trial, completely unblinded to patients, physicians, and GSK.

I’m relieved that Avandia will remain restricted and continue to doubt the wisdom of spending taxpayer dollars to try to resurrect this drug.”

GSK issued its own statement. James Shannon, GSK’s Chief Medical Officer, said:

“We appreciate the committee’s thorough examination of the RECORD results and will continue to work with the FDA as it considers the recommendation of the committee. We continue to believe that Avandia is a safe and effective treatment option for type 2 diabetes when used for the appropriate patient and in accordance with labelling.”