FDA officials divided on Avandia safety as panel meeting looms

Last week, a new analysis of Avandia trial data yielded positive results for the restricted drug. Now, the FDA's staffers have spoken. Ahead of this week's advisory panel meeting, staff reviewers backed a readjudication of the controversial RECORD trial, which found that the GlaxoSmithKline ($GSK) diabetes drug does not increase cardiovascular death risks.

As advisory committee members sift through the latest assessments this week, they will be considering whether to lift heavy restrictions placed on Avandia the last time RECORD had an airing. Since the FDA limits went into effect, Avandia's sales have plummeted to just $9.5 million last year.

After the 2010 restrictions went into force, the FDA required Glaxo to bring in independent researchers to take another look at RECORD. Duke Clinical Research Institute (DCRI) scientists did so and concluded that the study's principal findings--that Avandia didn't present a higher risk of heart attack or cardiac death--were solid.

And now, FDA staffers have examined the Duke researchers' results. Some of the trial's original limitations couldn't be addressed by DCRI's after-the-fact analysis--the fact that it was an open-label study, for instance. Still, the readjudication dispensed of some major concerns about RECORD, the staff reviewers said, including allegations that researchers incorrectly interpreted adverse events in that trial.

But not all of the FDA officials who reviewed the analysis are on board with the findings. Staff scientist Thomas Marciniak contends "the readjudication is not independent," citing design flaws carried over from previous trials and GSK's financial responsibility for the Duke team's research. The memo also contains a previous review of the Duke readjudication, signed by Norman Stockbridge, the FDA's cardio division head, among others. That review asserts there is "no amount of analytical rigor that can compensate for a weak trial design" and emphasizes that "RECORD's design irreparably hampers its ability to characterize definitively the CV risk" of the drug.

The topic of Avandia's safety has obviously been a heated one for some time. The FDA's decision to revisit the case is no different. It has drawn heavy criticism from those who believe the drug is dangerous. Among the most outspoken has been Dr. Steven Nissen of the Cleveland Clinic, whose independent meta-analysis in 2007 showed the drug increased heart attack risk. Nissen recently accused the FDA of reopening the case to absolve itself of responsibility for deaths he believes were caused by Avandia. 

But the FDA is stressing its impartiality. The memo acknowledges the "strongly held views" on the drug and notes that some of these views are represented in the memo and review documents. Some, it says, may also be expressed during presentations and comments at the meeting. This time around, the panel comprises new members and is absent of those holding some of the strongest views--including Nissen, who claims he was also denied the right to present at the meeting. Still, the WSJ notes, two former panelists who voted for Avandia's removal and three who voted for severe restrictions are part of this week's 28-person panel.

A new determination on Avandia could be lucrative for GSK, as Avandia once brought in $3 billion a year. Glaxo emphasizes it has not asked for more freedom to market Avandia but says it will work to make the drug more available if the FDA decides to open up its use. Will Avandia get that reprieve? Only time, the advisory committee and, ultimately, the FDA can tell.

- see the memo (PDF)
- get the WSJ story (sub. req.)
- read Bloomberg's take
- get more from Forbes

Suggested Articles

The future may be uncertain for AZ’s Imfinzi in first-line lung cancer, but its targeted med Tagrisso now boasts a green light in that setting.

Ultragenyx is back with another FDA nod, this time for Crysvita to treat X-linked hypophosphatemia in patients one year and older.

Roche got a two pieces of good Hemlibra news early this week—and what's good for Hemlibra must be good for Roche.