Skip to Main Content

Following the Food and Drug Administration’s polarizing authorization of the Alzheimer’s therapy Aduhelm on Monday, a member of an agency advisory committee that recommended against the drug’s approval has resigned.

Neurologist Joel Perlmutter of Washington University in St. Louis, a member of the FDA’s expert panel for nervous system therapies, told STAT in an email that he had quit the committee on Monday “due to this ruling by the FDA without further discussion with our advisory committee.”

The advisory committee, which convened in November, couldn’t have been more openly skeptical of the drug, also known as aducanumab. Ten of the 11 panelists found that there was not enough evidence to show it could slow cognitive decline. The 11th voted “uncertain.”


But the FDA still approved the treatment on Monday. On top of the potentially massive implications for patients, clinicians, and health care spending, the decision also raised questions about the role of the advisory committees — and what it meant that the agency, in its final adjudication, bucked the very panel it had convened.

“This isn’t the first time when I was on a committee where the committee voted one way and the FDA decided another,” said biostatistician Scott Emerson, a professor emeritus at the University of Washington, who has served on many advisory committees in different disease areas. “This was the first time that nobody voted for approval of this drug — nobody — and they went against that.”


The FDA does not have to follow its advisory committees’ recommendations — one study found the agency went against the experts 21% of the time from 2008 to 2015 — but those overrulings generally came when the votes were closer. The FDA often convenes the panels when there is uncertainty around whether or how to approve new products, and there are different “ad comms” for different types of medications.

With Aduhelm, the story is also more complicated than the FDA just rebuking the expert panel. The FDA granted the Biogen therapy what’s called an accelerated approval, based not on firm evidence the drug slowed cognitive decline — which even the FDA acknowledged was not clear — but on a “surrogate endpoint” that the therapy cleared toxic protein plaques in patients’ brains. In the FDA’s view, that finding “is reasonably likely to predict a clinical benefit to patients.” It also said Biogen would have to run another trial to confirm a clinical benefit, though the results of that aren’t due for nearly another decade.

But during the advisory committee meeting last November, FDA officials explicitly said they were not considering approving aducanumab based on a surrogate endpoint. Billy Dunn, the director of the FDA’s Office of Neuroscience, made it clear: “We’re not using the amyloid as a surrogate for efficacy.”  That meant the panelists were not asked to consider that possibility.

So on Monday when the agency announced it had indeed given Aduhelm the OK based on amyloid levels as a surrogate endpoint, it left experts on the panel scratching their heads. Trials of other amyloid-busting therapies have found that they didn’t result in clinical benefits.

What had happened, Dunn explained in a letter to the chair of the panel Monday, was that after the panel hearing, there were “further discussions” within the agency that “raised further consideration of the accelerated approval pathway.”

“We recognize that there has been tremendous public interest in aducanumab and differing viewpoints on the extensive and complicated data supporting the application for aducanumab,” Dunn wrote. “Our discussions leading up to the decision to grant an accelerated approval for aducanumab considered a wide range of views, both external and internal to FDA. We appreciate the comments from the advisory committee members and can assure you that we listened carefully and viewed the meeting proceedings as an important source of input as we discussed the appropriate action.”

To Aaron Kesselheim, an advisory committee member and the director of Brigham and Women’s Hospital’s Program on Regulation, Therapeutics, and Law, Aduhelm’s approval didn’t just set “a dangerous precedent” for what kind of evidence an Alzheimer’s therapy would need to show to get the green light, “but even more broadly for the idea that a company can turn around and at the last minute seek [accelerated approval] when their primary clinical endpoints in their trials don’t reach the level needed for FDA approval,” he told STAT in an email.

It’s no surprise that some members of the panel denounced the FDA’s approval, and not simply because of their votes. During the November hearing, some of them criticized how the data was being discussed and the evidence Biogen had submitted. Earlier this year, Emerson, Kesselheim, and a third panelist published a paper in JAMA outlining what they saw as the flaws of the therapy.

Still, Emerson, the biostatistician, said he would serve on another advisory committee if asked. (The panels have some permanent members, but for individual hearings, they invite “temporary members,” which included Emerson in the case of aducanumab.)

It is important, Emerson said, for outside experts to have a voice in the decision-making process.

Matthew Herper contributed reporting. 

Create a display name to comment

This name will appear with your comment