Expert panel review on Alzheimer’s drug aducanumab becomes a one-day trial by fire as critics turn their guns on Biogen — and the FDA
Two days after an enthusiastic FDA insider review of Biogen’s aducanumab provided a $15 billion pop for the big biotech’s market cap, a panel of outside experts provided some fresh fireworks on the debate, with a big majority turning thumbs down on the controversial pitch that there are data to prove the drug is effective — as well as the agency’s perplexing endorsement.
Most of the experts slapped down 3 questions regarding whether there was any evidence to back the efficacy of the drug — with a lone vote in favor on one of the questions.
The money question:
Considering all the trial data, the FDA asked, is it reasonable to consider the positive study 302 “as primary evidence of effectiveness” for the treatment of Alzheimer’s disease?
Ten said no, none said yes and there was one abstention. Several said the FDA presentation was heavily weighted in Biogen’s favor, with no discussion from the agency of the harshly critical statistical review from FDA staffers.
Anyone expecting a dry and dispassionate review of the data for Biogen’s Alzheimer’s drug aducanumab at the FDA panel meeting Friday was in for a rude awakening.

Both Biogen execs as well as Billy Dunn, the head of the agency’s neurosciences division, offered a full-throated endorsement of the drug, outlining the reasons why they found the data backing the controversial application “compelling.”
That argument, though, invited a sharp shove back from members of the advisory panel asked to weigh in on the application, especially from the statisticians who repeatedly probed why Dunn was so supportive when the FDA’s statistical review took the completely opposing view, concluding that the data were conflicting, offered clear evidence that the drug doesn’t work and that Biogen’s argument in its favor was hollow.
“We brought innovative thinking to this unusual situation,” argued Dunn, constructing support for aducanumab that discounted the early termination of the study due to futility, offering reasons why the 1 positive late-stage study offered “extremely persuasive” reasons for an approval, why the other failed study could be explained as providing supporting data and why a small cohort in an earlier study also provided encouragement for an OK — especially as Covid-19 demanded regulators to work cooperatively in salvaging trials.
Several members of the panel were clearly outraged by that.
G. Caleb Alexander, a Johns Hopkins professor, blasted the FDA’s review as “strikingly incongruous” which offered conclusions that were at best compellingly conflicted, raising questions on how the FDA could determine there was evidence of effectiveness.
“It just feels like the audio and the video on the TV are out of sync,” he said.
For every point that suggests support for the drug, he added, there are more that raise concerns.
At several points, Alexander harshly criticized Dunn’s team for skewing the briefing documents and questions in Biogen’s favor. “I’m calling out the fact … that this question is open to eliciting selective information,” he noted at one point.
The University of Washington’s Scott Emerson was another persistent critic with pointed criticism of the drug and the data used to back the application.
And Joel Perlmutter, a professor of neurology, warned the panel that if the FDA approved something without strong evidence of efficacy they would prevent a good and effective treatment from being approved “for more than a couple of years, for many years.”
The friendliest question the panel faced concerned whether trial 302, counted as the only positive study of the 2 Phase III efforts, provided strong evidence of efficacy — if you also ignore 301, the failed trial.
That went down hard, with 8 ‘no’ votes, 1 ‘yes’ and 2 abstentions. Even the abstainers, though, offered objections.
How about the smaller study 103, the Phase II trial Dunn cited for offering some supportive data for the drug? Was it supportive in some way?
The experts shot that one down as well, by a vote of 7 no, 0 yes and 4 uncertain.
The one upbeat result was on pathophysiology, but several noted questions whether the drug’s ability to reduce amyloid would ever translate into efficacy or an impact on symptoms. That earned some support: 5 yes and 6 uncertain.
“I view the question narrowly,” said Chiadi Onyike, saying there was clear evidence it mopped up amyloid beta. And others noted questions whether that was enough to gain clinical efficacy, or if it’s the correct biomarker to use at all.
The FDA doesn’t have to consider the panel vote in making its decision. In this case, given the heated objections, it would be hard to ignore. But given Dunn’s earlier decisions supporting Sarepta, there is still a distinct possibility of an approval.