A divided FDA advisory panel voted 13-10 to recommend the oral antiviral molnupiravir for emergency use authorization (EUA) for adults at high risk of progressing to severe COVID-19 on Tuesday.
While votes at the Antimicrobial Drugs Advisory Committee (AMDAC) were largely split, similar questions were raised on both sides of the vote about modest efficacy, especially in light of other available treatments. Very few committee members offered strong "yes" or "no" votes, as most were in the middle. Ultimately, the positive votes felt that despite a number of questions, the benefits of the drug outweighed the risk.
The committee went half an hour over its scheduled time, which AMDAC chair, Lindsey Baden, MD, of Brigham and Women's Hospital in Boston, said was a first in his years of chairing the committee, "which speaks to the complex issues we had to deal with," he added. Indeed, most members when discussing their vote prefaced their comments with how difficult a decision this was for them.
"I agree with all that's been said by the 'yes' and 'no' voters," said Baden, who voted yes. "There are many, many more questions than answers. However ... overall, I trust our practitioners that if we educate them properly, we can deploy this properly."
Of particular concern to the "no" voters was not only that efficacy against hospitalization and death declined from the interim analysis submitted to the FDA (48% relative risk reduction) to the final analysis (30% reduction), but that when examining only the data from the post-interim analysis enrollment, there were fewer placebo patients who were hospitalized or died by day 29 versus patients receiving the intervention (4.7% vs 6.2%, respectively).
Baden said that spoke to the drug having a benefit in the "right population" in the "right time," but others weren't so convinced.
"I was struck by a modest benefit in a highly adherent trial population and then the unclear efficacy in the latter half of the trial when we had increased circulation of the Delta variant," said Daniel Horton, MD, of Rutgers School of Public Health, who voted no. "I think there's the potential for increased pressures for viral evolution in the setting of lower adherence in the real world."
However, the lack of options in the real world is what propelled many members to vote yes.
"There's a potential concern for a lack of availability for an alternative therapy for those at high-risk, perhaps considering the loss of efficacy with monoclonal antibodies with variants," said Michael Green, MD, of University of Pittsburgh School of Medicine, who voted yes.
Interestingly, even those who voted yes argued that it may only be temporary, given other similar drugs in the pipeline that may have better safety and efficacy profiles.
"If another medication becomes available under EUA, this [EUA] should be revisited and have the potential to be withdrawn," said Sascha Dublin, MD, PhD, of Kaiser Permanente Washington Health Research Institute, who voted yes.
Safety was of chief concern among many committee members. While the committee did not vote on the subject, they discussed how the drug should be handled among pregnant women, given that data from preclinical studies appeared to show embryo-fetal toxicity.
The consensus appeared to be that the drug should not be recommended for pregnant women, but that healthcare providers didn't have the right to deny a woman a drug if the benefits would outweigh the risks, based on shared clinical decision-making.
"As a woman of childbearing age, I don't think I'd want to take this drug," said consumer representative Roblena Walker, PhD, of the public health non-profit EMAGAHA, in Georgia, though she did vote yes.
Others suggested a negative pregnancy test prior to administering the drug to a woman of child-bearing age, as well as a pregnancy registry to track potential adverse events. Nearly all agreed that it should not be given to a woman in the first trimester.
For other committee members, potential mutagenicity was an issue, which the committee also discussed, but did not vote on. Baden said that "in the face of efficacy, the real risk is in prolonged replication ... rather than short-term replication." He added that "optimal sequencing and sampling" would be crucial to detect mutations.
However, Peter Weina, MD, of the Defense Health Agency, pointed to a number needed to treat of 34, which "means potentially a large amount of virus is going to be exposed to the drug."
"It plays into the question surrounding mutagenicity of the spike protein, and the potential of creating a new variant," said Weina, who voted no.
James Hildreth, MD, PhD, of Meharry Medical College in Nashville, was unequivocal: "It's an easy vote for me to vote no. Genotoxicity and mutagenicity data raise more questions than answers," he said.
Committee members agreed that the criteria for a "high-risk" individual should adhere closely to how patients were defined in the clinical trial, with definite benefit shown among unvaccinated older adults and those with obesity, although it was possible that those with "suboptimal response to vaccination" would benefit from the drug too. More study was needed, they said.
The FDA does not have to follow the advice of its advisory committees, but it often does.