SILVER SPRING, Md. -- A divided FDA advisory panel voted 9-7 to recommend withdrawing approval of the 17-α hydroxyprogesterone injection (Makena, 17-OHPC) after overwhelming no-confidence votes in its efficacy in preventing preterm birth or clinical benefits on neonatal outcomes.
The FDA's Bone, Reproductive and Urologic Drugs Advisory Committee also voted 13-3 that the findings from the most recent trial did not verify a clinical benefit of Makena on neonatal outcomes, and 16-0 against substantial evidence of effectiveness of Makena in reducing the risk of recurrent preterm birth.
Those members not wanting to see Makena taken immediately off the market preferred that it remain available while the manufacturer conducts a new confirmatory trial. Of note, no panelists voted for a third option offered by FDA staff -- leaving the drug on the market without a new trial. Many members admitted to conflicting feelings over their votes, with a few saying they wanted to abstain (though none did).
But ultimately for many, it came down to the data, and committee members felt that only a new trial could provide the hard scientific evidence for effectiveness of the drug's benefit.
Indeed, Daniel Gillen, PhD, of SLUCare in St. Louis, argued for "increasing the prevalence of truly beneficial drugs" in the marketplace. "Our job is not only to give patients choices, but well-informed choices ... we can stand behind," he said. "The horse is out of the barn on this and the only way we can get it back in ... is to remove that approval."
Dissenting voters were equally passionate, in some cases arguing that removal of the drug was "untenable" and echoing the arguments of the sponsor, AMAG Pharmaceuticals, and several clinicians during the public comment section -- that if the drug was removed, patients would simply use compounded versions of 17-OHPC.
"One of the things I struggle with is tomorrow, I'm going to be seeing patients and I have to give them some guidance," said Michael Lindsay, MD, of Emory University in Atlanta, who voted to keep Makena on the market with a new confirmatory trial. "A randomized controlled trial would be ideal, but in the real world, as clinicians, we don't deal with ideal."
AMAG argued against conducting another confirmatory trial because Makena had been recommended by medical societies as standard of care, and any trial would "lead to bias towards a lower risk population."
Sean Blackwell, MD, of McGovern Medical School at UTHealth in Houston, and lead investigator in the PROLONG study -- which the FDA had required AMAG to conduct as a condition of Makena's original approval -- told the panel that U.S. physicians won't accept another placebo-controlled trial and that there would be difficulty finding a combination of high-risk women with no access to Makena, as well as the "research infrastructure" to conduct a major trial.
AMAG also hammered home the fact that preterm birth is still classified as an orphan disease with no alternative treatment options, which FDA staff acknowledged. But the latter also reiterated skepticism about AMAG's "composite" risk profile, and upon performing their own analyses, found no evidence of treatment effect in the co-primary endpoints (preterm birth prior to 35 weeks gestation and a composite neonatal morbidity index) favoring Makena over placebo.
"I think [withdrawal is] a big step backwards," said Uma Reddy, MD, of Yale School of Medicine in New Haven. "I would only vote for [withdrawal] if there was a safety issue."
Perhaps the most striking moment came from the committee's patient representative Annie Ellis, who not only had an experience with preterm labor herself, but shared the story of her daughter with the same experience. However, she ultimately voted for withdrawal of Makena from the market, and tearfully explained her vote:
"This vote may prevent my own daughter from accessing this drug -- however I got lucky ... [but] mothers and babies shouldn't have to rely on luck. We need data," she said.
The FDA does not have to follow the recommendations of its advisory committees, but it often does.