乌鸦传媒

BETHESDA, Md. -- In a joint meeting of two FDA advisory committees, members voted 3-21 against recommending approval for stannsoporfin injections for infants who are at risk of contracting severe hyperbilirubinemia.

Across the FDA's Gastrointestinal Drugs and Pediatric Advisory Committees, not one panelist voted in favor of recommending approval for stannsoporfin, a tin-based heme oxygenase inhibitor, without a Risk Evaluation and Mitigation Strategy (REMS).

However, three members favored approval with a REMS, and 21 members were against approval with or without a REMS on the basis of the drug's risk-benefit balance.

"There are many unanswered questions" in the drug's trial data, said Wael Sayej, MD, a pediatric gastroenterologist at the Ochsner Health System and Clinical School in New Orleans.

"The sample size is too small ... to calculate a number needed to treat. In addition to that, to calculate a number needed to harm is also even more difficult," Sayej added.

“There’s probably a very narrow population who could benefit from this drug,” said K. Sarah Hoehn, MD, MBe, a director of the Pediatric Palliative Care and Comfort Team at the University of Chicago Comer Children’s Hospital.

Hoehn said her vote in favor of stannsoporfin with a REMS was contingent on restricting the drug to infants who first fail phototherapy and to use in Level 3 NICUs. In that way, she proposed, the drug might give parents an option to avoid exchange transfusion.

One challenge highlighted by panelists is that while hyperbilirubinemia is common, the implied threat sponsors would hope to avoid, severe bilirubinemia encephalopathy or kernicterus, is relatively rare.

Severe hyperbilirubinemia affects only 7 to 40 newborns per 100,000 live births, noted Stephanie Omokaro, MD, FDA's lead medical officer in the Office of Drug Evaluation III.

Another obstacle for the committee was that while the FDA had received two investigational new drug applications related to clinical testing, only one was deemed "compliant with good clinical practice standards."

Preliminary data from pooled long-term extension studies showed "a numerically higher rate of both speech and hearing adverse events" in children treated with stannsoporfin as neonates compared to those treated with phototherapy alone, which worried a good portion of the joint panel's members.

So the core question for the committee, even with its limited data, was whether the threat of potential brain damage or other adverse events due to severe bilirubinemia exceeded similar potential harms posed by the drug itself.

In a second vote of 6-17, the panel, with one abstention, found the sponsor did not provide "substantial and persuasive evidence of effectiveness" for using stannsoporfin as an adjunct to phototherapy in neonates who are 35 weeks of age or older and show evidence of hemolysis and hyperbilirubinemia -- the indication, the drug's sponsor, InfaCare Pharmaceutical Corporation (a Mallinckrodt subsidiary), was seeking.

On a question related to preferred dosage, the committee did not achieve consensus.

InfaCare had sought an indication for treating "neonates greater than or equal to 35 weeks of gestational age" who show signs of of hemolysis-rupturing of red blood cells -- and are at risk of developing severe hyperbilirubinemia or jaundice. They proposed a dose of 4.5 mg/kg.

Hyperbilirubinemia is a common condition in newborns, manifesting as jaundice and seen at some level in more than 80% of all neonates by some estimates. In about 8%-11% of infants the condition is severe enough to warrant treatment, briefing documents noted, and these cases are "the most common cause of hospital readmission in the neonatal period."

FDA staff noted that the agency generally requires two adequate clinical studies. However a single trial can support approval when there is "data from a large multicenter study, internal consistency across study subsets, evidence of an effect on multiple endpoints evaluating different events, [or] statistically very persuasive findings."

In addition to assessing the drug's overall risk-benefit profile and determining whether the applicant provided "substantial and persuasive evidence of effectiveness" as an adjunct to phototherapy, the joint committees also discussed whether the "long-term and short-term" safety profile of the drugs support approval; and whether the given data on long-term safety assessments were adequate to characterize the potential risk of adverse neurodevelopmental outcomes related to the drug

On each of these points, a majority of the committee responded in ways that did not favor of the sponsor.

In a separate discussion of the “clinical meaningfulness” of the sponsor’s primary endpoint in its pivotal study— a reduction in total serum bilirubin (TSB) at 48 hours post-treatment with stannsoporfin compared with the TSB levels in placebo— a few members voiced skepticism.

Two members — Thomas Newman, MD, MPH, of the University of California San Francisco, and P. Brian Smith, MD, of Duke University Medical Center in Durham, N.C. — wondered aloud whether the sponsors might have created an “artificially high rate of rebound” in the placebo group.

They suggested that neonates who received phototherapy may have seen a subsequent spike in bilirubin because phototherapy was stopped “sooner than most [clinicians] would,” in Newman’s words; or, Smith said, because a second phototherapy light wasn’t added.

In that case, the primary endpoint is not just “clinically meaningless,” said Peter Havens, MD, of the Medical College of Wisconsin in Milwaukee, ”it’s potentially misleading.”

While Omokaro pointed out that there was no prespecification to add additional lights, the FDA’s Julie Beitz, MD, also noted that the sponsor’s protocol specified using 30 microwatts of light, which is the standard for intensive phototherapy in the American Academy of Pediatrics’ guideline.

“So, I’m not certain that we could say that that placebo group did not get the standard of care,” Beitz said.

The committee was also asked by FDA to discuss a potential REMS study (if the drug were approved) or other additional studies that the sponsor might pursue.

A REMS would be “an end-run around a phase III study,” said Havens. “If the sponsor wants to continue with drug development, they should do it in a standard way, which is a phase III study… and spend the money instead of hoping the healthcare system and parents will undertake the burden.”

Kelly Wade, MD, PHD, of the Children’s Hospital of Philadelphia, who voted against approval, suggested soliciting feedback from parents. A core argument in public testimony, made by clinicians and mothers, as well as the sponsor in its presentation, is that the drug shortens the duration of phototherapy by roughly 10 hours, giving mothers more immediate mother and infant bonding time.

One question she suggested the sponsor ask of parents: “How much risk of hearing loss, speech, [and] neurodevelopment, are you willing to take for shortened hospitalization and improved time for bonding?”

In a press statement issued Thursday, Mallinckrodt said it appreciated the committees’ review and that it will continue to work with the FDA as the review process continues.

“Given the outcome of today’s meeting, the company is evaluating alternatives for this development program,” the statement said.

Comment