Retsef Levi Explains His Vote against Routine RSV Monoclonal Use in Newborns

Retsef Levi Explains His Vote against Routine RSV Monoclonal Use in Newborns
by Maryanne Demasi at Brownstone Institute

Last week’s meeting of the CDC’s vaccine advisory panel was unlike any in recent memory.

It was the first full session of the new-look Advisory Committee on Immunization Practices (ACIP)—a panel now entirely handpicked by Health Secretary Robert F. Kennedy, Jr., and tasked with restoring public trust in vaccine policy.

Over two days, the committee tackled thorny topics such as Covid-19 vaccine safety, thimerosal in flu shots, and preparations for future infectious disease threats.

But it was a single vote—on whether to recommend a newly approved product called clesrovimab—that split the committee.

Clesrovimab, marketed as Enflonsia, is not a vaccine. It’s a long-acting monoclonal antibody designed to prevent severe RSV illness in newborns during their first RSV season.

The CDC proposed offering it to all infants under eight months of age whose mothers hadn’t received an RSV vaccine late in pregnancy.

Two members of the advisory panel—Dr Retsef Levi and Dr Vicky Pebsworth—voted ‘no.’

Levi is no stranger to weighing risk. A professor at MIT with deep experience in data analytics and risk-based decision-making, he had combed through five clinical trials of RSV monoclonals, including clesrovimab and its predecessor, nirsevimab.

What he found gave him pause—in every trial, more infant deaths occurred in the treatment groups than in the control arms.

Levi acknowledged that the trials were small and not powered to detect ‘statistical’ differences in mortality—but the consistency of the imbalance was difficult to ignore.

Among the deaths were rare cases of gastroenteritis in otherwise healthy babies—events that defied easy explanation.

“Four different trials all show deaths going in the same direction,” he told the meeting.“Should we not be concerned that maybe there are some potential safety signals?”

What unsettled him further was that the CDC had not presented this data to the committee.

Shortly before the vote, Levi spoke not just as a scientist, but as a father. “I’m a scientist but I’m also a father of six children…If I had a premature baby, I might consider it. But giving this to a healthy newborn? I would be concerned to use that.”

The rest of the committee voted in favour of recommending the product.

In the days following the meeting, co-chair of the committee Dr Robert Malone, elaborated on his reasoning in a public post.

He acknowledged the risks of RSV in infants are low—affecting about 2 to 3 percent—but said most severe cases occur in babies with no known risk factors. That makes it impossible to predict who will be hit hardest.

“This may seem like a small number—unless your baby is one of them,” Malone wrote.

“The risks of the antibody product injections are much lower than the risk of your otherwise healthy baby being one of the unfortunate few that happens to develop severe RSV disease,” he added.

To him, the broad recommendation was a pragmatic response to an unpredictable illness. With no way to know which infants might become critically ill, a single, long-acting dose before RSV season offered a measure of protection.

Levi, however, saw it differently.

He didn’t feel the FDA, CDC, or Merck—the product’s manufacturer—had adequately addressed the trial imbalances. The findings weren’t statistically conclusive, but in Levi’s view, they were too consistent to dismiss.

Instead, Levi advocated for a precautionary approach for an intervention aimed at healthy full-term newborns.

In this exclusive interview, Levi breaks down the data that shaped his dissenting vote, and reflects on the responsibility that comes with going against the majority.

👉 Read the full interview with Dr Retsef Levi – edited for brevity.

DEMASI: Is it difficult to vote against your fellow panel members when the chair calls on you all to vote one by one?

LEVI: I don’t think about it as voting against my fellow panel members. Each of us is there to make the best decision we can based on our own scientific understanding and judgment. I actually believe in the value of debate—it’s healthy for a functional team. We don’t need to agree all the time. In fact, disagreement is important.

And I accept that, as part of a team, sometimes the decision won’t be what you personally think it should be. But in the long run, I really believe that a functional team that debates and brings different opinions to the table—that’s actually a winning strategy.

DEMASI: So you don’t think it puts pressure on individuals to conform to how the other panel members are voting?

LEVI: I think you’re raising a very good point. In functional teams, that should not happen—there shouldn’t be pressure. Because otherwise, you end up with groupthink, right? And that’s a very bad thing. When you look at some of the major failures of organisations and systems, they stem from groupthink. So I think a functional, high-performing team, with the right culture, should allow people to disagree—and still work together.

DEMASI: Give me a glimpse behind the scenes. Do you go into a meeting knowing how each of you will vote? Do you discuss how you’re going to vote before the public meeting?

LEVI: We certainly discuss the issues—but we focus on the facts, the data. We don’t discuss things with the goal of making a final decision beforehand. And we’re definitely not trying to arrive at consensus behind closed doors and then put on a show at the public meeting. In fact, to be honest, I wasn’t sure how my colleagues were going to vote.

DEMASI: Do you think you could enter a meeting thinking you’ll vote one way, and then get swayed by the evidence on the day and change your vote at the last minute?

LEVI: Absolutely. Because the discussion isn’t just between the ACIP members. All of us participate in the discussion with the staff and researchers from the CDC, the FDA, and many other organisations that aren’t voting. They bring opinions and data to the table that could potentially impact our vote.

DEMASI: Can you explain why you diverged from most of your fellow panel members on the RSV monoclonal issue?

LEVI: So let me step back. When deciding whether to recommend a product, you have to look first at the benefits. And in this case, it’s really important to remember that we’re talking about giving this to healthy babies—not high-risk babies. These are not terminal cancer patients. These are healthy babies. So the benefits have to clearly outweigh the harms.

DEMASI: Right, got it.

LEVI: The benefits were presented to us as a reduction in RSV-related hospitalisations. Which sounds important—no one wants babies in hospital—but the hospitalisation data wasn’t sufficiently detailed. A baby might go to hospital just to be supervised. But what we really need to know is, how many actually needed intervention? Oxygen? ICU admission? How serious does it get?

DEMASI: Ah, ok… I understand.

LEVI: Because when you’re weighing up serious adverse events—things that are life-changing, require hospitalisation, or result in death—you have to make sure those don’t outweigh the severity of the disease itself. Again, I know I’m repeating myself, but these are healthy babies we’re recommending get a new product—the benefits must far outweigh the harms.

DEMASI: But my impression, listening to the CDC, was that it’s healthy babies who end up hospitalised…

LEVI: Yes—because most babies, luckily, are born healthy. But the more important question is to look at what the risk of is of a healthy baby to actually end up with severe lung disease or die from RSV, it’s actually quite rare—thankfully, in developed countries. So you have to be careful that serious adverse events from the monoclonal antibody don’t end up outweighing the actual disease burden.

DEMASI: That makes total sense. So you obviously thought the serious adverse events reported after the product were alarming?

LEVI: Yes, I did. When I looked at trials of a similar product—Sanofi’s product, which is essentially the same—I saw a repeated trend; that is, more deaths in the treatment group than in the control group and also more serious adverse events of the nervous system.

DEMASI: I remember you mentioned there was an “imbalance” of deaths…an upward trend across four or five trials?

LEVI: Yes. Each of the trials was small, but they showed the same upward trend in the absolute number of deaths. In Sanofi’s pivotal trial, for example, there was a 2:1 ratio—intervention versus placebo—and there were five deaths in the intervention arm, zero in the placebo. In another 2:1 trial that focused on preterm babies and ones with comorbidities, they tested the new intervention against the current standard of care—another monoclonal antibody that’s short-acting—and again, there were five deaths in the monoclonal group and only one in the control. Even in Merck’s trial with clesrovimab, also 2:1 – the one we voted on – it had 7 deaths in the treatment arm and 3 deaths in the placebo arm. They also have an ongoing 1:1 trial against the standard of care and they reported on 8 deaths in the intervention versus 4 in the standard care. That is, double the number of deaths.

DEMASI: But you yourself acknowledged that the trials were too small to get an accurate statistical calculation. Someone from the FDA even said the numbers were too low to get anything statistically meaningful…

LEVI: Yes—but it wasn’t just the deaths that worried me. There was also consistent excess of nervous system issues in the intervention arms across the trials, for example seizures. And these serious adverse events showed up again and again in different trials. When I looked at the post-marketing safety analysis that the CDC conducted, they didn’t address any of those adverse events properly—because they only looked for seven days after the intervention, and did not consider deaths at all.

DEMASI: Wait, what? Are you kidding? Only seven days?

LEVI: Yes. That’s why I felt that moving ahead with a broad recommendation for all healthy babies was not a sufficiently precautionary approach. It didn’t properly address the concerns or the potential safety signals I think we’re seeing.

DEMASI: I heard them say during the meeting that they analysed all of the deaths in the trials and determined that *none* of them were related to the product. None. What did you think when you heard those proclamations?

LEVI: I think those things need much more detailed discussion. In my experience, the regulatory data—the narratives—don’t include enough information to make those decisions confidently. So yes, I heard them say the events weren’t related, but I can’t see how anyone can be so sure. There are things like sudden infant death syndrome (SIDS)—how do you know for sure it wasn’t related? If there’s an imbalance in one trial, maybe it’s chance. But if you see that same imbalance across multiple trials—how unlucky can you be?

DEMASI: So, humour me—why not just say to the manufacturers, “These trials aren’t big enough or long enough to support a recommendation. Go back and do the trials properly”…can’t you just demand that?

LEVI: I think you’re touching on something much more fundamental than what we discussed at ACIP. Our current system for approving products—especially those intended for healthy babies—needs to be improved. That starts with the FDA, and continues with NIH, as well as CDC.

In this situation, there’s no reason we couldn’t ask manufacturers or the government to conduct a large trial that would quickly clarify both harms and benefits. There’s a lot to rethink in the system, and ACIP is part of that.

DEMASI: Ok. So I get that this product was already FDA-approved—But why didn’t the FDA demand that bigger, longer trials be conducted before approving the product?

LEVI: That’s a very good point to raise. And I’m sure people at the FDA are thinking about it. I suspect the FDA is going to go through some changes—and I think we’ll hear more about that in the future.

DEMASI: I know the panel voted for thimerosal-free flu vaccines, but it surprised many that you also supported recommending the influenza vaccine for everyone over six months of age, given their very poor efficacy.

LEVI: I think there are some nuances there. At the meeting, I did say we need better evidence of efficacy—something more than just immunobridging data or the current efficacy analyses based on retrospective observational data.

DEMASI: Right, but there has been all this talk about demanding “gold-standard” trials—so why are we still relying immunobriding data?

LEVI: We’re a new group. We had very limited time to deeply understand every issue. Even if you want to change things, it’s important to first grasp the rationale behind the current state. We can’t rush in and start rewriting everything. There will be a working group on influenza, and I think we’ve already signalled the need for stronger efficacy data. But it has to go through due process.

DEMASI: I think some people expected ACIP would just walk straight into the first meeting and demand an immediate ban on mRNA vaccines. Was that just fanciful thinking?

LEVI: Let me put it this way—some people thought we’d come in and immediately change everything. But that would be naive and, frankly, irresponsible. We need to take the time to study, deliberate, understand the current state, and then work with CDC and others to make improvements.

DEMASI: But to play devil’s advocate on the mRNA vaccine issue—don’t we already have enough data? Why not pause mRNA vaccine recommendations now, so more people aren’t harmed?

LEVI: I don’t want to pre-empt future meetings or votes. People know my personal views on mRNA vaccines. But when you’re part of a committee like ACIP, it’s critical to engage the right stakeholders and follow the proper process. Some issues are clearly time-sensitive as you point out, and we’re committed to addressing them proactively. But we also need to do it the right way and we will address these issues in the Covid vaccines work group over the next few months.

DEMASI: There’s a lot of pressure – from both sides….

LEVI: Yes, there’s pressure and noise—some of it justified—but the committee has to stay focused and run a sound process.

DEMASI: OK, thanks for your time, Dr Levi.

LEVI: You’re welcome.

Republished from the author’s Substack

Retsef Levi Explains His Vote against Routine RSV Monoclonal Use in Newborns
by Maryanne Demasi at Brownstone Institute – Daily Economics, Policy, Public Health, Society