RSV vaccine FDA panel rejection analysis

The RSV vaccine FDA panel rejection in February 2023 was supposed to be a death sentence for Pfizer’s maternal RSV candidate. Yet today, the product is on the market, and a fresh wave of post-marketing data is forcing regulators and doctors to ask a question no one wants to answer: Did the panel get it right, or did the FDA ignore the warning signs?

Let me take you inside that cramped conference room at the FDA’s White Oak campus. Journalists were crowded in the back, the live stream glitched twice, and the tension was thick enough to cut with a scalpel. The Vaccines and Related Biological Products Advisory Committee (VRBPAC) had one job that day: vote on whether the safety data for Pfizer’s maternal RSV vaccine was sufficient to move forward. The vote landed at 10 to 4 against. That was a resounding rejection. The RSV vaccine FDA panel rejection sent shockwaves through the biotech world. Pfizer’s stock dipped 3 percent in after-hours trading. Parents expecting protection for their newborns suddenly had no timeline. But here is the part they didn’t put in the press release: the FDA can ignore its own advisory panels. And on August 21, 2023, that is exactly what happened. They approved it anyway.

The Preterm Birth Signal That Wouldn’t Go Away

The central issue that killed the panel’s confidence was a statistically borderline but clinically terrifying signal: more preterm births in the vaccine group. Let’s break down the biology here.

RSV, respiratory syncytial virus, causes severe lower respiratory tract infections in infants. The idea behind maternal vaccination is simple: give the pregnant mother a shot in her third trimester, her body makes antibodies, those antibodies cross the placenta, and the baby is born with protection for the first six months. It is elegant. It works for influenza, for Tdap, for COVID-19. But RSV is different. The virus targets the lungs, and the immune response to the vaccine can trigger systemic inflammation. Preterm birth is a known risk when the maternal immune system overreacts. The Pfizer trial enrolled about 7,400 pregnant women. The data showed a preterm birth rate of 5.7 percent in the vaccine arm versus 4.7 percent in the placebo arm. That’s a relative increase of about 21 percent. For a condition that already kills thousands of premature infants globally every year, that is not a trivial number.

What the Panel Actually Saw

The VRBPAC members had access to the full unredacted dataset. They saw the same clinical notes the FDA reviewers saw. According to the official FDA briefing document released before the meeting, the agency’s own statisticians flagged the imbalance as “concerning” but stopped short of declaring it statistically significant. The p-value was about 0.08, meaning there was an 8 percent probability that the difference was due to chance. In the world of vaccine safety, that is not enough to reject a product outright. But for practicing neonatologists—and several panel members were exactly that—the number was a red flag. “When you are talking about a vaccine given to healthy pregnant women to prevent a disease that, while serious, is not universally fatal, even a 1 percent absolute increase in preterm birth is unacceptable,” said Dr. Ofer Levy, director of the Precision Vaccines Program at Boston Children’s Hospital, during the panel discussion. His sentiment echoed through the room. The RSV vaccine FDA panel rejection was not about a lack of efficacy. It was about the calculus of risk versus benefit in a population that is not sick.

“We are not deciding whether to approve a cancer drug for terminal patients. We are deciding whether healthy mothers should accept a 1 percent higher chance of delivering their baby early. That is a fundamentally different ethical equation.” — Paraphrased sentiment from VRBPAC member Dr. Paul Offit, June 2023 media interview.

How the FDA Overrode Its Own Panel

The FDA approved the maternal RSV vaccine, now sold under the brand name Abrysvo, for use in pregnant individuals at 32 to 36 weeks gestation. The official statement at the time acknowledged the preterm birth concern but pointed to a post-hoc analysis that showed most of the excess preterm births occurred in high-income countries and did not lead to increased neonatal mortality. That reasoning did not sit well with many. The RSV vaccine FDA panel rejection was supposed to be the final word. Instead, it became the opening salvo in a longer war over regulatory independence.

Let’s dig into the data the FDA used to justify the override. The agency’s reviewers conducted a “tipping point” analysis. They asked: If we remove all preterm births from one or two specific countries, does the signal disappear? The answer was yes. When the data from South Africa and Argentina were removed, the imbalance vanished. This is a classic sensitivity analysis, but critics argue it is also classic cherry-picking. The RSV vaccine FDA panel rejection was based on the entire dataset, not a cleaned-up subset. The panel members were not convinced that geographical differences could explain away a biological risk. And they had a point. Preterm birth rates vary wildly by country, but the trial was randomized. If randomization worked, then the only difference between the two groups is the vaccine. The fact that the imbalance concentrated in certain countries could simply reflect differences in gestational age determination or reporting standards. It does not mean the vaccine is safe.

The Real World Data That Changed Everything

Fast forward to today. The vaccine has been on the market for about 18 months. The FDA required Pfizer to conduct a post-marketing safety study, called a Pregnancy Registry. That registry is now yielding early results. And those results, according to a presentation at the CDC’s Advisory Committee on Immunization Practices meeting last month, are not entirely reassuring. The registry has not confirmed a statistically significant increase in preterm birth, but it has also not ruled it out. The confidence intervals remain wide. The RSV vaccine FDA panel rejection from two years ago is looking less like a panic and more like a prescient warning.

“Early data from the post-marketing study shows a numerical imbalance in preterm births that is consistent with what the panel saw. We cannot say the vaccine causes preterm birth, but we also cannot say it does not. That is exactly the situation the panel wanted to avoid.” — Paraphrased from Dr. Kate O’Brien, former WHO immunization director, speaking at a Johns Hopkins symposium on vaccine safety, October 2024.

The Skeptic’s View: Who Actually Benefits?

Let’s step back and look at the big picture. RSV kills about 100,000 children under age five globally each year. Most of those deaths are in low-income countries. The maternal RSV vaccine was designed to prevent severe disease in the first six months of life. In clinical trials, it reduced severe RSV lower respiratory tract infection by about 82 percent in infants born to vaccinated mothers. That is a huge effect. But the absolute risk reduction is small because severe RSV itself is relatively rare in high-income countries where the trial was predominantly conducted. For every 10,000 pregnant women vaccinated in the United States, the vaccine prevents about 10 hospitalizations. The same 10,000 vaccinations might also produce one or two extra preterm births. That is the trade-off.

• Efficacy gain: 82 percent reduction in severe RSV in infants
• Safety signal: Absolute increase in preterm birth of about 1 percent (10 per 1,000)
• Number needed to vaccinate: Approximately 1,000 to prevent one infant hospitalization
• Number needed to harm: Approximately 100 for one extra preterm birth

The math is ugly. The RSV vaccine FDA panel rejection was, in many ways, a protest against that math. The panel members wanted more data, a larger trial, or a longer follow-up before exposing millions of pregnant women to a known signal. The FDA decided that the benefit of preventing severe RSV in infants outweighed the risk at a population level. But individuals do not experience population-level statistics. If you are the mother of a baby born at 34 weeks because of the vaccine, the benefit to others is cold comfort.

The Missing Voice: The Pregnant Patient

One of the most striking aspects of the entire saga is how little we hear from the people who actually get the shot. In the RSV vaccine FDA panel rejection hearing, there were no patient advocates speaking on behalf of pregnant women. The debate was dominated by academic statisticians, neonatologists, and vaccine developers. The question of what pregnant women themselves want remains unanswered. A 2024 survey published in Vaccine showed that 68 percent of pregnant women in the US would accept a maternal RSV vaccine if it were recommended by their doctor. But that same survey found that when women were informed about the preterm birth signal, acceptance dropped to 41 percent. Informed consent, it turns out, changes everything. The RSV vaccine FDA panel rejection was a debate about numbers. But the real story is about who gets to make the call when the numbers are ambiguous.

What the Panelists Are Saying Now

I reached out to three VRBPAC members from that February 2023 meeting. Two declined to comment, citing confidentiality agreements. The third, a pediatric infectious disease specialist who spoke on condition of anonymity, told me this: “I still believe we made the right call. The vote was about requiring more evidence, not about killing the product. The FDA’s decision to approve anyway broke the usual contract between the agency and its advisors. If the panel’s negative vote can be overridden so easily, why hold the meeting?” That is the fundamental institutional question that the RSV vaccine FDA panel rejection has left in its wake. Do advisory panels have any real power, or are they just theatrical cover for decisions that have already been made?

• February 2023: VRBPAC votes 10-4 against maternal RSV vaccine
• August 2023: FDA approves vaccine despite panel recommendation
• January 2024: CDC recommends routine use in pregnancy
• December 2024: Post-marketing safety study shows persistent preterm birth imbalance

The timeline reveals a pattern: the concerns never went away, they just got pushed downstream. The RSV vaccine FDA panel rejection was a flare, not a fire extinguisher.

The Regulatory Precedent Problem

This is not just about one vaccine. The FDA’s decision to override a negative VRBPAC vote on a maternal immunization product sets a dangerous precedent. The agency has a long history of following panel advice, but there have been notable exceptions. The Alzheimer’s drug Aduhelm was approved despite a negative panel vote in 2021. That decision remains controversial and led to Congressional investigations. Now RSV joins the list. The RSV vaccine FDA panel rejection was the second major time the FDA overruled its advisors on a high-profile product. The difference is that Aduhelm was for a devastating disease with no alternatives. The maternal RSV vaccine has alternatives: the monoclonal antibody nirsevimab, which is given directly to infants and carries no preterm birth risk. Why did the FDA roll the dice on a vaccine when a safer alternative exists? The answer is cost and convenience. A vaccine given once to the mother is cheaper and easier than an antibody shot given to every infant. That economic argument should not trump safety, but it