Angry US national vaccine advisers say government scientists and Moderna concealed data on a new Covid-19 booster last year during discussions on whether the shot should be authorised, and that the missing information showed infection rates among trial participants receiving the new vaccine actually increased.
The early data had a number of limitations, suggested the updated booster might not be any more effective at preventing infections than the original shots, they told CNN.
US taxpayers spent nearly $5bn on the new booster, which has since been given to more than 48.2m people in the US.
“I was angry and disappointed that we were not showed other data relevant to our decision,” said Dr Paul Offit of the Vaccines and Related Biological Products Advisory Committee, a group of external advisers that helps the Federal Drug Administration (FDA) make vaccine decisions. “Decisions made for the public must be made based on all available information – not just some of it.”
At meetings of this FDA advisory group in June and in September of a panel that advises the US Centres for Disease Control and Prevention (CDC), the experts received information indicating the new vaccine worked better than the one already on shelves.
Those immunogenicity data were based on blood work done on study participants to assess how well each vaccine elicited antibodies that fight off the Omicron strain.
The data NOT presented to the experts looked at actual infections: who caught Covid-19 and who did not, and revealed that 1.9% of the study participants who received the original booster became infected.
Among those who got the updated bivalent vaccine, which scientists hoped would work better, a higher percentage, 3.2%, became infected. Both versions of the shot were found to be safe.
However, the infection data were far from complete. The number of study subjects who became infected was very small, and both the patients and the researchers were aware of who was getting the original shot and who was getting the new booster.
Despite these imperfections, the data were included in a preprint study posted online in June, again in September in an FDA document and then later that month in a top medical journal – but advisers to the FDA and the CDC said the data should have been shared with them, too.
The six FDA and CDC advisers interviewed by CNN said that this infection data wouldn’t have changed how they voted, because the data had such limitations, but still should have been presented to them.
The FDA and the CDC convene their advisory board meetings and make presentations to the advisers. At last year’s meetings, Moderna executives made presentations as well. The advisers then make their recommendations to the agencies, which decide whether to authorise the shots.
Moderna spokesman Christopher Ridley told CNN the company shared the infection data with the FDA and posted the study manuscript before the agency’s panel meeting in June “in response to requests that we share an update from the ongoing study”.
That study preprint was posted online on 25 June, three days before the FDA advisers met.
Michael Felberbaum, an FDA spokesman, said the FDA received the preprint less than a day before the advisory committee meeting, and that “the information was therefore not provided in an adequate timeframe for it to be included in the meeting materials … generally the FDA only discusses data … the agency has had opportunity to substantively review”.
He said Moderna should have presented the data at the meeting.
Kristen Nordlund, a CDC spokesperson, said that “due to the many limitations involving this clinical data, it was not featured” in the CDC’s advisory committee discussion.
Infection data not included in presentations
Last summer, when the FDA’s and CDC’s outside advisers considered the updated boosters, they knew that within months, in winter, Covid rates could spike. They also knew the original vaccine was becoming less effective with each new variant and that the addition of the Omicron strain in the updated booster might help battle the virus.
The FDA advisers – 21 voting members, including infectious disease experts and vaccinologists from Stanford, the University of Pennsylvania and Harvard – met on 28 June.
In his presentation, Dr Stephen Hoge, Moderna’s president, frequently referred to information from the preprint study, funded by Moderna and posted three days before. It was led by company scientists, and had not been peer-reviewed or published in a medical journal.
The data he shared with the advisers demonstrated that blood tests on about 800 study participants indicated the new bivalent booster was “superior” at increasing antibodies to the Omicron variant compared with the original vaccine.
Hoge did not mention another part of the study casting a less positive light on the shot.
In that part, the researchers gave some participants the existing vaccine and other participants the updated booster, then tracked who became infected with Covid.
Among the hundreds of participants receiving the original vaccine, with no evidence of a prior Covid-19 infection, over the period of the small study, 1.9% became infected. Among the hundreds who received the new bivalent vaccine, a higher percentage, 3.2%, became infected. The preprint did not indicate whether these findings were statistically significant.
A 22-page FDA briefing document given to the advisers did not mention these infection data.
Dr Jerry Weir, director of the Division of Viral Products at the FDA’s Office of Vaccines Research and Review, also did not mention the infection data in his presentation to the advisers.
At the end of the 28 June meeting, the FDA advisers voted 19-2 to recommend the inclusion of an Omicron variant for the booster vaccine. Offit, a professor of vaccinology at the Perelman School of Medicine at the University of Pennsylvania, and Dr Henry Bernstein, a professor of paediatrics at the Zucker School of Medicine at Hofstra/Northwell in New York, voted in opposition.
Over the next month, the US Government agreed to purchase the bivalent booster from Pfizer and Moderna: a contract for $3.2bn with Pfizer and for $1.74bn with Moderna.
Limitations of Moderna’s infection data
The infection data not included in the Moderna and FDA presentations had several serious limitations, advisers told CNN.
First, the numbers were very small. The study analysed only hundreds of patients, and only 16 became infected. None ended up in the emergency room or hospital.
Dr Eric Rubin, a professor in the Department of Immunology and Infectious Diseases at the Harvard TH Chan School of Public Health and an FDA adviser, said vaccines are meant to prevent severe disease and death. These are “impossible to measure in such small studies” and a “better antigenic match could result in decreased numbers of people who get hospitalised or die”.
Second, participants were not randomly assigned to receive either the original or new vaccine, and the study was not double-blind, meaning the participants and the researchers knew who was receiving which shot. Lack of randomisation and blinding can bias study results.
Third, the study’s primary purpose was not to study infection rates but to do immunogenicity analyses, taking blood from participants and examining their antibody responses to the vaccine.
Nearly $5bn for updated boosters
On 31 August, about two months after the FDA advisers’ meeting, the agency authorised the Moderna bivalent vaccines. The Moderna infection data were included in the agency’s written decision, but it wasn’t posted online until 13 September, said Felberbaum.
On 1 September, the CDC advisers – 14 voting members – met to consider whether to recommend the bivalent boosters for Americans. Several CDC scientists presented data but did not include the infection data.
Nordlund, the CDC spokesperson, said the “CDC was aware” of the data that would later be published in The New England Journal of Medicine but that assessing infections was an “exploratory objective of the study”, which was “not designed to assess vaccine effectiveness”.
She added that in assessing infections, researchers used different durations and points in time among a very small group of people, and because of the limitations of the data, they was not featured at the meeting.
Dr William Schaffner, a non-voting member of the CDC’s advisory committee, called Nordlund’s argument “weak.”
“The data are limited but informative. A complete presentation would have included them,” said Schaffner, an infectious disease specialist at Vanderbilt University Medical Centre.
Dr Jacqueline Miller, a senior vice-president at Moderna, who presented to the advisers, also showed material suggesting the bivalent vaccine was superior. She did not show slides detailing the statistics about the infection rates.
During the meeting, a CDC adviser, Dr Sybil Cineas, an associate professor of medicine at the Warren Alpert Medical School of Brown University, asked Miller about Covid-19 cases among study subjects who received the original vaccine versus the updated booster.
Miller said that among subjects with no evidence of prior infection, as well as those with evidence of prior infection, the disease incidence rates were 2.4% for the group who received the original vaccine and 2.5% for those who got the bivalent booster.
Those data would be published two weeks later in The New England Journal of Medicine.
When she answered Cineas’ question, Miller did not mention other aspects of the infection data indicating that among hundreds of participants with no previous Covid history, infection occurred in 1.9% of those who’d received the old vaccine and in 3.2% of those who’d received the new bivalent vaccine.
Neither did she mention breakdowns of whether participants were sick with Covid or had asymptomatic illnesses. The study, like the preprint, did not mention whether any of these findings were statistically significant.
After the meeting, the CDC advisers voted 13-1 to recommend the bivalent booster, with Dr Pablo Sanchez, a professor of paediatrics at the Ohio State University College of Medicine, voting in opposition. CDC Director Dr Rochelle Walensky signed off on the recommendation.
More than six months after the FDA advisers met, Moderna still has not released data from a randomised Phase 3 trial comparing infections in participants who received the new booster with those who received the old shot. The company expects to release such results “shortly”.
Pfizer does not “currently have data on incidence of infection post bivalent booster”.
“However, we continue to monitor real-world data and collect data from our own studies,” said Jerica Pitts, senior director of global media relations.
About a month after the CDC advisers met, studies from Harvard and Columbia University researchers suggested the new vaccines didn’t work any better than the original.
Those studies, which were small and only in preprint and not in a medical journal, measured immune responses after people got the bivalent vaccine compared with the original version of the vaccine.
“We see no difference between the old boosters and the new about a month after the shot,” said Dr David Ho, a professor of microbiology and immunology at Columbia whose team authored one of the studies.
Offit, the FDA vaccine adviser, said the Columbia and Harvard studies convinced him even more that the infection data and all the related caveats should have been given to the advisers from the beginning.
“I understand we’re building the plane while it’s still in the air, but you can’t do this,” he said. “It shook my faith in how these decisions were being made.”
BioRxiv preprint – Immunogenicity of the BA.5 Bivalent mRNA Vaccine Boosters (Open access)
See more from MedicalBrief archives:
Only ‘marginal benefits’ from second COVID booster – Israeli study
FDA and Fauci eyeing a second COVID-19 booster in coming months
Moderna/Pfizer update two vaccines against newer variants