Faust Files: Paul Offit on His Vote Against Bivalent COVID Boosters

In the first part of this two-part conversation, Jeremy Faust, MD, editor-in-chief of MedPage Today, talks with Paul Offit, MD, about his recent decision to vote against the bivalent COVID-19 booster shot at a recent meeting of FDA's Vaccines and Related Biological Products Advisory Committee (VRBPAC).

Offit is director of the Vaccine Education Center at Children's Hospital of Philadelphia (CHOP), attending physician in the division of infectious diseases at CHOP, and Maurice R. Hilleman Professor of Vaccinology at the Perelman School of Medicine at the University of Pennsylvania.

The following is a transcript of their remarks:

Jeremy Faust, MD: Hello, I'm Jeremy Faust, medical editor-in-chief of MedPage Today. Thanks for joining us.

Today we are joined by Dr. Paul Offit. He is a member of the FDA Vaccines and Related Biological Products Advisory Committee, the VRBPAC, which is the committee that publicly discusses, and debates, and ultimately votes on recommendations related to the COVID-19 vaccines and their authorization for use.

Dr. Paul Offit, thank you so much for joining us.

Paul Offit, MD: My pleasure.

Faust: Let's start by talking about an article you wrote in the New England Journal of Medicine, 'COVID-19 boosters -- where from here?' You argued, essentially, that these mRNA vaccines in particular were designed to do something very specific, which is to prevent severe disease and deaths, and that they happen to temporarily do what everyone else wants them to do, which is to decrease infections.

It seems to me that a lot of people are looking at boosters and they don't realize it, but they're treating them like they're almost prophylaxis, and I think that, in your view, is not a sustainable way to do things. How do you think this is going to play out?

Offit: I think probably you can trace the confusion back to December of 2020.

In December of 2020, the FDA vaccine advisory committee was asked to consider both Pfizer and Moderna vaccines. Those were two-dose vaccines given 3 or 4 weeks apart, and protection against mild, moderate, and severe disease was 95%. Those vaccines were roughly 95% effective at preventing not only severe illness, but also mild illness.

Then what happened was 6 months later, the CDC did studies showing that protection against severe disease holding up, which is good, but the protection against mild disease was fading -- as you would've guessed -- because protection against mild diseases, for the most part, is mediated by neutralizing antibodies present at the time of infection, and those faded.

So why was it 95% protection against mild disease in December of 2020? Those were 3-month studies. Those participants had just gotten their second dose. So therefore, you sort of had this unrealistic sense of what this was all about. That was the way it was handled, not only by the media, but I think in many ways by public health officials; officials who use words like 'fading immunity' or 'breakthrough infections' when you were seeing exactly what you would expect to see.

Even 1 year later, when studies were done by Mark Tenforde out of the CDC, with just two doses you found that protection against severe disease was holding up.

Then Omicron hit, which was an immune-evasive strain, and caused mild illness even in those who were vaccinated. I think this sort of led to a lot of confusion and ultimately led to a third dose and then to a fourth dose. And I think we used terms like 'fully vaccinated' as compared to 'fully vaccinated and up-to-date,' and I think it would be hard to find 10 Americans who would agree on what it means to be protected against this virus.

Faust: I think that's fair. Now, I actually think this next question is sort of out of order, but I think it's important that we do it first. That is that you are, as I mentioned in the introduction, on the committee -- VRBPAC -- that looks at these questions of what we should do next and makes recommendations, and you voted against the bivalent booster. That is, using a booster that's got some of the old strain and some of the Omicron [strain] in it. You voted against that for the fall. I'd like you to talk us through your thought process and why you think that doing it your way would be better.

Offit: The studies that were presented to us on June 28 of this year, 2022, by both Pfizer and Moderna were done the right way, which is to say they gave people three doses of the ancestral strain vaccine and then the fourth dose was either the ancestral strain vaccine or the fourth dose was a bivalent vaccine. The bivalent vaccine in both cases, for Pfizer and Moderna, was the ancestral strain plus BA.1, the original Omicron strain. They then looked at the difference in those neutralizing antibody responses based on whether you got the ancestral boost or whether you got the bivalent boost.

What they found was -- depending on how the study was done because Pfizer did 30 mcg or 60 mcg and Moderna did 25 mcg for the ancestral and then 25 mcg for the Omicron strain -- they found that the difference in neutralizing antibodies was either 1.5-fold greater against neutralizing antibodies for Omicron if you got the bivalent strains compared to if you just got the ancestral strain, or 1.75 greater, or 1.97 greater.

The implication was that that's good enough; that those statistically significant differences were also clinically significant differences, when we knew that back in December of 2020 -- when you looked at the neutralizing antibody differences between Moderna and Pfizer -- there was roughly a twofold greater neutralizing antibody response for Moderna than Pfizer, but that didn't really matter in terms of protection against severe illness.

So why were we making this difference of 1.5-fold or 1.97-fold greater with the Omicron boost than the ancestral boost? Why were we making that seem as if that was going to be the answer to all our problems? Plus, the BA.1 strain is essentially gone. It's been replaced by BA.5/BA.4 and now BA.2.12.1, which are just Omicron subvariants that are somewhat distant from BA.1.

So where this is settled out is that I think the [Biden] administration is interested in having a bivalent vaccine where one of the strains is the ancestral strain and the other strain is BA.5. And we'll see, but I certainly hope that before they launch that kind of program for the American public, they show clear evidence that there is a dramatic increase in neutralizing antibodies with BA.5 associated with that boost. Because remember, even with boosting the ancestral strain, you do get somewhat of a broadening of immune response as Linda Saif and coauthors showed in a New England Journal Medicine paper at the end of June.

So I think that's the burden that it should be held to because otherwise it's just marketing, right? Otherwise it's like, 'I'm going to give you BA.5 because BA.5 is circulating and therefore it's going to be better,' which theoretically makes sense, but you need to have clear evidence, preferably clinical data, but at the very least compelling immunological data.

Faust: Now, would moving to a bivalent booster just in general be progress against this concern about 'imprinting or antigenic sin'? The idea is that whatever your body sees first, it will sort of narrowly respond to that. Therefore in the future, your body doesn't have as good of a response because of what it initially saw. Does giving bivalent anything actually have some benefit just on those grounds?

Offit: Again, it should be proven, but the imprinting thing was originally defined by a researcher named Thomas Francis, who was working with influenza back in the 50s, and he used the term 'original antigenic sin.' The thinking was this, you can tell when children were born based on the way they respond to influenza -- either natural infection or immunization -- because they respond as if they're responding to that original infection. The original infection that they had when they were children, if they then get a vaccine or a secondary infection, they're responding as if they're responding to that first infection and not recognizing the subtle differences with that either vaccine or natural infection.

That's what you worry about here, that you sort of lock people into this response against the ancestral strain and don't have the chance then to have a broader response.

I will say this though, the ancestral strain is the original strain. It's the Wuhan strain. The one that first raised its head in Wuhan in October/November 2019 has served us well. I mean, all the way through, even including Omicron and the Omicron subvariants, it does appear to continue to protect against serious illness, which is the goal of this vaccine. At least it was the stated goal of this vaccine, but I feel like we're drifting a little bit from protection against serious illness to trying to protect against all illness.

And I would argue that there are some people who can't handle a mild illness. They can't handle it because they have severe heart disease or severe lung disease or because they have out of control diabetes or whatever, so when they get a mild or moderate infection they're more likely to develop a severe illness or because they're immune compromised.

So that's different. That's really not fading memory, it's someone who either never had a very good immune response because they're immune compromised or because of their age they don't make very good so-called cytotoxic T-cell responses, so they're not as able to crowd the infection and keep it as a mild or moderate infection. You could argue that that group may benefit from a boost, if you will, to keep neutralizing antibodies high, to the extent they can make neutralizing antibodies for a few months during the times when these viruses are at peak circulation, you could make that argument. But I think we're sort of not making that argument; we're making the argument of boosters for all. And first of all, I think people are a little weary of boosters at this point.

Faust: Yeah. I mean, the way I'm looking at it is that people who, as you mentioned, have another risk factor for hospitalization probably need to stay what we'd call 'up-to-date' on their vaccine so that they don't encounter mild illness because mild illness has implications for them that they don't have for others. It's not that they'll get severe disease; severe COVID, it's that that mild illness will make something else severe. That's what we keep seeing, and the data are bearing this out. So, absolutely, I think saying up-to-date for older populations makes sense.

I don't think the mRNA vaccines are optimized for this, but they're the best we've got, so that's what we're stuck with for now. But I agree with you that we could kind of spend all of our effort on that group and have huge benefits instead of diffusing the efforts, the money, the signal, for very little in return.

As I think about younger people, some younger people obviously really want doses. They've bought into this idea that these vaccines can eliminate infection. I think you and I agree on this, the benefit for boosting for young people is very transient, if any, but over time I think arguments have been made that say, 'OK, look, there's a lot of good reasons to boost younger people.'

Where do you stand right now? Do you still think that two doses of the ancestral vaccine is sufficient for most people, and have you softened on third doses for younger people? I guess the third question of this three-part is: who do you think right now needs more than two doses?

Offit: I still would argue that for a healthy young person, two doses appears to continue to protect against severe illness. What that third dose does do though for Omicron, meaning BA.1 and the Omicron subvariants, is you do get an increased response against those variants and subvariants.

The question is, does it matter? Because still it looks like you're protected against severe illness, so does it really matter to get that third dose? I would argue 'no.' [But] I think that ship has sailed. I think this is now a three-dose vaccine for the most part, because that's the way the press and the public have handled it, it's the way public administrators have handled it.

But I still would argue, if you looked at that Mark Tenforde paper that was published in Infectious Diseases, what he did was he looked at two doses for 1 year, up through December of 2021 so the vaccine had been out for a year -- so you're looking through the D614G variant, the Alpha variant, and the Delta variant -- and what he found was that protection against serious illness was holding up. And this was for...80% of the people in that group had at least one comorbidity and a significant percentage were over [age] 65. So that was holding up pretty well.

But then Omicron hit, and we clearly have shown that three doses is better than two for Omicron in terms of preventing hospitalization. My question is: who are those people? And I think that's not everybody.

I still think it would be helpful for the CDC to sort of sub-stratify that. To some extent they have; they've shown us that at least 75% of those who are getting hospitalized despite having two doses were in these high-risk groups that we just talked about. So focus on them. I think that is what makes the most sense to me.

Faust: Yeah. If you look at the CDC's readouts -- I think we've discussed this before offline, but I'm not sure we've discussed it in public -- it's actually very difficult to tell the difference between a person who got their second dose a year ago and a person who got their third dose a year ago. So the idea of the three-dose vaccine kind of only makes sense in that in the period of time for which a third dose increases those neutralizing antibodies.

So yes, you could tell the difference between someone who just got a third dose, but you also can't tell the difference between someone who just got a third dose and someone who just got their second dose. What really matters is how far are you out from your most recent dose, and I think that the public is basically not going to keep doing this except for high-risk groups. Is that a fair read?

Offit: Very fair. And I think when we compare vaccinated to unvaccinated or got a two-dose versus three-dose you need to also account for who has gotten naturally infected in the meantime, because that does alter things. And it is knowable, I mean, you can know that by looking at say antibodies against, say, the nuclear protein, which is only going to happen if you've been naturally infected. So I think that also is one of the problems -- controlling for natural infection, which now is common.

I remember when a couple months into this pandemic, Jon Yewdell, who was the head of virus research at the National Institutes of Health, said 3 years from now you're going to have two choices: be naturally affected or be vaccinated.

If you look at some of these recent seroprevalence studies, there was one in JAMA recently, showing that as many as 90% of people have either been naturally infected or vaccinated when they look at the thousands and thousands of lots of people who have donated blood. Now, that's limited by who chooses to donate blood, but I still think it is striking.

Faust: Right. And I think maybe a cognitive shift that I made at some point was that you might not necessarily choose between infection or vaccination, but you have to choose infection with protection or infection without protection. In other words, you walk into that infection or that exposure with a vaccine on board or not, and you want to do that, obviously.

I want to make sure that we don't make incorrect news here. You said that you think it's a three-dose vaccine, but you said that because you think that basically the scientific debate is over because no one is having it. Not because of science support stuff but basically due to marketing, is that correct?

Offit: Yeah. I just got an email today from a mother of a college student who was upset that her child can't go to a certain East Coast school because he hasn't had his third dose. She argues reasonably -- he's a healthy young person, he's gotten two doses, he's a boy who is in an age range where he is at some risk for myocarditis, and I don't think the benefits outweigh the risks here. A perfectly reasonable argument. I agree. I agree with her.

Faust: But I also think that, as my friend Carter Mecher pointed out to me when we discussed this like a year ago, that most people who get a third dose, even in that group of a young, healthy male, most people will be happy with their decision to get a third dose because for 99% it makes no difference. For a small, tiny group it would help; for a small, tiny group it would hurt.

So most people would just be happy with their decision because whatever they did, they think, 'Well, I got a third dose. I did my best.' Therefore the administration is saying, 'Well, look, if it breaks even we might as well just make everybody happy.'

Offit: Yeah. There was an old study that was done at a Canadian racetrack where they asked people to rate their horse while they were on the way up to making the bet, then they asked a different group of people to rate their horse after they'd already made the bet. Obviously, after you already made the bet, you rated it much higher.

But what I loved about that article was the title, which was "Post-decision dissonance at post time."

Faust: Right. Post-vax something or other -- we could do it with that.

The mix and match, the idea of starting with one formulation of Pfizer and then boosting with Moderna, do you think it matters? Certainly, it has a little bit more of an antibody response and more side effects, but on a population level do you think it matters? If someone asks you, 'What should I do?,' what do you tell them?

Offit: Wait for data. I mean, it is interesting to see some of these early reports now of boosting with Novavax, which is a purified protein vaccine that's adjuvanted with the same adjuvant that's used in Shingrix, which is an exceptional vaccine.

If you've already gotten, say, two doses or three doses of an mRNA vaccine and then you get this dose of Nova, the antibody studies, at least the small ones that have been reported so far, look promising.

But again, you ultimately need clinical data, and hopefully the CDC can help generate those data.

Comment