FDA Commissioner on COVID Vaccines: 'I'm Incredibly Proud'

John Whyte, MD, MPH; Eric Topol, MD; Stephen Hahn, MD; Peter Marks, MD, PhD


December 29, 2020

Editor's note: Find the latest COVID-19 news and guidance in Medscape's Coronavirus Resource Center.

  • The initial assessment of the COVID vaccines gave robust safety and efficacy data. Emergency use authorization was granted instead of a full biologics license application approval because that would require greater evidence of safety and efficacy, as well as longer duration of follow-up. However, an EUA still requires clear and compelling evidence of safety and efficacy from at least one randomized clinical trial.

  • A 2-month median follow-up allowed a large majority of adverse events to be seen, as well as confidence that the vaccine gives at least a 2-month median duration of immune response.

  • The next step is to see whether the blood samples that have been obtained indicate immune correlates of protection (eg, antibody levels, T-cell response or humoral response), so that clinical endpoints won't be needed.

  • The vaccine development schedule was compressed by decreasing the time between the phases of development, moving seamlessly from animal studies to the various phases, not by cutting corners. The FDA worked on a rolling basis for the regulatory assessment to significantly reduce the time needed for review.

  • Findings in adults cannot be extrapolated to children. Clinical trials can be done in a manner similar to other vaccines, with age de-escalation.

  • Though we are seeing mutations, additional clinical trials will not be needed for potential booster shots; there may be a small immunogenicity study.

  • There are three things we really need to know: (1) Do these vaccines stop asymptomatic transmission? (2) How long will protection last? (3) How effective will they be against mutations?

  • Natural vs vaccine-induced immunity doesn't seem to be an issue. CDC recommends waiting 90 days before vaccination if you've already had COVID-19 or if you have received monoclonal antibodies.

  • The FDA is receiving daily reports on allergic reactions to the vaccine, which have generally been controllable.

This transcript has been edited for clarity.

John Whyte, MD, MPH: Welcome, everyone. I'm Dr John Whyte, chief medical officer at WebMD. Welcome to Coronavirus in Context. We're going to do the show a little differently today. I'm delighted to be joined by a co-host, someone many of you know: Dr Eric Topol, editor-in-chief of Medscape. And our guests today are Dr Stephen Hahn, the director of the U.S. Food and Drug Administration (FDA), and Dr Peter Marks, the director for the Center for Biologics Evaluation and Research (CBER) at the FDA. Gentlemen, thanks for joining me.

Stephen Hahn, MD: Thank you, John.

Peter Marks, MD, PhD: Thank you.

Whyte: Eric and I had a coin toss and I won, so I get to ask the first question. And I'm going to give it to Dr Hahn. I want to ask you what has been on many people's minds. A lot of folks have been talking about how the [COVID vaccine] data are so compelling in terms of safety and efficacy. So then the question becomes, if that's the case, why an Emergency Use Authorization (EUA) and not a full approval?

Hahn: Well, John, it gets to the heart of the issue of what an EUA is. It's a special authority given to us by Congress after 9/11 and the attacks to expedite medical products. Very simply, the major difference between an EUA for a vaccine and a biologics license application approval is the amount and complexity of manufacturing data that we would need to see in a biologics license application, as well as the duration of follow-up for safety.

But I want to make one thing clear, as Dr Marks and his team were very clear in the guidance that we issued, both in June and in October: We need to see very compelling evidence. The actual term we used was "clear and compelling evidence" of safety and efficacy from at least one randomized clinical trial for a vaccine. Those criteria are very similar to what we would use for a biologics license application.

In terms of at least the initial assessment of efficacy and safety, very robust data were received, very similar to what we would expect from a biologics license application. And what we'll do moving forward is work with the companies for that full application, get additional manufacturing data, and also get additional safety and follow-up data. Dr Marks and the Centers for Disease Control and Prevention (CDC) have put in a terrific postmarketing pharmacovigilance surveillance system to pick up any additional safety issues that might arise.

Eric Topol, MD: The EUA and the two vaccines that are now out there in the United States are really extraordinary. I want to start off by congratulating you, Steve and Peter. What you've done over the past couple of months is just extraordinary, considering all that you were up against. We had nothing, and then you also had all sorts of external, unnecessary pressures. You were basically in a fishbowl to try to get this thing through.

I think it's extraordinary, and I know how hard you all have been working to get where we are, which is just momentous. So, it doesn't go without appreciation. I want to say on behalf of the medical community, and I'm sure the public, that to see the FDA and their crowning achievement — transparent, everything vetted with the external experts and internal experts at the FDA — is just remarkable.

Now, the EUA was tightened along the way, I think back in October. Peter or Steve, could you give us some insights, because that was an important step that you all took to make sure that you got where we actually wound up, which is two complete trials that demonstrated replication of such extraordinary efficacy.

Marks: I think the issue here is that we tried to be very clear about what we needed to see. It's something that we've been thinking all along, but we thought it would be very important to make clear that if we were going to have data on these vaccines that made people feel comfortable using them, they needed to have the kind of safety data that allowed us to have confidence that the great majority of adverse events would have been seen.

We know from studies that have been done before that it takes in general about 42 days for most of the adverse events occur. And if you look at various studies that have been done, 2 months of median follow-up allows you to get the large majority of adverse events associated with vaccine administration. It's true that there can be some later events that come along, but the great majority occur within that period.

So, by putting that condition there, it allowed us to give people confidence about the adverse events that we might see. It also allowed us to know that these vaccines didn't just cause some transient immune response, that at least there was a median duration of immune response of 2 months or longer. And that was another piece of this. Because you could imagine, especially when using new classes of vaccines, that you could see some unexpected things. I think that the combination of safety and efficacy with a median 2 months of follow-up for the population was something that allowed us then to have confidence in what the output was.

Topol: Yeah, you got the full efficacy endpoints for both the Pfizer-BioNTech and the Moderna trials. Everything was just remarkable there. And another point is that I think a lot of people don't understand the contribution you made with respect to this concept of Operation Warp Speed.

My understanding, Peter, is that this was your brainchild, and you're a Star Trekkie.

Whyte: I don't know he's going to take credit for the name.

Topol: And you had this kind of incredible, crazy idea that we could all do this pretty quickly. Can you tell us about the eureka moment that turned out to be Operation Warp Speed?

Marks: Operation Warp Speed started from something that we started internally at FDA, which was Project Warp Speed. It was about how we could internally speed up the regulatory advice that we would give to product developers.

Then it gradually morphed into something more when it became apparent that a number of the vaccine manufacturers were talking about timelines that were a little bit less aggressive than we were comfortable with, in terms of when they thought they would have a vaccine. There were timelines being talked about that had us not having a vaccine until summer 2021. And when looking at modeling exercises, we were able to see that that was not really going to be a great thing, because we could already predict that we were going to have issues this fall and winter.

That led us to feel the need to escalate this and have conversations with the companies, and it ultimately snowballed into Operation Warp Speed. And thanks a lot to Commissioner Hahn for supporting moving ahead with this and advancing this up through the secretary. I'm very grateful to Commissioner Hahn for his leadership in supporting this. I think it helped wake us all up as a community that we had to do something, and I'm glad that it took off.

Whyte: I do want to ask you, Dr Marks, a little more about the safety and efficacy data. Because even when we talk about Operation Warp Speed, in some ways we should be celebrating innovation. But people are concerned because they're not hearing enough about how we didn't cut corners, how we were able to innovate and no steps were cut in terms of safety and efficacy.

You addressed the issues of safety, that it's consistent with other studies, that most adverse events or serious adverse events are going to be seen in 60 days, and we're continuing to evaluate that. But on the efficacy side, I wanted to ask you, because there's been an argument by some that it wasn't the best measurements: We're looking at symptomatic infection at a certain number of days postvaccine. Should we be looking at different types of antibody? Should we be looking at other outcome measures? I recognize the studies are ongoing, but can you explain to our listeners about the efficacy data and how those endpoints were chosen and how they were the correct endpoints in your decision-making?

Marks: I think the idea here of using a clinical endpoint of disease prevention that you could come to reasonably, in a reasonable amount of time, was a very... These were event-based trials based on that, and I think it worked out actually quite well.

In general, what we've seen with vaccines is if you can prevent mild to moderate disease, you generally have a beneficial effect on severe disease. I mean, think about it with influenza vaccine. That's how it works there, too. Even if you're not completely protected against flu by flu vaccine, you generally can show an effect of lessening severe disease. The same goes for any number of other diseases.

So although some would say, well, we should have had a more severe disease endpoint, some would have said, well, you haven't done an asymptomatic carrier study. The problem was to try to actually swab people's noses daily for 30,000 people; that was going to be a real challenge. So I think this was a balance of a somewhat pragmatic approach with what I think is a very valid endpoint, a disease endpoint, which I think we'll follow up.

Now the most important place to build next is, can we see immune correlates of protection from the blood samples that have been obtained that help us for our next trials moving forward? Do we have an immune correlate of protection so we don't have to use clinical endpoints?

Whyte: And what is that immune correlate that you're looking at?

Marks: I think the immune correlates that we'll look at will obviously be the antibody levels and, in some cases, the T cell. The immune responses that we're seeing, whether it be a T-cell response or a humoral immune response, will hopefully be things that will correlate with protection.

Topol: Can I go back to the timing issue, because one of the things that I think a lot of people don't realize is this hyper-accelerated schedule from a sequence of the virus on January 10 to the beginning of phase 1, and phase 2 trials of Moderna in March, and then the seamless [move] to phase 3. Everything is moving.

And then, on the other hand, there is critique that how come the United Kingdom got their approval a week or 10 days earlier? You pulled out all the stops to move this so fast, and then you get this flak about the United Kingdom getting a review earlier. Can you balance this out for us?

Hahn: That's a terrific question. You really nailed it, Eric, that Peter and his team worked very early on with the developers of vaccines to reduce the time in between the different phases of development to compress that time schedule — not to cut corners in the development, but to go seamlessly from animal studies into phase 1 and then phase 1 to phase 2, and then of course had to wait for the events for phase 3. And on the side of the regulatory assessment, working on a rolling basis with the companies, having frequent communications also significantly helped reduce the time that we would need with respect to review of this.

So I'm going to [say] this up front. We at FDA did not see this as a competition among regulatory agencies around the world. I can't speak to what any other regulatory agency looked at. What I can tell you is that Dr Marks' team, in the past, for an application of this size, it would normally take 3 or 4 months to fully review that and vet all the data.

Remember, FDA, unlike any other regulatory agency, goes line by line in the data. We do our own number crunching. We do our own statistical analysis. Eric, you know how this is, and John, in medicine, if you don't verify and actually do your own analysis, you're relying upon other people's assumptions and the way they look at the data. We have to get this right for the American people. So we did that.

And so when people ask questions about pregnant women, people who had renal failure and other issues, we were able to look at the data and come back with answers or not, and say, look, there's not enough data there. If you look at the issue around allergies, that was something that we looked into in the data and we could provide great confidence about what the data from the clinical trials showed.

So this is what distinguishes FDA, and I am incredibly proud of the heroic efforts that the people at CBER did. I mean, three weeks, basically, for the Pfizer-BioNTech vaccine application, and two and half weeks for the Moderna application — really record time. But at the same time, they did their incredibly thorough review. They are truly heroes.

Topol: I've been doing clinical trials since the 1980s, making me very old, but I've never seen anything [like this]. These are a couple of the largest trials in medical research history, with 74,000 people. It's really quite extraordinary how you got all this done so fast. You've set a new standard that will not be easy to replicate.

Hahn: That's true Eric. And, you know, the other thing that should instill confidence is that these two trials had 30,000, 44,000 [people]. When you look at the average trial for a vaccine that Dr Marks and his team has reviewed in the past couple of years, it's in the low 20s. I think 21,125 is the average. So it's really a lot of data.

Whyte: Let's talk about children. I asked Dr Fauci about that yesterday. As you know, Dr Hahn and Dr Marks, even in my time at FDA, we would say that children are not mini-adults; we don't necessarily extrapolate findings in adults to children younger than 12, recognizing that there are changes based on age.

So as we talk about the reopening of schools, as we talk about the symptomatic and asymptomatic journey of the disease in children, tell us your thinking about what needs to be done in terms of studies in children: whether that needs to be a priority, or we have enough information and we should focus resources on the adult population.

Marks: This is a great question. We have to study children, because the only way we're going to stop the COVID cycle is to immunize enough of the population where we have individuals who are mainly asymptomatic, although children do get COVID-19 symptomatically. But we're going to need to stop the spread, so we're going to need data in children, and we're going to need data that we can trust.

That means we're going to have to do clinical trials in the manner in which we've done them for other vaccines, which means we'll do age de-escalation. So in the very near future, we'll probably see trials starting where you'll see 12- to 15-year-olds enrolled in the first cohort, followed by 7- to 12-year-olds, followed by younger children.

It will be critical to have those studies, especially because it's not totally clear that these vaccines will be absolutely safe from extrapolation in those younger populations, because we do see this inflammatory syndrome, and we want to make sure that we're not going to see something there that is unexpected. So we do have to study them. We can't treat them as mini-adults. And those studies will get started.

Now the hope is that we'll be able to use some of what we've gained, in terms of immune correlates of protection, to accelerate that study. But we need the safety data in those populations for sure.

Topol: One of the things that has come up post-EUA is the single-dose story. As you both know very well, there was a sign of efficacy that was somewhat anticipated starting about 10 days after the first dose in both the trials, Pfizer and Moderna. And there are some people advocating studying that. More worrisome is that people are advocating that you should just take the one dose because then you won't have to deal with side effects or logistics or whatever.

Can you speak to this? Because I think there are some real concerns about potential drop-off of efficacy. The only thing that was tested was two doses.

Marks: You know, I understand the temptation here. But we spent all this time and effort doing really good science, right? We actually tried to do the science right, and then, oh, we'll just throw it all out and just —

Whyte: One's better than none, isn't it, Peter? One's better than none.

Marks: Well, one's better. But actually, there are social scientists who will tell you that one could be worse than none if what one does is makes you throw out all of the protection you're taking and rely on something that might not be true.

So let me just tell you why you need the science. It looked like that there was about 50% efficacy between dose one and dose two with the Pfizer vaccine, for example. And the problem is that nearly everyone got both doses of the vaccine. So we don't know if after one dose you could have 50% protection, but maybe it only lasts for 2 months.

Topol: Or less.

Marks: We just don't know what that answer is. So I think, as Dr Topol noted, it's not crazy to actually study this. It's a reasonable question for study. But to just go off and do this, I think it's — I don't want to say something too critical, but I think it's not exactly responsible. The reason I see healthcare providers sending me pictures where they're smiling, despite the fact that you can see it hurt when the vaccine went in or they felt kind of crummy the next day, is because they know they're doing so with the risk for that discomfort balanced by a benefit of a vaccine that they know to be 95% effective.

Topol: Yeah.

Marks: That kind of calculus gives people confidence. If you have something that's maybe 50%, but we really don't know how long it lasts, I think that's not a good risk-benefit calculus to give people, and I think that undermines vaccine confidence.

Whyte: But I think the problem too is that people are misinterpreting the data.

Marks: Yes.

Whyte: Some people are saying it's 90% effective after dose one. They're not taking those points that you're considering. I know Dr Topol got the vaccine. I'm getting the vaccine next week. May I ask you, gentlemen, are you scheduled to get the vaccine? Have you gotten it already?

Hahn: John, I have not gotten it, and I'm waiting for my provider to tell me when it's appropriate for me to get in line and get it. But I can tell you, there's no hesitation on my part. When it's my turn, I'm going to take the vaccine. I do have an adult child who has gotten it because he's a healthcare worker, and he tolerated it very well.

Marks: I'm in the exact same position; I'm lined up. As soon as it's my turn, I'm there. Those of us who are essentially office workers, we're a little lower on the list here.

Whyte: I want to ask all of you, including Dr Topol. Dr Topol, you've been such a great communicator on your Twitter page. So many of us [turned to you] during the advisory panels to see what you're talking about. I don't know how you always have these great images and charts and graphs. I mean, I can't pull it together.

So having said that, how do we interpret — and let's be honest, it's on everyone's mind — these mutations in the strain?

Topol: I'm really glad you brought that up, John, because I did want to ask Steve and Peter about this. Because we have this so-called B117 variant in the United Kingdom and other countries, but predominantly seen in the United Kingdom. It has 23 mutations. There are a couple in the spike protein, which are particularly incriminating. And one of those mutations is shared by a South African variant, this so-called N501Y.

I just posted a really deep discussion about this with one of our guru structural biologists. He talked about the molecular high-five, the two tyrosines, and how this could really make this virus entry into the cell facilitated, which is more transmission.

But what I wanted to ask Steve and Peter was, we're going to probably see further drift and evolution of this virus. We're already starting to see it beyond the pandemic-dominant D614G. So there's going to be a need for booster shots at some point, whether it's at 1 year or whenever. Do we have to go through trials at that point? What's going to be your stance regarding these adjustments or tweaks on the vaccine to take into account the drift of the virus? Because this virus unfortunately is not going away.

Hahn: No. Eric, I'm going to let Peter answer that last question from a regulatory point of view. But the other thing is that this is important for the diagnostic tests.

Topol: Yes.

Hahn: Are the tests going to have the same level of sensitivity, specificity, and predictive value that they do now? It's something that our teams at the Center for Devices and Radiological Health (CDRH) are all over and are actually looking at right now. So, Peter, with respect to vaccines —

Marks: Great question on vaccines. I think, unfortunately, this virus is starting to look like it wants to become like Flu 2, Son of Flu, or Father of Flu, with this kind of strain drift.

The nice thing about the flu though, is that it's set up for us a very nice paradigm for strain change supplements. And so we will not need... I think especially with the mRNA-based vaccines, we know how they work. We will probably just have to set up essentially what will be strain changes.

It might be that the first time we do it, we'll check with an immunogenicity study. But it's not going to have to be another 30,000-patient clinical trial. Those immunogenicity studies are usually 400 patients, just to make sure that we have the right check of what's coming out. And even that may not be necessary after we check it the first one or two times. So I think we'll have a way of evolving here with these.

And that brings to mind that the three things that we really need to know right now that I'd love to know: Do these vaccines stop asymptomatic transmission, how long will they give us protection, and what will be their spectrum in terms of dealing with these various mutants that seem to be coming up?

Topol: Yeah.

Marks: You just wonder where it's going to go now. It keeps me up at night.

Whyte: I got to ask the first question. I want to give Dr Topol the option to ask the last question.

Topol: I had two quick questions for you before wrapping up, if I could. One is to just give us your view on these rare, severe allergic reactions. The word out there is that it could be the polyethylene glycol, or it could be the lipid nanoparticles. That's one question.

And the other one is about superhuman vaccine immunity that is better than the typical person — the concept that you had 3% of people in the Pfizer trial and 2.2% in Moderna who turned out to have been previously infected with COVID. So you had a readout of people who had a prior infection, then got placebo or got vaccine. What does that tell us about vaccine-induced immunity vs the native intrinsic infection response?

So if you could just give us a little response on the allergy and this concept of vaccine vs natural infection immunity.

Marks: This concept of natural vs vaccine-induced immunity — I think the most important piece for me that I'd draw from that is that we're not seeing this antibody-dependent enhancement. If you got vaccinated and already had the virus, we're not seeing an adverse effect there. And that's consistent between both of the vaccine trials, which is reassuring to us.

Topol: Right.

Marks: It would be very cumbersome to screen people for infection before vaccinating them. I think the CDC has recommended 90 days, if you've actually had COVID-19, before you go get vaccinated.

Whyte: Or received monoclonal antibodies.

Marks: Exactly. I think that makes sense. So that's where I'll leave it with that because of the time.

The allergic reactions are something we are working very closely on with the CDC and with the companies to make sure we keep very good track of. I get an update every day. The good news is that we're seeing them settle out; it's not like the number of them is increasing wildly. I think we're going to see these as events that are going to be in the range of 1 in 50,000, 1 in 100,000.

Thankfully, they have been generally controllable with typical measures, such as epinephrine or hydrocortisone. We're going to have to keep looking at them, and we're going to try to delve into what is causing them. We will be looking into seeing whether these could be IgE antibodies to polyethylene glycol. We will look into that and try to determine what we can to try to help make these vaccines as safe as possible to use.

Topol: Great. Thank you.

Whyte: Dr Hahn, I want to give you the final word.

Hahn: I just really appreciate being able to do this. We're very concerned and continue to work on the issues of hesitancy and confidence. I'm incredibly proud of CBER in the work that they did and the transparency associated with it, and the 17,000 employees at FDA who have been heroic throughout the pandemic. So, thanks for the opportunity for us to join you today.

Topol: You've got to be so proud of what [you have] accomplished, getting these two over the goal line so fast and changing the whole way that we get out of this mess, which is a real mess. We're so grateful to you and all the people you work with to make this happen, and in such an extraordinary way and speed. It really paid off. We thank you both and everyone.

Whyte: Let me echo that on behalf of WebMD and Medscape. You really have provided us now the opportunity to crush the pandemic, and vaccination is that pathway to return to normalcy. Thank you both.

Marks: Thank you so much.

Hahn: Take care. Thanks, John.

Follow Medscape on Facebook, Twitter, Instagram, and YouTube


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.