Transcript of interview with Dr Richard Hatchett* by Dr Lara Marks and Dr Ankur Mutreja, 18th June 2021

(this transcript has been edited for clarity and brevity)

*CEO of Coalition for Epidemic Preparedness Innovations (CEPI)

Lara

What do you think the best vaccine bet is for the SARS-COV-2 coronavirus given the emerging new variants? What model should we be developing? Should we be creating a booster, third doses with the current vaccines? Or should we be going for a multi-variant kind of vaccine? What do you think's the best bet going forward?

Richard

I don't think we actually know specifically yet. This is really our first engagement with this virus, obviously, and we've watched it expand and unfold over time. We've seen the emergence of variants in other domains [such as] with flu. I don't know if you're at all familiar with a little micro-discipline called antigenic cartography, which [enables] flu virologists to understand the ways in which flu can evolve over time. They can track that evolution and map it in multi-dimensional space, and you can see sort of clusters form and evolve away from each other. And, at a certain evolutionary distance, the vaccines against the previous cluster begin to lose effect against the new cluster.

Flu is a protean virus that has a great deal of evolutionary space into which it can evolve and yet retain its characteristics as a pathogen. So you get this continual evolution that's just ahead of immune pressure from the population and evolving immunity in the population.

We don't know yet how much flexibility SARS-CoV-2 has to evolve. What we are seeing right now and accumulating evidence around is the evolution of the new variants. There's evolution in a couple of dimensions. One is, of course, towards more transmissible variants. That could occur in a naive population. If the virus happens to mutate in a direction that makes it more transmissible over time, even with no immune pressure from the population, that virus would sort of emerge from the pack. But then also the virus is now beginning to evolve in the presence of people who've been previously infected or people who've been vaccinated. So there's actually pressure on the virus that will increase over time to evolve away from the immune responses either by natural infection or vaccination. We're already seeing some evidence of that with a few of the variants where the vaccines appear to have less ability to elicit neutralising immune responses to some of the variants but not completely. I mean, it's only a partial reduction. What we don't know yet is whether the virus has enough space to evolve completely away from the vaccines and [whether we will] require new vaccines.

There was initially concern certainly when the Beta variant, the so called South African or the 351 variant, emerged, that it reduced the efficacy of the vaccines that happened to have been tested in South Africa. So, we actually had real world clinical trial data showing reduced efficacy against that variant that we might be rapidly moving to [the need for a new vaccine]. That hasn't happened yet with the other variants that have emerged. Observational studies suggest that the vaccines are still working against the Alpha and Delta variants. There is some evidence, I will speak very euphemistically here, that the Beta variant is evolving in one direction and the Delta variant has evolved in a different direction. But they're both still close enough to the original variant, that the vaccines against the original variant are still providing coverage against both. So, in answering your question, which is the question that we're all seeking a response for, is it's not clear if the initial generation of vaccines that we have will continue to serve us well with the virus changing on the margins and fringes maybe, but not changing so dramatically that we need to have multivalent vaccines, or [whether] we need new variant specific vaccines.

We are quite interested, as you know, in a broadly protective coronavirus vaccine that would protect against SARS-CoV-2, SARS-CoV-1, MERS and future coronaviruses. That is a goal that's going to take time. We are also interested in [whether] we need to have broadly protective SARS CoV-2 vaccines. I mean, do we need vaccines that are slightly differently constructed that would provide broader coverage against the emergence of the [new] variants? We're still exploring that, but it may turn out that the vaccines against the original Wuhan strain provide enough breath that they'll be effective for quite some time. We're still gathering data and gaining experience [on this]. I think we need to have some humility about what we know currently and what we can know. We just have to be vigilant.

Lara

How realistic is getting an all in one CoV vaccine?

Richard

One thing to say about the first generation of vaccines is that they all target one particular protein, the spike protein on the surface, or a portion of the spike protein called the receptor binding domain that's very important for the actual binding to the ACE2 receptor. Most of the mutations that we have seen have been either in the receptor binding domain or in a couple low [down] in the spike protein. And so because we had strong reasons to believe that the most important antibodies against the virus were directed against the spike protein that's why all of the developers have focused on those.

But let me step back just slightly. The Chinese vaccines and a vaccine being developed by a company called Valneva are inactivated vaccines. They [contain] killed virus and are actually using either the whole or a fractured form of the virus to elicit immunity. That presumably presents not just the spike protein but also other component proteins or parts of the virus. So it's possible you could potentially develop vaccines on the newer platforms that present both the spike protein and other parts of the virus, either non-structural proteins or the nucleocapsid, to the immune system. Perhaps by presenting both the spike protein and these other parts of the virus that might be evolutionarily conserved there may be very little pressure on those parts of the virus to change. [So] you might be able to create a vaccine that elicits a broader immunity and thus provide for greater durability of that vaccine against a virus, whereas the spike protein itself continues to evolve. Of course, there may be conserved portions of the spike protein that have locked in a certain sequence that essentially might be important, or where you might be able to take advantage of those highly conserved regions to produce immunity.

There's a lot about this virus that we're still exploring and still understanding, and it's the hope that we can find these combinations of weak spots that would allow us to potentially develop these vaccines that provide that broader immunity. Underscoring all of that, as of right now, the original vaccines are providing enough breadth that we're not in a situation where we absolutely need to have something new.

Lara

Talking about finding the conservative part of the virus, is that what we do in terms of looking for [a universal flu vaccine]? Is that the same kind of approach you're talking about?

Richard

Yeah, exactly. So in the stalk of the hemagglutinin [being the surface protein of the flu virus which is somewhat analogous to the spike protein on the coronavirus] there are some highly conserved [regions]. The hemagglutinin is the part of the flu virus that binds it [to the host cell]. It sticks up like a flower on a stalk. And the immune system mostly sees the hemagglutinin head, so most of the natural antibodies are directed against the head. But the head is constantly mutating and it can change shape so it can evade the immune response. The stalk appears to have some regions that are very highly conserved.

The idea is that if you could have antibodies that [can] sort of sneak in and hit these highly conserved regions, that even if there are significant changes in the head you might be able to have vaccines that protect against multiple different serotypes of flu, so H1N1, H3 and H2, H5 and whatever a new pandemic flu virus would be. The search for a universal flu vaccine has been so far not completely successful. I mean, they have ideas, they're pursuing them. There are different strategies for how you might be able to elevate the percentage of antibodies to target these conserved regions as opposed to the head of the virus, which is the thing that's most immunogenic. The experience with flu so far would caution us to be conservative [in thinking] that we might be able to elicit broadly protective vaccines or develop early protective vaccines against coronaviruses.

But there's another example from a different family of viruses that I think is a counterexample to the flu where it's been very different. And the counterexample to flu is with the viral family called the orthopoxviruses, which is the viral family that smallpox is the most famous representative of. Now orthopoxviruses are very large, double stranded DNA viruses, they're not RNA viruses like SARS-CoV-2 or flu. And that's important, because RNA viruses tend to be more mutable than big DNA viruses. But even having said that, for orthopoxviruses the vaccine that vaccinates against smallpox is not a variola based vaccine. It [uses] the vaccinia virus, which is thought to be related historically to cowpox. And the vaccinia virus, which vaccinates very effectively against variola also vaccinates very effectively against monkeypox and mousepox and every other known orthopoxvirus, So we have a single vaccine that effectively protects us against the entire class of viruses.

Lara

Interesting.

Richard

Yeah. I was just having a discussion yesterday about why that is and I don't know if we know why that is. The other interesting thing with the orthopoxviruses is that in the US they invested a lot of money in developing an antiviral therapeutic against smallpox as part of the national bio defence programmes and so far that antiviral therapeutic, which is a small molecule obviously not a vaccine, also appears like the vaccinia virus to protect against every known orthopox virus. And so you have one antiviral that protects against an entire class of threats, at least in animal studies and in very limited human experience treating people with monkeypox.

Lara

But presumably with coronavirus we don't yet know at this stage whether we'll get to that point where we have the same issue do we?

Richard

No, we don't, that's the point. I cite the counter example just to say that in some cases this is possible. In other cases, as with the flu, it's been very difficult. It varies from virus to virus. Another important thing, when you hear about a flu virus designated H1N1, the N stands for neuraminidase, which is a different protein [from hemagglutinin]. And the antivirals for flu are not targeting the hemagglutinin, they target the neuraminidase. So you have neuraminidase inhibitors for flu that work across multiple strains or serotypes of flu. So while it's been very difficult to develop a broadly protective vaccine, the therapeutic for flu is actually broadly protective. It really depends on the particular virus and the biology of the virus and we don't know enough yet about where the coronaviruses and in that range of possibilities. Obviously, our hope would be that we can come up with a broadly protective therapeutic and ideally a broadly protective vaccine.

The thing about the coronaviruses is that we do know that it frightens us and it made us invest, even before SARS-CoV-2 emerged, in developing vaccines against coronaviruses [especially beta coronaviruses]. It's important to [note] that the beta coronaviruses are an entire class of coronavirus, including SARS-CoV-1, SARS-CoV-2 and MERS. At least some of the beta coronaviruses are known to have a high mortality rate.

Lara

I'm very aware of that. So the issue is how we can get to that point where we have the vaccine at hand in case that emerges even more.

Richard

In other coronaviruses, including at least one of the beta coronaviruses, actually including COVID as well, [there] are high rates of transmissibility. What we haven't seen are viruses with the kinds of mortality rates associated with SARS and MERS that have the transmissibility of COVID or some of the other coronaviruses that we know about.

Lara

So what platforms do you think are available at the moment that would help us to prepare an all in one CoV vaccine?

Richard

The platforms that we've got are really platforms for delivering antigens. The whole definition of a platform is that you can put whatever antigen into it that you want. Now, some of the platforms likely will elicit different kinds of immune responses. The quality of an immune response [are] not uniform. Some may be better at eliciting an antibody immune response, others may have more ability to elicit cell-mediated responses, and particular pathogens may require a particular quality of an immune response. It's not even certain if we had a broadly protective vaccine, whether it would require antibody responses, cell-mediated responses, or a particular combination thereof.

Ultimately the real question is are there antigens conserved across coronaviruses as a viral family to which you can develop immune responses that effectively protect you? You can develop an immune response against any antigen, but not all. Antibodies, for example, are neutralising antibodies, so just having an antibody response doesn't necessarily protect you. The question is can you find a set of conserved antigens that elicit protective immune responses, not just a futile immune response? That's going to require [detailed] understanding the virus. Obviously, there's an immense amount of research going into understanding the virus and looking for those kinds of targets. There's an emerging discipline called ‘systems immunology’, which looks holistically at the entirety of the immune response: the antibodies that are produced, the cell-mediated components, the innate components of the immune response, and how they work together to provide protection. So, I think we're going to need to understand the systems immunology of natural SARS-CoV-2 infection.

Then are there ways to construct vaccines that would provide that protective response and provide broadly? This is a multi-year effort. Our desire is to begin investing in broadly protective coronavirus vaccines. That's the Holy Grail. It may take many years of investment to find that. But you know the saying is ‘if you want to grow a tree, the best thing to have done is to have planted it 20 years ago. And if you didn't do that, then the next best thing is to plant it today’. And you need to have these broadly protective coronavirus vaccines because I think we now have to regard coronaviruses as a true pandemic threat. It was a hypothetical pandemic threat until we encountered COVID-19. We need to get started as quickly as possible exploring these different approaches. Fortunately, it's not a completely cold start. There were a number of groups who took the threat of coronavirus seriously before COVID-19 and who even began exploring the potential for broadly protective vaccines.

Lara

One thing that is interesting about the mRNA vaccines is that they [are designed to elecit] both an antibody response as well as a T-cell response. Is what you're saying that what we need is a [vaccine that generates a] broadly protective [immune response]? It's not just about going for the antibodies, it's going for things like T-cells?

Richard

We don't completely understand the correlates of protection for vaccines yet. If we did, we would be able to licence vaccines just by doing immunogenicity studies, but that is what I was talking about. I mean the T cells being the cell-mediated component of the response and the antibodies. Virologists think that neutralising antibodies so far have [probably ] been a reasonably good predictor of protective efficacy. Those vaccines that have elicited the highest neutralising antibodies seem to be producing the highest rates of efficacy in the vaccine studies. I'm not a virologist, but virologists do think it's a little more complicated than just the antibody.

Lara

It was my sense reading up about Sarah Gilbert's work and Adrian Hill and the malaria vaccine that's how they came to the platform that they developed.

Richard

The chimp adenovirus is the main viral vector technology Sarah and Adrian work with. But you're right, malaria is a good example. The problem [has been] developing vaccines against particular pathogens. Malaria, as you probably know, has been really tricky. Historically, tuberculosis is [also] really tricky [and so is] HIV. So far not all pathogens are susceptible to successful vaccine development and it's because of the complexity of the immune response required.

I think one of the things that we began to learn about HIV and then particularly in efforts to develop vaccines against RSV, respiratory syncytial virus, which has been applied to SARS-CoV-2 [and] which is what enabled such rapid vaccine development is for the spike protein you have what's called a ‘pre-fusion’ and then a ‘post-fusion’ conformation. Once it binds to the ACE2 receptor on the cell, the shape of the spike protein changes. What appears to be very important is the prefusion form of the spike protein, the form of the spike when the viral particles are just floating around before it's latched on to the cell. [This was established by] the work that had been done on MERS vaccines prior to the emergence of COVID-19 (Kirchdoerfer et al).

Barney Graham, a researcher at the vaccine research centre at NIAID in the US had worked drawing on the lessons from HIV and from RSV about the importance of the prefusion form of the antigen.[He] had worked to develop a stabilised form of the spike protein that locked it into the pre-fusion stage and you had to insert some mutations in the sequence in order to lock it into that state. Barney and his colleagues at the vaccine research centre figured out how to do that. And as soon as they got the sequence for SARS-CoV-2, they were able to take what they knew from the MERS-like protein and bring some of those mutations over into the proposed vaccine for SARS-CoV-2. And what's remarkable, I mean, this is almost science fiction, is they were able to do that in 24 hours.

Lara

That's amazing.

Richard

And that is the Pfizer and Moderna vaccines. So literally within 24 hours of having the sequences, they had designed the Pfizer and Moderna vaccines. The rest of it has been figuring out how to produce it at scale, doing clinical trials and taking it through a regulatory process. But the antigen discovery had been done, they knew where to target in the immunogen design, you know, changing that spike protein, so that it was locked into the pre fusion state was all done in advance. As soon as they had the sequence they were able to make those changes and then [they were] off to the races in terms of trying to produce a vaccine.

Lara

Is that the kind of thing you're looking for if you're trying to go for an all in one CoV vaccine? You're looking for that research that's been happening for years ahead. Can you give me some examples?

Richard

Yeah. There's a nice article in science that reviews some of the efforts (Cohen).

Lara

What centres are you particularly working with in this area?

Richard

We're just started. We put out our proposals and we're just getting proposals in. Like you, we've read about the centres that have begun doing work on this even before COVID. We're hoping, obviously, that they'll respond to the call for proposals. The most advanced vaccinology work that's being done is at Scripps, Harvard, Oxford and elsewhere. There's a lot of attention and focus on this emerging discipline of immunogen design. And some of that is exciting as we gain greater and greater computer modelling capabilities and ability to model antigens at sort of microscopic scale. Some of this work can potentially be done in silico. As we encounter new pathogens, [we can] do that work in silico before making the investments and doing the actual wet bench lab research or the animal model testing. Hopefully that can accelerate the appropriate identification of the antigens and even accelerate some of the work on the immunogen design, which hopefully would compress the vaccine development timeline.

Lara

In terms of CEPI’s target of 100 days preparedness, where do you see it fitting into that?

Richard

I would differentiate between the broadly protective effort, which is trying to take an entire class of threat off the table, and the goal of having the capability to respond very rapidly to a new threat. Because, one, the broadly protective [approach] is specifically targeting coronaviruses, because they're really, really dangerous. The 100 days is targeting anything that emerges (CEPI). But I think they're related because at the core the basic scientific problem is you have to solve vaccinology problems for different classes of threats. It’s actually a really good analogy. Just as we were prepared for a new coronavirus threat, because of the work that Barney and others had done to solve these vaccinology problems using known coronaviruses, we need to now extend that kind of work.

Viral taxonomy is constantly in flux. Sometimes viral families get combined into one, sometimes it gets split out. We're discovering new viruses [all the time]. There are probably between 25 and 28 viral families that we know can cause human disease. We've got approximately 260 viruses that are known to cause human disease coming from those 25 to 28 families. We currently have vaccines against 15 of those families, and of the 15 families only two have vaccines on anything that could remotely be called a rapid response platform.

What we need to do is develop the capabilities on these rapid response platforms like mRNA, to be able to develop vaccines against any of the 25 to 28 viral families. We need to recapitulate for each of those viral families. Some we have lots of information about, [but] others comparatively little. But we need to solve specific vaccinology problems for all of the families in the same way that we solved it for coronaviruses in advance.

To achieve the 100 days, we have to be at least as well prepared for any viral family as we were for coronaviruses. We have to be able to do that vaccine design as soon as we get the sequence. ...and to be able to compress the clinical development, the manufacturing, and the regulatory parts of the problem, as much as we possibly can to enable us to rapidly manufacture and test vaccines and assess those vaccines to a point where we are comfortable using the vaccines. Even if we don't have absolute certainty about their level of efficacy, we want to be comfortable using those vaccines much more rapidly than we were able to do.

Lara

Would you say that we're better prepared now because we have genomic tools to sequence very quickly now?

Richard

Yes, absolutely. Just as I broke down the 100 days of the vaccine development process into the immunogen design, clinical manufacturing and regulatory pieces, I think you need to think about the response to a new virus. There's three big buckets of response that need to be tied together. The three big buckets are on the front end. You need much more capable global bio surveillance techniques and that's where the sequencing comes in, with rapid front line genomic sequencing so that you get earlier detection and warning about new threats. If we'd known at the beginning of December about the new virus, rather than a month later, you could have started responding a month earlier. 1) there would have been fewer cases, 2) you'd have [had] an extra month of time for the vaccine development. The second piece of that response is where the sequencing comes in and that hopefully would trigger the third piece, which is the countermeasure development.

The second piece, which we've all been living through for a year and a half, is we need better, more scientifically based, more tailored public health interventions to suppress disease transmission. And if that had started at the beginning of December, rather than the middle of January, we [would have had] much less disease, much more chance to contain it if not to completely suppress [it], even before you get to countermeasures. If you're starting countermeasure development, and you can do it in 100 days rather than 300 days, and you're starting it earlier in a setting where suppression is rapid and you're trying to keep the disease from spinning out of control before you even get a chance to start, then the combination of those three elements potentially puts within reach this idea of eliminating pandemics as a risk to humanity. It's those components on the vaccine development and the countermeasure development, therapeutics and diagnostic development that need to be compressed. Earlier detection, earlier sequencing, and earlier more tailored public health responses. If those three things come together, that's how you radically reduce epidemic and pandemic risk to the world.

Lara

The problem is, obviously, that's reliant on society, and you've got different countries with different governments and different ways of looking at the problem.

Richard

It is. I think the opportunity that we have now is that we actually have the technology. The technology to do that is no longer fantasy. We have the tools that we need and we know what we need to do. And we've got, at least for the foreseeable future, let's say at least the next five to 10 years, this very salient event that has been globally shared. There's a global appreciation of the risk and the consequences of failure. Hopefully that can be converted into political will, to make the kinds of investments in the first piece, the surveillance and sequencing piece and the countermeasure development piece.

Society and the social response is critically important obviously, and we've seen how divisive the response has been in many countries during COVID-19. I would argue that part of [division has resulted from the] open ended [response]. The duration of the requirements for lockdown and the differential risk across a population that COVID-19 presented didn't help with high compliance rates. In the future, if it could be more geographically restricted, [where] you respond with these things in a tight geographic region and a promise that it actually only [will] have to be in place for 75 or 100 days [will enable you to] bring the public along, to have confidence that this is contained, and that we have to do this so we don't all have to live through a 2, 3, 5 year pandemic again, I think it changes the equation. There's a huge piece of public communication, public education, public buy-in for this global project to reduce that risk. If the public believes in the capabilities and governments are willing to make the investments, then I think an effective community might be willing to tolerate things and even make some adaptations in the future to make it go smoother.

References

CEPI, J (10 March 2021) 'CEPI launches plan to tackle risk of future pandemics and epidemics'. Back

Cohen, J (15 April 2021) 'Vaccines that can protect against many coronaviruses could prevent another pandemic', Science, 342/6165, 1484-90. doi: 10.1126. Back