Novavax, Inc. (NVAX) Earnings Call Transcript & Summary

Hello, everyone. Welcome back to the second webinar of this May COVID-19 series. This is Wing, the Congress Director of the World Vaccine Congresses. And thank you very much for tuning in today. And I want to take this opportunity to thank our webinar sponsor, DNA Genotek, for supporting and helping us deliver today's webinar of Novavax. So just a gentle reminder before we kick off, you can also make questions using the control panel, please do. And we will do our best to bring these up during the Q&A. And this session is video recorded, and you will get a link to it tomorrow, 24 hours after the session. So I'm very, very excited to have Dr. Gregory Glenn today from Novavax to present on their COVID-19 vaccine. You may have seen the news on the investment from CEPI. And he's also part of the World Vaccine Congress Scientific Advisory Board and a wonderful speaker. So we are definitely in for a very interesting session. I know that you're all familiar with our moderator by now as well, Dr. Philip Krause, Deputy Director, OVRR, CBER, FDA. So I will leave the introductions there. And so we can hand it straight over to him. Thank you so much, Phil.

Yes. Well, welcome, everybody. It really is a pleasure to introduce Greg Glenn, who's President for Research & Development at Novavax. He has a huge amount of experience, almost 30 years in vaccine development and, of course, has been leading the programs in vaccines at Novavax for quite a while, which have focused on a lot of pathogens, but recently, RSV, which actually has some structural similarities to COVID-19 as well as a number of other pathogens, Ebola, other emerging infectious diseases. So he's going to talk to us today on the topic of recombinant nanoparticle COVID-19 vaccine platform technology for emerging infectious diseases, which I think will be a good contrast with the talks that we had yesterday on DNA vaccines and one that we'll have tomorrow on RNA vaccines to give you a flavor of the different approaches that different people involved in the fight against COVID-19 are employing. So Greg, it's a real pleasure to have you here, and we all look forward to your talk.

Thank you very much, Phil, and thanks, Wing. I think this seminar -- these webinars are really very valuable. As Wing mentioned, I'm one of the advisers for the World Vaccine Congress, which I think is an extremely good format. It allows speakers to creative review and go through their topic quite thoroughly. And so I hope to do this, this morning. But I'd first like to just -- also just thank the public health officials that have been so helpful and their leadership, Dr. Redfield, Dr. Fauci, Dr. Rick's that's just -- and of course, the leadership of the FDA. So I think this is really such a historic time for us at Novavax. My team has been focused on type 1 fusion protein vaccines. And the theme is common for how they infect in terms of RSV and flu. And so -- and we have been working on SARS and MERS previously. So it really has been gratifying to be able to work with the people that I have at Novavax, extremely good scientists, Gale Smith, for example, who first identified the insect cells and expression system. So I feel very confident that we can bring forward a vaccine. I think when you look at the interventions that are being brought in, and all I can say is I have with humility, it's not entirely clear what to do, and I -- hats off to all the huge efforts that have been made to try to address the problem. But certainly, a vaccine -- working vaccine soon would really be important. So we wake up every day, my group is extremely motivated to create for the public safe and efficacious vaccine. We're trying to work with widely and collaborators. Hats off to many other companies that are making good progress here. I think it's -- I'm optimistic that the field is going to be able to address this big problem, this historic problem in a historic way and bringing to bear all the great science and development technology we have today. So I'm just going to jump right in. We have a prefusion, stable recombinant spike protein nanoparticle vaccine. It's what we've been making for flu, RSV, MERS, SARS, Ebola. And one of the hallmarks, I think, of our technology is it's highly immunogenic. And I'm going to press some definitions on that in my talk because I think it's important to kind of define those terms, robust immunogenicity and highly -- because I think we can. And I think it's important. We do have a mature platform technology. We just completed a Phase III trial with NanoFlu. Flu needs a differentiated product. I think we have -- we just had a pivotal trial. We met all of our endpoints from an immune standpoint, from a manufacturing standpoint. That vaccine -- the adjuvant there is same. It was in an older adult population. So there's a lot of maturity to our program. And that allows us to take risks. And many people are doing this -- doing things in apparent development risk, I would say. And so many companies are employing the same. We have a Phase I study that's up and going. We have -- we'll have data in July, maybe some earlier. And we're looking to be a player here. We -- I think the other thing about our technology -- it's very immunogenic. It's very stable. It's a mature platform. We can scale up. This is something that is not theoretical. And so we are now looking at being able to produce up to 100 million doses by the end of the year and beyond that 1 billion doses. And so CEPI, this recent funding grant, that was fantastic for us, is allowing us to plan that specifically. And so we're working to have the kind of data that could be considered for deployment. And we're taking -- I think a lot of the leadership on that will be -- obviously, will not be up to us. It will be up to regulators, funders and our government. So we're looking forward to trying to be a participant in a major way in the solution here. So let's just talk briefly about our pipeline. As I mentioned, we have completed a pivotal Phase III trial successfully for [ improve ]. It's an adjuvanted nanoparticle, seasonal, quadrivalent vaccine. The data looks very good. You can see our coronavirus program. We've been working extensively with RSV in 2 major areas. We've got promising data, and we intend to continue to develop a vaccine for RSV setting for infants and older adults. And we had a very good Ebola vaccine. I'm going to talk about that because that was -- that's been a blueprint, I think, for us. A lot of commonality. The subjects are immune naive. It's an emerging disease. So speed is of the essence. And we have very good data through Phase I, and it's been helpful as a blueprint for us going forward for our development year. So you can see on the right. This is a nanoparticle. It's an RSV nanoparticle. This is a fantastic image taken by NIST using small-angle neutron scattering, small-angle x-ray scattering. And that allows you to take a full-length glycoprotein, which is -- has a transmembrane domain and ectodomain. So in a normal solution, you just simply can't do this. Our technology allows us to manufacture a full-length surface glycoprotein. And what you can see here is these are trimers that they stick out from a PS 80 -- well-organized PS 80 nanoparticle, which has about 300 detergent molecules in there. So it's a protein-detergent micelle, 30 nanometers. It looks like a virus to the immune system. It also allows presentation of -- you can see here physically the sides, stem, other epitopes that are normally hidden in a highly packed surface glycoprotein on a virus. So we have a unique way of presenting antigen. You get the native configuration, but it's not -- it's presented in a more robust way. And all this to say is it's a good technology. We have a lot of experience. And especially the work we do early in development, we know it can project to our clinical experience very accurately. And so that's been helpful. And I'm going to walk you through a couple items here. So just some highlights from our seasonal influenza vaccine. It's -- we just released our results in March. We met 8 co-primary endpoints. We are statistically significant with our secondary endpoints. We had good cellular immunity. What you can see is what we're trying to do is improve on the -- especially on the drift strain responses. So the top bar -- this is from our Phase II trial. I have the reference there. It should be out on the web now. And you can see we're increasing -- improving the sort of functional immunity using HAI by using an adjuvant, using the nanoparticle compared to the market leader, which is Fluzone High-Dose. On the H3N2, where there's been a lot of antigenic drift and really very disappointing vaccine performance due to the fact of the evolution of this particular strain with flu, resulting in a lot of morbidity and mortality. So older adults, adjuvanted vaccine, nanoparticle, better response. You can see on the bottom right, we think in the flu -- having functional T cell immunity is going to be important, and we're here comparing ourselves to a couple of licensed vaccines. And again, you can come back and review the data. What you'll see is the adjuvanted flu vaccine with nanoparticle makes a very robust, CD4 polyfunctional T cell response. And we think that could be important here. And so that experience, safety, immunogenicity, dosing, many -- the formulation, many of those things could be applied here. I just want to say a couple of words about Matrix-M. It's the adjuvant we're using. It's a saponin-based adjuvant. It's sort of -- that whole field has evolved. We think it's -- that it has found a good balance between adjuvanticity, reactogenicity. We did a lot of work with this with Ebola. This just shows you our CD4 and CD8 responses compared to aluminum adjuvanted, which you can see here on the bottom is the key. But you can see here, we tried this compared to aluminum, which is in yellow. And you can see we get very robust T cell response. These are all Th1-like cytokines when we adjuvanted the GP with Matrix at a couple of different doses and compared that to alum. So we have a good sense that we have the Th1 like response that is one. And you can see here that that's published, and you can go look for that data. So great adjuvant. And just a couple of other points. It does improve the immunogenicity with/without adjuvant, and it seems to give qualitatively better responses than aluminum in this case. That's not always the case, but it's important to take a look. But here, you can see the neutralizing response and it was fully protective in that challenge model. So if I may step into baboons, so for us, baboons, we have a great collaboration with the group at the University of Oklahoma, James Papin there. And they were able to pivot very quickly. So when Ebola was identified and the sequence was published, we literally went from the sequence to having clinical data in about 90 days. And one of the bridging studies we did there was to do this in baboons, which represented for us essentially something like a Phase I trial. And the lessons we saw, the immunogenicity we saw in that trial was translated perfectly so as a blueprint for what we can expect in human trials. So what did we see? We did see robust antibody in T cells, and I won't have time to show you here the key role of the adjuvant. So here, we gave a 60 microgram dose of unadjuvanted nanoparticle and really it was much less immunogenic. Although it's not nothing, but you can see when we added our Matrix-M adjuvant, you can see that we had a very nice response, really, really strong. And actually, what you can see here is dose sparing. So 5 and 60 micrograms, it didn't matter in terms of the magnitude of the response because I think we're at the top of the response S-curve. So my definition of immunogenic is when you have a low microgram dose and you get a very robust response, and I'll talk about robustness in a minute. And so you can see it persists. We also just boosted the animals out -- about a year out. You can see very nice boosting. So persistent high responses. In this case, it was better than alum, and that was a very nice, very helpful study for us to know before we got into the clinic. It was done rapidly. And so here, this has been published now, but just to show you more -- sort of more color in humans. So this is a human trial. We gave a dose at day 0 and 21. You can see the immune response at day 21 with the adjuvants. These are unadjuvanted regimens. These are sort of the levels -- there's a lot of debate about what might be protected, and I'll talk about that in a minute. But you can see the second dose gave us a very good response. And now you can see with the 2-dose regimen, really great persistence. And in fact, out at day 365 as good as a single dose regimen. We can talk about the dosing regimens and options for COVID. But this -- my point is good blueprint for what we're going to see, and I'm going to show you some data in just a minute with our COVID vaccine on this. So again, this is going to be an issue for COVID. What assays do you use to judge the relative immunogenicity. So there's a large collaboration. They came up with a centralized lab and a centralized ELISA, very well controlled. And when you look at our vaccine, you can see, in this ELISA, very robust responses. Again, you can see the time frame down here. This is in humans. You can see it kind of dribbles up over time and then it persists a very robust response. And a good marker for this is this -- the red vaccine is the VZV, which I'm very thankful to see that's got licensed. That's a big breakthrough to have an emerging disease vaccine license. And relative to the immunogenicity of the vector vaccines, we find that our vaccine is really quite immunogenic, and this is a good way to look at it. So I think -- looking forward, I hope there's going to be a moment when the immunogenicity of the vaccines can be looked at. And relatively speaking, with something that might be functional, this is considered to be functional and sort of predict protection. So that was a very useful comparison. So just now back -- getting back to COVID, I think our -- I made these points, our previous experience has allowed directional development, confidence in our early development with COVID. Our vaccine was engineered for immunogenicity, stability and productivity. Now going to that. And we have this spike protein structure. We believe -- and frankly, we've taken our cues from people like James Crowe, McClellan. There's been really good work done on type 1 fusion protein structure as appropriate for vaccines. And we've taken advantage of very good characterization tools. So what do we do? So when the sequence was -- actually, when the virus -- when the clusters of pneumonia first were described, we're -- because we're a flu company and we're an emerging disease vaccine, we had an internal conversation and said, oh, there's another pathogen out there we should be interested in. It's our ammo to make a vaccine. We've made many, many vaccines. As you can imagine, with flu, we're constantly watching the environment for new flu strains. And so we often will take that, get the gene sequence, download and make a vaccine at some level. And so we just -- we started on that. It was around -- actually around January 20. We were a little slow, thinking about it that it might be important. But we had experience in the past where we'd taken from gene sequence to Phase I data, made a vaccine and have been and gotten into the clinic with Ebola and H7N9. And so those are both published studies, very good data. So we've been through this drill before. So we have this gene synthesized. But we felt it was important because this is a fusion protein, and there's lots of issues around the structure, function, immunogenicity, et cetera, to evaluate a large number of instructions, sub domains, full-length proteins, different mutations. We screened these for binding to the receptor to the H2 receptor for stability, for productivity and immunogenicity. And we landed on this -- now our candidate, this very sexy name, NVX-CoV2373. But it had the characteristics we were looking for. It looks very stable. I'll show you data as prefusion. We know that from a number of now characterization studies. H2 binding, which is really critical. And so this has now become our candidate. And I'll just note here, this is now the primary structure. There's a couple of issues here. So these proteins have a furin cleavage site. When that's cleaved, it allows it to advance into a post-fusion form. As you know, the fusion proteins act like syringes. They essentially -- in this case, it's both binding and then also go through a fusogenic confirmational change, which allows it to inject, if you will, the RNA into the whole cell and the whole sell becomes a virus factory. So the structure of this is really critical. You want to try to reflect what might be seen out in nature. And so by freezing it in this upstream prefusion form by taking out the furin cleavage site and then there's a 2 proline mutation on the S2 around the heptad repeat, which keeps this from unfolding. And so we have a prefusion stabilized form, which I think is fairly common. We have a lot of data that shows that, that is optimal here. And so we took the time to evaluate these structures. So here's our structure, and just to show it interacts very nicely with H2 receptor. And then the other point I want to make is Matrix-M as an adjuvant is really critical. It causes many things, increases affinity maturation; better, higher, longer-lasting immune responses. We've got a good safety profile in the older adult trial in the Phase II trial where we compared our vaccine directly to Fluzone High-Dose. We could not distinguish sort of the reactogenicity profile between the two. So we think it can be advanced as a candidate for this. And so it's very important. We've done a number of things. This is just one of our own internal studies where we looked at 2D class imaging. And you can see it's in the prefusion form. You can see the tail. The nanoparticle itself is up here. As I mentioned, we are collaborating with others to get and have seen the crystal structure. And we're using sacsands to actually look at the drug substance, which is a full nanoparticle, very tough to do. It's a really great technique that NIST has. And you can see -- to Jason McLellan and his group, I think they created a very nice road map with this paper. And it's something I think we're all grateful for. And I think it does -- again, just to give credit, it's great to be in the part of a very active scientific community, and I think this is going to bear out to be an important observation done at this level. Okay. So when we make our spike protein, one of the things we want to know is does it bind to the receptor? That should be needed as that is going to reflect it being in a proper configuration. So here, we've taken recombinant H2 receptor. On the left, it's an octet. We put it on a chip. We put -- we see it go on. And then normally, you would try to wash it off and you would see this down. So this sticks. It's very high-affinity binding. We can't actually measure it because there's no off-rate. And we could have an orthogonal method with the same principle, binding to ACE receptor in the format of ELISA. Here, you can see the MERS receptor. So it's a very specific, very nice tool for looking at structure -- function of the spike protein. So just to make the point, some of the constructs we made, this is one of the screening tools, we said, look pretty good at the beginning, but when you put them under stress conditions, which we do here. So stress conditions include low and high pH time, agitation, et cetera. The oxidation should do something and it did. So here you go, no changes in all these very harsh conditions where our spike protein is subjected to stress conditions. Whereas this other construct -- previous construct looks pretty good. If you just leave it alone 2 to 8 over time, there's some change. And then when you do these stress conditions, you can see it change. So this is very nice. This H2 receptor binding assay is a very good way, I think, to interrogate the stability of your construct. Okay. So let's talk about immunogenicity. Very important, in my view, to think about the assays. So just to show you here, we have 3 assays. We measure after immunizing mice in this particular experiment with Matrix-M with our adjuvant. You're seeing now the immune responses and -- measured as anti-spike IgG. We also have an assay -- now taking advantage of the principle, I just showed you with the binding, where we can now block this with antibodies. I think it requires a very high-affinity antibody. So this is now a functional assay and in classic neutralization. We believe that you need to do this with wild-type virus, with Vero cells. It's a 3D assay. It's very traditional. And it gives you a benchmark to what's been done in the past. And I think when you see neutralizing antibodies in this sort of assay, you can make some judgment as to the robustness. So here, 4-parameter fit curve, not an optical density reading, not an endpoint titer. And you can see with 1 and 2 doses really quite good responses. They're proportional. These 2 functional responses look a lot like disproportionally. You can see after one dose and two dose. And then I would just say, quite remarkable neutralization, titer of around 1,000, and this is a titer of around 10,000. Thanks to Matt Frieman, University of Maryland, who has a BSL-3. He's a very well-respected investigator. And he's been a great collaborator for us. So this binding assay is helpful because it seems to translate into neutralized antibody when you can detect it. I can't show you all that today. But this is now a study where I think we can comment on how -- that our construct is highly immunogenic. In my view, that definition will be, if you can take nanogram amounts and create a functional immunity, your vaccine would be, in my view, highly immunogenic. That doesn't mean that you can't get a good immune response with higher doses, and that can be fully appropriate. But when you talk about relative immunogenicity, this vaccine is highly immunogenic. You can see the adjuvant effects of 10 micrograms. This is in mice versus 10 with Matrix. So the adjuvant effect is clear. And even at 10 nanograms in its 2-dose regimen, you can see these functional responses. And by the way, we can detect anti-spike protein a lot more robustly, if you will. And you can see this very nice dose response. So very immunogenic construct. So now turning to baboons, which I talked about before. Again, we collaborated with University of Oklahoma. They are able to get animals for us. And we've measured anti-S IgG and the receptor inhibition and neutralization. And here, you can see, if you step back to proportionality, again, this is an endpoint titer. You can see we had 1, 5 and 25 and -- the anti-spike response is really quite good. Thousands after 1 dose of 1 microgram. And -- so we thought about, do you need a 1-dose or 2-dose regimen, and maybe we can talk about that more in the Q&A. But this manifests itself as really quite good neutralization, again, conservative assay. So this is baboon. So we should expect to see something approximating this in our human study, and this is very encouraging, I think, to see this type of immunogenicity. We've also looked for T cell responses. There's a lot of interest in sort of the profiling. So when you do this by ELISPOT, you can see we make with our vaccine and adjuvant quite respectable interferon-gamma-producing cells 10^6 in this ELISPOT. And we did have an animal that was seen to be ill, so it kind of dropped out. But over here, you can see IL-4. And just to be clear, these are controlled. So we know that the assay was working, really not seeing IL-4 when they're stimulated with antigen, which is good. So very much a Th1-looking profile. We also did these polyfunctional factor T cell assays with double or triple cytokines. You can see really quite nice responses here. We think these should be functional and help with resolution of disease. And so this is a nice feature of the vaccine with the adjuvant here. So -- okay. So now when people talk about robust immune response, I think it's now possible to make that claim based on data. So here, we have convalescent sera. Thanks to collaborators. They were all PCR positive, had illness. And you can see here, this is now the individual sera, measuring the spike protein IgG. So you can see across the line, they're somewhere in the hundreds to 1,000 with a maximum of 10,000. You can see with our vaccine, with one dose, we're certainly reaching at the upper limits of what we're seeing from convalescent sera with the second dose, where at least a lot better than the best responses for convalescent sera. So I would contend that if you can make this kind of comparison, you can say that your immune response looks quite robust. So just -- turning to the functional assay. So this is interesting that people that have had same sera, same people have had responses, not all of them seemed to make functional immunity. And I think that's being talked about in the field. This may be overly stringent. But certainly with the vaccine, we see this very nice ACE2 receptor inhibition blocking antibodies. And they're as good as the best one seen in convalescent sera. So I think we're seeing robust functional immunity from our baboon study. Very helpful for me. It is validating that the vaccine should produce this in humans. And so on the go, the human studies will be really important. Just to say that we are moving into multiple animal models for safety and efficacy to support the deployment of the vaccine. And so there's been a lot of work in this area. We've taken that into advisement, and most of these studies are actively being done. The data should be completed well before we intend to go into Phase II. And then, what do we do in the clinical plan? We have -- we began to enroll. We have a Phase I trial. It's being done in Australia. Its subjects 18 to 59 years of age. And the key immune measures we've seen before, ELISA, receptor binding inhibition, neutralization and cell media immunity. You can see on the right, it's a placebo-controlled trial with 25 of antigen given twice by itself without adjuvant. Obviously, we need to demonstrate the need for an adjuvant. And so group C and D, however, are of greatest interest, 5 and 25 micrograms of antigen with 50 of Matrix given twice and in control with the single dose of Matrix. And as I mentioned earlier, we expect to see dose sparing, which is convenient here where the lower dose of the antigen is still quite effective. We have a very robust surveillance system setup and safety monitoring committee. We should see the data in July, and that should allow us to pivot into the Phase II trial since this is a single protocol, and this will be based on the day 35 results. Right now, proposed to be 2,200 subjects with older adults, U.S. and Australia. And we may look at some dosing regimens in that trial. And obviously, we'll be collecting disease endpoints, multiple severities. And we hope to be -- have the data that could trigger the next stage Phase III or EUA. And again, we're not going to be the ones, I think, driving those initiatives. We're just trying to be prepared to participate. We are doing other parallel trials and other geographies are considering that because we think that, that can add value to the data package. We have just -- as we mentioned and that recently announced CEPI funding, it's a very large, I think, affirmation. The data you saw suggests the vaccine could work. It's going to support us. It's allowing us to really scale up the manufacturing. And I think meet the goals -- the production goals that I outlined earlier, both Matrix and adjuvant and gets a large global scale manufacturing capacity. So just to summarize, we have a very good candidate. It's a prefusion, stable immunogenic spike protein and a nanoparticle, used Matrix-M. We're scaling up in parallel. We think we will deliver 100 million pieces this year and more than 1 billion in 2021. And we're -- that clinical [indiscernible] underway. We would like [indiscernible] vaccine in a year [indiscernible] regulatory authorities. I think that's important. Our [indiscernible] under emergency missions. And I'll stop. I'll take -- there is a little glitch in the sound. Maybe if I missed anything, I have controlled the [indiscernible].

A number of questions about your previous vaccines and the outcome of those. And people are pointing out that there have been a number of vaccines developed using this technology that haven't yet been FDA approved. And -- but presumably, [indiscernible] to immune responses. So the question is, I guess, it can be put [indiscernible] [Audio Gap]

Look, yes, those are good questions. I think, first of all, the insect cell-derived platform, which we make our antigen is the basis for flu block. And so our lead scientist, Dr. Gale Smith, invented that expression system in his Ph.D. time. So there's been a lot of vetting of the antigen expression system. And so that's important. We don't have licensed products. We do have a lot of experience in the late stage. For example, with RSV on the CMC side, we've done the PPQ, et cetera. So we have a lot of information that I think could support a BLA or a licensure package on the CMC side. So -- and obviously, flu in -- was a pivotal trial for Phase III. So I think that that's helpful. I mean -- and it's super -- very directional and so many aspects of development, all the different disciplines. We're not dealing with an empty inbox there. So -- but that being said, it still requires meeting stringent standards that I think are going to be set out by EMA, the FDA to take it to licensure. But I think those -- one thing I think we focused on -- in RSV, there's been a lot of debate about assays and it somewhat relates to our data, our construct, et cetera. And I just would say those debates have helped us really sharpen our pencil here and help us think about what's important. Assays are really important. Making -- in this case, we're going to end up making some projections based on immunogenicity. I think there will be an opportunity here to collect efficacy data, so the combination -- but it may not be the normal pivotal efficacy data. So getting animal data, getting good assays, validated assays, meaningful assays, I think, is going to be a really important part of this package.

Thanks, Greg. One thing that you have a lot of experience with this RSV. And of course, if you go back to the 1960s, RSV is where a lot of our concerns about enhanced disease evolved from. And so you've obviously spent a lot of time thinking about enhanced disease. And so what polls can you give us about enhance disease, both in general, as well as in the context of your specific construct?

Yes. I think, first of all, it is the one place where it actually is a clinical phenomenon. It's a theoretical risk in the context of these coronaviruses, but it really happened in RSV. So there was an infant child study. The whole virus vaccine was formed and then activated, used an alum, there is an 80% hospitalization rate in the -- in children that have received the vaccines compared to placebo. So that has spawned, as you can imagine, a large attempt to define exactly what happened. I don't think we can fully, exactly, precisely reproduce that. But there have been some useful things. For example, I think that the cotton rat model was useful. So the actual Lot 100 that was used in that trial was available at a vendor. So one could perform a study and in contrast, if you immunize with this, it was actually quite immunogenic. And so you could contrast that with your construct. And in the -- I would say that the striking feature that you could see with a Lot 100 in a cotton rat, which then if you immunize good immune response but not functional and pathology is that it was digital. There was no all no functional antibody, you could detect and you immunize with that inactivated vaccine. So that's a good -- to me, that's a good marker. So I think we would all agree that neutralization is likely to be important. In the context of RSV, the strange thing is that babies are actually born with neutralizing antibodies, but they're not fully protected. So it's a bell curve. But I think with enhanced disease story there, this signal is very dramatic. It was 80% -- These are very sick infants. And so it won't be subtle, I think that the proposals to do several animal models, looking at its pathology is important. Now we did -- we also addressed one of the big questions, is that the thoughts for one of the hypothesis is that if you make antibodies, and they're not the greatest that by the time they become low, you should see enhanced disease. So that was one of the hypothesis that actually, I think our trial debunked. We did -- we immunized mothers. So the infants receive the bolus of our vaccine, and then we follow them out through the year. And through the year, the overall -- and so the phenomenon should manifest itself as an SAE, a respiratory hospitalized SAE. Through the year, we saw a 50% reduction in all hospitalization clinical respiratory SAEs in the vaccines compared to placebo. So even though the antibodies ran out, you continue to see something in favor of the vaccine. And I just think that that's a hypothesis, in my view, if you -- make you utilize antibodies, you're very unlikely to see enhanced disease. Other thing that's been touted, and I'm not sure about it is Th1 versus Th2 and use of alum. And I just -- I know there's a group in China that had a inactivated vaccine using alum, got good protection, no histopathology so it's not -- it's a picture we have to take seriously. We have -- you have to do a number of animal models, I think, is reassuring. But at the end of the day, I think you need to have a good safety monitoring committee and assess the risk benefit. And here, I think this is such a dire situation. We should not let, with prudence around showing utilization, functional immunity. We should not let that stop us from being forward aggressively. We'll see it if it's very doubtful. It's a clinical phenomenon. And so just, I think, a relative level. I think that's the guidance we're getting from partners as well. And so that would be my wisdom. I think RSV was a place in our setting where I think we addressed some of the questions. And I hope that this is helpful. So functional response should be unlikely to have enhanced disease in the clinic.

Thank you very much. Maybe I just at vaccines and [Technical Difficulty] right? Are you hearing me? Yesterday about DNA labs [Technical Difficulty]. The question that comes up. Parent contrast -- the differences between the way in which the antigen is actually presented in the immune system using these different platforms, which may give people a sense of differences in the way in which the platforms induce immune responses?

Yes. I think we're kind of in a classic in some sense a classic setting I am injection give a danger signal a form of an adjuvant, had something that looks like a virus. I mean those are all themes we know are important. And I -- one of the things that take some this is that, [Technical Difficulty] me, a vaccine older [Technical Difficulty] efficacy. That's an adjuvanted and [Technical Difficulty]. Can you hear me? Okay. Can you hear me?

Now we can.

Can you hear me? Okay. So for some reason, okay. It's a signal coming from you that seems to be very weak. So can you hear me?

Yes.

Yes, I can.

Okay. Now can you hear me. So on the part [Technical Difficulty] vaccine is very -- lines up, presenting looking for less likely to [Technical Difficulty]

Hello?

Yes, I can hear you.

All right. You can tell me, this is right, because I'm having a hard time hearing you. I'm imagining that your platform is producing the virus antigen in something that may be presented to the immune system in a more native confirmation, while you claim [indiscernible] has these vaccines are presenting the engine in a way that puts them on the surface of cells, which might or might not look as similar to the virus is what you're producing. Is that a fair?

Yes. I mean, it's possible. I think that the issue is you don't know precisely what you're making when you're objecting the construct and expressing it in VIVO. So it's possible. The presentation I -- look, I think it's doubtful you're going to see enhanced disease in anything. And so it's a little hard to assess the risk of one over the other in this. I don't subscribe to that. I just -- I think it's -- we can see what to do, look in animals, look for enhancement and monitor carefully out in the clinic. I think that's the best answer to this.

Sounds good. On one of your earlier slides, you were showing what appeared to be strong CD8 responses, which bought my eye because typically, one thing, so CD8 response is being a result of presentation in the context of MHC Class I, which is at least most typically occurring when the antigen is expressed inside of a cell. Do you have any comment on the breadth of the cellular immune responses and why it is that your construct might also be inducing some CD8 responses. In addition to the sort of more classic CD4 responses.

Yes, yes. That's a really good question. So look, this is a feature of adjuvants. I've been working with adjuvants and delivery system for some time. So normally, you think of this is a cross presentation is not as robust, but requiring, obviously, the generation of antigens in an infection intracellular. But you can see cross presentation with properly configured antigen. It's a little harder to do this assay in the clinic. And so we haven't done much of it, but I don't think it's a surprise that you can get CD8. You need the epitopes to be present. The assays are a little different. You need either antigen presentation or different ways of looking at it. So my experience, years of working with adjuvants and delivery systems, this is not a big surprise. It doesn't require peptide, It doesn't necessarily require the cross presentation can occur. And I think it's something we haven't looked at. We don't know whether or not this is going to be a key factor for COVID. Nobody knows right now. So we're just trying to show that we do support the immune response with the kind of T helper response that people think is important and we'll measure this out in the field. And someday, maybe they'll be immune correlate, but I just think it's a characterization of the type of immune response is what we had in mind here. And it is and it's a classic way for making cross presentation with adjuvants to make CD8 responses.

Thanks, Greg. So there are a number of questions that anticipate the successful demonstration of efficacy of the vaccine and the deployment of it in the field. And these questions sort of or have to do with the deployability. And one question sort of asked about the trade-off between a higher dose, which might give a better immune response, but being able to make enough doses another question asked whether this might actually also work at intranasal vaccine, which we might be able to avoid needles and syringes. And there was one more question about the thermal stability of the vaccine, how deployable is it around the world?

Okay. So I'll try to go backwards. So one thing the Ebola vaccine taught us is that we seem to be at the top of the dose response curve at the lowest dose. So that's good. That's the top of the s curve. So that's convenient. We can use the lowest dose. I think that -- so you can see that here, all those lines pop together, that's a dose range between 50 and 6.5 micrograms of antigen. So that's a feature of the subjects being immune naive, the vaccine is immunogenic and especially the adjuvant really brings that out. So that's convenient because that allows you to consider using the 5-microgram dose or even 25 is still quite good. And I think from a capacity standpoint, we're going to be very strong there in terms of adjuvant antigen supply. So here, is it thermostable? What we're finding is with the stabilizing mutations, these are really stable antigens. And under stress conditions, we can't -- we -- this is predictive, we think. And we are now having -- because we've manufactured, we now have some real-time stability to support our program. So it looks like a very stable, doesn't -- obviously, it does not have to be frozen, our drug product would be likely handled within the cold chain at 2 to 8 degrees. And of course, it needs to be able to come out and be at room temperature. So we do those studies other sort of stress studies. So intranasal is interesting. I think that just in general, with these types of vaccines that are not live, it's hard to develop a strong immune response in the array of things that are being presented. The nice thing about IM injection or getting ID is you are just delivering antigen into an environment that's looking for it. And so you get a very focused specific response. So I think I foresee IM. I mean one of the things that we're going to have a debate about, frankly, is 1 or 2 doses. And I just throw this up here, for example, you can see with 1 dose, we're getting what we think are neutralizing antibodies, it look like convalescent sera. The second dose is quite a bit higher. One lesson we learnt from RSV with motavizumab, palivizumab, our vaccine, cotton rats, is that if you had a higher affinity, high level response, you could expect to disrupt the infection both in the nose and the lungs. And that might be something for us to consider. That is, I've got friends who teach school, what are they going to do in the fall? Do they go back? So if you get them a vaccine that protects them against severe disease, that's very good. But maybe you also want them to be protected against transmission. So they don't become transmitters. And so that might be -- those types of folks might really go for a 2 dose regimen, whereas in a setting where you have mass deployment, you're trying to prevent hospitalization, you just give them a single dose. So I think that's going to be an important consideration for the future.

Great. A question about -- and you probably haven't looked at this yet, is the epitope range on the spike protein that your vaccine is inducing immunity to. And this question also in relationship to the fact that obviously, protein made in insect cells may be glycosylated differently from one made in mammalian cells. And is there -- do you think there's any potential difference in the range of epidural neutralizer or not that one might get using the -- approaching that's made with slightly different glycosylation.

I do. I think I throw up this picture I think -- so first of all, glycosylation is important for structure, right? It's the soluble end of an insoluble protein, and it's part of the structure. So it's important for it to be glycosylated. Insect cells have relatively short glycosylation result in relatively short glycosylation patterns on the trimers. So I think that's a -- I think what you really want here is access to the epitopes by B cells. And what happens in a nanoparticle is instead of 300 tightly packed trimers on the surface of a virus, which gives you mostly access to a head we know from our other programs that the side in the case of flu, vestigial esterase, the stem the interfacial antibodies. So now we're presenting -- these are proteins are like strings of pearls. They're mobile. It is an issue when you start to stabilize or fix, you can lose epitopes or epitope access, if you will, and so by having this highly mobile, nonfixed full lake glycoprotein in a nanoparticle, you do expose other epitopes. And if you look, we just published a very nice paper on our flu vaccine, characterizing the fact we had SIM side, if you will, head conserved epitopes, interfacial epitopes, so in between the trimers. So those all now get exposed in a highly mobile, highly flexible, full linked protein. And we think that that's important. So I know, for example, James Crow, Scripps, everybody is working hard to -- and they have stables of monoclonals. And so we're looking forward to the antigenic characterization of our vaccine, the sum total of which should manifest itself in functional immunity. We have 2 assays, which I think are telling the receptor binding inhibition assay, which really focuses on that very small area as well as neutralize the antibodies. But I think that's a tool that's going to be very important. It will be deployed. It's just taking some time to make the monoclones. But I think shortly, you'll start to see a lot of work on defining the epitopes and maybe dissecting what's important in terms of immunogenicity.

Thanks. Greg. So we're at noon, which is the official end of the webinar, but I'd like to continue on a little bit while longer in fairness to you all because of some of the audio difficulties we've had so we'll go on with a few additional questions. But for those of you who need to leave us, I just wanted to remind you that there will be another webinar tomorrow at the same time, which will feature Dr. Tal Zaks from Moderna, we'll be talking about their RNA vaccine. So another question for you, Greg. So far, I think you presented mostly immunogenicity data from the animals. But what can you tell us about additional animal experiments or planned animal experiments, in particular, in terms of animal challenge and perhaps for efficacy in animals as well as then perhaps any other distinctions among the various animal models that people are proposing for use in evaluating COVID-19 vaccines?

Yes, okay. That's a good question. So we're looking for enhancement model. So we have nice nonhuman primates, Syrian Golden hamsters, ACE transgenic mice, and so these animal models all have various features that are -- may be helpful. So the group at University of Maryland has an ace 2 adeno vector based mouse model. So they obviously create a situation where the mouse has a human receptor they develop some pathology, and you can see where you're protecting. So we're looking at that. We know there's a couple of different nonhuman primates that are kind of at the four, the cynomolgus and REIS MCA. And so we're doing both of those. We'll do a live viral challenge and look for histopathology. The Searing Golden hamster is interesting. Obviously, the -- since they're a smaller animal, you can do larger studies. They seem to get quite sick. And so we're doing those challenge in Instapot Logy. We have access to the ace 2 transit device recent saw paper there, and they seem to have a clinical illness. We're looking for the Th1, Th2 profile and BALB/c mice, specifically with our antigen and adjuvant. So I think there's a big package looking at all the sort of proposed models to look for protective efficacy in challenge, look for histopathology that may suggest that you have enhanced disease I think we're looking to see enhanced disease means it's worse than infection. So I think in some of these models, it's just going to be protection, that is loss of no virus, but some inflammation as the immune response clears it out. So -- but I think that's a very robust package. And we intend to have this data in time to pivot into our Phase II study.

Fantastic. A number of questions still about sort of funding and international collaboration. People noticed your Phase I studies were down in Australia. And wondered why you didn't do them in the U.S. but also noted that the obviously, you've got this very nice funding package from CEPI. And what we're wondering whether there were other international collaborations and fundings in the offing and then also, I guess, is a part of the funding question, somebody asked about the overall cost of goods. It's when it's thinking about a vaccine that's going to be used around the world. What -- how expensive is it likely to be? Is it feasible to actually deliver this to huge numbers of people?

Yes. So that -- those are great questions. I can answer that to a certain degree. Let me just say about manufacturing. We move ahead at risk. So we've been working, as you can imagine, for some time, on the manufacturing piece. We've made many, many lots in the lab. We know it's a fairly standard process because it's a platform technology. Obviously, some of the conditions on the gels, et cetera, the columns might be a little bit different. So that's a big leg up for us. We're making something that really is recognizable in the process that's really recognizable. So as you may saw, we have a collaboration with the emergent biosystems. They have the ability to scale up in a big way. They're making our clinical lots. They've been making -- we've made several lots in their setting. They're very capable, as you can imagine, they have the potential to scale up and one of the things that CEPI is doing for us is giving us a global footprint. So we intend to have acquisition of manufacturing capacity in the U.S., Europe and Asia, and so because this is a -- it's an imminently tech transferable process, we're in -- we're very much in the throes of seeing that implemented. So I think our projections were vetted by our collaborators. We know it's not to discount the huge amount of work going forward. But in terms of risk I guess we've done this enough that we think that scaling up, based on what we know from our -- as I mentioned earlier, with the development experience we've had, we can project from our preclinical discovery lab, what will look -- what we'll need to go -- what can go forward with success. And I think we're confident that we can make 100 million doses this year that could be deployed. And go to a billion doses. That is not discounting the need for a lot of work. A lot of collaboration, I think, with regulators guidance from them, the CMC package I do think at the basis of our platform technology just gives us a leg up. And I think the CEPI funding reflects that they're convinced that it's possible and we're very grateful. They're experienced people. They -- you know this. And so in every aspect. So their assessment, it can be done. Now our goal is to deliver. And we know it's going to be a big challenge. Our group is committed, experienced, and we think we can deliver on this promise. So cost of goods part of the CEPI, it enables the manufacturing capacity. It comes with a commitment to have equitable access for global citizens. We like -- we're all -- we're very much aligned with that. Right now, I mean, just right now, we -- the cost of goods just as theoretical, but we know because it can be scaled like this that it should be acceptable pricing both for global citizens and for a company to survive. So I think between the 2 we're again confident that we can work with these organizations to provide vaccines for the globe and very much committed to doing that for our own country here in the U.S.

Sort of a little bit of a follow-up on that, Greg, is that, of course, one of the models for a vaccine development in general is to identify vaccines that are effective. And then figure out how to make use quantities of them, which may be more than can be made by the individual manufacturers, which then, of course, involves some level of technology transfer, which would have to be done to possibly plants around the world that would be able to rapidly scale up and make the billions of doses of vaccine that may be necessary. And so when you said that you already have another company who's doing some of your manufacturing, it sounds like you already have developed some tech transfer experience. But is there anything else you'd like to say about that?

Yes. I think that, as I mentioned, I think our contract with CEPI is enabling this. So we are, in fact, doing very actively looking, identifying specifically large-scale manufacturers that -- where we can tech transfer in Europe, U.S., Asia. Those are going on specific places with specific capacity estimates that were vetted by CEPI. So -- and they're -- they've been enabling for that. They've helped us identify places and given us advice. So we're aligned with that. So emergence has been -- is one of the manufacturers we work with. There's others in the U.S. and others around the globe. And I think that we -- right now, those are somewhat to some degree, business negotiations as well as planning so I think we'll be able to disclose that in the not-too-distant future because we intend to really ramp up manufacturing as soon as it is being ramped up and deliver as many doses as possible, 100 million doses this year is obviously no small feat. And so we need to get going, and we are. So I think we'll have more information on specifically sites we're working with in the near future.

Fantastic. And it sounds like on one of your slides, you said you thought you would start to have the Phase I immunogenicity data in July. Is that right? Or assuming that?

Yes. We can have -- we have some potential to show some of the data sooner than that, but right now, our target is the immunogenicity in July, which allows us to pivot in the Phase II?

Well, fantastic, Greg, this has been an excellent conversation. We're at about 12:10 so I think we'll close now. I apologize for not getting to all the questions, but I really appreciate the broad variety of questions that I think have enabled a pretty interesting conversation about this vaccine and indeed about the whole COVID-19 problem. So we really appreciate you sharing your data and your information, your plans with us. And it's, of course, an exciting time to be in the middle of all of this. And I think we're all rooting for success. Any closing remarks you would like to make.

Just thank you, Phil, for supporting the world Vaccine Congress as one of the organizers. I think it's a great format to allow extensive public discussion of technology and inquiries through questions, just -- I appreciate -- I know -- I see you almost everywhere so your effort here to moderate, appreciate it. And just again, thanks to our colleagues, CEPI. I just threw up the picture of a couple of our key investigators, we just have a super group here at Novavax. Very proud of them. It's an honor to represent them and I hope we can see a vaccine deployed this year. That would be great. So that's what we're thinking, what we're hoping and we appreciate the cooperation of everyone to make that happen. And we wish the other manufacturers well because I think this is going to require multiple platforms to address this huge problem. So thank you very much.

Wing, any final word. Of course, all of you are interested, please tune in tomorrow to hear Tal Zaks from Moderna, complete this trifecta of talks on different vaccines that we're having this week. But thank you all for hanging in there and listening. The response has been overwhelming with a very, very large number of people. I think I saw a peak of almost 1,000 people listening to this webinar. That's very impressive and shows a high amount of interest in this topic. Anything you'd like to finish with, Wing.

I think you've summarized that so well. There's not much more, but I would like to say thank you so much for everyone's patience. We do apologize for some of the audio that was cutting in and out. I hope we've made up with some of the time on the Q&A at the end there. Thank you, Phil, for moderating that. And like you said, please join us again the same time tomorrow. And just a final thank you to our sponsor, DNA Genentech as well for supporting us. But thank you, and have a great rest of the day, everyone.

Bye-bye.

Thank you. Bye-bye.