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

Good afternoon, everybody, and thanks for joining us for our penultimate fireside chat session with -- and here today with Novavax. I'm Seamus Fernandez, the global [ biotech ] analyst here at Guggenheim. And I'm also joined by my biotech colleague, Evan Wang. He's our CoV-2 expert. We're also joined from the Novavax team here by Greg Glenn, President of R&D; and John Trizzino, Chief Business Officer and Chief Financial Officer. Novavax is really one of the frontrunners in the development of COVID vaccines. The company has efficacy trials ongoing in the U.K. and South Africa, and plans to begin its 30,000 patient U.S. trial in the next month or so. The company has rapidly developed its vaccine and has received over $2 billion in funding from CEPI and OWS. In flu, the company announced results from its Phase III trial in March and is preparing to submit its BLA. And with that, let me turn it over to the Novavax team who will provide a brief overview of the company and the platform to just help orient everyone, and then we'll jump right into Q&A.

Good afternoon, everybody. Thanks. John Trizzino here. Just -- we'll make it quick, and then I'll turn it back over to you to kind of orchestrate the conversation. Recently, we've had a tremendous amount of focus in energy level that all of our coronavirus vaccine activities, having recently reported our Phase I results and the start of the Phase II trial. And so we actively continue to move forward with all the development activities, all the large-scale manufacturing activities. And of course, critical to the ongoing success of the company. There's a presentation deck available for those of you that are online and want to access our website or we can be presenting the slide deck as well along the way. But there's a couple of highlights here on Slide 3 of over $2 billion of funding coming from CEPI and Operation Warp Speed over the last couple of months. Also highlighting here earlier in the year before we got into all of the COVID vaccine activities, our success from our NanoFlu Phase III trial. This is having achieved all of our primary objectives during the course of that trial. And so we're very excited and pleased by the progress we've made there and continue to place a lot of value on that asset as well as the COVID vaccine asset. And our balance sheet is strong, allowing us with the partnering activities and with the funding activities, to continue to do whatever work needs to be done from a development standpoint. On Slide 4, there is a little bit about our pipeline. I think this is really just a recap of what I just said. Obviously, our COVID vaccine candidate leads the charge of all of our activities and prioritization. Flu continues to be a focus, somewhat affected a bit this year by all the COVID activities, but an important asset to the company, of which we're going to continue to aggressively pursue. And then, of course, some other candidates that we continue to monitor, our RSV program as well as combination vaccines. So I'll leave it there just kind of a brief introduction and turn it back to you for some Q&A.

Perfect. So maybe just to start off, in terms of what really separates the Novavax platform from others in terms of your recombinant technology and then just the work with your Matrix-M adjuvant in particular.

Yes. Thank you. I'll take that. This is Greg Glenn. I'm the President here at Novavax - President of R&D. Look, I came to the company about 10 years ago. I've been in vaccine development for 26 years. I was very, very intrigued by potential for the making of a nanoparticle and use of an adjuvant. And I think we really do have the best vaccine technology out there as a platform and so you're seeing the manifestation of that, I'd say, through the funding we've seen with our COVID vaccine and our flu success. So that's the fundamental basis for all of our vaccines to make a nanoparticle. Why is it so interesting? What we can do is on a vaccine, you're trying to replicate an infection without having an illness. And what we do with our nanoparticle is essentially make a part that looks very much like a virus to the immune system. And so it recruits the kind of cells and the kind of immunity you want. And specifically, we make the proteins defined on the surface of these viruses in a way that they look like the native configuration within nature. And so you get a very nice, very specific immune response to the spike protein or the hemoglobin or the F protein for the RSV. What we do, however, when you create the nanoparticle the way we do, which is a very pure protein, you also create a need for a danger signal, and this is where Matrix-M, the adjuvant comes in. And again, I've been working in vaccines for a long time. It's very gratifying to see the fruit of these -- of all the work going on with novel adjuvants to get to a point where they're going to play a critical role. So the adjuvant provides the danger signal. We have a lot of data that show it's very effective at doing that. You get, therefore, a much better immune response, a much higher quality immune response. And so we have these partners, if you will, in our vaccine where you have the nanoparticle. It looks like a virus as the surface protein, very pure. And in addition to that, added to that, is this Matrix-M. It's a staffing and it's a soap. It's an additive to many oral products like root beer. But it's made into a particle. And so by being in a particle, when you inject it and immune cells see it as dangerous and they get or if they get activated, and they really just go to -- the other, I think, really nice feature of our technology is the immune activation that you see, which can cause some side effects. It's very local. It's in the drained lymph node where you want it. So you have a really, I'd say, very good safety profile, modern adjuvant recombinant technology that can faithfully replicate what you'd see in infection. And therefore -- and I would say the distinguishing feature of our COVID vaccine, our coronavirus vaccine is this immunogenicity. And it's really very, very good. It's hard to imagine that this vaccine will not work very well.

Great. And maybe you can just walk through some of the specifics around the protein itself, the subunit protein and your selection of the prefusion confirmation. Maybe just how you designed the construct would be helpful to discuss.

Yes. I mean this -- if you go back one slide, this is -- this -- I would say, we've been studying the very problem that came up in January for more than 10 years now. My team -- we are experts in this type 1 fusion protein. So this mechanism of infection by viruses -- so viruses carry a small amount of genetic material, in this case, it's called RNA. And they're inside their protein shell, and to infect the cell, they have actually bind to the cell, inject that genetic material into the host cell, and then the host cell starts making copies of the virus. So that's how it works. So if you can block that process with immune response, you can prevent infection. And that mechanism is common to flu, to RSV, Ebola, other coronaviruses and of course, SARS-CoV-2. So we knew a lot about how one would need to do this. One of the features from a vaccinology standpoint that you have to take into account is these proteins act like a syringe. So they're made in the host cell, they're assembled, they're processed and they get clipped. And that clipping process creates a very unstable, wound up like a rubber band, high energy state, and that state then changes when it infects a cell like a syringe injecting the RNA. And so you've got to deal with the protein that is inherently unstable, has different forms. And so a lot of work has gone into stabilizing that protein, and it's called the prefusion form. So it's the form that is upstream of this injection event. And if you don't make it in the prefusion form, you can make it in for an immune response that may not be protective at all, even though it's a spike protein vaccine, but the 3-dimensional configuration would have changed. So that was a big consideration. We -- you see here in the slide, I think 7, we've both described the structure. We've also have a very good piece of work done in Scripps Institute, looking at the crystal structure. So we know it is in prefusion form. And then again, you see the manifestation of it being in the proper form because the immune response we get is so functional. You have to have this -- the spike protein in its proper configuration to get the kind of immune response that we've made. So we've made 2 mutations. It's a full link protein, so it looks a lot like this, spike protein in the virus. Again, it's just in this nanoparticle. So the foot is in this detergent micelle, it's a really great way to present this protein to the immune system.

Can you talk a little bit about the immune responses that you've seen with your construct in Phase I. And then also, any color on the T cell responses and the importance of the T cell responses in those data sets.

Yes. Let's start with that very last feature. I mean I think the T cells, what they do for you is they create a long-term memory response that allows you to have recall if you have a viral infection and then have a stable, a pretty stable immune response. So -- but the feature that I would just point to that we know, I just kind of gave my little mini lecture on type 1 fusion proteins, we know that the other side of the vaccinology issue there is that you need to induce, neutralize the antibodies and they work. And that's been shown, I would say, definitively for type 1 fusion proteins with RSV. 5, I think maybe 6 randomized clinical trials using monoclonal antibodies as prophylaxis, where the only feature which those monoclonal antibodies were selected are neutralizing antibodies. So there is no doubt. There's likely -- there is a lot of redundancy in the human immune system. But I think that there's very strong evidence for protecting against a respiratory disease that infects by a type 1 fusion protein that neutralize the antibodies, will be a correlate. So if your data -- and that's why I think you've seen the field very focused on inducing this type of immunity and also the level at which you induce. So I'll throw up here, I've got Slide 13. From our trial, the way this trial was conducted, we had a placebo group, we had an antigen-only group. You can see them kind of going by columns from the left. So the antigen-only gave us some neutralizing antibodies. And remember, our dosing schedule was to give a dose and a booster 3 weeks later. And then you can see after the second dose, without antigen, you got a 41 titer. It's decent. With 2 doses of the adjuvant, you got a titer of around 4,000 and those are very high neutralizing antibody in my experience. This is a conservative assay. It uses a 100% utilization readout with a wild-type virus. There's a couple of reasons to think this is a very good immune response. First of all, we don't see a dose response. So normally, if you -- if there's room to increase immune response, if you go up on the antigen dose, you'll get a bigger response. So we've gone from 5 to 25 and we don't see a greater neutralizing antibody response. That tells me that this response is on the plateau of the S curve, the dose response curve, and so it's maximum. The other evidence for this being maximum is on the far right. You see we are able to assay convalescence serum from subjects who are from Baylor College of Medicine. They were -- we were blinded, of course. We didn't know how they would soar. But we got -- we asked -- we paid people who were clinically ill. These are patients who were pretty sick. They're either in the ER or they were hospitalized and intubated, and there's a couple that are mild or asymptomatic. And you can see that there's a very nice stratification of the neutralizing antibody response and the severity of illness. And I would contend that with these type 1 fusion protein respiratory viral illnesses, a severe illness generally confers protection. And you can see here on the right, that our -- that this is -- and these are severe patients. I wouldn't expect that the blue and red, kind of be near maximal and be protected. And you can see that our geometric mean titers in the yellow and the 5-microgram given twice with Matrix-M is hovering there at the upper end of those convalescence serum, who are quite sick. So to me, this is extremely encouraging that our vaccine neutralize the antibodies, which should be functional, should be protective, are very high. If you go to the next slide, one last point on this data event. What we worry about in vaccine trials are the people on the low end of the bell curve. There's always a bell curve of responses. But what you see here is a reverse cumulative distribution and the way this is graphed is on the Y-axis, we have the percentage of subjects. So as you go up, more people, and you go across the right, this is the higher titer. And you can see the placebo, the unadjuvanted, the -- and then you see our 2 dose adjuvant 5-microgram group, where 94% of the people have a neutralizing antibody titer, 1,000 or greater. That's very unlikely to be a vaccine that will not work well in the field. So we are -- we're very eager to see our data because from what we can see also from our nonhuman primate data, which you may have seen, very encouraging. This vaccine looks like it's going to work and work well.

Great. Well, let me turn it over to my colleague, Evan Wang, to drill into some of the Phase III questions that we've got.

Thanks, Seamus. So yes, so Novavax is in multiple efficacy trials ongoing in South Africa and the U.K., with planned trial shortly in the U.S. There's been a push to be transparent with trial protocols and minority recruitment. Does Novavax plan to share these protocols across all studies as well as total and minority recruitment?

Yes, yes, yes. So absolutely. I mean we've -- I would just say that we've always been transparent. When we go out into trials, we generally share the key issues, which are the endpoints, the success criteria. So -- but we -- the reason they're not out, first of all, in South Africa, just was both -- all the protocols had a little tweak. Maybe you go back 1 slide. And if you look here, let's talk about the clinical development. I'll weave the answers in, if you don't mind because I think the context would be helpful. So the first bar there is a New England Journal paper. That trial result triggered the start-up of essentially 4 trials. The first trial -- and so the FDA reviewed the data, the safety immunogenicity data and they asked us to increase the population of older adults. So that second study 1 part 2, increased the population of older adult at-risk populations receiving the vaccine. So that data, they've asked us for a subset of that information to trigger the start-up for the U.S. Phase III. So that's the 2 there. So we've recruited. We expect to provide that data in mid-October. So obviously, it's October, so shortly. And that would be limited safety and immunogenicity in older adults, and then that would trigger the start of the Phase III trial below. Now the Phase II trial, we started in South Africa. We did that because there was a high transmission setting. We had good investigators. It was an opportunity for us to expand the clinical experience. So that is an efficacy study. The primary endpoints will look just like the U.K., and I'll describe that in just a minute. We have just recently increased the number of subjects in that trial. We're expecting to go up to 4,000. So that requires a protocol change. And so we're doing that. So once that change is made, and I think it's just about done, we'll post the protocol for that trial. It does include 240 HIV positive subjects that the group there have a lot of expertise in vaccinating in this population. We felt it would be really good to get the safety in that population out there. The U.K. trial, which, as you know, started last week, I think now we're well into recruiting this. This trial is going to -- again, same thing, we changed the protocol a little bit. So we have a protocol version floating around and about to be finished. We will post this. But this trial, and I'll talk about the endpoints in just a minute. They are also, is -- we'll do up to 10,000 subjects. We went there because when we were planning these trials back in April, we had a suspicion that the virus might come back in a big way. And if you go look today actually at the Johns Hopkins COVID virus tracking site, you'll see there's a very large upswing. So that makes, for us, to [ track ] on the really attractive place to be doing this trial. So the U.S. trial, as you noted, we expect to start sometime in October. This trial would be 30,000 subjects. So there's a lot of harmonization between these trials. First of all, they're all going -- we're all looking for prevention of COVID disease, PCR positive, symptomatic and there's a lot of harmonization between these 3. The first 2, the South Africa and U.K. trial, if you wouldn't mind going to the next slide, I think we have the design there. So it's a placebo-controlled trial using 5 micrograms of our antigen and 50th Matrix, given at day 0 and 21, ages 18 to 84. We expect to get -- so now to start answering some of your questions, 25% of the subjects would be greater than the age of 65. And if you go to the next slide, just to talk a little bit about the endpoints. You can see there with South Africa and the U.K., we have 2 -- we have primary endpoint is in or. So it's symptomatic COVID disease or it's moderate to severe disease. And so one of the sort of critiques out there has been, could you collect the moderate to severe events and have an endpoint based on that, and so we do here. So these are the endpoint-driven analysis. In the case of the U.K., we will have an interim analysis at 50% and 75% of the desired number of cases. In the U.S., we'll do the same. The U.K. trial, we project, but these are approximate, that the final analysis would require about 109 asymptomatic or 43 mild and moderate, and you can do the math, 50% or 75%, there'll be some interim analyses done there. So in the U.S. in particular, would you mind going back 1 slide, going 2 slides? So the U.S. in particular, this is being done in collaboration with the NIH COVPN network and obviously, Operation Warp Speed. And that trial, there's a lot of really good information on who gets the disease, where the at-risk populations. So there's a huge push there, of course, to increase the minority, the Latinos and African-American population in these trials. So I think you'll see there's a very good overlap. Of course, we want to do the trial in populations that are higher risk because the attack rate's higher and there's a huge amount of work going into ensuring that, that will happen. So I think I answered your questions, and maybe we can pause here and see if we see there's any more information on the clinical trials you want to talk about.

Yes. So from the South Africa study, is it fair to assume that the original 2,900 or so patients were enrolled and that's what caused the expansion?

No. I think it's more a matter of -- we thought it was -- we became more, I'd say, intrigued with the opportunity to demonstrate efficacy in this setting. So look, so let's talk -- it would be good just to contextualize a little bit about the success criteria. So I think you'll see there's been a pretty, I would say, a good definition of what success would be in these trials, which is, as you probably know, is to have a confidence interval around the point estimate, where the lower bound of that confidence interval is at 30 or greater. That's a pretty high bar. So we have 3 shots in goal here. And we felt that South Africa. And that's -- that, by the way, that criteria has recently kind of been projected by the U.K. and EMA. So they've dialed it up from a lower bound of 20 to a lower bound of 30. So each of these trials could actually meet that goal depending on the attack rate and the vaccine efficacy. So the South African trial, we wanted to give a little bit more robustness to it in terms of the numbers. And really, that's what drove that increased numbers. So we haven't typically historically projected recruitment numbers. And so it's something we could debate a little bit here, but it's not fully recruited, I can tell you that. And the 4,000 won't just -- is there to augment the signal. We did the same in the U.K. We felt that if we turned up the numbers just a little bit, that was somewhat -- also, in both cases, taking into account what sites could manage. So it's somewhat of an operational decision. But again, our goal in the U.K., we think we have a very good situation where we might be able to, in fact, determine a vaccine efficacy and meet the criteria of a lower bound of 30% or greater. And that, in the case of the South African trial, I think it's a little uncertain as to where that could be the basis for licensure. But certainly, it could be in Europe. We know that EMA is open to a trial result from any geography. And so that's a sort of a shot on goal for that in addition to creating more information. The -- obviously, the U.K. trial is a pivotal trial. So we have agreed with MHRA, and that could be a pivotal trial that gives you -- gains licensure in EMA as well. And of course, the U.S. trial is designed to be the pivotal trial for U.S. licensure. And again, the FDA has defined the criteria for success, and that trial is designed to beat it as well.

Great. And I think we're approaching the end. So I just wanted to wrap up with the final question on these trials and timings and the anticipated disclosure results. Now as these different trials hit their interim or final analyses, how does Novavax or the DMC plan to disclose the efficacy? Will it be some kind of harmonized result? Or will it be disclosed as these trials conclude?

Okay. Historically, it's -- I would expect it to be trial by trial. So -- and they all will come off different times depending on recruitment, attack rate. So I think that, that's probably how we would do it. I think if we were at all to try to pool the data, that would be post and blinding. So I expect -- I fully expect us to unblind the trials as the result comes in. And they are event-driven trials so it's hard for us to faithfully project out a time frame. I think that, that's a mistake that there are people who are projecting it. So we're not going to, but we're going to do our best to recruit subjects. I think it is a really unique situation where we know so much about where the disease is and the geography and the population that I'm very optimistic this can be done extremely efficiently.

Great. And I think we're out of time. So I want to thank Greg and John for joining us, and best of luck with your efforts.

Well, thanks for your interest.

Thank you.

Thank you very much.