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

Good afternoon. My name is Gregory Glenn. I'm the President of Research and Development here at Novavax. I am proud to come to you with data from our Phase I trial. It has come to our attention that a reporter with STAT News misunderstood and misreported our safety data regarding hospitalization. To be clear, there were no hospitalizations. I need to state this upfront because of this confusion. This article has been corrected and posted. So what I'm going to do today is walk you through our data from our Phase I trial. And let's go to the next slide. We have a safe harbor statement, which I will not read. And let's go to the next slide. So Novavax has constructed a recombinant nanoparticle vaccine, and as illustrated here how we've done this. This -- by the way, this information is posted in our Tian, et al. article site below. We have a fully stabilized SARS-CoV spike glycoprotein vaccine. It forms a nanoparticle. The nanoparticle is then adjuvant with Matrix-M, which is our partner in creating the very strong immunogenicity. You can see on the right, an illustration of the primer and that primer is then inserted into a nanoparticle. The vaccine is a liquid formulation and vial and it's stable at room temperature -- sorry, at 2 to 8 degrees as standard for vaccines to use the cold chain. So if you go to the next slide, I'd like to start with some recent data we generated in an important challenge model. The vaccine, the spike protein vaccine is called Novavax CoV2373. we immunized cynomolgus macaques with our vaccine with the doses we're using in the clinical trial and a fractional dose. The animals are immunized twice a day 0, 21. We then measured antibody responses prior to challenge -- 2 weeks prior to challenge and then the animals were challenged with live wild-type SARS-CoV2 virus. What you see here on the immune responses on the left, you can see these are robust wild-type neutralizing antibody responses. And then the second panel, which you can see is we have then looked for the presence of subgenomic RNA in the lungs of the animals. And you can see the placebo had -- clearly had virus at day 2 and day 4. In the vaccinated animals, in the clinical doses, the vaccine-induced sterile immunity and you can see the same phenomena in the nasal swabs. So the animals were also challenged in the nose. You can see there was presence of virus in the nose and in the vaccinated animals, we had complete sterile immunity. This is quite remarkable. We think that that's important because this suggests that the vaccine can not only protect against the lower respiratory tract illness, that is the cause of much of the morbidity and mortality to this virus, but it may also disrupt the transmission through neutralizing the virus in the nose. So we start -- we wanted to start this program with this data. And now if you go to the next slide, we'll get into the details of the Phase I study. Okay. So let's talk about the study design. This was a Phase I trial as part of a -- part 1 of a Phase I/II protocol. This is a randomized, observer-blinded, placebo-controlled trial. And it was designed to evaluate the immunogenicity and safety of our vaccine, Novavax CoV2373. And the trial regimen focused on assessing the 5- and 25-microgram dose levels with and without Matrix-M. You can see the groups here. You can see there were 25 placebos receiving placebo vaccination on day 0, 21. The group B contained no adjuvant. It was 25 micrograms of the spike protein nanoparticle. Group C had 5 micrograms of our nanoparticle vaccine with Matrix-M, and you can see Group D had 25 micrograms of the nanoparticle with Matrix-M. And then Group E was a single dose, and it's notable that, just to remind everyone, they have a dose at day 0 and a placebo dose at day 21.

It's a booster dose.

It's a booster dose, but the placebo dose was given at day 21. So in that group, I should say. And what you can see here is where our focus designated by the red box here was to look at the 2 doses that we thought were the most likely prospects for going forward in development. That is the 5 micrograms with Matrix-M and the 25. And so we are testing the hypothesis if this 2-dose regimen could induce good immune response and have a good safety profile. So if you go to the next slide, so study objectives for the Phase I trial was the -- primary outcomes were reactogenicity, safety lab assessments and the immunoglobulin anti-spike IgG protein responses. Secondary outcomes include adverse events, wild-type neutralizing antibodies and T-cell responses. If you go to the next slide. This shows you now the anti-spike IgG in ELISA unit per ml on the log scale on the left. What you can see here on the far left, we were able to obtain human convalescent sera from a colleague at Baylor College of Medicine. This was from subjects who were PCR positive and had relatively clinically significant illnesses, and I'll give you some details in a minute on that. So as an anchor for our data, you can see that we had a -- the -- you can see the geometric mean titer and the scatter of the anti-spike IgG responses. And what we've provided on the right-hand in the blue area is some of the details, some of the actual numbers, so you can refer to that as well. So you can see convalescent sera gave a response of 8,000 approximately. The placebo, you can see on the bottom there at day 0, A0, which it says there, A21 and A35, you can see there's no responses. At day 0, the unadjuvanted 2-dose vaccine provided some level of IgG. Now it's important to compare that with the adjuvant group. So at day 21, B21 and C21 are the comparators and D21, you can see the addition of adjuvant to the nanoparticle gave a very robust response. And you can see that also with E21. So 25 micrograms with our Matrix-M adjuvant, you see a marked increase in the immune response. Then going back to B35, which would be the immune response seen after the second dose, you can again, compare that to the D35 where the 2-dose regimen, again, adjuvanted is a much higher response. So it was very clear the adjuvant gave a profound effect. The other thing you can see, clearly, is the second dose provided a more than 1 log increase in the anti-spike IgG protein. And when you look at these, these are quite robust responses, you can see relative to that seen in the geometric mean of the convalescent sera. With the 5 micrograms, we have 63,000 -- 63,000 titer compared to 8,000. And so these look like very robust responses. You can see the 1-dose regimen, Group E on the right that resulted in about 2,000, 3,000 titer. So again, you can see the benefit of the 2-dose regimen. So the adjuvant effect was clear. The other notable feature here is that you can see there was no difference between the responses in 5 and 25 micrograms. In other words, this is a dose-sparing regimen here. And you can see very robust responses with this 2-dose. Okay. So if we go to the next slide, we are very grateful we are able to collaborate with the University of Maryland School of Medicine, Matt Freeman, who is a coronavirus expert. He has -- he has an assay that's done in the VSL3 facility for wild-type neutralizing antibody. You can see here now the results. And again, I'll start with the human convalescent sera. You're noting that these subjects were relatively ill and clinically significant illnesses. You can see they had a titer -- geometric mean titer of 983. Now the patterns you saw that I described in the IgG are again evident here, and you'll see why there's a strong relationship between the induction of spike antibodies and the wild-type neutralizing antibodies. So again, you can see the value of the adjuvant at 1 dose and 2 doses. You can see the value of the second dose. So the immune response goes up very significant with the second dose. And you can see that they're, again, dose sparing. So very similar take home messages. This is a very high neutralizing antibody. So you can see the response at day 35 after 2 doses with the 5-microgram, with 50 micrograms of Matrix-M gave a titer of 3,906, very robust, approximately fourfold greater than the response -- than the antibodies detected in the human convalescent sera that we attained from Baylor. Now if you go to the next slide, there's just a little more detail on this. So this, again, just to remind you, these are subjects who are PCR positive in a hospital setting. And you can see that there were some that were asymptomatic. Some that had quite significant clinical illness such as hypoxia, shortness of breath, cough or cluster symptoms that were clinically significant and required outpatient care and then several who are also hospitalized. And you can see there is a strong relationship between the severity of the disease phenomena and the microneutralization titers. So we think that this is a very good anchor for our data for understanding our data. We would expect that a clinically significant illness is likely to lead to the sort of titers that should -- could protect people. And so without overstating that case, we think this is a very good anchor to understand that our immune response here as measured at the day 35 titer is quite high and certainly in line with what you might expect to be protected. Okay. So the next slide, as I mentioned, it's very important to see if they're -- in the case of these types of viruses, we want to have functional responses. And here, what we're looking at is the relationship between the measurement of anti-spike IgG antibodies and functional neutralizing antibodies as we measure the wild-type assay. And this is quite interesting. You can see on the left, the convalescent sera, there's a very tight relationship. And the way you measure that is a correlation. Correlation of 1 is perfect. Correlations of 7 are very good. You can see the correlation here is 0.958. So a very tight correlation between the measurement of a spike antibody to our vaccine and a neutralizing antibody. Now this on the left is from the convalescent sera from the patients at Baylor. So they did not receive the vaccine. You would expect them to have a functional immune response from a natural infection and you see that, and it shows that our assays are detecting the right kind of antibodies. Now on the right side, the far right, you can see our adjuvanted vaccines. And you can see there's a correlation between the anti-spike IgG and neutralizing antibodies of approximately 0.95. So very close relationship. This tells us that we have -- a significant portion of our antibodies is highly functional, able to neutralize the virus and presumably block the infection. You can see in the middle, this is dose Group B. They received the vaccine without Matrix-M. You can see there's still a good correlation, but not nearly as strong as when we add the adjuvant. So this, again, is further evidence that the adjuvant allows us to induce not only high quantities of antibodies, but very high-quality antibodies resulting in neutralizing antibodies. So going to the next slide. There is another arm of the immune system that's quite important in viral respiratory diseases. So we also looked at T cells in a subset of our subjects here, and we were looking for antigen-specific CD4-positive T cells. These are in subjects who are immunized, and you can see the different groups, the placebo, the vaccine alone. Group C was given the vaccine with 5 micrograms of Matrix and 50 of the placebo. And we've drawn these cells now after the second immunization, looked at the PMVCs, and they are now measuring the antigen-specific CD4 T-cell population as a percent of the total T cells. And what you can see in the upper left panel, as you might expect, there's no reactogenicity in the placebo groups. And the way this graph is designed, you can see what's called the Th1 response on the left and a Th2 response on the right. So the Th1 response is considered to be a desirable qualitative response that could result in protection. If there's an imbalance, where there's too much Th2 responses, it's felt this might lead to an adverse type of antibody that could lead to less protection. So you can see in Group B, no adjuvant, very little T cell responses. In Group C, where we used 5 micrograms with 50 Matrix, you can see very strong T cell responses at day 28, and reflecting what we polyfunctional CD4 cells. So it includes the secretion of IL-2, TNF-alpha and interferon gamma, all of which are indicative of a T cell that would be very active. And then you can see, we've also tabulated any cell that has NE2 and NE3. And so we do see these polyfunctional T cells in this population, which is very desirable. By contrast, the Th2 response is quite low. We looked at IL-5 and IL-13 and really quite a bit lower than what we see with the Th1 response. So this is a Th1-bias response, very desirable. The magnitude is high. You can see with the 25 microgram with Matrix-M, in Group D, we also got a robust response, again indicating the adjuvant is important for making these polyfunctional T cells. And again, the control group was very low in induction of the cytokines associated with Th2. So very robust T cell responses, has a desirable phenotype, has a Th1 response, and this will be important as we move forward. If you go to the next slide, okay. So this is -- I'm going to now talk about safety. What we do is we look for solicited localized symptoms and solicited systemic symptoms. These are then -- with the way this slide is designed, you could see vaccine 1 on the left and vaccination 2 on the right. On the top, you just have any detection of any kind of local symptoms in the right -- at the top and you have vaccination 1 and vaccination 2, and then it's broken down by different facets of this data. So you can see with the, first of all, the tabulation with vaccination 1, the vast majority of these reports are mild or not at all. So -- and then we have a small number of moderate reactions, you can see, with erythema, which is redness, and swelling at the site are very, very low or nonexistent. And there's some level of mild and moderate pain. So very desirable profile in terms of local reactogenicity of the vaccine. The second dose is a little higher, again a few more moderate events. And you can see they're kind of clustered around the local pain and local tenderness. And again, I would say with perspective of lot of vaccines, this is a very good safety profile for a vaccine. If you go to the next slide, these are now solicited systemic symptoms. Again, they're graded through mild, moderate, severe. These are functional gradations. These are not -- these are, of course, not treatment, as I mentioned at the beginning. Just to be clear, there are no hospitalizations in this trial. And these are patients who are reporting their functional reports after vaccination. And again, you can see the vast majority of the reactions that we recorded are mild or nonexistent. There's some moderate and it's very scattered few that are considered severe, which it means that this has disrupted the activity of the subject. And so you can see on the right, slightly increased level and severity. But again, these are -- remember, this is a placebo-controlled trial. You can see a number of placebo events reminding you that the second dose of the 25 with Matrix at the bottom is also placebo. So we see a number of placebo events, including a few severe events. This suggests that these are really sporadic and no particular pattern in terms of adverse events. We do expect there will be some level of adverse events such as headache and myalgia, and we do see that here. But again, overall, it's our assessment, this is a very good safety profile for the vaccine. So on the next slide, so just to summarize, we do see a dose independent response. Both dosage levels induced high and comparable levels of IgG. So it's dose-sparing. You could not -- I'm sorry, there's a little error there. But this response appears to be the same for 5 to 25 microgram. For a vaccine like this that's going to be deployed, this is very convenient. This allows us to extend the supply of vaccine enormously by using such a low dose. We did see IgG responses that were compared favorably with those from convalescent sera from people who really had quite an important clinical illness and the adjuvant was required for the optimal response. We did see wild-type neutralizing antibodies that were numerically superior to convalescent sera, which is encouraging. Both dosage levels induced high and comparable wild-type neutralizing antibodies and we did see 100% neutralization and seroconversion rate after the second dose. Again, it's very important not to have people who are essentially left behind and don't have neutralizing antibody responses after the vaccination. And again, as I mentioned, the response -- the neutralization response is tightly correlated with IgG. That's a very desirable feature of a vaccine. We did have a strong T-cell response, multifunctional CD4 T cells were induced. They're largely the Th1 phenotype. And we have reassuring safety. We had no serious adverse events, that means no hospitalizations. All the unsolicited events were mild to moderate, and the local and systemic reactogenicity was not dose limiting, and we think, generally well tolerated. Okay. And I think with that, we're going to turn to questions.

Before you do that, this is Stan Erck, CEO. And Greg, thank you very much. That is saying it's a presentation that you've spent a lifetime getting ready to make, and it's really a big deal. We've completed the Phase I trial. I think that we could say that we couldn't have asked for a better profile of a vaccine for pandemic, particularly in the time frame that we've done this. And so the Novavax team has done it. And we've got a good safety profile, as Greg just mentioned. We've got all the immune responses that we could ask for, particularly in the Phase I trial. So where do we go? This is -- the pandemic rages on. We have -- we worked 24/7. We've got 2 sets of tasks now before us that we've been working on. The first is actually, you typically do this after you get your clinical data, but we're trying to figure out how to make vaccine at super large-scale globally. And we're doing that. We have now started production. We started tech transfer and production in over 7 countries of both the antigen and the adjuvant and what they call fill and finish packaging of the vaccine globally. We're doing it in Europe. We're doing it in India. We're doing it in the U.S., and we're doing it in Asia. And so that is going as fast as humanly possible, so that we can make a target which is to have tens of millions of doses available toward the end of this year and 100 million doses available early in the first quarter of next year in the U.S. and a similar amount in Europe. And that's our goal. We have ambitious goals for capacity for 2021 and we think we can make somewhere in the order of 1 billion to 2 billion doses of vaccine in that time period. Then we have to show that we have to extend these results that we got in Phase I and we take it into Phase II trial as soon as possible. We delivered these data to the FDA yesterday or today, and the FDA will look at them and give us feedback and a thumbs up on proceeding into a trial of approximately 1,500 people in the U.S. and Australia, which will extend these results into the older adult population and expand what we've done in the 131 people in this trial. And in parallel, we are making plans for global trials that show efficacy. And so our goal is to have efficacy trials in multiple countries and with the target of getting into our first efficacy trial by the end of September. And we'll keep everybody posted on the details of those plans as they become clearer to us, both for the manufacture and the clinical development and the ultimate licensure of this product on a global basis. So with that, I'd like to turn it over to -- for Q&A.

[Operator Instructions] Our first question is from the line of Eric Joseph from JPMorgan.

I guess, first question is, to the extent it's possible, there's going to be a lot of interest in making -- being able to draw comparisons across competitive landscape. I appreciate that there are different -- little differences in the particular assays used to look at neutralizing antibody titers, but to the extent it is possible can you sort of characterize your convalescent sera samples here in terms of titer and how that sort of establishes the bar that your -- what that represents as a bar that you're referencing to when looking at the activity of your candidate being somewhere around 3 to 4 x more potent? And then secondly, as we think about the Phase II study design, can you just kind of give us a little more color on sort of the doses that are being moved forward there? At what point you -- what point of follow-up do you think you'd be able to advance the dose into the Phase III?

Thanks, Eric. I appreciate the questions and the interest. So -- yes, so I think we, like all the companies, have attempted to anchor the immunogenicity on convalescent sera from humans that have been collected in different settings. And the rationale, as you can appreciate, is that, generally speaking, in the context of viral infections, there's an expectation that an infection that is clinically significant should result in immunity. In fact, if you have someone who's really sick, it kind of defines the ceiling of what might be achievable. And so we were able to collaborate with Baylor College of Medicine who had -- was obviously being overwhelmed with subjects. And we are able to obtain the clinical history, ensure that they were PCR positive and sera samples that will allow us to assess the immunogenicity. Frankly, we -- this process has all moved very fast. I know all the other companies and investigators have struggled a bit to obtain convalescent sera. One reason is that most of it's being used for -- as plasma for therapy. And we were able to do this, as I said, through Baylor College of Medicine. So we, frankly, didn't know where this would fall out. We knew that we had clinically significant illnesses. So we felt that was important and identified the sera. We were, I'd say, gratified to see that, in fact, there was a good relationship between the severity of the clinical outcomes that we've collected and the titer, the neutralizing titer, which fits with the, I would say, the normal scientific paradigm of recovery from a viral infection. So I think it's a good anchor. It's in the same assay. Our assay, by the way, is conservative. It's a wild-type assay. It's 100% neutralization. Obviously, all this is done blinded at the University of Maryland. We're very gratified or have a lot of gratitude that they were able to do so much work for us in such high-quality and get such clear answers. And so -- and you can see, if you look at the slide there, Slide 10, if you wouldn't mind putting that up or we can't see it, Slide 10 has the geometric mean, and it really sits at the top of this set of convalescent sera. And so again, it looks like we're really -- we've really pushed the immune response to a good place with these neutralizing antibodies. So that's gratifying. I missed, sorry, help me with the other piece of the question...

It was regarding the Phase II.

Oh, Phase II. So yes. So just to be clear, as Stan mentioned, this data has been delivered to the FDA for review. That's going to their view. We believe it should be very positive, it will then be important for us to then kick off the next stage of our trials. Our Phase II trial is slated to -- so as you can see here, we have a really convincing answer on dose and that the 5 micrograms is going to be the suitable dose to go with. And so however -- these are healthy young adults, are 18 to 59, not exactly young, but maybe can still play soccer. So it's important to extend our data -- our safety and immunogenicity data into older adults and we felt it was wise for us to use the higher dose. We're having a small set of subjects who are going to receive the 25-microgram dose, again in the older adults just to confirm this result that 5 micrograms is the dose going forward. But for planning purposes right now, this is such good data. And so clearly, digital that we are, in fact, planning to use the 5-microgram dose in our efficacy studies.

A follow-up, if you don't mind. I guess given that you don't have a dose-responsive relationship from 5 to 25 micrograms, appreciating your point about this safety profile here being -- your comfort with the safety profile, I'm still curious to know whether you'd see the need to evaluate perhaps a lower dose, a lower adjuvant dose in the Phase II?

Yes. So what you see here is you see the evidence that we're at the top of dose response curve. So we're still at the maximum dose that the immune response can be achieved with this vaccine and formulation. So clearly, it's possible we could go down in the dose. Now we're on a very fast trajectory out to clinical trials. So we may investigate this in one of our trials by going down the dose. And that would be, of course, another positive thing to see that we can further dose spare. You may have seen in the primate study, we used a fractional dose, which is something we wanted to know about. And again, in that study, that we saw sterile immunity, that was the one -- there was one very low titer virus in one animal in that study with the fractional dose, but it was half Matrix -- half the Matrix dose of 25 and then half the nanoparticle dose of 2.5, and we still saw a very robust response. So we think there could be room to explore down, however, we're on a very rapid trajectory. We've got to make some assumptions today. In fact, we made this guess that this would be likely 2 months ago based on our non-human primate preclinical data, our previous experience with Ebola and it looks like we chose wisely. We felt that it would be likely to be dose sparing and are very glad to see it so clear in terms of the fact that the 5 and 25 really are hard to distinguish.

Our next question comes from the line of Charles Duncan from Cantor Fitzgerald.

Congrats on a great result from our perspective. I wanted to ask you a couple of questions. So first of all, with regard to convalescent serum, I'm wondering if you could provide any additional color on the clinically significant disease and perhaps looking at at-risk patient population beyond just age. So do you have any additional insights that you can make on the comparison? And then secondarily, with regard to the clinical results, you did study a little bit broader sample in terms of age, up to 59 year olds in some of your -- some of the other data sets that we've seen. And I'm wondering if you have any data from some of those older patients, say, 55 to 59 in your sample set?

Yes, thanks. Good question. So we haven't stratified by age the analysis. So we unblinded last week. So that would be a kind of exploratory analysis, and I'll take that up. It's an interesting thought. We -- as you know, we have recently unblinded a Phase III pivotal trial for influenza, and that is focused on the older adult population, 65 and above. So we have a lot of experience with our safety and the immune responses. As you know, we met all 8 primary -- coprimary endpoints. Every secondary endpoint, we had a significant result. In the exploratory endpoints, we saw great T cell responses. So actually, the agency will allow people to bring some of that experience to bear on the current product. So they're looking for a platform like this. And if you have that kind of data, I think that's going to be relevant. So that will help us extend into this older adult population. So we're very confident that we can induce a good response, a good T cell response, a good antibody response in older adults, based on our experience in Phase I, Phase II and Phase III with our NanoFlu, which is a very -- obviously, a very similar platform technology. So just breaking this down a little bit more in terms of the convalescent sera, the guys on the top are the ones that were in the ICU. There's a cluster in the middle. These are people that were in the ER, and they had really actually syndromes that you would expect to essentially be moderate COVID disease in the clinical trial. So they had cough, fever, shortens of breath, hypoxia but just were not sick enough to be hospitalized. And then we had a smattering of a few asymptomatic subjects. And you can see they had -- some of them had titers, and you know that story. That, of course, you can see the immune response, the way this assay works, you start at a dilution of 20. So you see all those dots at the very bottom. Those are no neutralizing antibodies, that just is a place where we actually start the dilutions of sera. So there are a few people in the convalescent sera who had no detectable antibodies by that measure. And you can see also, if you look back at the IgG, the same -- it's a very concordant story. The neutralizing antibodies and the spike IgG really go together very nicely. Again, more severe have a higher spike IgG and I showed you the correlation. So I will tell you, there's a lot of work that goes into establishing good measures. We think our measures in immunity are extremely sound. We've worked hard with our collaborators to make sure that data was good. We've been gratified. We really saw a very reliable picture on the sort of the quality information that came out of our collaborators at the University of Maryland School of Medicine. So we're quite confident this is really good data. We know these are really very good functional immune responses. And it's our view that this trial has been a fantastic advance forward for pandemic vaccine.

Yes, that's very helpful, Greg. We definitely appreciate substance over speed. One quick question regarding the Phase II and timing of it versus the Phase III. I guess I'm wondering if you see the Phase II data with a little bit older patient population for sample to be enabling for the Phase III, is that a nice to have, the need to have, or could you anticipate actually starting the Phase III before you got the Phase II results with that little bit different sample?

It's a great question. So all issues that we discuss with our partners at the Operation Warp Speed. I will tell you, right now, the plan is to extend the safety data in adults over -- 16 over. That data can be collected fast. We -- and then we'll use that. Again, it's going to be an interim look at the data, and we'll use that to allow us to move into the broader population for the Phase III trial. So I think what we -- we're on track to do that soon. And we think that will be an important part of what -- trigger for the Phase III trial.

We do have more questions online now. The next question comes from the line of Mayank Mamtani from B. Riley.

And appreciate you putting out this data in such a tight time line. Also helpful clarification upfront on the safety around hospitalization, so good to lay that out. So 2 questions for me quickly. So Greg, as you think about the older population and you think about the adaptive immune system -- immune response, obviously, you talked about CD4 Th1. What other elements do you think like CD8 T cells also that you may or may not have yet? Could you comment on that just broadly as you think about the older population?

Yes. So that's really good. So let me just harken back. As you know, the flu vaccine, the seasonal influenza vaccine, where the results were very good, one of the features we have there in immune senescence is an almost absence of antigen-specific T cell responses. Matrix-M is a very good adjuvant for inducing CD4 and CD8 responses. And so what we're seeing today CD4 responses, it's not just important to show that it has a Th1 phenotype, I would say, more importantly, we know that these cells are involved in enhancing the antigen -- sorry, the antibody maturation and maintaining durability of responses. So this T health does, in fact, manifest itself as better and longer antibody responses. It's really critical. And we know that, that could be a real issue with the older adult population. And so we expect that our Matrix-M adjuvant will generate very robust T-cell responses, CD4 polyfunctional T cell responses in that population. Now we know that CD8 responses, we've -- in our animal studies, and again, in our preclinical paper, which is posted and out there under review, we do show we get very robust CD8 responses. It's been -- today, we're showing some preliminary data. We expect to bolster that and pursue showing the CD8 responses. It's a little more technically difficult; when you do that in mice, they are a very narrow genetic background for the MHC Class 1 pockets. So with generally diverse humans, it's a little more involved, but I think in the future, we will be able to show CD8 responses. And just 1 other comment, you can -- we just provided the scatter plot. You can see the magnitude of these responses are really quite robust compared to what's been reported out there. So we're very happy with these responses. These are in a small subset of subjects, and we're continuing to expand on that data. So I think you want a vaccine that does recruit both arms of the immune system, both the T cells as well as the antibodies. And you can see here, we have a very good profile. Again, that's why we're very buoyant about this vaccine because it looks so promising. And I would say that our -- to date, our challenge data has looked very good. In fact, we see sterilized immunity, which is really encouraging in more than 1 animal model. And so again, prospects for a vaccine that really has good efficacy seem quite high. We're very eager to get out into the phase -- the efficacy studies.

Absolutely. I mean the volume of evidence to date across mice, baboon and challenge studies and this look fantastic. Just quickly on the 5-microgram dose, if I may. A 2-part question there. Can you comment on the reactogenicity that you saw here maybe relative to the NanoFlu, which also looks pretty good? Just kind of observations since the Matrix-M adjuvant is kind of the common denominator. So if you could comment on that? And then the -- it definitely seems like a sweet spot where you've landed with this lower dose. If you could comment on the scalability of this? And what is specifically your assumption? You've said 100 million doses by the end of the year. Does that include an assumption of 5 microgram? Or is it a dose higher when you think about delivering on that dose goal by the end of the year?

This is Stan. Those -- my forecasts are based upon a 5-microgram dose. And so that's the most likely dose that we're going to go forward with. As Greg said, we will evaluate at some point in the future trial, in probably not-too-distant future, at least an arm that tests the level below 5 micrograms to see if there's a smaller dose available than that, which obviously automatically expands potential capacity. So we'll look at all of that over the coming months.

Great. And any...

Of course, we are using the baculovirus insect-cell system, that's the basis of 2 licensed products. I think we have the nanoparticle, the Matrix-M, really are the culmination of years of thinking and work to come up with an ideal vaccine platform, but it does -- it is the manufacturing is based on a well-established manufacturing platform and so I think it's given us confidence in having dose sparing that is really convenient. That's given us confidence that we can as we scale up and make a lot of doses.

Any color on the reactogenicity of what you saw in NanoFlu versus...

Yes. Yes. So NanoFlu is in older adults. I think when you get into older adults, things quiet down a little bit. Again, there, our safety profile was compared to the Fluzone High-Dose, where you really couldn't distinguish the reactogenicity between the Matrix-M adjuvanted nanoparticle vaccine and Fluzone High-Dose. So I think the -- we're giving a lot of detail here. But you can see that the -- there's a lot of subjects who report no reactogenicity. I'll point out some of the really positive things. You can see, there's almost no fever. We had 1 subject who had a very low-grade fever. So that's all -- that can be a very difficult thing to have in a trial when you're out in COVID. So our view is the reactogenicity looks quite good, mostly mild and moderate. In fact, frankly, most of it is mild to nothing. So that as a profile for a vaccine is really quite good. And given the fact that we have so immunogenic in terms of the immune response, I think together, this is a good profile. And as I mentioned, the fact that it's a 2 to 8 stored liquid formulation is also very attractive. So we're celebrating. We think this is a tremendous result. We're eager to get the FDA's view on this. We have, I think, a very positive outlook. A lot of people that we work with in terms of scientific advisers, et cetera, they're looking at this. I saw with some gratification that Ed Belongia, who's a very experienced vaccinologist at Marshfield Clinic. He's an expert in these respiratory diseases. He was impressed with the data. And I know him from social meetings, et cetera. When he's impressed, you know the data must be very good. So we're extremely excited about this data set.

It's great to hear. Congrats again and look forward to more updates.

And we do have 1 more question on the line currently. [Operator Instructions] Next question we have on the line comes from the line of Seamus Fernandez from Guggenheim.

Great. So I guess the focus of my question, I think most of the questions have been asked. But one that hasn't really been addressed is distribution. Can you just talk a little bit about -- as you progress towards the broader manufacturing capabilities, what your storage conditions are and what you envision as some of the complexities? And then separately, just on the commercial side, as we think about potential for additional negotiations, we know that you have the sort of robust plan and agreement with BARDA, but expanding beyond that, where do you see the opportunities to kind of add additional capabilities? And would you anticipate needing any distribution partners in that regard? Or is this something that Novavax is prepared to distribute and handle on its own?

Yes. Thanks, this is Stan. So one of the highlights, I think, and distinguishing features of our vaccine is that it is stable. It is -- you can -- we've got data that says you could store it at room temperature for days and perhaps weeks. It has long-term stability as much data as we have because you can't have too long-term stability working on for 6 months, but 2 to 8 degrees, it looks very stable. This is really important. I mean it's important in a normal commercial vaccine, but a pandemic global, they could be able to distribute it worldwide with using the established cold chain, and it's formulated as a liquid. We are coformulating it, the adjuvant antigen. So it's a single vial. All the characteristics of a vaccine for pandemic are contained in this product profile. And so we're really, really happy about that. And I think the people who have funded this program like Operation Warp Speed and CEPI and the Gates Foundation, really appreciate the character. And maybe it's not unique to our product, but it might be. So we're very happy with that. Going downstream to your question of how are we going to get it distributed globally, We're going to have partners. And as I say, we're trying to manufacture it in -- we've got 7 different countries going right now, and there's the process of getting the antigen and adjuvant made at large-scale and then find fill/finish capacity in many, many different countries globally, and we'll seek out partners. We have -- once we make product and deliver it to the U.S. government under a contract with Operation Warp Speed, that will be distributed under established vaccine distribution systems. Ultimately, we'll get to a past the pandemic period in the next however long, whether it's a year or 2, and we'll get into a commercial distribution system that we're yet to define as we try to commercialize both this product and, frankly, our flu vaccine down the road. So we're going to be the respiratory vaccine company of the world.

And maybe just as a follow-up question. Can you -- would you mind just providing an update -- and I know this is a little off topic, but other efforts beyond flu, I know that you've proceeded on the RSV side to some degree. I was just wondering what -- where we would see additional vaccines, other catalysts moving forward? I know you guys are well in the weeds on CoV2 and thanks for that, of course, from all of us, but just love to know about some other catalysts as well.

Well, so we actually -- everybody who has followed the company knows that we had 2 Phase III programs in RSV, one in the older adults and one in maternal. In my view, both of those vaccine worked quite profoundly. We dropped hospitalization from pneumonia in kids globally by 50%. Nothing else has ever done that in the first year of late. And we also had an effect in the older adult -- high-risk older adult population. So I think those are 2 vaccines that we could go back, we could restart those trials with maybe a bit of a different design, use our Matrix adjuvant, which we didn't use before. And so those are 2 very high unmet needs. So look at the respiratory world, you've got RSV. You've got flu, you've got COVID. And then, of course, we have to be ready for -- down the road, new pandemic threats. And ultimately, there's an opportunity to combine these vaccines and have a pentavalent vaccine with flu and COVID something. So there's lots of opportunities. We're not a one-stop shop.

Yes. You may have seen, we published our RSV Phase III trial in infants was published in the New England Journal last week. So that was, I think, a good outcome, a great vaccine.

Yes. Well, again, congratulations on these data and look forward to seeing the Phase II and the Phase III advancing.

And we do have 1 last question in the queue. This is from the line of Vernon Bernardino from H.C. Wainwright.

Congratulations on the results. You give hope that we could have a safe vaccine next year. And that's really something to celebrate. Question I had is, will you be expecting to continue to follow the Matrix-M adjuvanted subjects who were given 2373 as far as their antibody levels? And I was wondering if you could remind us the -- if you did follow, and if you could remind us the antibody levels in the nonprimate challenge studies? What the antibody levels were afterward? And that's it.

Vernon, good question. So we will follow-up on the Phase I subjects for safety after a year. And we'll have several blood draws to look at immune kinetics. Right now, everyone's going to be interested in durability. We don't have a lot of data on that because everything is fairly fresh. But we will -- we have a couple of primate studies where we're going to look for the long-term immunogenicity. I would say this is very reminiscent of our Ebola vaccine. From an immune standpoint, the immune system doesn't care whether it's an Ebola glycoprotein or a COVID glycoprotein. So in that context of that development program, we did follow subjects out to 1 year. And we measured the immune responses out at 1 year. We had done work on the Ebola vaccine like this, the ELISA using IgG to the GP protein, was associated with protection. And in fact, titer of around 5,000 was predominantly protected. And we can see in the humans at the -- end of the second dose, we had 50,000 out at 1 year. It was still -- it was 5,000. So we see persistent responses, and that's going to be a topic of a lot of interest. One thing I would say, as Stan mentioned, this -- we are in the view this virus is not easily going to go away, and there probably will be an appetite for boosting schedules. And so that is completely something that's up in the air, but we can demonstrate by this trial, that the vaccine is boostable, which could be a very important quality for COVID-19 vaccine. So nanoparticle, Matrix-M has that feature that you can boost and I think that's a very desirable feature.

There are no other questions in queue at this time.

All right. So this is Stan. Let me wrap it up. Thanks, everybody, who's stuck around for this amount of time. We're -- couldn't be more excited about where we are. We've been working for this day for the last decade in trying to get vaccines against emerging infectious diseases. This one became very apparent. And as recently as late January and February that this was going to be -- this level of effort was going to be needed. We worked our way through the mice and the nonhuman primates, and now we've got our first human data. Everything is consistent. All the studies are consistently showing that the vaccine is -- has the power to prevent COVID disease. And so we're going to be working as hard as we can to continue this into efficacy trials and ultimately, where it needs to be in the marketplace. So thank you.

Ladies and gentlemen, that does conclude today's call. You may all now disconnect.