Moderna, Inc. (MRNA) Earnings Call Transcript & Summary

Over time?

Over time, over the course of -- maybe it takes maybe [ a day, maybe 60 day ], 70 day, 90, with the South African variance. It looks like what day 35 looks like with the Wuhan variant, like because we don't have sequential serum.

So I don't think so, and let me tell you why. So I think we're confusing 2 different things here. The -- that sense of improvement over time in the first few weeks, I think, is unique to the adenovector platform because the adenovector stays around in the body for longer when you give it -- remember, it's a competent virus until you clear it. And so think of the immune system's response, it's a response to a more lingering initial infection as a result of vaccination, if you will, compared to an mRNA vaccine that really, within days, comes, focuses the attention of the immune system just on that spike protein to a very high level and then completely dissipates. If you look at our data, our antibodies peak, and then they start to slowly decline, as you would expect with a vaccine. When we come in and give our booster shot, we then get at least a log, if not higher, of a boost because what we're doing is we're coming back later and telling the immune system, hey, that thing you thought you cleared, you didn't, again, with a very short, high burst of focus on a protein that's foreign, which then completely dissipates. Now you can boost with an adeno platform, and to some degree, you can be successful, but the magnitude, that delta you get between the first and the second is less than a log. And with an mRNA platform, it's higher than that. And I suspect if you come in with a third dose, whether it's matched or mismatched to the strain, but remember, these mismatches are still 99% homologous, you're going to see that third dose give you again a log of a boost, just like we saw with our CMV vaccine. And so with an mRNA platform, what you're going to get is you're focusing the attention of the immune system, you're getting a boost, that boost gets to a certain max within the weeks after infection, and then since there's no longer any antigen, you should expect it to wane over time. I think with an adenovector, it may be that, that peak happens later, but it will still peak at some point and then wane because you don't have any antigen around. Now teleologically, if what you're saying was true, which is you just gave a vaccine, a transient infection, and then immunity continues to build, well, by the time you were my age, you would be walking with your blood chock full of antibodies against all the things you've seen in your life, right? So biology can't function that way. You got to get to a certain peak. It's got to be temporarily somehow associated with the antigenic exposure, and then you're going to see a decline over time.

That's useful. I think another question that we received a few times is, so we've seen fairly clean safety profiles for most of the vaccines, relatively a small number of rare events, a couple of sort of more concerning exceptions in the AstraZeneca trials [ where it ] all results in a brief pause. To what extent can we draw conclusions around the safety of delivering a polymer-modified spike vaccine by pooling the data across all of these programs? And to what extent do we have to be more wary and really think of the safety of asset-specific and platform-specific?

So it's a great question. Let me break it down in 3 parts. First, I think the emerging safety initially is super reassuring if you look both individually and the totality. So CDC just published their initial look, millions of people dosed in the first month. That was in the Morbidity and Mortality Weekly Report last week. And basically, their conclusion is, look, even the anaphylaxis is rare and is seen at the rates that we're used to seeing with other vaccines, and at least in early days, in the first month, we haven't identified any significant safety concern that we hadn't known about in terms of the reactogenicity of the Phase III trials. So millions of people in, and that's the reassuring picture. And I can tell you that we have not seen any significant safety concern since then. So I think as it relates to the platform, the 2 mRNA platforms that so far have gotten the most exposure and ours, including now with a total of over 60 million Americans exposed, I think we're in a really good place, and that should be reassuring to everybody. Now that speaks to a couple of elements. It does speak to the mRNA technologies writ large. I think it speaks specifically to ours and specifically to Pfizer's. Now your question about the spike protein, I think, is sort of point #2, which is yes. This has been the most astounding scientific replication in my career in the sense that if you just look at us and Pfizer, you've got 2 companies who are operating completely independent. By law, we can't collaborate, right? And yet, we've taken the same scientific principle and the same antigen, and we've run huge Phase III trials. And then we've gone with the U.S. government, we've treated millions, tens of millions of people, and you see amazing concordance. The safety -- the efficacy came in within 1% of each other, and the safety profile is looking concordance as well. So that replication from a simple -- from a lens of a scientist that always, okay, one is a fluke maybe, can you replicate it, my God, we've just seen the most beautiful replication you could ask for. And that's true of the emerging real-world evidence data, whether it's from Israel, the Mayo Clinic recent report. So those high rates of efficacy hold in real world and the safety, as I've described. So that's point number two, there is good scientific replication. The third point, and this is coming back to the spike protein. I think that the lack of a safety concern tells you that there's no emerging concern for the spike, nor frankly, for anybody's lipid nanoparticle, and we know that the mRNA itself is completely innocuous. Now can the spike protein per se generate any adverse immunogenicity? Well, that's theoretical, perhaps. The truth is, if that would be the case, then the highest risk would be when you're exposed to the spike as part of an infection by COVID, right? So a COVID infection is still going to be worse than any vaccine would be. But in that case, I do think that if we start to see data from other platforms, we would be informed. So if we have safety signals that occur now and they occur in a different platform, and they are not seen across, then you would draw the logical conclusion that to the degree that there's causality there, it probably has to do with the platform rather than the spike protein being encoded. And I think that's the beauty. All of us are encoding the spike protein. Most of us are encoding probably the same proline-substituted spike protein for the prefusion conformation. So I think that should give all of us comfort, not just in the existing safety, but that we're well positioned to understand the safety profile in the future.

So drawing on that the same realm, as we look at the data sets from Novavax and J&J, whatever you -- however you want to interpret the puts and takes on nuances of the vaccine themselves, they do give us an initial look at the baseline characteristics of how people who are seropositive at baseline, presumably for the ancestral variant, appear to have, for some level of protection in the J&J study, which is the larger and more robust of the 2, apparently 90% implied VE, if the vaccine was [ direct with your ] viral infection, against the so-called South African strain, [ maybe they will ] emerge in many places simultaneously, can we take from that implied cross-protection of ancestral infection? Can we take from that some comfort that vaccines driving protection against the ancestral version should also have some level of translating robust protection against this variant? Or is that a step too far, using natural infection as a proxy for vaccination, not the right way to think about it?

So I -- people will have different beliefs. Where I land is I do believe that natural infection is a proxy, especially for any vaccine, whose level of immunity exceeds that of natural infection. So I feel that I'm uncomfortable ground-extrapolating because I know that an mRNA vaccine, will induce higher titers of neutralizing antibodies than natural infection. So if natural infection is protective, then vaccination should, for sure, be protective, again under the assumption that the neutralizing antibodies are indeed the correlate of protection. I think there's good data from preclinical species to demonstrate, and even humans with the monoclonal antibodies, that show that if you can transfer protection with antibodies, you indeed protect. So it's the antibodies that matter. We achieve higher levels than natural infection does on average with immunization. And since natural infection, to some degree, is protective, then the vaccine should be at least as protective, if not more. And the last piece is I look at the preclinical data for our vaccine and note that while the in vitro -- there is some loss of neutralization against the variant. When you look at in vitro assays, the level that you maintain of neutralization, I think Tony Fauci has made the same sort of comment, you still have a cushion of protection. That level of neutralization that you maintain is still sufficient to protect a nonhuman primate from infection in the lungs. So if you connect all the dots, both the natural epidemiology data, standing on the legs of natural infection and your relationship to that and the nonclinical species, my belief is that a primary series with an mRNA vaccine should be protective against these variants of concerns certainly in the following months of -- after vaccination. Now is immunity in the long run going to wane and will it wane more quickly against the variance of concern? That is something that does worry us, and it's one of the reasons why we put these variants into production. But I think where we stand today in the midst of this pandemic, the best thing we can do that should be protective is immunize everybody as fast as we can with these ancestral vaccines.

And I'm going to pivot to something other than COVID for a moment. I know that's -- everyone -- I know everyone is dropping their phones in shock at the moment. One of the things that was brought up in the morning -- on the call this morning was the potential to expand the pipeline with an incrementally greater focus on autoimmunity. Obviously, that's a very different space than vaccines. And in some ways, you're attempting to do the inverse, the chronically modulating downward immune and unwanted immune response as opposed to acutely modulating upward immune response. Talk a little bit about -- how you think about are there unique technical challenges in translating what you've learned in a large human population [ and ] vaccines to applications in the autoimmune space? What lessons are translatable? And what lessons, what technical challenges have to be addressed for autoimmune indications, specifically that don't necessarily show up in the vaccine world?

So I think the 2 challenges that need to be addressed have to do, one, with the technology and the other with your targets. So the one thing -- vaccine was always a low-hanging fruit for this platform because you need a smidgen of protein, your delivery vehicle as a simple syringe and you sticking it into somebody's arm, you've delivered your medicine. For the autoimmune applications, what we need is the ability to encode larger amounts of protein that would be systemic or in the target of the organ of interest. And you would need to know that those proteins are expected to have a therapeutic benefit. And it's easy to know what to vaccinate against. In fact, it's not a surprise that all the COVID vaccines are all using the same spike protein. Science was already -- knew what the antigen was. I think for autoimmunity, and that's why you see us take a few interesting shots there, whether it's an IL-2 variant that is predicated on science, or whether it's trying to do something completely unprecedented by encoding a PD-1 ligand in hepatocytes, we're taking different approaches, but the biology there is not as simple. As it relates to the technology, I think that entire modality stands on the shoulders of what I think will be the coolest scientific thing I will have ever done, which actually isn't COVID, it's the demonstration that you can encode a monoclonal antibody, an mRNA encoding monoclonal antibody, given in a lipid nanoparticle-injected IV and get potentially therapeutic levels in your blood. And we showed that back in 2019 with our monoclonal anybody against chikungunya. The reason we chose the monoclonal antibody was because we knew the protein was going to be innocuous. So there wouldn't be a question on safety of the protein you encode. A monoclonal antibody is as easy as it gets from a safety perspective. It was a test for the platform. And we showed there that, indeed, we are able, and we did this a couple of times, so even with a repeat dose, we could get to a potentially therapeutic level of a monoclonal antibody. Just for comparison, we reached 15, 16 microgram per ml in the blood. The concentration, if you look at the label for REMICADE is between 0.5 and 6 microgram per ml. So you clearly are in the ballpark of what should be a therapeutic level. You've done that once or twice already and showed it in vivo, in man. So I think from a technology standpoint, we'll get there. The challenge, of course, is that while the safety in the vaccine platform has been terrific, we're talking about doses -- a dose level that's going to be a magnitude of order higher and one that's given IV as opposed to into the muscle. So there's more work to be done there as we bring these forward. But the foundational concept that you can turn a messenger RNA into a medicine and that medicine in the body will lead to translating a protein you care about and that will lead to a clinical benefit, I think that question with COVID has been answered in a way that's unimpeachable.

That's really helpful. And we are -- I'm being told we're now actually over time. Thank you so much for taking this time. It's been a really helpful conversation. Congratulations on everything you achieved at Moderna, and glad we're going to have you around for a couple more quarters.

It's been a real pleasure. And I couldn't have asked for a better run as a CMO of a biotech than the past 6 years. Thank you.

Thanks, Tal. Have a good day.

You're welcome.