Regeneron Pharmaceuticals, Inc. (REGN) Earnings Call Transcript & Summary

All right. Welcome, everyone. This is Josh Schimmer from the Evercore ISI biotech team. Absolutely thrilled to have George Yancopoulos, President and Chief Scientific Officer of Regeneron. We also have Justin Holko, I think -- Vice President of Investor Relations. I think before we start, Justin has some forward-looking commentary to make for us.

Sure, Josh, and thanks for the invitation to participate in this year's conference. Before we begin, I'd like to remind you that remarks made on the webcast today will include forward-looking statements about Regeneron, including those relating to every great topic that George is going to speak about, mostly science. Each forward-looking statement is subject to risks and uncertainties that could cause actual results and events to differ.

Excellent. Well, George, one thing I'm finding is that the volume of innovation is far more expansive than ever. You've always been an expert in everything you've looked at and touched upon, how are you keeping up with all the science? And is it starting to affect Regeneron's strategy in terms of just the vast nature of it? And if so how?

Well, it is true. It does seem like technology is accelerating in terms of the rate of progress. But the thing that I have to say I'm proudest of, after more than 30 years here at Regeneron, is that I think that early on we were recognized as being at the bleeding edge of science and technology. And here we are more than 30 years later, and I think the argument could be made that we're still at the bleeding edge. I think that we're still leading in many forms of technology. Look at our Regeneron Genetics Center. We have sequenced more human beings than anybody else on the planet. And what makes our database different is that each and every one of them is linked to a detailed electronic health record, creating the largest big data set in the world having to do with the biologic sciences. And we're the ones who are driving and leading that. And when you're leading a lot of these technologies and a lot of these revolutions, and it doesn't matter whether we're still leaders in antibody technologies, where we're creating the bispecifics field, we're creating the human genetics space. It makes it a lot easier to keep up when you're the ones who are pushing the envelope. So yes, I think that the rest of the world, obviously, I mean, progress is being made and so forth. But the advantage that we have is, I think that in so many areas, we're still the leading technology group in the industry, and that makes it easy to keep up because it's us.

Got it. I'm hoping we'll spend a fair amount of time on the oncology portfolio. Just a quick question more thematically in oncology. It feels like maybe it's been 5 or 6 years for the focused Regeneron oncology effort. What have been your big lessons, takeaways, surprises or otherwise as you pivoted into oncology?

Well, I think the big lesson, not only for me, but I think for the world is that you were just talking about progress and how everything seems to be accelerating and so forth. But I think many of us, most of us, 10 to 15 years ago when the checkpoint inhibitors burst on the scene, Nobel Prizes were awarded and so forth, it was assumed that we'd be making a lot more progress that cancer would be cured by now, that you just put maybe the first checkpoint together with the second checkpoint. And literally, the challenge of cancer would have been met. And I think what we've learned is that's not the case. In fact, the first checkpoint inhibitor, for which Nobel Prize was awarded is literally disappearing from the marketplace and as a standard-of-care. And even the second checkpoint target, the PD-1, people thought it was going to be commoditized. They thought that there were going to be so many -- so many PD-1s out there, and I think most people reflecting back would be astonished that the first PD-1 blocker, nivolumab, still has not been able to demonstrate efficacy in first-line lung cancer. It just shows how hard and how challenging this field is. And I think that it goes back to the fact that it all depends on the science and the technology. It's not so easy. And that we need to continually be thinking ahead and innovating in terms of coming up with new approaches and new solutions because, unfortunately, it's never that easy. And so I think that, that's the lessons. And I think that -- I'm pretty proud of how we've positioned ourselves. In a world where everybody thought that PD-1s would be commoditized and anybody could come up with one, okay, the first anti-PD-1 itself, nivolumab, couldn't demonstrate first-line lung benefit. And here we stand now with some of the world's best first-line lung cancer data, pending our approval. But not only we have behind that a whole entire pipeline of leading and first-in-class agents in terms of our bispecifics, both the CD3 class and the co-stimulatory class that have all shown efficacy on their own, but most importantly, in preclinical models, they've shown incredible opportunities for synergies when combined. And so I think that, that's the lesson. The lesson is things aren't usually that easy. Things don't usually work out as people usually see them. And you've got to be prepared. You've got to be prepared for failure by having a large pipeline of exciting new classes of agents that you can bring to bear on problems, and that's what we try to do in all fields, and that's what we will try and do in cancer. And I think that we have a chance. We have a very legitimate chance of really changing the landscape here, of really becoming the leaders in this space. And just like we have the leading approach to blindness, the leading approach to all these allergic type 2 inflammatory diseases, I think we have a chance to be doing the same in many settings of cancer as well with our pipeline.

As I look at the new molecular entities in the pipeline, very heavily weighted towards oncology now. Obviously, you've got kind of the Dupixent pipeline and product opportunities as well. But as we think about Regeneron in 3 and 5 years from now, should we think about kind of this continued balance very oncology heavy? How do you kind of know or decide what the right balance is for oncology and non-oncology?

Well, I think, obviously, as we were just discussing, I mean, cancer remains a major challenge. I mean there's so many areas of unmet need. And we are obviously trying to do a lot to make a difference in so many different settings of cancer. But we continue to think that there are so many opportunities outside of cancer as well. And you mentioned Dupixent, but Dupixent and our anti-IL-33 strategies, we see them in terms of the COPD space as being game changers there. These are opportunities that unfortunately, I just recently had a dear aunt who was like a second mother to me die of emphysema. It's a horrific disease. There's really so little available for patients, and there's so much need out there. And we have an opportunity now with different segments of the population to maybe make a difference with both Dupixent and our anti-IL-33. Those could totally change the treatment landscape there and make such a difference for so many patients. I think that we have so many opportunities based on our genetics and based on some of our own, but also with our leading collaborations with companies that are at the forefront of gene therapy, I think we have opportunities there to really make differences. We're very excited about our collaborations with Alnylam. We are doing our combination studies. We're combining. We're doing the first combinations of antibodies with siRNAs, which I think could create a whole new modality that could combine the benefits of both in ways that nobody else could really combine them because other people don't have both of these tools in their toolkit. And so one of our first indications there is a proof of principle could by itself be incredibly important in terms of our C5 opportunity, where we really think that, once again, we can -- we can change the practice of medicine by combining our -- what we think is our best-in-class C5 antibody with a best-in-class siRNA against C5 and really change the treatment paradigm, and that could also be a huge opportunity. And behind that, there's a whole pipeline of combinations of antibodies and siRNAs. With Intellia, we just announced that our first program using systemic CRISPR therapy, the first CRISPR therapy now entered into the clinic, and we have a whole pipeline behind that of opportunities. So as you said, we're not standing still. We're trying to build the future. And the future is, to us, it's everything. It's very broad. Every place where there's disease, that doesn't have an optimal solution today, we're hoping to go there, whether it's gene therapy, whether it's CRISPR, whether it's siRNA or whether it's these new classes of bispecifics, that just a few years ago would have been viewed as science fiction. Imagine that, not only a bispecific that brings a T cell to a cancer cell but a second bispecific that actually stimulates that T cell to ramp up its killing. And we actually have applications of our bispecifics outside of cancer as well. So these are all incredible proof of principle opportunities that we're doing that each one of them could unlock much larger and very exciting opportunities behind them, whether it's our bispecifics unlocking a whole new field or whether it's combinations of antibodies and siRNAs unlocking a whole new field or whether it's CRISPR-mediated systemic gene therapy unlocking a whole new field.

But what does the combination of RNAi and an antibody get you that you're so excited about? What does it do that either one individually can't? And where is that best applied?

Yes. And I think that this -- obviously, a lot of people know about PNH, paroxysmal nocturnal hematuria, and a lot of people already know about the antibody approach there. And the beauty of antibodies is, of course, that we know they can be rather safe and well tolerated depending on the target they're attacking and so forth. But their limitation is on what you call target load, what they bind to. And if there's a very large amount of the target that they bind to, then you have to give a very large amount of the antibody and the antibody gets used up very quickly. siRNA, in some cases, can do a good job of reducing target load against whatever target you're trying to address. But very often, it cannot reduce that target load in activity to 0%. And so as you may or may not know, already with our partner at Alnylam, in their clinical data, they showed that they could reduce target load of C5 by 80% to 90%, but that had no benefit because you really need to reduce it to 0, okay? And so you can get the best of both of them now. So we can reduce the target load using the siRNA to 80%, 90%, which by itself would have no efficacy as they've shown. But we have the antibody, the antibody brings you down and you get complete efficacy, but now it lasts a lot longer because you don't have all that target load. So getting the benefit of the siRNA for the target load reduction and then getting the benefit of the antibody biologic that can really reduce things to 0, you get the best of both and allows you to have long action with complete blockade. And I think this, for us, is a powerful approach which we already are -- we have proof of principle with the individual compounds, and we're now initiating our combination studies, we really think that this could provide a whole new approach, and we are in a prime position in that we have a great collaborator on one side, and we have the expertise on the other side, and this could be a whole new franchise of opportunities for us. And I think C5 is the perfect example.

Where else beyond C5? I mean I'm surprised we're not talking about VEGF intravitreal approach as doing the same thing?

Well, I think that we've announced a very broad collaboration with Alnylam across 3 different sets of disease area targets. And I think that -- hopefully, I'm not speaking out of turn Justin, but we announced that our target areas are liver targets, of which they are numerous, where you could use the combination approach. CNS as well, where you could also imagine the same thing, as well as the eye. So yes, we think that there's a broad set of applications and using these combination approaches together in many of these things could change the practice of medicine.

Interesting. Maybe sticking in ophthalmology, maybe the latest on the high-dose EYLEA program. I guess we've -- Beovu has run into its safety issues. Is this still an important and relevant program and if so, it's for what? Are the other programs that you feel like you just need to keep EYLEA ahead of the curve relative to?

Yes. As those of you who've been following this field now for many years, EYLEA has really been the standard for about 10 years now. And it's obviously been under attack and every year there is another -- somebody else who is going to outdo EYLEA. The problem for all the competition, but also for us, is that EYLEA sets such a high bar, not only for efficacy but for safety as well. And this is unfortunately what we're seeing with the brolucizumab, was that passing that safety hurdle is not so easy. I mean the eye is not some sort of protected space, it's actually a very sensitive space, where immunological reactions and side effects are actually more of a concern or problem because they so easily disrupt vision. There has been a fallacy. There has been entire fields that were devoted to the notion that the eye was somehow protected vis-à-vis angiogenesis and protected vis-à-vis immune reactions and it turns out that those 2 fields have largely been disproven. There is no protection in the eye. And this is why you have to be so careful. And this is the beauty of our EYLEA program from the beginning. That's what we focused on. We think that there is an opportunity now to enhance and extend the duration of action of EYLEA by going up to a higher dose. And I think what we have announced based on our Phase II is that so far this high bar of safety was met, that we're able to achieve by, obviously, concentrating but also making sure we had highly purified and well-formulated form of EYLEA, that by increasing the concentration of the dose level fourfold, at least at the scale of initial Phase II, that we're able to clear that safety hurdle. And so this would allow us to now give this more concentrated form and maybe we could get all of the benefits of EYLEA, well, may -- in terms of efficacy, extended duration of action while not raising additional safety concerns. And as I said, safety right now we believe for all of these approaches including EYLEA are the biggest hurdles because EYLEA has been such a safe and well-tolerated approach, having saved so many eyes over the years with such a gratifying safety profile. So that's going to be our hurdle. If EYLEA meets it, then maybe we can provide additional benefit in terms of less frequent dosing and longer duration for patients, but it's going to be a challenge because of the high bar that EYLEA sets.

How do you think about the optimal trial design then for high-dose EYLEA to showcase its potential?

Yes. We've, I think, understood and developed a lot of approaches that we think really allow us to more efficiently than others, determine whether things have benefit or not. And I think, like I said, I mean, I don't want to review the graveyard of challengers to EYLEA, where they based results on small Phase IIs, and they saw a few letters difference one way or the other. And they got a lot of people on board. They got a lot of market cap value and so forth only to fail in the larger Phase IIIs. Well, I think we really learned how to design the small Phase IIs and to really understand and show whether or not we see additional efficacy or not. And I think we did it beautifully. For example, when we combined our Ang2 antibody with EYLEA, and we were able to convincingly show that there wasn't this dramatic benefit of that mechanism, there wasn't a benefit to that mechanism. And so we're doing the same sorts of things. And I think that when we release our initial data, which I think is going to be released in the first half of next year, hopefully, it will answer the question about whether there is additional benefit, particularly in terms of duration for patients, and it will continue to also amplify on this question of safety: is it meeting that very hard to meet safety bar that EYLEA has set?

Got it. Maybe toggling back to the oncology portfolio on the bispecifics, and you've got the CD3 bispecifics and the CD28 bispecifics, and I think you've talked about maybe even -- you can use one, you could use the other, you can use both together. How do you figure out like which technology to apply for which target in which patient? Maybe talk a little bit about what differentiates the CD3 from the CD28 approach.

Right. So the CD3 approach brings -- by just binding to the CD3 complex on T cells, brings a killer T cell to the tumor cell. And obviously, based on our clinical data, I believe, our CD20 bispecific was the first-in-class, and it demonstrated really profound efficacy in various late-stage lymphoma settings. So just doing that is enough. The second class of bispecifics, the so-called co-stim bispecifics, take advantage of the most powerful co-stimulatory pathway that is this CD28 pathway. And what they do is once that T cell is brought in contact with the tumor cell, you now add a second bispecific that also binds to the tumor cell but now further stimulates that T cell to do even better at killing, and we've seen really profound synergy. The advantage of these approach is that, if anything, you can actually get focused specificity, you can do this, you can get enhanced killing with better specificity because the tumor targeting agents for the 2 bispecifics can be 2 different targets, but both of them shared on the same tumor target cell. So now you don't amplify non-tumor toxicity, that is you're not going to increase your chances of killing off cells that you don't want to kill off in your gut or in your skin or anywhere else, but you are going to enhance killing specifically just of the tumor cells, which are the only cells that have, both of them, the targets on them simultaneously. Now the way we think about this is that we pair every -- so when we have the CD3 class, we have a whole pipeline, actually 3 of them are in the clinic now. And we have a whole additional pipeline targeted to different tumors based on tumor antigens that are only on those tumors, i.e., let's say, on certain classes of lymphomas or in certain classes of plasma cell tumors like myeloma or on certain classes, for example, of perineal tumors like ovarian cancers and so forth. So the CD3 targeting agent bispecific targets you there. We pair each one of them with a second targeted co-stimulatory bispecific that's targeted to a different tumor antigen but shared on the same tumor cell. And we try to pair them so that these antigens are not expressed on any normal cells simultaneously. They're only expressed simultaneously on tumor cells. So we get enhanced targeting and enhanced activity, but we pair them based on the tumor target. So for lymphomas, we have our CD3 bispecifics and then we pair them with specific CD28 co-stimulatory bispecifics that are also targeting the same tumor cell. The same thing for the myeloma-like tumors. We pair the sets of bispecifics together. The same thing with the ones that are targeting, for example, ovarian cancer or lung cancer or other cancers. So it's a very targeted approach. It's not sort of an endless mix and match where everything could be matched with everything else. They come in these directed pairs with a specific CD3 bispecific against a specific cancer and a second co-stimulatory bispecific that also targets via different tumor antigen the same cancer. So it's a very logical targeted approach. It's predefined. And so we have this whole pipeline that we're now rolling out, where we start with the CD3 bispecifics against these specific cancers. And then we add on the co-stim to each one of them in a paired fashion. Now we do also think that in a relatively nontargeted way as we've shown in animal models, some of these cancers will also benefit from PD-1 inhibition from combining with our anti-PD-1 antibody, Libtayo. So for example, we've shown that, for example, prostate tumors, which on their own, do not respond very well, if at all, to PD-1 inhibition. When you add a PD-1 blocker to either of these classes of bispecifics, you get specific enhancement of T cell killing of these tumors. So it's all in a very matrix but logical way, which really directs us as to what to use. And so we really think that the right combinations could really provide the sort of enhanced immunotherapy effects that we were all hoping for by combining checkpoint inhibitors, we think we can get there now. But instead of just combining random checkpoint inhibitors, we're combining the right checkpoint inhibitors with the right targeted bispecifics and we're hoping and we believe it's quite possible we're going to get to exactly where everybody was dreaming we could get to about 10 years ago, which is enhanced immunotherapy, but very specific. So you're not going to be enhancing the side effects. You're not going to be enhancing the autoimmune sort of diseases that you can induce with these therapies by using these very targeted co-approaches.

Got it. Does it always have to be a separate antigen for the CD3 and the CD28? Or can you perhaps use the same one?

You can -- we've shown that you can use the same one. You don't then benefit from focused specificity by combining -- it's sort of like having a latitude and a longitude, you get to a single point. Whereas if you focus on both on that same longitude, you don't have a point, you have a lot. So you don't get the same specificity which is why we try to combine 2 different tumor antigens. Either one of them have to be absolutely specific because they intersect at only one point on the tumor cell.

Is this kind of the crux of Regeneron's oncology strategy? It's not -- because I know it's -- as we look at Libtayo, you found room to navigate in -- with Libtayo, it alone amongst the other PD-1 antibodies may not jump off the page as a better PD-1 antibody. So the same, as we've gone through all the ASH abstracts, you clearly have bispecifics that are active. Are they clearly better than some of the other bispecifics that are active? Hard to tell. But now what you're talking about is something that I don't hear anyone doing. And it makes me wonder if this is kind of all part of a -- almost a chess game of oncology that you've been putting in -- putting the pieces in place for years, and we're finally going to start seeing that these moves play out maybe in the next 2 to 4 years, and that's where Regeneron now jumps off the page versus the field. Is that kind of the way to think about it? Or am I missing something?

No, I think you summarized it perfectly. Our goal was, because we believe we have the leading state-of-the-art technologies, that we could make the best PD-1, okay? And we can make the best CD20 bispecific and the best myeloma-specific bispecific. And we could create the first co-stimulatory bispecifics. We could create the best-in-class, but by themselves, as you said, they may not stand out enough to really entirely take over different fields in different areas, especially in cases where they may not be -- such as with Libtayo in lung, we have some very impressive lung cancer, but we may not be first-in-class there. But it's exactly as you put it. If we have best-in-class competitive reagents in all of these classes, but we have positioned ourselves to be one of the few people who can now combine all these. And this is, as you said, exactly what we saw from the beginning, that solutions weren't going to necessarily be there. We weren't going to have cancer cured by 2020, like a lot of people thought. And we were going to need to have exactly these sort of innovative combination opportunities, and that's what we've been working so hard to set ourselves up, to have these components where each individually can be argued that it's maybe best-in-class. But putting them together, particularly in ways that you can maintain specificity and get tolerable safety profiles could be the key to really unlocking the potential of immunotherapy in cancer. So I think you put it very well. That's been our strategy. It's been our dream. And the individual reagents are proving themselves. And we really hope that putting the combinations together in intelligent, targeted ways could really take anti-cancer therapy, immunotherapy to the next level.

Got it. And so then sticking on the CD28 bispecifics, maybe give us a sense of what we should be looking for. I guess PSMA is one of the leading indications. When are we going to see data? What is it that you're looking for to know that you're on the right track?

Right. So of course, what we're hoping -- and the PSMA by CD28 is a great example. That's being combined with Libtayo. We also have coming down the pike, CD3 bispecifics to combine with it. But in the ongoing trial with Libtayo, as you know, PD-1s do not have profound efficacy or activity against prostate cancer. We are hoping to unleash profound anticancer activity. So we are hoping to see responses and durable responses in settings where normally with PD-1 alone you don't see that. So that's what we're hoping to see. And similarly, in our lymphoma program, where we already have some of the first-in-class and best-in-class results, we are adding on to that CD3 bispecific and CD28 bispecific. And we're going to hope to now show that in either patients who don't respond to the CD3 bispecific or who don't respond long enough, that we're going to now get response where you don't see it and/or enhanced durability of responses. Same thing in our myeloma program. We already have a CD3 bispecific. We're going to be adding a complementary co-stimulatory bispecific there to enhance responsiveness in those patients who don't respond to the CD3 alone or extend the duration of response. Same thing with our ovarian cancer one. We're going to be adding a co-stimulatory bispecific there. So each one of these programs, we're hoping to see responses in the individuals who don't respond to the individual agent on its own as well as enhanced durability of the responses. So I think that these are going to be efficient programs in terms of getting that information. And I really think that we could be seeing combination effects that could really unlock -- I don't know if we've announced it publicly, but we have, for example, we've already released our PD-1 lung data, which is as good as any data that's out there and for a PD-1 agent, but we have a bispecific that we're going to be targeting in lung as well, allowing us to maybe take it to that next level. And all of these things are working in animal models. And so if any of them work in the human setting, they could be real game changers.

Got it. And which of these are we going to see first half next year in terms of read-out versus later?

Justin, maybe I don't know what we've announced publicly, maybe I'll turn that over to you in terms of timings.

Yes. So I think when it comes to PSMA, we're still in dose escalation. We've been through several dose-escalation cohorts at this point. We're moving fairly timidly with this target initially, given the history of CD28, but we're hopeful that there may be some insights into data next year at some point, not quite a guarantee yet. But that's the leading on the co-stim side. The other targets that George was mentioning in lung, in particular, is EGFR by CD28, that is expressed across multiple other tumors as well. That's just now entering the clinic. And then we have the MUC16 co-stim that we're pairing with the CD3 that George referred to as well.

That's super helpful, because I -- it gives us kind of a new set of milestones to watch. Maybe just kind of quickly, beyond CD28, I'm guessing you have a whole bunch of tools in your toolkit that you're thinking. What -- CD28 seems to be one that's really risen to the top in terms of strategy? What -- are there a couple behind it that you're keen to explore?

Yes, we tested essentially all potential co-stimulatory pathways. Based on their activities in preclinical models, we've chosen to highlight and move forward with the CD28 co-stimulatory pathways. But we have versions of co-stimulatory molecules targeting other pathways as well, which right now, we're considering sort of as backups based on the benefit that we see with the CD28 pathway in our animal model data. We also have other classes of bispecifics as well that do different things. So we have classes of bispecifics that, for example, increase target destruction. So they can be used essentially to turn down pathways, whether that be in oncology or in other settings. And so we have many, many classes of bispecifics that can do different things, both in oncology and outside of oncology. And it's really a -- for us because we have optimized the generation of these and the manufacturing of these. So essentially, for us, it's like making regular antibodies, whereas I think for the rest of the world, it's still much more challenging that we are considering bispecifics as a whole new platform for targeting all sorts of things. We're using bispecifics to do targeting, for example, to target gene therapy. We're using bispecifics to target siRNAs. So there's all sorts of future combinations that we're combining all of our technology. And I think that, that's part of our vision and part of the reasons we've established some of these great collaborations we have with companies that we see equivalent to sort of younger, earlier versions of ourselves, whether I'm talking about an Alnylam or an Intellia or companies like that, or a bluebird, where we think that combining our sets of technologies, whether it be anything in the biologics field, whether they would be conventional antibodies, whether it be these bispecifics or either other forms of biologics that we're creating right now and combining them with these other platforms from our collaborators, such as how I described for C5 with the Alnylam siRNA. But in all of these other fields, we can be combining our bispecifics with a T cell therapeutic. We can be combining our bispecifics with a way to target to a specific cell type, an siRNA or a gene therapy approach and so forth. So I think that we really are pushing the envelope, and we're positioning ourselves to be inventing the new modalities of therapy in the future. And what we're seeing now, we're testing the great ideas in the clinic now that we had 5 to 10 years ago. We are now inventing the technology that I think that you'll be seeing in the clinic rolling out over the next few years. And I think that, that's what I'm proudest about in terms of Regeneron and where we stand, which is that when we started, people thought we were a great technology company, great innovators, we didn't know how to develop drugs and get them across the finish line. Well, we've proven that we can get a lot of important game-changing drugs, blockbusters across the finish line, but we haven't lost our edge. We're continuing to push the edge. And I think that our capabilities with our genetics, databases, with our Regeneron Genetics Center, our ability to look at now millions of humans, we sequenced over 1.5 million people linked to the electronic data, that information we utilize every day in making our decisions. These things are allowing us to keep pushing the edge further and further forward, and I think that combining that with all of the treatment modalities that we're working on, which are going to create the next generation of great therapeutics is -- I couldn't be more excited, obviously, as you can probably tell, about not only what we have going on right now, what's going into the clinic now, what we're going to be seeing over the next few years, but in the years following that, the new classes of modalities that we're going to be putting in interesting combinations going forward.

Maybe in the last few minutes, we can pivot to the COVID and the monoclonal antibody mix for Regeneron. Given the evolving data, who's the right patient to be treated with the Regeneron cocktail aside from the President of the United States, which is quite impressive?

Well, I think that the -- the EUA right now targets the right patients based on the data that we have right now. So basically, we are doing studies in 3 populations: people who are at high risk of getting infected but not yet infected, people who are in the early stages of infection with mild-to-moderate disease with high risk of getting much sicker but have not gotten to that point and then to the people who are hospitalized who are suffering from more advanced disease. So these are the 3 separate sets of studies. One is prophylaxis for high risk, one is outpatients in earlier mild-to-moderate disease and one is in hospitalized patients in later stages of disease. The first data came out from the middle set of studies, the ones in the outpatients. And what we showed there was that particularly in people who we identified as having the highest risk of proceeding to more serious disease and requiring further medical attention, we could dramatically help those patients by rapidly reducing their viral load and reducing their need for requiring future medical attention. And we did this. We're the only people who have now done that in a prospective manner. People probably don't realize it. But for example, remdesivir has not yet been able to demonstrate any definitive antiviral effect, okay, prospectively or nonprospectively. We've done it all prospectively now in these patients. And I think that this represents an important collection of patients. As we all know, the numbers are now staggeringly up to about 100,000 new people a day are getting infected. If you have a high risk, then -- if you're in the high-risk category, then you have a significant chance somewhere in the orders of 5% to 10% of requiring further medical attention at that point, including hospitalization. And what we've shown is that those are exactly the right patients that you can treat and dramatically lower the virus and reduce their need for further medical attention. That's where the EUA is currently targeted. We are hoping in the near future to also show that we can protect people who are at very high risk. Now as you know, we're all desperately hoping and waiting for the vaccine to be widely available and to really prevent disease in a lot of people. But for example, you can't give that and protect somebody today, you have to wait on the order of a month or so for the person to get protection. So God forbid, you are in a high exposure setting now, you're a health care worker or you're in a nursing home or there's an outbreak or in some other setting, where you might need to protect people who are all close contacts, you can't do that with a traditional vaccine. But if our prophylaxis study works out as we are hoping it could possibly work out, then you could protect those people, that will be a second group. And finally, the third group are the hospitalized patients. We hope to, once again, over the course of the next month or so, be getting our first data on hospitalized patients. And there, again, there's different classes of hospitalized patients, 4 different cohorts that we've identified and the world has identified that we're studying it in. And once again, we're hoping that at least in some of those stages of hospitalization, we might actually be able to benefit patients. So right now, it's patients who are very much like President Trump was, meaning that they are in their early stages of infection, as was publicly reported with high viral loads and without their own antibody response, but at high risk due to all their risk factors, those are exactly the sort of patients that in our clinical trials we were able to show the most -- the strongest evidence for consistent benefit. And that's who the EUA is indicated for now, and we're hoping to extend the treatment population, as I said, backwards to the prophylactic setting, forward to the hospitalized setting, pending data coming from our ongoing clinical trials. And of course, we're all hoping that sooner rather than later, there's going to be broader protection that is going to decrease the need for these approaches. But for sure, for the foreseeable future, there is crying, desperate need in all of these settings. And right now, as I said, we're indicated as is the Lilly antibody for that middle setting.

Excellent. I think we're at the end of the time, George and Justin. Thanks so much for joining us and giving us a little glimpse of the future, it feels like, what Regeneron has cooking, looking forward to the next updates over 2021 and well beyond.

Yes. Well, that's what we're thinking in 5 to 10-year time frames here at Regeneron, and we couldn't be more excited about the future.

Yes, sounds like it, excellent.

All right, thanks, Josh.

Thanks, Josh. Thanks, George.