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

Ladies and gentlemen, thank you for standing by, and welcome to the virtual fireside chat with Regeneron Conference Call. [Operator Instructions] Please be advised that today's conference is being recorded. [Operator Instructions]. I would like to hand the conference over to your speaker today, Evan Seigerman. Please go ahead.

Thank you, operator. Thanks for joining us this morning. It's my pleasure to have with us on the phone Regeneron. And from Regeneron, we have Bob Landry, CFO; Israel Lowy, SVP Translational Science and Oncology; Neil Stahl, EVP of R&D, currently focused on the COVID-19 program, Justin Holko from the IR team. And just before we get started, I want to read a quick disclosure statement for everyone on the call. By accepting this invitation and dialing into this event you consent to having your contact information being provided to the corporate presenters so they may better understand their audience and ensure the content of their presentation is appropriate and relevant. If you do not agree with that, please disconnect now and send me an e-mail. And of course, [ submit the 2 ] notice, this event is only open to CS clients who are in scope for MiFID II and pay for interaction. And of course, those who are not in scope for MiFID II, it's your responsibility to ensure you are entitled to attend and not receive an inducement. Again, if you do not meet this, please disconnect and let me know. Now that that's out of the way. Thank you, everyone from Regeneron for joining us. Are you guys on the phone?

Yes.

We're here, Evan.

Excellent. So with that, we have a lot to get through. I want to hand it over to Bob just for a quick introductions and forward-looking statements. And then I'll speak to Izzy, Neil and Justin, I'm sure will chime in, but I'm very much looking forward to the conversation with everyone. Bob?

Sure. Thanks, Evan, and good morning, good afternoon, everybody. So our statements today, we will be making forward-looking statements that will involve risks and uncertainties relating to future events and the future performance of Regeneron, and actual events or results may differ materially from these forward-looking statements. As always, I encourage everyone to go to our first quarter 10-Q or our year-end 10-K to find the risks and uncertainties that are outlined within those documents. Back to you, Evan.

Perfect. That was a quick introduction. So I want to start off with Izzy, who runs -- essentially is the SVP of Translational Science and Oncology, responsible for Libtayo, but also all the novel bispecifics and co-stimulatory assets that have been introduced recently. So I want you to kind of frame the oncology program. Tell us where you are and kind of what the overall strategy is?

Sure. Thank you very much, Evan, and good morning to everyone who's listening, and thank you for getting up early to listen. So I'm pleased to say that I've been now at Regeneron for just a little over 10 years. And after years of sort of sowing the crops, we're really excited about now starting to reap our harvest. I -- before Regeneron, I was at a company called Medarex, which pioneered some of the immuno-oncology revolution, where I led studies in anti-CTLA-4 and anti-PD-1, which was then acquired by BMS. And then I came to join George and Len and all the research colleagues after the BMS acquisition because of the scientific foundation that drive the company that was sort of like Medarex on steroids. And basically recognizing at that time from the beginning that the key to really building a strong immuno-oncology portfolio lies in combinations. And we began to sort of build a whole effort in wide range of models and looking at various targets and concepts and recognize first that an important component of any credible immuno-oncology portfolio would be a PD-1 inhibitor, not necessarily because it's going to be better than the other ones out there, but to provide a foundational element that can combine with other agents. And on top of that, we also sought to look for what would be appropriate combinations. So we looked at targets that fall into the regular sort of checkpoint inhibitor class, the TNF receptor agonist class. But we also, as Regeneron has always done, innovated with a new technology and started developing bispecifics in a way that I think really stand out from the rest of the field. And it stands out because the bispecific antibodies that have been created are very close, essentially indistinguishable from normal antibodies. And the reason that's so important is because, ultimately, when you bring these things into the clinic, a -- it's easy to sort of make all these Rube Goldberg type of combinations of this binding region in that binding region. But at the end of the day, you need to turn it into a drug that can be administered to patients that behave with appropriate PK, that have appropriate stability and ability to manufacture. So being able to build a bispecific program on the antibody framework meant right away that in the long term, it would have many attractive features that we think will play out in the long term. We also developed, and by we I'm taking credit for -- or I'm talking about the phenomenal research that has gone on for 30 years. Building the VelociGene mouse that's resulted in VelocImmune and now Veloci-Bi, which is basically an altered version of the velocity in mouse that makes making Veloca -- and making bispecific antibodies, very [ facile ]. And there are 2 flavors of bispecifics that have been generated. One targeting CD3 and one targeting CD28 and called the costim bispecific. So let me get back a little bit more into where we are with our PD-1 inhibitor, Libtayo. So from the beginning, we chose to divest target anti-PD-1 rather than anti-PD-L1. We put our candidates through a very rigorous testing program to feel confident that we were coming in with our best foot forward with a really potent and promising agent. And we were able to actually, despite sort of coming in a bit later than everybody else, we were able to identify 2 different indications that had been missed by everyone else in the non-melanoma skin cancer space. The first cutaneous squamous cell cancer, advanced CSCC, which we have our first approval, which we achieved in essentially record time, a little bit over 3 years from first, the patient dose in our first in-human study to getting an approval. And more recently, we announced very promising results in the second line, basal cell carcinoma study for patients who have now progressed or intolerant to the Hedgehog inhibitors. And at that point, really have no approved options. And so that's where we want to start, with competing in areas that haven't been the -- sort of the open space. But we've also gone and weighed it into the big market of lung cancer and have been very pleased recently to announce our ability to stop our pivotal study in patients in first-line lung cancer with PD-L1 positive tumors greater than 50% by the independent data monitoring committee, who basically after only 50% of the planned events had occurred, said, look, you guys have won, go forward. So we have stopped the randomization in that study. All patients are able to get Libtayo. And we've met now with the FDA for both of these programs, the BCC and the lung cancer. And we're working hard towards filing our applications for these this year. So the basic core of the approach has been, let's get our PD-1 approved. Let's show that it's capable of performing as well as any other PD-1 out there. And I think everyone knows that particularly in this monotherapy setting in first-line lung cancer, the field -- despite all the other antibodies out there, the field doesn't have many people, many [ tarks ] and many agents in there. We're only the second anti-PD-1 to announce a monotherapy positive study. And we're looking forward also to our chemo combination study for all patients with any PD-L1 positivity of -- where we basically should be finishing enrollment shortly and are expecting perhaps some initial results in 2021. So...

Sorry, I'll let you finish. So I wanted to just stop you right there and kind of focus on the lung cancer data that you just referenced. Specifically, the strong pipeline data you presented, I believe, back in April. Then you had a modified intent-to-treat update on the coattails of ASCO. Can you explain to us what the difference between these 2 hazard ratios, the first one being 0.676. The second being an impressive 0.566 in that front line monotherapy lung setting, kind of what are the differences? And how does this shape what you want to do with Libtayo in non-small cell lung cancer?

Well, of course, we were very thrilled with the initial observation of the intent-to-treat analysis. But when we started the study, we were not -- we were blocked from being able to use the PD-L1 assay in the fashion one needs to use it for a pivotal study under investigational use settings. And until we were able to sort that out, we had to use a somewhat less reliable testing system. And therefore, the FDA asked us and from the outset said, well, look, we want you to check your data with the tests on any patient that had the test done per -- up to snuff. So when we did that, what we found was that of the original 710 was that we analyzed for an intent-to-treat, there are about 560 that had either initially done the test up to snuff or upon retesting, we found that they actually truly were PD-L1 positive. So by definition, then some of the patients turned out to not test positive when we retested them. And so in that, what we call now the modified intent-to-treat analysis, which is the intended population that we wanted to study, namely the patients with PD-L1 greater than 50%. We did that analysis and found that it was an even stronger hazard ratio. And we've talked to the FDA about this. We're hopeful that this analysis will be part of the label. And what it does tell to us, I mean, even our initial intent-to-treat analysis was in line with the best data that are out there. And of course, the modified intent-to-treat analysis is even stronger. So what this tells us in this study that was quite large as well, that we have a good antibody, that we have something now that physicians, hopefully, if we get this approved, they will have a reasonable choice. And what we've heard from all the physicians that we've spoken to and when we started to share some of our information about the positive outcome, they are excited to have choice. So we think this is basically what it needs. I hope I've answered your question.

No, that's very helpful. And along those lines, where do you believe Libtayo can play a role in treating patients with non-small cell lung cancer? You had mentioned that this is really only the second PD-1 to have a positive monotherapy trial. How does that data set shape where physicians could potentially use this asset?

Well, if we get the approval as we hope, then patients that are coming in, in the first-line setting who have a PD-L1 test that shows that they're greater than 50%, and the physicians are willing and think that it's appropriate for these patients to start on a monotherapy agent and delay or avoid chemotherapy, we hope that Libtayo will be an option. And as you know, that's a very large market, a very large patient population. And so we're looking forward to many physicians getting experience with Libtayo and seeing how well it performs. And as I also said, we have a study ongoing to test the combination in both non-squamous and in squamous non-small cell lung cancer in the first-line setting, regardless of PD-L1 positivity in combination with chemotherapy versus chemotherapy. And so we hope if that study is positive also, and obviously, if the monotherapy study was positive, we're cautiously optimistic that, that too should become positive. That we will have for any patients coming in, in first-line lung cancer, they'll have a choice.

Excellent. And then just shifting gears slightly. One of the themes from your ASCO call was the importance of combinations. Can you at a high level talk about how Libtayo could be the backbone for many of these combinations? And what indications do you think would be most suited to a combination approach? And I know that's a very broad question. We can talk about the bispecifics in a moment. But on a high level, you've kind of introduced the topic for us.

So on a high level, any tumor, and would you think there's any opportunity to consider a immunotherapy approach, or bringing to bear a T cell-mediated and maybe in the future, other arms of the immune systems, it's fair game. Because one of the things that happens is what we've learned is that the immune system plays in a very important role in managing the emergence of potential tumors throughout life. And the fact that tumors become clinically apparent basically, typically occur because the tumor has outstocked the immune system. What we've learned is one key mechanism that they do this by is by using the checkpoint blockade system inappropriately, if you will, to basically turn off an immune system that probably protects us for many years from tumors actually ever emerging to be clinically apparent. On top of which, if there are tumors that don't have antigens, either neoantigens or molecules that are being expressed in the wrong place and the wrong context that the immune system is able to identify as something fishy going on here. Then we can think of artificial ways, "artificial" using the bispecifics to really bring T cells to bear on these with other approaches. What we will talk more about the bispecifics. Now in terms of combination agents, we view this very broadly. We view this. We've learned that chemotherapy actually can combine with checkpoint blockade. I think people at the very beginning were concerned whether that would work or not. It does seem to work. There are other molecules in the checkpoint family or in the TNF receptor family. So we have LAG3 program underway in combination with PD-1. We announced that we'll be bringing our own GITR antibody into the clinic this year. The IND for that has been allowed. But we hope to start that soon. Obviously, things get a little slow with COVID but hopefully, that will move forward. And then we have the whole combinations with the bispecifics. But before we talk about the bispecifics, I'd also just add that one of the unanticipated, to me anyway, surprises and benefits us developing such a strong antibody has been that many companies have come to us with seeking to partner with us. And it's put us in a position to really look at companies that have complementary technologies that are not in our wheelhouse. But where we look at companies to make sure they're like-minded scientifically and they have compelling platforms as well as lead products. And so we've partnered now with several different companies through making vaccines, oncolytic viruses, techniques to deliver cytokines. And we are continuously looking and evaluating opportunities in that way as well. So that was sort of an unexpected benefit that we had as another way of going forward. But our main excitement lies around our bispecific combination platform. And I can go into that if you'd like, or...

Perfect. Well, along those lines, can you kind of walk us through your bispecific platform? You've mentioned a lot about the costimulatory bispecifics. How do those bispecifics differ from CD3 targeting bispecifics? And kind of what is the potential incremental benefit that you could see in these patients?

Sure. So when T cells are triggered and activated to learn how to respond to specific antigens and generate and expand a T cell response that proliferates, expands and is activated to actually kill, target infected cells, which is what we think was all originally man for, but also tumors. It requires several signals and requires several signals, I think, that many people think it's part of sort of the balance so that you don't get overactive immune responses in autoimmune disease. So it requires a Signal 1, which is the T-cell receptor seeing an abnormal antigen in the context of the HLA molecules. But that signal, one by itself, if it is not sufficient to really activate a T cell. In fact, if you initially expose a T cell like that to Signal 1 alone, it can actually turn the T cell off, and we think that's partly how we learn not to attack our own tissues. So there's Signal 2, and Signal 2 is mediated by any one of a family of what are called co-stimulatory molecules, but the most important one on the T cell surface is CD28. And it binds with molecules that of the B7 family, that are typically shown -- presented on antigen presenting cells. So dendritic cells, macrophages, even b-cells, these are the cells that sort of speed up antigen, present it and educate and teach T cells to expand and get fully activated. And then after that, there's the class of Signal 3, which can be cytokines that are also produced, IL-2, IL-12, IL-15, et cetera, that drive further the expansion and activation. So what we've learned with the CAR-T cells, in particular, was that if you change -- if you have a T cell that doesn't know how to recognize a tumor and you give it something to substitute for a T cell receptor, the chimeric antibody receptor, the CAR, you can redirect T cells to kill tumors. And that's been successful. We made CD3 bispecifics because we didn't think it's great, the CAR-T cells are great, and we've partnered with some CAR-T cells. But in many events, many settings that probably doesn't need to be that complicated. So we've made antibodies that are off the shelf, able to bind to the T-cell receptor, the CD3 component and bring it to bear on a tumor that has a particular antigen of CD20xCD3 or BCMAxCD3 in myeloma and -- on MUC16xCD3 and -- et cetera, and more to come. But what has been noticed in the field is that these CAR-Ts and bispecifics so far have been active and successful and more in hematologic malignancies than in solid tumors. And one possible explanation and I'm sure there are others, is that the hematologic malignancies share their developmental history with the same types of cells of b-cells and myelodysplastic cells, they share the same kind of lineage as authentic antigen presenting cells. And they typically express some of these B7 molecules. So it's possible that the bispecifics of CAR-T cells that target the hematologic malignancies are also getting the costimulatory signals. In contrast, solid tumors, which are generally epithelial or mesenchymal, like sarcomas or breast cancer, lung cancer, in melanoma. They don't make B7 on their surface. And so the only way for the B7 signal is if there are dendritic cells or macrophages in the tumor and the tumor figures out ways to keep them out or to not make them work too well. So what the costim approach is, is that we say, "Oh, let's make a bispecific that binds to a tumor target and also binds to CD28. And therefore, turn an epithelial cell into a essentially pseudo-B7 positive cell that is like an antigen presenting cell and can deliver Signal 2. And the cool thing about this is that the way we've worked on these targets is that they seem to not work that well -- they are not that active by themselves. And in fact, if you deliver signal 2 alone in general, it doesn't do anything. There was in the past a very unfortunate experience with the TeGenero antibody -- the CD28, that actually ended up being broadly stimulating for a variety of reasons and almost killed a few normal volunteers, so people shied away from this. But we recognize that CD28 is key. And we made bispecifics that only activate and only transduce the CD28 activation signal in the context of the tumor. So as a consequence, what we found is that these costim bispecifics synergize beautifully with either CD3 bispecifics, making solid tumors now accessible to these T cells that are redirecting and they also synergize with anti-PD-1. So if you have a T cell that does recognize a tumor but has been inhibited by the PD-1 pathway, if you have PD-1 there and it's still not enough, we can augment that by using the costim bispecific. And so these are 2 pairwise combinations. There's also the CD3 and PD-1 pairwise combination. And in the future, we could entertain all 3 if it's required. So this is the kind of plug-and-play combinatorial opportunities we have that we are really excited about.

So I want to focus a little bit on the MUC16xCD3 bispecific in ovarian. And can you comment about why MUC16 in ovarian? And how do you get that potential utility of a CD3 in solid tumors, given that we've seen it usually work better in liquid tumors?

So MUC16 is the cell surface molecule that when it's clipped, gives rise the CA125 in the blood, which is a typical marker that is used to follow ovarian cancer. And so it's very broadly expressed on ovarian cancer. It's also expressed on a number of other tissues as well as some other tumors like some pancreatic cancers. So we've directed our bispecifics to target the region of MUC16 that's on the cell surface even after clipping off the CA125 and linking it to CD3. We're in dose escalation now of both monotherapy and in combination with PD-1. And we -- so far, so good. We're seeing -- we're able to deliver this, and it's been well tolerated. And we hope maybe in the coming year in 2021, maybe we'll have some information to share. But on top of that, we've just received permission to go forward with a MUC16 costim as well. So MUC16 targeting CD28 -- a bispecific with CD28. So this was an example where if the MUC16xCD3 by itself as monotherapy is insufficient to really mount a meaningful anticlinal antitumor response, we will have the ability to further augment it for whatever it's able to do. We'll be able, we think, to further augment it with either anti-PD-1, Libtayo or our MUC16 costim. So this will be an example of the ability to bring this to bear in one solid tumor. We already have our first costim in the clinic. It's a PSMAxCD28 costim for prostate cancer. And that study is underway in combination with Libtayo, and that dose escalation is proceeding. And we hope also next year to have some initial clinical data from that. And we have -- we should be advancing more costims this year, in both actually solid as hopefully liquid tumors as well.

And along those lines, your 2 kind of most -- assets most furthest along are REGN1979 and REGN5458. Can you -- for 5458, this is the BCMA targeting bispecific. That's a pretty crowded space. How does this asset differ from the whole host of other drugs being developed, targeting BCMA?

So I think the BCMA field, right now, there are BCMA CAR-Ts, there are BCMA CD3s and BCMA ADCs. We've been very pleased with the data that we've had to date with our bispecific CD3xBCMA. We've seen right from our first cohort some of the first patients we've dosed a meaningful response in a patient who has all the patients in our studies have failed all 3 classes of agents: [ iMabs ]; proteasome inhibitors and an anti-CD38, typically multiple different combinations of those. And one of our first responders that we described at ASH last year actually had extramedullary disease, which is an even tougher setting and this patient has actually been -- remained on treatments and is having embraced response. So it's -- we recognize it's competitive. That's good for patients to have options as usual. But we think we have a very strong agent and I think what we'll see as we go forward is that we will be looking to also explore combinations. So if, again, I think our BCMAxCD3 will be competitive. We are rapidly expanding our dose escalation. Hopefully, we'll be initiating additional trials that will have registrations in temp. And so stay tuned for that. I mean, it's all I can say on that for now.

And before I -- I want -- if Justin, just comment on the opportunity back for Libtayo and basal cell carcinoma one of your -- the indications that you're pursuing. Any final comments on the oncology program? I know we could talk for hours and hours and hours because of a variety of combinations you have. But anything that you think people need to fully understand better or you want them to internalize when it comes to your work on oncology?

So, we are happy to, Evan.

Basal cell in particular or in general?

Well, I'm just talking about, in general, on kind of what we've been talking about, then I want to pass it over to Justin just to comment on kind of some of the commercial opportunities.

Perfect. So we basically have come forward with trying to bring forward a deck of cards, basically where we can have multiple winning hands. And that we can -- we have -- we know the agents. We vet them internally. We have great technology underlying them. All of the antibodies are simply not commodities. I mean people sometimes think of them, oh, it's easy to make an antibody. Well, yes, it's easy to make an antibody, but it's pretty hard to make a very good antibody. And we make very good antibodies. And we have really, I think, been able to now develop a tripod, at least of foundational technologies, checkpoints, CD3 bispecifics, CD28 co-stim bispecifics, and there are more in development looking for other ways to engage the immune system, to bring it to bear on tumors. And so we think we have this combinatorial flexibility that as we go forward, we think we will have really winning combinations to bring forward to treat cancers.

Excellent. And then Justin, so from the commercial perspective, I know you're making a lot of progress with Libtayo. Can you talk about how -- what you think about the market opportunity for Libtayo in basal cell carcinoma, following the data you toplined a couple of weeks ago? And then also talk about where you think Libtayo can compete in the long setting?

Sure. Happy to, Evan. And I'll try to do so quickly given that we have Neil and Bob on the call as well. I think the first thing to add is just that basal cell carcinoma is on the order of a few thousand patients in that second-line setting for patients who have failed a Hedgehog inhibitor. And the data is really remarkable in that, not only do you see response rates in 20% to 30% range, but just the high proportion, 85% or so who are maintaining a response after 12 months is just -- it's staggering as you think about just the opportunity that patients can have post a Hedgehog inhibitor. But I think if I were to leave you with just a couple of thoughts here on more broadly, commercially speaking here, but I think one is that, as you heard from Izzy, we are very much focused on the long-term here. We have a tremendous pipeline that is going to offer us flexibility to get after cancer in ways that a lot of other companies have not been able to, to date and really believe that our pipeline positions us very uniquely in that regard to help more patients, both in cancers that respond to PD-1 blockade, but also in patients and in cancers that don't necessarily respond. So that's very much the long view here. And as you heard from Izzy, we're beginning to reap some of those benefits. But I think there's always the focus on the short-term and to just round out your question on how do we think Libtayo compete in the short term. I think it's just important to make sure everyone is clear that physicians, by and large, are fairly heterogeneous. They make decisions that are appropriate for their specific patients and the type of disease that they see or how they believe certain drugs will work in certain patients. And so there's definitely a personalization aspect to treatment that I don't think can be understated. And so when you can bring a new drug into a large market like lung cancer, and offer physicians a choice for a monotherapy that they really have not had to date, that is going to be meaningful, particularly when you can bring along with it a very strong overall survival benefit. And so while we do have a long-term view, we do believe that we can be competitive on the merits of the strength of the data and what we see in terms of being a very strong drug within lung cancer and some of the other skin cancers that we've shown to date.

Excellent. With that, I want to turn the program to Neil. Neil, are you there?

Yes, I'm here.

Perfect. So I don't believe people have met you or have been introduced to you much on the phone. Can you give us a little bit of your background? And kind of talk about what you're doing right now with the COVID-19 program? And then I want to just jump right into questions.

Sure. I've been at Regeneron for -- this is my 29th year there. So I started when it was quite small. And I started in the discovery research area and did a lot of work on the mechanism of cytokine receptor signaling. And I actually was the inventor of our Trap technology, which turned out to be the basis of our first 2 approved drugs, before we developed the capability to make fully human monoclonal antibodies. I also built the entire preclinical group that has responsibility for making the antibodies and making cell lines out of the antibodies, doing the process development that's required to scale them up and give them to our industrial operations group, our IOPS group, the PK Group, the formulations group, the toxicology group. And I have to say that I think that we put together the best operation end-to-end of anybody in the business. And at Regeneron, we've always focused on, as Justin mentioned, the long term, but also on two things: very strong and deep basic research, coupled with technology development that allows us to innovate around bottlenecks in the whole process of discovering a new drug and getting it rapidly to patients. And I think the perfect example of that is our efforts on these emerging infectious diseases. So we've made fully human antibodies with our VelocImmune platform, which is our mouse that was engineered using VelociGene and it's the largest genetic engineering project in the history of the world, where we humanized 6 million bases of the mouse genome to allow them to make fully human antibodies. And they have a wild-type immune response and it's our go-to platform that we've made every single fully human antibody that we put into the clinic up till now. So for Ebola, we did the entire project of starting the project and isolating thousands of antibodies that bound to the external protein, the glycoprotein of the Ebola virus. And we ended up picking 3 very strong antibodies with complementary mechanisms that bond independently to the virus. So they could all 3 bind at the same time, which, as I'll get to is a really important feature. And so we did that whole project in 6 months. And then when the coronavirus outbreak happened, the viral sequence of the genome was actually published on June -- sorry, January 12. And we started the project that day. And within 5 months, we had screened through thousands of fully human antibodies and isolated hundreds of very strong neutralizers, both from our VI mice as well as from convalescing humans. So we had antibodies that we got out of people that had COVID. And so we were able to look at both modalities and pick the best antibodies from each to combine together as a cocktail, so to speak. And so the spike protein for coronavirus is a little smaller than the glycoprotein for Ebola. And we ended up with 2 complementary antibodies that bind independently to separate places on the virus. They both independently neutralize and together, they will hopefully prevent virus' escape, that I can talk more about.

Perfect. So why don't you -- where are -- where is Regeneron currently with the development of the antibody program? I know that there was a recent announcement of the initiation of clinical trials. But where are we? And when can we see initial clinical data from this program?

Right. So we announced that we started trials a couple of weeks ago now. And that is in the setting of patients that already have COVID that are either hospitalized with -- because of their severe symptomology or their ambulatory and their symptomology, although they're positive, is not bad enough to make them be enrolled in the emergency room. And so we have both of those settings right now. And at least for the patients that are in that setting, I think we're going to start getting data that's meaningful in the next few weeks, and we should have data, I think, this summer, for sure, because there is no shortage of patients in the hospital still or with COVID positive that are ambulatory, telling us that the antibodies are safe with some initial readout on effectiveness. And we're looking both at the ability to inhibit viral shedding from positive people as well as to resolve symptomology in people.

And you had mentioned the importance of studying viral escape and kind of how the antibodies prevent that. Can you expand on that and talk about why that's so important? And what it means for this program? And how we actually get out of this pandemic?

Okay. So you know that for HIV, the experience with HIV is very clear-cut that the initial efforts against HIV were with single drugs and the virus, which is an RNA virus, rapidly mutated and was able to apply mutations that would inactivate the mechanism of the drug, but still allow the virus to be functional. And so it turned out that a combination for HIV of 3 independent drugs was the best way to suppress the ability of the virus to generate escape mutants. And the same thing is true of the coronavirus, we believe. And I believe so far, we're the only people that have done real experiments using a pseudotyped virus. So what it is, it's a VSV, vesicular stomatitis virus that you can use in the lab, that we put in the spike protein so that infects cells just like a coronavirus would. And then we're able to do very rapid experiments then to see it with a single antibody that virus can generate mutations in the spike protein that will allow it to still be functional, but escape the neutralization of the antibody. And we published in science papers a week ago today, actually, in the Journal of Science, that with a single antibody, we were able to bind escape mutants, but that with a combination of 2 antibodies that we did not find any escape mutants. And moreover, we also made separate viruses with individual mutations from escape mutants and showed that our cocktail of 2 antibodies would actually also neutralize those. And so I really think it's a really hugely important thing. Because we don't want to have treatments generate escape viruses out there that become inactive for the treatment we give them to. And so I would just implore other organizations that are looking at single antibodies to do these escape experiments and make sure their antibody is not subject to escape. But so far, we have not found any antibodies that are able to suppress escape mutations on their own.

Excellent. I want to just pause right there, and I'll come back in one moment. Operator, can you see if anyone on the line has any questions? Can you read the instructions?

[Operator Instructions]

And while we're putting together the question queue, Neil, can you talk about the difference between your primary cocktail and the backup antibody cocktail, kind of some of the differences between them?

Yes. So because we had so many antibodies available to us, we were able to make multiple cocktails. And we've actually have looked at structures of where they bind on the viral spike protein. And so we picked the ones, to start with, that we thought were a very strong cocktail with strong neutralizations. But should that one show any escape or stumble in any way or we're not happy with the efficacy, we have a backup cocktail and with more antibodies coming behind those that we could switch to that have different binding sites and would, I think, have independent properties.

And when you look towards the clinical trials, what's the bar for efficacy that you want this to be? How do you -- how quickly do you want to see symptoms resolved? Do you want to see a mortality benefit? And how do we think about what type of benefit you can measure in the prophylactic setting?

Well, for Ebola, what we saw was that -- and you may -- some of you may recall that last year in our Ebola efforts that we were involved in the study in the Democratic Republic of the Congo on patients with Ebola that was a randomized-controlled trial under incredibly challenging conditions, that these people out in a remote part of Africa where there was actually some local violence going on. And so they deserve a huge amount of effort for getting that done at all. But in that setting, they stopped the trial early because our drugs provided a -- it hit a prespecified boundary for superiority over another triple antibody cocktail, which was ZMapp. And in that study, the amazing thing was to us was that in patients that were treated within 5 days of developing symptoms, there was a 90% survival rate compared to about a 30% to 35% survival rate in untreated people. So at least in that case, we were able to get a really resounding increase in survival. So in this case, we'll have to see what happens. And it may depend on how early we get treatment. We certainly believe that the earlier you get treatment, the better. And so it may turn out to be a mix depending on how progressed the patients are when they actually get treatment. [indiscernible] I should also add, I can add one more thing, is that we did have also an antibody cocktail against MERS, which is a related coronavirus to the current CoV-2. And in animal studies, we were able to totally cure mice that were already infected with MERS and totally prevent disease in mice that were given the drug before they were infected with the virus. So at least that's one data point we have with another coronavirus.

And operator, I believe we have 1 question in the queue.

We have one question coming from the line of Diana Bilski with Viking Global.

I was wondering if you could comment on the dosing that might be required for hospitalized cases versus early patient versus prophylactic. And if you have any updates on guidance for how much -- how many doses you can provide in the fall and beyond?

Yes. We haven't really disclosed the hard details about the dosing. But certainly, we are using 2 different dose regimens in both the treatment and the prophylaxis setting, and we're using a higher dose in treatment than we are for prophylaxis. You may know that the virus -- once there's virus on board, will actually bind to antibodies and accelerate the clearance of the antibody. So in general, you have to give more for treatment. But we haven't talked about the specific doses yet. We should be able to crank out as many as a couple of hundred thousand doses a month right now for the prophylaxis type of setting. And we're trying to shift some of our manufacturing to our facility in Raheen in Ireland, so we can make more doses here in the States. And we're also looking to see if there's more capacity out there that could make more drug in anticipation that it's actually going to provide some benefit.

Excellent. And I have one more question on the COVID-19 program. And then I want to just turn the call over to Bob for the final few minutes of our conversation. This has been very helpful for all of you on the call. Can you discuss the rationale for the current adaptive clinical trial design for the program? I know that was something you highlighted in the press release. What does that mean? And how does that potentially speed up the development effort?

Well, I think the first part of the adapted design is, obviously, these antibodies have not been in people before. And so we have to get some basic safety information. I think we're all very optimistic that the antibodies will be safe. Because our experience is, and the experience in the field in general is that, if antibodies are aimed at nonendogenous targets, like this is a external viral target and it doesn't bind to anything in a human that they are very safe unless for some very rare reason, there might be some infused reaction or something like that. And so we're optimistic on safety. But we have to establish safety and get some initial data, I think, on frequent viral clearance measurements, and give us maybe a better idea of the doses we need to go into a second Phase II. And so that's the general idea of the adaptive study.

Excellent. Any final comments on the COVID program before we shift gears to Bob?

No, I think that's the top important line. So let's go to Bob.

Bob, I would never forget about you. Are you on the line?

I am. I'm here, Evan. Ready to go.

I'm ready to go. Perfect. Well, you've -- as Israel and Neil have had a very busy first half of the year, so have you. You've restructured the accounting presentation for the -- all the collaborations, you've executed a share buyback. So can you just kind of walk me through, now that you've done the share buyback, the Sanofi secondary has already occurred. What's your capacity for business development? And how are you thinking about capital allocation going forward?

Yes. Thanks, Evan. And I think Izzy gave me a good setup for this. When he began the conversation with regards to coming over from Medarex and then obviously everything that Neil built, all the precleaning stuff that Neil talked about, I mean there's no -- we see, we see no better investment than within ourselves right now. And people look at percentage of revenues and they look at our R&D number, and sure, it's higher than most, but that's for good reason. That's because we made the conscious decision that we really like the technology that we have, and we like investing in ourselves. We think there's no better return on invested capital. So despite what people saw in the last couple of months with regards to the buybacks and all that stuff. First and foremost is within R&D, within the Regeneron Shop, is where we are going to apply our capital first and foremost. And then secondly, we are seeing good opportunities with regards to BD. But we look at BD, kind of not at the M&A initials, but more as kind of tuck-in technologies. I think Izzy or Neil said like-minded science people. I mean we try to find companies that can be additive to all the platforms that Neil built, right? What do we not have that 1 plus 1 will give us 3 with regards to our antibody technology. And we haven't been afraid to do stuff. Recently, we're coming off of the heels of doing an Intellia transaction. Izzy talked about Vyriad on the oncolytic viruses. Certainly, we're really pleased with the Alnylam transaction that we did in the technology. They're kind of intracellular technology that we didn't have access to that we will now have access to, and we're going to be bringing a C5 program into the clinic very, very shortly on that end that we're very positive on. So BD will be second. We're not out there trying to get in a Tesla for accretive cash flow or fill a Phase III because we're missing this or missing that or a Phase II, really kind of early tuck-in technologies is really what our focus is. And then lastly is return to shareholders. When we see opportunities like we've seen, particularly back in November and then even this $5 billion buyback with regards to tacking it on to the secondary offering in knowing that the price that we're going to do this at was going to be at [ $510 ] or [ $515 ] or [ $505 ] handle. I mean we liked that. We like that valuation in terms of the intrinsic work we've done with regards to where the public is seeing us. So we kind of put our money where our mouth is, went out with the $5 billion buyback. And on top of it, we were able to kind of use a little leverage, which is a word that Regeneron hasn't used before with regards to its balance sheet, take advantage of the capital markets that are out for right now, we have a fully financed bridge loan that we will replace either in Bank Club loan or we'll do 5s and 10s. And we're working hard to kind of get to the rating agencies and take advantage of good market conditions.

Right. I was pleasantly surprised when I did hear the word leverage come out of your mouth with regard to the loan. So looking -- can you comment as to kind of the strategic rationale behind the $5 billion share repurchase. That also seems to be a deviation from some of the -- from the Regeneron capital allocation strategy from years ago?

It did, Evan. And I think kind of first and foremost, you need to have conviction in business fundamentals. We had nice kind of tailwinds coming off of, unfortunately, the Bayview situation. What Izzy has been able to accomplish on non-small cell lung, stopping that trial early BCC. And we really like the future prospects for our commercial and pipeline products. And therefore, we like the valuation. Certainly, it provides immediate accretion. Rarely does the CFO have an opportunity to kind of take 7% kind of shares outstanding off the table in a nice smooth transaction. So that was great. And again, it was -- it did allow for a kind of easier placement of shares by us being part of it. We're able to get these shares into a really sticky hands. So we're really kind of pleased with that transaction. And I think, Evan, the way to look at it is it was kind of a -- I hate using the word one-time because one-time becomes old when you do it the second time, but kind of a one-time opportunistic trade that we had good tailwinds to do it. The capital flows were good in the biotech business, right? IPOs were going well. We knew the secondary offering was going to get placed. We had this damn overhang that we needed to get rid of. Paul Hudson, the CEO of Sanofi, was very clear that this event was going to come upon our existing shareholders, and we just didn't want this overhang out there. So kind of a culmination of all those facts led us to our decision to do that.

Right. And then talking about the outward leverage, how do you expect to continue using leverage to invest in the business, form these early-stage partnerships? Is that going to become part of your balance sheet management strategy going forward?

Sure. I mean, certainly, my background as Treasurer of Pfizer and Wyeth, I mean, leverage is your friend if used properly and in moderation. So I'm pleased that we're going down that path. And again, we'll be good stewards of our caps and marketable securities, of our shareholders' spend going forward. So kind of rest assured on that. And the other use of the word leverage is certainly getting good operating margins and the like. And we do expect with DUPIXENT going forward, all the indications, all the geographic locations, all the different age groups that as more and more gets put out there -- again, as you know, we had an important end of May PDUFA filing that we -- our FDA acceptance with regards to AD for the younger kids. We expect to get better and better leverage as it associates to our DUPIXENT and our alliance brand P&L. So kind of 2 good uses of leverage. One, for the balance sheet that we'll have to do it in moderation. And then certainly, one for the P&L in terms of just getting better leverage as more and more sales occur, and we have the infrastructure set up from an SG&A perspective.

And we are about 2 minutes past the hour, but I have two final kind of questions for you. One, has -- have you started to see kind of the benefit of the collaboration restructuring in terms of understanding -- investors understanding the business better? And do you think that there is still things that you can do? And two, 6 months from now, are -- with kind of the business going as it is, is there anything else that you'd like to change or kind of evolve in terms of the way you manage the P&L, the balance sheet or kind of any other -- the way you look at capital allocation?

I think on the first question, when we redid our accounting and kind of recast the numbers, we were just trying to ensure that there was transparency, like people that go deep into Regeneron, they know our operation, but there's a lot of people that don't. In today's day and age, old biotech companies don't have the time and energy to be able to go really deep. So I'm happy that we changed our accounting and provided what we think is a really transparent scene going forward. But with that transparency, Evan, comes exactly to the question you're asking, I mean, what do we need to do now? We need to execute. Because we're laying out within our MD&A section of the 10-Q or the 10-K, we're telling you what our kind of margins are. You can see them within the alliance. And I'll tell you, in Q1, we did still have KEVZARA and PRALUENT. We had losses in both. So again, back to the leverage word, we hope to get better leverage on our margins going forward. And 6 months from now, it's the continuation of delivering, right? Ensuring that what we put into the clinic are fully tested and viable antibodies that are coming from Neil's Shop. I mean it shouldn't be a change of pace from anything that we're doing. We feel good in the position that we're currently in, and need to continue to execute on all fronts. All right. So why don't we stop there, Evan, I think we're over our time.

Sorry, I was on mute. I just wanted to say thank you to everyone on the call. I really appreciate it, Bob, Izzy and Neil and Justin for the time today. And if you need anything, I'm in the virtual office today. Have a great one, everyone. Bye now.

Thanks, Evan.

Thanks, Evan.

Bye-bye.

Ladies and gentlemen, that does conclude our conference for today. Thank you for your participation. You may all disconnect.