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

Okay. Good afternoon, everybody. We're just going to wait 2, 3 seconds, make sure everything is clicking in. And so good afternoon, everybody, and thank you for joining us for the Fourth Annual IO Summit that annually coincides with the Citi Conference. I'm Yaron Werber, biotech analyst here at Cowen, and it's a great pleasure to moderate the session today with Regeneron. The team here really needs no introduction. We have with us today, Izzy Lowy, who is the Senior Vice President for Translational Sciences and Oncology; and also Andres Sirulnik, who is the Senior Vice President for Translational and Clinical Sciences, Hematology, and everybody, of course, knows Justin Holko, VP of IR. So we're going to pass it on to Justin to give the quick forward-looking statements, then Izzy is going to make some opening remarks. If you want to ask any questions, just look below the webcast window. There's a little box there, you can kind of punch in your question, and we'll be happy to read it for you. So Justin, over to you.

Thanks, Yaron, and thanks for the opportunity to be with you today. There's a lot of things going on at Regeneron that get a lot of attention, but happy to talk about oncology today, is an area that we're very much excited about, both near-term and longer term. Before we begin, I would like to remind you that remarks made on today's webcast do include forward-looking statements by Regeneron, including those relating to, in particular, oncology. In this case, every forward-looking statement is subject to risks and uncertainties that could cause results to differ materially. With that said, Izzy, do you want to take us through a brief introduction?

Sure. Good afternoon, everyone, and thank you also for the opportunity to present. So just to set the stage for the actual questions that I think people -- you would like to get some details on. I'd just like to step back and just refresh the audience's understanding of where -- how we are approaching oncology. And over the last nearly 10 years, we have made the decision that this is a long-haul kind of effort that at the core of the effort is the ability to bring combination therapies into immunotherapy. And that we made a commitment to use our significant, we think, best-in-industry antibody making prowess to come up with a suite of agents that would afford us combinatorial flexibility to bring to bear potent combinations to treat tumors using the immune system. And as a foundation for that, we decided we needed our own anti-PD-1 to support this. I think what we've done with that is to demonstrate over the data that has come out over the last couple of years that we have indeed succeeded in making an antibody that stands up to anyone else in the class. And that will serve and is serving as a first foundation. And where we've moved from that is to generate now a series of new bispecific antibodies, both CD3 targeting, costimulatory targeting that affords us the opportunity to make progress on those. What we're not going to talk about today is new things that are in our lab that will, I think, afford us even more combinatory or flexibility in the years to come. And of course, one of the really nice things about our success has been we've been excited to get interest from partners that have complementary technologies that we can synergize with also. So I don't know, I am fortunate now to have a colleague, Andres Sirulnik, who is taking on the focus of the hematologic malignancies as well as nonmalignant hematology. And so our leading bispecifics are actually in that space.

Terrific. Thank you, Izzy. So maybe let me ask the first question. And you -- Regeneron's approach is really to develop full antibody bispecifics, right? So the scaffold is a full human antibody developed from the VelociSuite technology. You can produce it within your manufacturing plants. It doesn't include any artificial sequences, any linkers. So it's very different and it's not complementary, variable F fragment or the fab platforms. What -- why is that important? What does that mean for yields? What does that mean for pharmacokinetics? What does that mean for chromatography and purification and obviously, immunogenicity?

Well, to start with any high school student these days can come up with recombinant DNA technology and make various constructs and put them into a cell line and make something in a laboratory. It's a whole different bowl of wax to come and make a product that actually ends up being a robust pharmaceutical agent that can be reliably produced and delivered without complications. And so that's been our approach. Nature has had millennia to develop antibodies as a preferred way of producing immunity. And we feel as closely as we can mimic that, that we are providing agents that start out from the gate with a lot of advantages in terms of what they would have to offer. So our format differs is as close to a natural antibody as you might ever expect. It's really hard-pressed to find the difference except that it has 2 different specificities. And what this does is it not only provides a confidence about how these molecules will behave as pharmaceutical agents, they're easy to manufacture because they're basically a bolt-on to our already robust antibody platform. And we expect them to have good properties, low immunogenicity, good PK. As we go forward and develop new things, we may make alterations on it. But right now, the core of our portfolio are these antibodies, these bispecific full-length antibodies that are, in fact, antibodies.

And Izzy, so what does that mean for the ability when you purify it to get the yields that you want to get, to get to productivity on the manufacturing cell line, to be able to make sure that you're getting high yields, but you're also getting appropriate folding of the different fragments of the antibody as well?

So if -- so if you want to get a little technical, if you were putting in 2 different antibodies into a cell line, to have 2 different specificities, there are multiple different ways they could combine in a cell, and you'd have like over 16 different combinations and it's only one that's really the right one. So we solved that by making a cell line by doing our immunizations in mice that have a single light chain. So everything uses the same light chain. And we developed techniques so that we could sort the 2 different heavy chains easily. And so we basically add a single column to the normal antibody manufacturing process that helps us selectively purify the heterodimer from the 2 different homodimers. And as a consequence, it makes it like rolling off a log in terms of making the antibody. And in fact, because these generally tend, we think, are going to be even more potent because they're engagers. We don't even need to have the same kind of super high yields that, for example, we've gotten with COVID antibodies so that we can deliver higher amounts. We don't need those for effective production. So we think we have a -- we call it now Veloci-Bi as our new speed to manufacture of these agents. And we're very pleased with them and think they offer a really good approach.

Great. Let's move on and talk about you've developed what we're all watching the CD28 targeting an antigen and also CD3x an antigen, bispecifics. The whole concept is with CD28 co-stimulation with PD-1, you can also boost the TCR and CD3 dependent T cell response to drive durable antitumor responses. There's a thinking that PD-1 accumulates in the cell signups and reduces the CD28 colocalization, right, and reduces anti-tumor response. And because of that, the CD28 bispecific as a single agent does not have activity, and you really need that co-activation with the PD-1 antagonist. Can you talk a little bit about the biology, how well-validated is this? And you obviously are not a superagonist. So how do you actually get the PD-1 blockade plus the CD28 activation?

So let me make some slight modifications to sort of the interpretation you presented. PD-1 interferes with the efficiency of normal TCR transduction. So that when a T cell sees its target and it has PD-1 negative stimulation, it won't be effectively activated and expanded and engaged. Normal process of T cell activation has 2 signals, TCR and CD28 is signal 2, 1 and 2. PD-1 really interferes mostly with signal 1. So what you're doing when you're separately engaging CD28 is you're augmenting the signal that you have now derepressed by blocking PD-1. And that is the reason why it works. Alternatively, you could combine CD28 co-stim with a CD3 bispecific, which now just directly activates any CD3 molecule and get activation. The issue is both of those combinations can work. But if you engage CD28 alone as Signal 2, if Signal 2 in the absence of Signal 1, whether it's from PD-1 or from a CD3 bispecific really isn't enough to get the T cell going unless you have an abnormal type of CD28 antibody like the TeGenero molecule. So we specifically were very careful to make sure our CD28 binders don't have that property.

And how is it then? I mean, it sounds like a TeGenero CD28 superagonist can overcome the lack of -- there's still a PD-1 negative effect and still missing the TCR activation via the CD3. So how does the superagonist then are able to overcome that? You're not a superagonist. So do you just...

I would have to speculate there about exactly. It probably has to do with the way it cross links and all sorts of different features of that. But essentially, what we specifically selected for were bispecifics that would bind CD28 and in the absence of the other cognate signal -- or the Signal 1 provision would not activate T cells. So we just -- we selected against that.

So how do you see ultimately this playing out therapeutically? You'll have a CD28 bispecific, let's say, that hits PSMA or hits BCMA but that's not enough, right? Concomitantly, you'll want to have a CD3 bispecific to the same target to then bring -- activate the T cell, bring it in while you're co-stimulating it at the site of the antigen. So...

I think the first thing is you need to have a good signal that gets link started, whether it's PD-1 or CD3 bispecific. And then to that effect, we've made a lot of good progress. And maybe Andres, you'd like to comment on the progress we're making with our CD3xCD20 and by BCMA, which is sort of getting us off the ground and making us confident in the platform.

Thank you, Izzy. I'll be happy to give you a bit of an update on where we are and the progress that we are making. I don't know which one, Yaron, you would like to focus first. And I'll be happy to answer questions on where we are today.

Maybe let's talk about -- I think PSMA is obviously more advanced and then eGFR is going into the clinic now or shortly. So let's talk about PSMA and then eGFR?

Oh, so you want to focus on the CD28. Is that what you're saying?

If you want to give an update on that, and then we can shift, and we'll talk about BCMA and can talk about specifically...

Okay. So yes, you're correct. We are in the clinic with our PSMAxCD28, we are -- and that is with in combination with anti-PD-1 Libtayo. And we have recently initiated open studies for 2 new ones, EGFRxCD28 and MUC16xCD28. All of them will have the same principle that they can combine with PD-1. MUC16xCD28 has the additional -- we have the additional opportunity to combine with our MUC16xCD3 bispecific. So basically, the first thing we do is just demonstrate that, in fact, these molecules are safe, and then we'd immediately start combining them with Libtayo.

[indiscernible] single-agent dose escalation to look at safety. And then you do a combination with Libtayo.

No, no. We just do a single agent running. We just give a couple -- because we don't expect there to be significant single-agent activity with these. So these are all from the outset designed to be in combination.

Okay. Got it. Okay. Okay. Let's -- there's so much that we can talk about, just to be expeditious on time. Let's move and talk about the BCMAxCD3 field. Obviously, a very competitive field. You have 2 different constructs, REGN5458, which per the last call, that was into the clinic first. That has a higher CD3 affinity. It sounds like that's the one that's getting prioritized to moving into a pivotal Phase II next. You also have REGN5459, which has a lower CD3 affinity. What did you learn about the importance of CD3 affinity? And maybe we'll talk about next, how is REGN5458 differentiated from other bispecifics in this area?

Yes. I would say that it's too early to come to a conclusion, as you know, with REGN5459, the low affinity which is behind the progress that we have made with REGN5458. Data is now emerging. We are confident on what this antibody can deliver. But I think it's a little too early to make any comparison or conclusions on what the low-affinity brings to the table vis-à-vis the high-affinity CD3. Of course, there is preclinical data and that strongly suggests that one could have a wider therapeutic index and potentially other opportunities with low affinity CD3 arms, but we are still learning. And as you mentioned, our most advanced compound today, at least is REGN5458, which, as you mentioned, is now advancing to Phase II following our ongoing dose escalation. We feel in terms of REGN5458, we feel very encouraged by the level of activity that we are observing, the safety of the compound. And again, you probably know that at ASH, we will be presenting updated data on where we are in our dose-finding and the outcomes that we have observed to date during the Phase I dose escalation portion of the study.

Yes. And the abstract is out. So we know across all doses, the response rate was about 37%, but it's about 60% at the higher dose, right?