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

That's great. There's a lot to unpack there. It sounds like you have a tree with a lot of different types of fruits, even some of the newer generation antibodies that's super interesting. But I did -- I want you to talk a little bit about Libtayo and just anecdotally what you've heard from like clinicians. I mean are there points of differentiation do you think around dosing or activity that distinguish themselves in the current market indications? Or maybe current read through to maybe a non-small cell population?

Well, I think in our -- we've had excellent feedback from our clinicians using it. They really like using it. The best experience, of course, so far has been in CSCC, where what's happened as a consequence of us getting this approval is that, whereas in the past, patients with advanced disease with -- for whom there was no really approved option and would still get chemotherapy of various types or other treatments, majority now are getting biologics anti-PD-1 upfront. The -- in terms of differentiating, what I would say is what I've said all along. Our approach was not to make a me too PD-1, not really clear that we could make a bio better PD-1, but we certainly wanted no one to have any suspicion that it was a bio worse PD-1. And we feel very comfortable and confident that we have achieved that. So this is as good a PD-1 as anyone out there and will stack it up. That's our confidence in it. Both the numbers that we've seen in our CSCC, what we've seen in BCC, what we've seen in lung cancer. Now there are, of course, all sorts of improvements and moves towards enhanced convenience and dosing and lots of people are trying these things, which we are working on that as well, be it extended dosing or subcu formulations that we've been thinking about that and working towards that for a while. Our first priority, though, has always been to like, can we prove in a field where not all PD-1s made it to the finish line in first-line monotherapy in lung cancer that we have a player, and we have a player.

Great. And do you -- how do you think about maybe some of the potential competitive pressure, maybe from line of Keytruda? I know you guys are very well established, but just how do you kind of think about those potential risks?

I leave that to my commercial colleagues. We just make -- we make the killer mabs. We make sure that we bring forward the best we can, and we're confident that we have been. We think that it's a big market. Right now, the entire PD-1 market is, I don't know, I keep losing, it's $25 billion, is that what it was? Or so even if we take a small piece of that, it's not chump change. So -- and like I said, that's part -- that's not the end goal. The -- if that -- we'll take that along the way and whatever we can prove is worth coming to us, and people may want options to try something different. Whether somebody starts noticing something out there in the clinic that is a differentiating factor, who knows? I'm not holding my breath on that. But the ultimate is going to be in our combinations, which are proliferating.

And heading into -- I would be remiss if I didn't ask, but heading into ESMO this weekend, can you just talk about the importance of this presentation in PD-1 high and then how it links to your strategy, obviously, with the PD-1 chemo study that you have ongoing?

So when we just -- when we first started out, we saw the initial results, KEYNOTE-024 was exciting. But then there was CheckMate-026 that fell on its face. So we thought, okay, I don't know why that should have happened. It was personally sad for me since I was involved in the early development of what became nivolumab. But I don't know what happened there, whether it was a clinical study issue or molecule issue. And we decided, look, we have this -- if we think we have a good antibody, it's got to prove itself there. And lung cancer is, unfortunately, a big market and a huge unmet medical need in general, even with the current treatments, even with chemo plus PD-1, we are very excited about the higher response rates. We're very excited about the improvements in PFS and OS. But we're not curing everybody by any stretch. And I don't know that we -- what the ceiling effect is going to do, but I think we can do better. And I think we are -- we needed, I think, what you'll see from the late-breaking data from the presentation is that no matter how you look at the data, you will say, wow, this is a potent anti-PD-1 and this is an alternative. And the same -- and what gives us, of course, if they work as a monotherapy, it certainly gives us confidence. I don't know what the data will look like. But if more confidence that it should work in combination with chemo, a bunch of it -- a bunch of drugs have shown that. It's the monotherapy that's the high hurdle.

And obviously, you're probably deliberate about developing the receptor versus the ligand, but just can you give us your brief perspective there? As there are obviously some ligands in the market?

Sure. So from the -- having been involved at the very beginning when I was at Medarex with programs, both the ligand and the receptor. At the time, we always thought there was a theoretical grounds for why targeting the receptor should be better. Initially, in the preclinical animal data, it wasn't so obvious. And initially, in the early clinical studies, it wasn't obvious. But from a theoretical point of view, PD-1 can engage with both PD-L1, which is the target of most ligand blockers or PD-L2. Moreover, PD-L1 is typically expressed within the tumor microenvironment on the target cells, tumor cells, sometimes other immune cells, antigen presenting cells, where it mediates inhibitory effects. And so for an antibody to make its way in there and achieve success, it has to get into the tumor microenvironment, whereas lymphocytes that are in the tumor microenvironment, often traffic in and out of the tumor microenvironment. And we know from the very early experiments and from our own data -- from early experiments from others and our own data that once we bind PD-1, that receptor occupancy is quite long lived. So we can sort of arm the T-cell going into the tumor microenvironment as well. And finally, in the tumor microenvironment, when there is an active component of T-cells targeting fighting cancers, they release gamma interferon, which further upregulates PD-L1. So I think the -- it's -- and if you just look at the dosing requirements of PD-1s versus PD-L1s, there are usually 3 -- 3 to 4 times higher in terms of mass amounts that people choose for their PD-L1 doses. And I think although within -- people can say it's Pepsi, Coke are the same. I do think that as you go forward in different studies, it becomes increasingly clear that it's very unusual to see a study where a PD-L1 blocker has been successful where a PD-1 hasn't, but the converse is true. So is it black and white? No. Is there -- and is it possible that as we go along, there will be more differentiating features that explain why one might work better than the other? I think likely. So all in all, if I had a choice from the outset, I would rather have an anti-PD-1. If it was with me or my -- anyone close to me.

Great. And just talk a little bit about we have about 5 minutes left. Your excitement around the costimulatory domain platform. And obviously, the paper from showing similarity with PD-1 as well.

Right. So I'm going to get a little technical here, so folks fasten your seatbelts. So T-cells recognize what we call antigens, right, on the surface of a cell presenting them. And these antigens are presented on HLA molecules, okay? That interaction with the T-cell receptor is called signal 1. In order to fully engage a T-cell and get it going and certainly get it primed and expanded so that it can seek out and kill these antigen expressing cells, the T-cell needs 2 signals, signal 1 and signal 2, costimulatory signal. And the main molecule driving that costim is CD28. That's the main signal. And it binds a receptor called B7. And all these molecules, PD-1, CD28, they're all part of the same superfamily. And it's part of an intricate network that the immune system has evolved to really make it -- make sure that you don't attack yourself, but you really attack specifically something foreign. Now years ago, there was a company called TeGenero that tried -- had an idea for binding CD28. So unfortunately, it went horribly wrong. And what happened was they -- the type of interaction with CD28 led to widespread T-cell activation and they almost killed a few healthy volunteers. And that changed dramatically how we approach Phase I studies. It also scared the heebie-jeebies out of many people from ever approaching CD28. But when we looked at this, we thought, gee, how come CD3 bispecifics or even CAR-T cells, why do they seem to be so much more effective to date in hematologic malignancies rather than solid tumors? Well, one of the potential reasons is that the hematologic tumors themselves are derived from the same lineage that express these molecules that engage costimulatory molecules. So B-cell lymphomas can engage CD28. So we had some very sharp scientists at Regeneron, Dimitris Skokos, is the senior author on the science papers. Can we really examine what the most important second signal is on T-cells to get full activity? And in a series of experiments, convinced ourselves that being able to bind CD28 is really important. So when you get signal 1, if you also add signal 2, you get a much stronger response. But if you only bind signal 2 without signal 1, you don't get much. It's like banging your head against the wall. So what we figured was that there are 2 ways we know of augmenting signal 1. One is directly linking T-cell receptors to cells through a CD3 bispecific. Another one is by using anti-PD-1, which takes a cell that recognizes an antigen by TCR, but doesn't generate a strong enough signal and by blocking PD-1, we augment that. So it made sense that both of these should combine. And indeed, they do beautifully. So one paper showed that you can combine a CD3 bispecific with a costim bispecific to get much better activity. And another one showed that combining PD-1 blockade with a costim bispecific worked very nicely. And so we are going forward with both those modalities into the clinic. So our MUC16xCD28, for instance, is going to probably will be tried with cemiplimab. But down the road, as we sort of understand how our MUC16xCD3 works, could be combined with our MUC16xCD3. So this is why we are so excited about this platform. We think it will also potentially unlock efficacy that has been hard to get at in solid tumors as well as probably being useful in some hematologic malignancies where we're not getting the full benefit we'd like from just a T-cell engager by itself with the CD3. So there's another little nuance here, which is when you use a CD3 bispecific, you don't care what the T-cell recognizes. Any T-cell on the street has a T-cell receptor on it, you grab it by a common part of it that all T-cell receptors express CD3, you say, come on in, join the party here, okay? The -- when you combine PD-1 with a costim, the only ones that you're going to be able to activate are the ones that already intrinsically recognize the tumor. Because if they recognize something else, then you're not going to enhance anything with PD-1. So on the one hand, there may be fewer cells in that setting. But on the other hand, what you create when you do that is deep and long-lasting immunity than memory. So they each have their nuance. And they each have their pros and cons. But the ultimate -- but the bottom line is both of them result in enhanced activity. And we have to do this carefully obviously, and we're doing that. But that's a major focus of our efforts, and we believe that because we have all of these agents under one roof and what I can't communicate here that sort of gives you the flavor of that is this really tight collaboration between our researchers and our clinical teams where we talk through things and decide what do we do next? How do we go forward? We are developing what we think is a plug-and-play type environment where we can, okay, let's take this and put it with that. No, maybe this will work and that will work. And at the same time, we're also trying to optimize further our ability to glean out of early clinical trials, the important signals that help us figure out that's a good combination. That combination unfortunately not working so well.

Awesome. Izzy, thanks for the chat. Justin, thanks for joining. And this concludes the fireside chat. Best of luck, and I'm excited to see what you guys have at ESMO for us.

Thank you.

Thank you, Alethia.