Arcus Biosciences, Inc. (RCUS) Earnings Call Transcript & Summary

Good morning, and we're just really excited to be here, and welcome to Cowen's 41st Annual Healthcare Conference. This is the Arcus Bioscience company presentation. [Operator Instructions] But today, I have the pleasure of introducing Terry Rosen, CEO; Juan Jaen, President; and Kaytee Bock, VP of IR and Corporate Strategy from Arcus. And with that, I'll turn it over to Terry and the rest of the crew to tell us more about Arcus. Thanks for joining us.

Thank you very much, [ Eve ] I was so excited I just jumped in right up there. So good morning, everybody, and thank you for your interest in Arcus and look forward to spending the next half hour or so telling you about Arcus. For those of you that are new to Arcus, we're an oncology company. We were founded in 2015. We focus on targets where we recognized early on that molecules would be being combined with each other. And so therefore, all things being equal, we focus on programs where we could combine molecules from within. It's a company that was founded with a vision in a very genuine way of being a long-term and independent company. And I hope you'll see that by our actions that's, in fact, what we are doing, have been doing. We've been saying for some time that 2021 would be a transformational year for the company. It is starting off that way, and we think that is the case. So when you think about where we are, when we founded the company back 5 years to go and change, this is really where we are looking to get for perhaps a first novo point. If you think about 2020 for us, it was a year of dose escalation, dose expansion, we started a number of randomized studies. And in 2021, I think you'll see that we have a continuous flow of data and a lot of randomized data that will be coming out. So where are we now, that's kind of highlighted in the blue box. We had 5 molecules in the clinic right now. When you think about the nature of our portfolio. Early on, we recognized that the checkpoint inhibitors were basically something that any company that had a long-term vision of being oncology, you need to have those in our toolbox. We brought in a PD-1 antibody, AB122, to the best of our knowledge it behaves just as does KEYTRUDA. And then early on, we invested in a big way in anti-TIGIT. We felt that could be the next potential backbone therapy. And we actually have 2 molecules. We're very excited. We are, we believe, amongst the leaders in the anti-TIGIT field, and I'll say a little bit more about that. We have in the clinic, 2 molecules that I would describe as both certainly best-of-class with opportunity first-of- class as well, AB928, AB680, both that hit important points in the ATP adenosine pathway. We have 6 ongoing Arcus-sponsored randomized trials, just a huge data flow. We expect to be presenting pretty continuously to almost every major oncology conference this year. And we have a very robust discovery group. So there's nothing virtual [ about ] Arcus, and we expect to bring HIF-2 alpha inhibitor, perhaps other molecules into the clinic later this year. Finally, we're very well capitalized. We have a great collaboration with Gilead. It's an all-in collaboration, really enabling for us, I'll say a bit more about that. And they recently took their equity position in Arcus from just over 13% to 19.5%. And it puts us in a position worth well over $900 million as we move forward. In addition to that, we continue to be aggressive in corporate development. We did a collaboration recently we announced with AstraZeneca, and I'll say a bit about more that as we go forward too. So why don't you go to the next slide, please, Kaytee. So this just gives you a sense that we do have a very broad clinical program. These -- every bar there has been very carefully selected. What I would say, if you think about the various trials that we have ongoing, clearly, we're playing in important settings, as you can see on the very left, but the way we pick these, there's generally a strong Venn diagram, if you will, that we apply when we think about settings. So generally, because of -- as I was talking about certain of the programs we have, we're picking programs or settings where there's high adenosine or high CD73, where there's a low bar in terms of the hurdle. And what we mean by that is things where there's great medical need. And finally, not surprisingly, in that sense, where that setting is not particularly responsive to anti-PD-1 therapy. So let's go ahead, Kaytee, please. So this gives you a sense that we have a lot of readouts coming in 2021. We promised that towards the end of last year that this would be a data rich, and as I said before, a transformational year for Arcus. And we started off with data from ARC-8. ARC-8 was our first look at our CD73 inhibitor, AB680, and it was in first-line pancreatic cancer. I'll say a lot more about that a little later. That's now in an expansion phase and enrolling phenomenally well and we'll get more data later this year. There's a study called ARC-3. It was our study in colorectal cancer that we spoke about last year at AACR. We'll be sharing the survival data from that. And it's certainly a very encouraging set of data. And as you see at the bottom of that slide, ARC-9 is a study that falls on -- that will be generating randomized data, particularly in the late-line aspects of colorectal cancer. ARC-7 is something that's been of intense focus. It's our Phase II randomized trial that looks at our PD-1 antibodies, zimberelimab. It has 3 arms. The second is our doublet that includes our anti-TIGIT, dom. And finally, he has a triplet arm that distinguishes it from anything really going on in the anti-TIGIT field now that has what we believe is that home run opportunity, zim plus dom plus AB928, our adenosine receptor blocker. And those are -- that study is being done in high PD-L1 non-small cell lung cancer, I'll say a bit more about that later as well. Then ARC-4 and ARC-6 are studies ongoing building on data that we generated in 2020 both with AB928, 1 in the setting of EGF receptor mutant non-small cell lung cancer, the other in the setting of prostate cancer. So partnerships have been an essential component for Arcus. We've been very targeted in how we thought about them. I'd say, I'd call it, smart corporate development, it's something that we continue to emphasize. And I'll mention, first, Gilead, so I'm going to talk about both Gilead and AstraZeneca as examples. So the Gilead collaboration was extremely enabling for us and really towards that vision of building a long-term independent company. First off, what it did is I talked about, we're really focused on combinations. And because of the all-in nature and great alignment of this 10-year collaboration, it allows what we like to say, maintaining the integrity of the portfolio. And if you look at our trials, you'll note that, again, we really mean that. You'll see that virtually every permutation of our molecules is in a study in one form or another. The key economic term for us was that maintaining 50% of U.S. commercial rights. Also, when Gilead opts in, it funds our 50% of the late-stage development. Gilead, invests in the other 50%. So what this does is it enables us to compete. If you look at our programs, they're all within these ubiquitous pathways, and we have competitors like Merck, Genentech. So to compete aggressively, it's really awesome having this collaboration with Gilead. With that said, even under the umbrella of that collaboration, there are places where it may be smart to do others. So in the case of this AstraZeneca collaboration on the right we are working with them with our TIGIT antibody. And later this year, we'll be starting a study in Stage III lung. It's a registrational trial. And really, they bring great expertise to that: execution, knowledge, a platform. We're excited to be working with them. And you can expect to see us to do other types of studies -- like other types of agreements as we move forward. So I'll start with our TIGIT program. We've been interested in TIGIT since the early days of the company. We feel it's an essential backbone. If you think about it, what TIGIT is looking like in these assets today as it turns anti-PD-1 therapy into the super anti-PD-1. We have 2 different molecules, 1 that's FcR silent. That's in 2 studies right now, ARC-7, I mentioned, and we've just and a registrational trial. I'll talk a bit about -- more about that. AB308 gives us optionality insofar as certain types of hematological malignancies, notable multiple myeloma where TIGITs are found on the tumor cells. And we could be in position as early as the second half of this year to start a registrational study there. Please go ahead, Kaytee. So this slide just shows an example. When we did the collaboration with AstraZeneca, we had a number of questions about what data AstraZeneca had seen. These are data that we shared at the end of 2020. This is a patient from our Phase I study. We're looking at alternative dosing regimens as we went along. And what's interesting about this is this particular patient had a very prolonged partial response in esophageal cancer. What particularly makes this interesting is this patient had progressed on multiple therapies, including pembro, was low PD-L1 patient CPS2, and this patient continues to do well even at cycle 12. What's also interesting is Genentech also reported a Phase I esophageal response. And as you know, they have a Phase III trial going on this setting. So the next slide summarizes the design of our Phase II/III arm study. This -- the data from this study, we will be doing an interim analysis that will happen toward the end of the second quarter. We're going to give an unambiguous thumbs up, thumbs down on the results from that interim analysis. And this also sits in an important place from our relationship with Gilead as this data set will represent the data of how much they'll make their decision for opting in on this program. We expect to present these data, then in a more formal sense, the first medical conference likely in the second half of next year. We also have just initiated, and we expect first patient in a -- for us, it's a big milestone. It's our first registrational trial. This study, you can see the design on the slide is 2 purposes. First, to seek approval for zimberelimab, our anti-PD-1 as a monotherapy as well as the doublet together with our anti-TIGIT antibody. Again, this is in high PD-L1 first line non-small cell lung cancer, a very competitive area. Next slide summarizes the ATP adenosine pathway. I'll just make the point that we have 2 molecules that I would describe, certainly best-in-class opportunity for first-in-class that [ hinted ] what are generally viewed as the 2 key nodal points in this pathway. So what you should think of is adenosine is something that tumors can vary generally cause the production of adenosine. Adenosine is highly immunosuppressive. That immunosuppressive action is well understood at a molecular level, and we have molecules that can both block its formation. So there's an enzyme called CD73 that's rate-limiting in that formation. We have an extraordinarily potent inhibitor of that enzyme. And then finally, you can block the action of adenosine and it's molecular target on immune cells. Those are adenosine 2 receptors, and they come in 2 forms or flavors, if you will, A2A receptor and A2B receptor. Like I say, in both cases we have certainly what would look like best-of-class molecules, and we think we have opportunity to be first of class. So I'll talk a little bit about what we're doing with those molecules. So AB680 is the first small molecule CD73 inhibitor to enter clinical development. A few things about its properties. It inhibits the enzyme with an IC50 of less than 10 picomolar. It has an incredible half-life. So, been working in small molecule drug discovery for more than 3 decades, haven't really seen a molecule like this. It can be given by IV once every 2 weeks, and still cover its target 100% at trough levels. This is a pretty hot area, AB680 distinguishes itself from the antibodies that are most advanced in development, most notably in its ability to inhibit the enzyme by 100%. The reason to this is important is, although the enzyme has found attached to cells, it's very easily shed. And hence, there's a lot of soluble CD73. So the antibodies were designed to internalize the adenosine -- obviously can't internalize the soluble enzyme. So we think we have not only the first small molecule, but certainly a best-of-class molecule with a great set of properties. This just summarizes the clinical protocol. We did a healthy volunteer study with this molecule AB680, showed it had the properties that we desired. And then instead of taking it into all-comer study, we took it into frontline pancreatic cancer in combination with the standard of care there, which is gemabraxane, together with our anti PD-1 antibody. Couple of things about that setting. It's really a poster child for what we pick, super high CD73, very KRAS driven. KRAS is generally a marker for high CD73. As you know, a terrible diagnosis and prognosis, so low bar and again, not anti-PD-1 responsive. In fact, there have been studies done with our anti-PD-1 therapy on top of gemabraxane, if anything, slightly lower response rate. The well understood response rate for gemabraxane is about 23% ORR. In that study with OPDIVO, the OR was 18%. I don't think there's much difference between 18% and 23%, but clearly, not PD-1 responsive. So here's the waterfall from the data that we presented in January to ASCO GI, pretty remarkable. This is essentially unheard of in pancreatic cancer. You saw shrinkage or stabilization across essentially every patient. So 15 out of 17 patients roughly almost 90%, 88%. Perhaps more importantly, though, what we saw from the standpoint of the spider plots, which would were very encouraging. Keep in mind that these data were presented in January, almost in real time. So this is an ongoing study as we speak. This gives a little bit more detail on what we saw there. As I highlighted, what we think is going to turn out to be exceedingly important about this molecule is that when you put AB680 on top of gemabraxane, you essentially don't see any added side effects, nothing different than you might expect to see with the chemo or the anti PD-1 alone. This is really important because when you think about new agents in this type of setting, oftentimes, what they bring is interesting biology. You see some initial effect but unfortunately, what happens is you're killing the tumor, you're also killing the patient. And basically, that toxicity makes it tough for the patient to get the full benefit of that initial relief of tumor burden. And so we think the great safety profile associated with this mechanism is really going to be a difference maker going forward. We saw an ORR of 41% across all doses that we looked at. If you looked at the 2 highest doses, at the time of that poster was 50%. If you looked at the highest dose, which is the go-forward dose in an expansion phase of this, it was 75%. So as I mentioned, 100 milligrams once every 2 weeks was selected as the expansion dose. Extraordinary investigator enthusiasm in the study. So we started enrolling this in December, the expansion phase, it's enrolled very rapidly. We expect to open a control arm very shortly. We'll be sharing those data towards the latter part of this year. But if things continue to look as they do, we expect to be speaking with the health authorities later this year, and we'll be looking to head towards a registrational trial as rapidly as possible, potentially as early as the start of 2022. Now I'll shift to AB928, our adenosine receptor blocker. It's, again, a unique molecule, the only molecule in the clinic that blocks both 2A and 2B receptors. Because of time, I can't explain right now the importance of that 2B but it is a very important aspect of this molecule. It also has incredible drug properties. It's a once-a-day oral dose, long half-life, we showed before we went unto patients that we could cover the target 24/7. You can see we have a very broad development plan. Again, I mentioned these trials a little earlier, ARC-4, ARC-6, ARC-9 that is just getting up and going, that's in late-line colorectal cancer. I'll use the ARC-3 as an example as to why we were so excited and moved into this platform study with ARC-9. This just points out that -- and I mentioned at the outset, 2020, we generated a lot of single-arm data, give us a lot of enthusiasm for the pathway. One thing you should note, and I think we'll end up looking back at that ARC-8 study with AB680, was a very seminal study not only for Arcus, but for the field because the demonstration of the importance of adenosine in that setting bodes very well for the field, because adenosine is something in CD73 associated with virtually every tumor setting. So sometimes I talk about there's low PD-L1 settings. There's no such thing as a low adenosine setting. It's a very ubiquitous pathway. And so any demonstration of the importance of the adenosine suggests there's a high likelihood that it will also play a role in other settings. So this is a complicated slide. I'll verbally highlight what's going on here. In 2020, we reported out about a study in colorectal cancer. It was across lines of therapy ranging from frontline through third line plus. We saw a very strong signal and activity across those lines of therapy. In frontline, a number of very deep responses. In fact, 5 patients that went on to surgery with curative intent. It also included a complete response, which is very rare in colorectal. But what was more intriguing to us and more intriguing probably because of the more facile pathway towards registration, you'll notice at the bottom of that waterfall or swim lane, if you will, what you -- what we saw in that late-line setting was a very durable effect. So we saw a couple of responses. And when you put us up near the 10% response rate, when you think about standard of care, it's anywhere from 1% to 2%. But what we also noted was a -- what looked like a PFS on the order of 4 months. And again, when you compare to the standard of care, that would be something that's 2 months or even less than 2 months. And it was these data that very much encouraged us to go into that ARC-9 platform study. And as I said, we'll be presenting the survival data from this study at AACR. Finally, if you go to next slide, Kaytee. This just emphasizes our -- some data on our anti-PD-1. It's being developed in China by a company called Gloria. We have essentially the rest of the world rights to that. We brought it in for strategic reasons. It gives us flexibility, whether it comes to controlling our own destiny in so far as our choice of development settings, how we commercialize or how we even price in China, Gloria has already filed the BLA. I think they're expecting approval later this year in classical Hodgkin's lymphoma. But as I mentioned, our experience with the molecule suggests that it looks essentially no different than KEYTRUDA. I think I have one last slide. It just emphasizes that Arcus has a very robust discovery organization. This was only enhanced by our collaboration with Gilead. And as I mentioned, we have these programs we're going to secure, any of which could be -- select a clinical candidate early this year. But HIF-2 is the one where we are most out front and committed and expect to take our HIF-2 alpha inhibitor into the clinic later in 2021. So that I'll final my final slide. I'd just like to thank you all for your attention. Thank you all for your interest. Again, we're looking forward to a very exciting year. And as I said, a data flow cadence that will almost be, if not on a monthly basis, certainly every couple of months or so. Thank you very much.

Thanks, Terry. We're looking forward to the additional data. We do have a couple of questions from the audience. The first is with respect to the TIGIT, what were your thoughts on the Roche data for tiragolumab and atezo in the Phase II CITYSCAPE trial, given that it seemed like the atezo control arm sort of underperformed in that trial. And how are your molecules differentiated?

Sure. So we actually thought there was a fantastic data set. Certainly, the atezo arm did underperform. But when you look at -- it was a very well done, Genentech did a great job in this space, very well done. Clear statistical difference on PFS as well as response rate. Also, we believe, was supported by data, although a different type of data that Merck presented. On the atezo arm, Genentech said that they believe those were the real-world data. We expect better performance out of our anti-PD-1 than at 22% response rate, but that 60-ish percent response that they showed with the doublet was clearly there. It was 29 patients per arm, a really excellent study. And so far as our 2 molecules, the AB154 is differentiated in that it has a silent Fc region. And AB308, the second molecule as a functional Fc. We think that functional Fc would prove important in the settings where you find TIGIT on the tumor cells. I mentioned certain hematological malignancies like multiple myeloma. Otherwise, we don't really expect any of the anti-TIGIT antibodies, just like the anti-PD-1s, just like the anti-PD-L1s, just like the anti-CTLA-4s, to have substantial clinical differences. Where we really think that the big difference is going to derive from what you combine it with. So things like AB680 or AB928. So we look at 2 different types of races in this field. Those that are the doublets that are a pure, who can get there faster, we can get their smarter, who can you pick the right settings? And then things where you might really differentiate by what you bring to that anti-PD-1, anti-TIGIT combination, whether it's something like 680, something like 928 or perhaps even something if we were to do collaboratively with Gilead something like [ atrudelvi ]

Great. And I think Juan mentioned on the prior panel that 2 of your assets and you sort of discussed here as well, Etruma the dual adenosine receptor antagonist and your CD73 inhibitor, both influence different aspects of adenosine metabolism. And can you sort of speak to how they're mechanistically differentiated outside of their targets? and any potential synergies between the 2 that may be [ in for ] sure?

Sure. And I'll use an example of prostate cancer. So in certain settings, all of the adenosine, like pancreatic is being generated in the CD73 mechanism. So what you have is -- where does the adenosine come from? When a cell breaks open, like a cancer cell, it spills ATP into the extracellular environment. And that ATP is converted by 2 enzymes, CD39 clips off the first 2 phosphates and CD73 clips off the third and you generate adenosine, which has that immunosuppressive effect. So in certain settings, as I mentioned, like pancreatic, all of your adenosine is coming from the catabolism of adenosine monophosphate to adenosine that's catalyzed by CD73. And in certain other settings, there may be other mechanisms that can convert that adenosine monophosphate into adenosine. So take prostate cancer, for example, there is an enzyme called PAP. And that PAP is a phosphatase. And in fact, PAP is so prevalent in prostate cancer that at 1 point, it was used just like PSA as a marker of prostate cancer. And so in a setting like that, since you have CD73 independent production of adenosine, you can think about wanting to block the actions of that adenosine form at its receptors on those immune cells. So that bodes well for something like AB928. But in fact, in our prostate platform, not only are we looking at combinations involving the adenosine receptor blocker AB928. But now that we have the going forward dose of AB680, we've even already started dosing patients. It was very straightforward, proving the platform nature of that study where we have both AB680 and AB928 in the same setting. So where you look really to squeeze all the water out of the stone of that adenosine action blockade, blockades formation, blockades action.

Okay. And so out of all of your portfolio of assets and which programs are you the most excited about going forward?

And I say this in a very genuine way. These are all very exciting pathways and clearly, they're at similar nodal points. So the fact that we're super excited on all of these is that the 2021 is the year where we'll generate the randomized data that we think will break open those fields. Clearly, the anti-TIGIT field, based upon the Genentech data, is very exciting. So we're investing very heavily in that. We've already -- we started the registrational trial in frontline non-small cell lung cancer. We'll be starting that Stage III lung cancer collaboration together with AstraZeneca later this year. You can expect to see us unfold a number of other TIGIT-based programs. So certainly, that's a bit exciting for us. And I have to tell you, we were so thrilled with the data from ARC-8, that pancreatic trial. It's such a dismal situation for patients there, and they've really been such exciting results out of something that we were really just looking for PK/PD safety data, and we got that early clinical activity. So it'd be hard to not be super excited about both that pancreatic setting and how we can advance it through 2021. And then more broadly, you can expect to see us looking at that CD73 inhibitor in other settings as well as we move through this year. So really, really exciting times for us.

Excellent. Thank you so much. And to the audience. Thanks for listening. Take care.

Thank you for including us.

Absolutely.