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

Hey, guys, thank you all for joining us. A pleasure to have the entire management team from Arcus join us today. We have a ton to talk about. But before we do, let me first turn over to Terry. Terry, what's on top of your mind, your top priorities as we head into the new year?

Sure. Thanks, Umer. I'll spend a few minutes just telling you how we're looking at things. So 2020 was obviously a year of a lot of execution, our collaboration with Gilead, collaboration with AstraZeneca. If you look across our programs, whether it's AB928; AB680 in the adenosine space; or AB154, our anti-TIGIT antibody, largely, you could look at this year as having been one that was focused on dose-escalation studies, dose-expansion studies and the early start of randomized studies. So we've been pushing, since the founding of the company in 2015, to really get to what 2021 will offer, which is randomized data across our programs. When we brought in Bill Grossman as our Chief Medical Officer, he really did steer all of our programs towards that type of information. So where are we? What should we expect to see? Well, first off, let me talk a bit about finishing this year. As you know, we're big in anti-TIGIT. We've got a lot of conviction around the target. We -- as I said, we just announced our collaboration with AstraZeneca. We're about to file an IND for a second anti-TiGIT, AB308. In all of our programs, the more we believe in a target, the more we're going to make sure we get a drug. Their AB154, which is in the midst of a trial, ARC-7, is moving along great. That's also the molecule that we'll be collaborating with AstraZeneca on that Stage III lung trial. So one key thing that we'd like to describe today, we've been talking about another registrational trial in that we would reveal those details prior to the end of this year, and that trial is with AB154 and zimberelimab. So keep in mind, there's 2 purposes to that trial. It's, one, to get approval for the monotherapy, zimberelimab, our anti-PD-1 antibody; and secondly, to get approval for the doublet. And so early next year, we'll be starting that Phase III trial. It's going to be in frontline non-small cell lung cancer, high PD-L1 patients. It's going to be a trial that looks at monotherapy, zimberelimab; the doublet, zimberelimab plus AB154, and the comparator will be chemotherapy. It's going to be a global x U.S. trial. We've vetted this with the FDA. We've also vetted it with a number of the European health authorities, and that's expected to start early next year. Really, it's going to be a huge year insofar as readouts from our adenosine program. So first, I'll just highlight one that's maybe a bit below people's radar, it's called ARC-8. It's our trial -- it's our first study in patients with our CD73 inhibitor. It's a dose-escalation study, but it's -- that's been going on in pancreatic cancer, and in fact, the molecule has behaved well. We've seen interesting activity. The investigators are excited. It's enrolling great. And we will be sharing activity from that at -- early in the year, in January at ASCO GI. A number of other trials that we'll read out with some randomized data, ARC-4 is our AB928 trial in the setting of EGF receptor mutant non-small cell lung cancer, TKI-failed patients. We generated very interesting data in colorectal cancer this year in what we call ARC-3. We saw very deep responses in the front-line setting. But most interesting to us, and particularly because of the more facile regulatory pathway, were our observations in the late-line setting. And there, we saw a very prolonged responses. So recall, we saw PFS on the order of 4 months. If you look at the standard of care, they're on the order of 2 months. So we're launching a platform study in later-line colorectal cancer called ARC-9. And finally, we've quietly been enrolling in our prostate trial with AB928 called ARC-6. And by the middle of next year, we'll have some randomized data from that program. We're moving towards our Stage III lung trial. That will be later in 2021. We really haven't shared the details, but we're excited about that. That's with AstraZeneca. One thing I would just mention, we get a question that comes up about that trial. Did AstraZeneca see all of our data? And of course, the answer is yes, with the exception of ARC-7. We've been treating that as a blinded study. They saw all of our other data. And when people ask, well, what are those other data? We've had an ongoing Phase I study. We reported the early Phase I data 2 SITCs ago. But we've continued to explore other dosing regimens, and those data have continued to look interesting. As everybody knows from what's already been disclosed, whether from us, from Genentech, from Merck, you don't see screaming activity in an all-comer, heterogeneous Phase I study with an anti-TIGIT. But we did see -- and I'll call it cherry-picked, but it's a pretty compelling patient in that study. And so I would just share that, that was probably amongst the most interesting pieces of data from that study and certainly in the eyes of AstraZeneca. We saw a prolonged confirmed partial response in a late-line esophageal cancer patient, was a patient that had progressed already on pembro. So you want to have expected activity there, and it's a patient that was CPS 2. So yes, AstraZeneca saw all of our data there. And so we're excited. We'll have 2 registrational studies that we've now talked about starting up in 2021. And I think you'll continue to see the anti-TIGIT strategy unfold as we work together with Gilead to further refine and begin on that. And then I think 2021 is going to be the year that we really blow open the adenosine field. The final point I'll make is entering into 2021, we're well fueled. We -- as at the end of last quarter, we had just under $800 million. So we'll be entering 2021 well capitalized with well over $700 million, and looking forward to a very exciting year and also developing further the collaboration with Gilead. So I'll stop there, Umer, and take your questions.

Sounds good, Terry. So Terry, I'll cut right to the chase. I know we have limited time, so I'll cut right to the chase. The question we've grappled with, Terry is, number one, we have proof of clinical efficacy from Roche's randomized trial on TIGIT. And right before this discussion, we were actually hosting Merck. They're really talking up their TIGIT now.

Good.

So all of that is fine. The question for us really is have we ever seen clinical evidence suggesting effector function's presence or absence would have a clinical implication in IO setting?

So insofar as any checkpoint inhibitors, if you look across the space, the answer would be simply no. We're not aware of any data that suggest any differences that have been observed to date.

Got it. Terry, why do you think preclinical evidence? Because Roche -- everyone talks about Roche's preclinical evidence on the effector functions's importance. What I find interesting is no one talks about how you guys had some preclinical data as well, suggesting it may not necessarily be helpful. Why so different? Or is that just standard in preclinical nonpredictive oncology cancer models?

I think that's -- I think your latter point is accurate. There's a lot of preclinical data in oncology that, given whether -- and when I think contrived, I don't mean that in a negative way, but whether they're in vitro studies or in vivo studies, they're just so different. And a lot of times, these things just don't manifest themselves in a clinical study. Keep in mind, in this particular case, you're talking -- people got into this mechanism because it was a very special mechanism insofar as the safety and the efficacy have to do with that CD155 interaction with TIGIT being converted into CD155 interaction with CD226. So you're turning -- essentially, you're removing a foot on the brake of the immune system, putting on to the gas-filled immune system. And no one would deny, including Genentech, that, that certainly seems to be operative. And then there's various other, what I would just call, anecdotal experiences that may or may not contribute to the activity. The one thing that's certainly been put to sleep, I think, by at least Genentech's data to date, is that they don't see Treg depletion in -- at least in their human studies. What I would say about our data, that -- the most profound preclinical data that caused us to not want to have effector function are the data that clearly everyone would agree with, that TIGIT is all over all sorts of immune cells. And you wouldn't want to deplete T effector cells, for example. Though the reality is we don't even think, and I think Genentech would agree, that you're -- with the functional Fc, that you're actually seeing effector function. They invoke other types of experimental data that they have preclinically as to roll for their functional Fc, but I don't think they're actually invoking effector function. So we think that the preclinical experiments just involve a number of different systems that may or may not manifest themselves or insofar as the actual human experience.

Got it. Hello?

Yes. Hi, Terry. So I have a -- so could you also remind us on the thought process of adding the triplet arm in ARC-7? Should we or should we not read across from the Phase I, where you show activity with pembro chemo and the A2A, A2B ARC -- A2BR triplet combo?

And Terry, just to add to Bo's question. I feel like the Street understands the confidence from your part is high on the TIGIT. It would be very interesting to understand if you would characterize the triplet arm as a high risk, which is how the Street perceives it. Or it will be just curious -- we're just curious to hear your take.

Sure. So I'd say the risk is greater than it is with the anti-TIGIT alone because we're data-driven. And there's not the clinical proof-of-concept that you have with anti-TIGIT. With that said, we have a lot of conviction not only from the biology, but as Bo pointed out, across any of the studies we've done in non-small cell lung, albeit them -- we haven't generated the randomized data yet, we've seen a very positive signal. And we think the positive signal tracks with the biology, and that's why we went with that triplet to begin with. So lung is a high CD73 tumor, and we view that triplet arm is offering 2 potential home runs in that study. And they're not 2 outs in the bottom of the ninth home runs. We think the biology really is strong there. Someone has to do the experiment. And we're the only ones that we believe have the molecule that can do that. And the first is that, just looking in the front-line setting per se, so recall that we have singlet arm with our anti-PD-1; doublet with the anti-TIGIT; triplet with the -- with AB928. So it's a high CD73 tumor. There's no question. It's high adenosine. So we feel like adenosine is going to be important just in the overall efficacy there. But the other interesting twist to that study is that we've set it up such that patients that progress on the anti-PD-1 antibody alone can also go into the triplet. And CD73 is a potential conspirator insofar as that progression. So we feel that's another important opportunity, and we'll be the first to generate those data. But we don't -- we feel it's risk because the data haven't been generated, but it's a very -- there's a very strong rationale. And frankly, all of our studies to date in the non-small cell lung setting, as Bo, as you pointed out, while there -- and is small, we've seen what looks like exciting responses, including in the EGF receptor mutant field. So it's an important mediator in non-small cell lung cancer, clearly.

Got it. Okay. And then, Terry, as we think about the competitive landscape, one thing that we're noticing is there's too many TIGITs popping up now, too many of them. Do you see a case for differentiation within TIGIT outside of just timing to market? I'm very curious how you're thinking about that.

So we think of TIGIT, and we -- and by the way, this hasn't changed since we started. We think of it similar to anti-PD-1, as a backbone that you need, but then it's going to be what you combine it with. The emerging data set, what's interesting about it, to us, is at least in the near term, we think of anti-TIGIT as making anti-PD-1 or anti-PD-L1 into a super anti-PD-1. So we feel if you only have one and not the other, if you don't have an anti-PD-1, an anti-TIGIT is certainly not going to serve you well in the near term. And we feel that if you have an anti-PD-1, you're going to be at a competitive disadvantage to -- if you don't have the anti-TIGIT to superfy it, if you will. So we feel, as to your point, there's a race in the field per se of that doublet. But we feel, in the long term, that this is going to be something that you want to have in your toolbox. If you're genuinely a long-term player in oncology and immuno-oncology, you're going to want to have that to combine with other agents like, in our case, where we're talking about the combination with our AB928. So we feel that's going to become almost like a black box backbone of that doublet, especially given the safety profile being so positive with anti-TIGIT. We think that the 2 will start to go together, certainly, in non-small cell lung, it's emerging and probably beyond that.

Got it. Okay. In the last minute or so, Terry, I know we're -- and these are short fireside chats. But in the last minute or so, one of the assets beyond your clinical-stage programs that's of high interest to us in light of all the Merck data is the HIF-2. Can you speak to the time lines for when is the earliest we could see some sort of initial clinical data for your HIF-2? And do you see any scope for differentiation versus Merck-Peloton molecule?

So I'm going to let Juan answer that, but I'll tell you the primary point of differentiation again, and that's why we got into it, is we think that might combine quite well, given the biology with molecules that modulate the adenosine pathway. So that's where we see the differentiation. We think they have a high-quality molecule. We think we have a high-quality molecule. But I'll let Juan take the last minute and talk about the program and its timing.

You're absolutely right. So hypoxia activates 2 pathways, HIF-1 alpha and HIF-2 alpha. In addition to a direct inhibitor of HIF-2 alpha, adenosine up-regulation is one of the main consequences of HIF-1 alpha activation. So by concurrently going after adenosine, the adenosine access plus directly inhibiting HIF-2 alpha, we feel that in a number of clinical settings that the combination of the 2 will be more effective than either one alone.

Excellent, excellent, excellent. Well, listen, we would normally love to have several additional questions, including with Jen and everyone, but we'll leave that for a future venue. Thank you, guys, so much for joining us.

Thank you, Umer. And thanks, everybody, who joined in. We appreciate it.

Thank you, guys.

Stay well. Take care. Bye-bye.