I tripped on the stair.
Hey, mic is hot. Terry, mic is hot. All right. Thank you very much, everyone, for joining us today. My name is Mohit Bansal. I'm one of the biotech analysts, biopharma analysts here at Wells Fargo. And I'm very happy to have the Arcus Management team. We have Terry Rosen, the CEO of the company, and Jennifer Jarrett, the CFO of the company. Thank you both for joining us today.
Thanks very much. Thank you, all.
Awesome. So maybe if we can get started with what is going on with the company right now, and what is the investment case at this point, and how investors and analysts like us should think about the company in next 6-12 months at this point?
Sure. So interestingly, we have three very near-term data readouts that we'll be talking about between now and the end of the year or the latest very early next year. And I think that encapsulates how people should be looking at the most important aspects of the company as we sit here today. So the first I would talk about is in our anti-TIGIT program. So on the heels of a SKYSCRAPER-01 inadvertent release or release, if you will, the certainly positive for the field and positive in a broader sense. I don't think it affected us in any way from a decision-making standpoint, but certainly it was a data set that I think people found very encouraging for the broader field.
So we'll have a readout from what we call, EDGE- Gastric. It's another study that will also been called ARC-21. It's a study that's looking in a patient population, where we're already running a registrational trial called STAR-221. Super exciting setting. We are first. We're years ahead in that. We can talk later maybe about scientific rationale, but that's in GI cancers, with the adenocarcinoma histology. So we're well positioned. We're first. We are positioned to stay first there. And this study was really designed to facilitate, as we bring on sites globally, although it's already really. I would say, this study is enrolling gangbusters. There's a large, trial need in addition to medical need, but this data set is substantial. It's in 40 patients.
It's single arm, but it'll tell you, it'll tell you a lot. You'll, you'll get a good feel. People are, are of course, gonna look cross-trial. You'll see how we compare on ORR at the, at this data cut. We may have six months of PFS, and I think you'll also get a very good look at the safety profile of our doublet. And when you look cross-trial, and you think about ultimately that key differentiation, where we, we do believe our biggest competitor in the broader field of anti-TIGIT is gonna be Merck. And so you'll see chemo, Dom, Zim. You'll have some measure in 40 patients of efficacy, particularly in the high PD-L1 patient population, where historically, that's where you see the best effect in this setting.
You'll have that ORR, but you'll also see the safety profile, and I think it's gonna be very exciting for Arcus and bode very well, not only for that setting but for the broader use of anti-TIGIT when you think about competing at that doublet level, particularly with Merck, and we can get into more of that later. The second data set that I would like to talk about has been one that's a little under the radar, and that's ARC-8, and that's our CD73 inhibitor in the setting of pancreatic cancer. And so you'll recall, we started a study. We saw interesting activity.
We did a dose escalation, we did dose expansion, and ultimately, we got into a randomized portion, where basically what we were looking at is Gem/Abraxane chemo plus our CD73 inhibitor, plus or minus in a randomized format, with or without anti-PD-1. And that was being done because historically, it's known that anti-PD-1 doesn't add to chemo in that setting. So the question that we discussed with the FDA that we were looking to answer was: In the context of an immunotherapy like CD73 inhibition, did that anti-PD-1 bring any benefit? Now, the take-home message, first off, is that the anti-PD-1 did not bring any benefit. The more important aspect, though, we had messaged to the world that we had taken a look at the data set. Wasn't clear that there was anything spectacular happening early.
This was even before it was fully enrolled, and we said we would wait till we got to overall survival. Keep in mind that what we'd said publicly when we talked about last patient in, at this point, our median time of follow-up in that patient population is on the order of 22 months. That's obviously tells you something about that durability and overall survival. So if you think about chemo, whether it's Gem/Abraxane or FOLFIRINOX, what you might expect to see for OS is it's dismal. It's between, you know, 8-11 months. So even a three-month advantage would be meaningful. I think we're gonna have a very exciting data set.
What also I can tell you, it's 120 patients that have seen 100 milligrams of Quemli, so it's very substantial. The randomized portion is 90. However, you cut the data, whether you look at the triplet, whether you look at the quad, whether you look at all the patients, whether you look at patients that had liver mets, you look at patients that didn't, you look at patients that had pretreatment, you did or not, you're always gonna beat the chemo. So we're excited to share that, and that'll also be either just prior to the end of the year or early into next year.
And then the final data set is one that I think there's a lot of people that have been excited a long time about our HIF-2α inhibitor, because that's a space where you have one approved drug, belzutifan. It's a validated target. Because the chemistry is so difficult, you don't see 20, and in fact, you know, you see single digit, if anything. And so we're coming along with a molecule. We did a dose escalation initially in healthy volunteers. Now we've completed that dose escalation in patients. We'll be sharing PK/PD and safety data that I think will clearly differentiate from the Merck molecule. I'll remind you that the limitation of that molecule is very well defined. It's known by Merck, it's known by the rest of the world. They have absorption-limited pharmacokinetics.
The dose that they use is 120 milligrams. What you're gonna see is the 100-milligram dose that we've taken forward into a dose expansion study in clear cell RCC patients. We'll also be taking it as a doublet, with a soon-to-be-announced partner. Within the next month or so, we'll disclose that. That 100-milligram dose delivers very clearly three- to four-fold the exposure of belzutifan, but with an incrementally more potent molecule. So whatever you can squeeze out of HIF-2α as a target, I think you're gonna see a substantial advantage, from the Arcus molecule. As I mentioned, we think we'll also have, an advantage with the potential for a best-in-class doublet once we, disclose what that is. You'll recall that Merck, uses lenvatinib.
That's their internal molecule, and they're stuck with it for good or for bad. It clearly has more safety issues than some of the other partners you might think of, so we think we have an advantage there. And then the final point I would just make, we have an awesome collaboration with Gilead. It's what enables us to compete in these very competitive fields, to compete in a right way, both the capital and the human element of that collaboration. Literally, Gilead has hundreds of people working in every function. We have over $1 billion cash right now, and that takes us into 2026. We feel we're super well-positioned, with a great portfolio and, you know, an early end on the R&D engine and, you know, a continuously growing development function as well.
Awesome. That's amazing, actually. So let's just take these programs you just, y ou talked about one by one here. So maybe let's just talk about ARC-8 trial a little bit. So, seems like you are trying to replace Gem/Abraxane, or how should we think about the bar for success? Do you want to improve upon that? Or even if it is as good as Gem/Abraxane, it's still chemo, so still could be a safety advantage there. So how should we think about that?
Yeah, we wanna improve upon it. We feel like if you look at the world right now, you're probably about 50/50 FOLFIRINOX, Gem/Abraxane. So I think our data will show that you will see a clinically meaningful difference. We think that is coming, t he other thing I would describe about the CD73 inhibitor, it brings about as close to zero added toxicity as you could imagine. And I think what you're gonna see when you look at this data set is when people talk about the aspirations of immunotherapy, where you're enabling and enhancing an immune response. So, removal of adenosine does not drive a T cell response, but it enhances one.
And what I think you're gonna see is that aspiration, where you can create that immune response in an individual, you can really see that tail that translates to OS, which is most important. And so then you might even start to think about a paradigm shift, where things like response rate, which again, when you're just looking at chemo, you respond, and then you progress. You may see patients that respond. A response is sort of an arbitrary number. 30% is a PR, 26% is stable disease, 42% is a PR. You may see that in certain patients that actually might have, with chemo, just achieved like a stable disease. Let's say they actually had a better immune response than that patient with 40% tumor reduction. They may even benefit more greatly in the context.
So I think you are gonna start to see something that looks like that, general concept that people tend to associate with immunotherapy, where you're enhancing an individual's immune response without bringing any added toxicity.
Got it. Even Merck did a deal this morning, right? I mean, with the adenosine antagonist. I don't know. Like, I'm sorry to put you on this, but yeah. I don't know if you have heard about this, but.
We did. So, there was like.
Portage.
Portage, yes. Yeah, we imported.
Whatever that means, we think that the general field is of high interest. You know, there's been another study that was under a lot of people's radar, a co-study that AstraZeneca ran in stage three lung, that showed a clear benefit with their CD73 antibody. We think we have very major advantages relative to oleclumab because despite the fact that they have that it's the same target, the antibody actually causes internalization of the enzyme but doesn't inhibit its activity fully, and there's always new enzyme that's being generated by those tumor cells.
So think about your most prolific source of cells, it's in the tumor, and that inability to inhibit the enzymatic activity, whatever they see. And by the way, they did see and report a signal in pancreatic cancer with oleclumab, gem/abraxane, and durva at last ASCO. So we think there's an interest in adenosine modulation, and we think after our data set, it'll probably get enhanced.
Got it. Completely makes sense. So let's just talk about the SKYSCRAPER-01 data, you know, the release, which was not supposed to come out, but it did come out. I think the most important question from my side is, when you saw the data, was it in line with your expectations? And does it make you more comfortable about that particular trial and then your own program essentially?
Yes. You know, definitely wasn't out of line.
Right
W ith our expectations. And maybe, like, our expectation wasn't as specific as maybe some other, you know, people had, investors. But, you know, we think more than anything, you know, if you look at the data, it clearly shows that TIGIT is active.
Right.
So, you know, that was our expectation. That's very consistent with what we've seen. You know, keep in mind, we've obviously put the ARC-7 data out there, but, you know, there's other data that we've seen in-house that we have not shared publicly that gives us even greater confidence in TIGIT. So we weren't surprised to see that TIGIT's active, and that Sky One showed what it did, and I think it's just more validation for the outside world. I think the most important point for us, you know, I know there's still this lingering question, like, are they definitely gonna hit or could they just miss? Like, either way, that's a win-win for us. You know, either they hit, and once again, everyone can say, you know, TIGIT clearly works, and it's a positive phase III study, or they just miss.
I think based on what we've seen, it's hard to imagine it would be a wide miss at this point. So either they hit or they just miss.
Right.
If they just miss, they lose really the only advantage that Roche was gonna have, which is their first-mover advantage over us and Merck, because they've got other things that are working against them. You know, having the Tecentriq as a combination partner, the SKYSCRAPER-01 study design was probably a little bit suboptimal because they used Atezo as a comparator.
Right.
They didn't use Pembro. So like I said, you know, we really view that data set and the outcome or whatever they ultimately show is a win-win for us.
Got it. No, that completely makes sense. So, talking about your own data from ARC-7 trial, I mean, at ASCO, we saw, we saw interesting results there. 33%, as a doublet, 28% as triplet. So, the data remained consistent over time. So when should we hear more about it? And then I think the ARC-10 trial is turning out next year, so how should we think about that?
Yeah, so the first question was on ARC-7?
So ARC-7. OS.
OS, yeah.
Yes.
So we haven't given any guidance. It's gonna be a while before we have it. Just if you look at the OS for PD-1 alone, and you can look at the SKYSCRAPER-01 overall survival data as well. But, you know, PD-1 alone is about 20 months, so it's gonna be a while before we have any data. It's probably towards the end of next year. We're just doing some internal math on it. So like I said, it's gonna be a while, but I think, you know, based on what we've seen so far with ARC-7, between the PFS data, you know, all the other data sets that we've seen, we feel very confident about TIGIT. We certainly don't need to see any other data for us to be very confident about TIGIT in the lung cancer setting.
You know, as far as ARC-10, you know, as a reminder, we redesigned that study about a year ago.
Right.
So we now have Pembro as the comparator, so that study is now enrolling. There's a lot of interest in this study. I think people really like the design. The fact that we are using Pembro as a comparator, we think SKYSCRAPER-01 will certainly help with enrollment and probably increase the enthusiasm for TIGIT.
Right.
You know, that said, I'd say that STAR-121 and STAR-221, so STAR-121 is our all-comer lung study that Gilead is operationalizing, and then STAR-221 is the all-comer setting in gastric cancer that we are operationalizing. From an execution standpoint, our focus is more on those two studies because the market opportunity is just much bigger. Those are easier studies to enroll because they're all-comers. We're not having to select out for just PD-L1 high. So while we're excited about ARC-10, that will certainly be a supportive data set. I'd say the studies that we probably spend more time talking about, both ourselves as well as with Gilead, are STAR-121 and STAR-221.
Got it.
And then just as a reminder, one other point on this that people forget about, you know, we have Taiho as a partner in Japan and other countries in Southeast Asia, and they have opted in to participate in both STAR-121 and STAR-221, so that will help as well from enrollment. They pay for their costs for patients that they enroll in the region, so that also helps from a cost perspective. So that's one other thing to keep in mind on those two studies as well.
Got it. That completely makes sense. So I know with TIGIT, lung cancer gets talked about a lot, but you showed some interesting data in liver cancer as well, and so did Roche as well. When you think about the opportunity for anti-TIGIT outside of lung cancers, I mean, how should we think about it? What are the opportunities there? And, and, yeah.
It's really broad. And I think one of the scientific and biological aspects that I think it's worth, and now that, like, I think people can settle in and think that anti-TIGIT, you know, how, what's the breadth of the opportunity? There's an important biological observation that we highlighted from ARC-7, and that was that what was known before that study was even done, if you get anti-PD-1 therapy and you have high CD155 expression, CD155 is the ligand for TIGIT, and when CD155 engages TIGIT, it's immunosuppressive. So blocking that enables actually converting a foot on the brake of the immune system to foot on the gas pedal of the immune system. What was known is that you have a much worse prognosis if you have high CD155, which isn't surprising.
What we showed, first off, and this was why we highlighted that CD155 data, we showed exactly that, that if you get on the anti-PD-1 alone, if you had high CD155, your prognosis is substantially worse than if you have low. So your hazard ratio is, like, 0.38. What we also showed is that with the anti-TIGIT doublet on top of the anti-PD-1, you basically bring those high CD155 patients back in line as if they had low CD155. Interestingly, if you go look, like, with a magnifying glass at the CITYSCAPE data, actually buried in there are the same data. So they showed a similar observation. For a number of reasons, it's less visible, but it's very clear. The reason that's important is that is a correlate with any other number of cancers.
So for example, one of the things that gives us so much confidence, but now we have the clinical data from ARC-21, but even biologically, that same negative prognostic exists in those GI cancers. If you have high CD155, you're gonna do worse on anti-PD-1 alone. It's also the same that exists in the PD-L1 all-comer patient population, where you're getting chemo plus anti-PD-1. High CD155 is a negative prognostic. So the places that you want to think about, at least as this field plays out in the near to mid-term, are places where you do see activity with anti-PD-1, where you know that CD155 is a player.
Those are the settings where you can expect to see activity, and I think that will be exploited, and it's a very broad, big opportunity on the order of what we thought when we first got into anti-TIGIT, that basically anti-TIGIT has the opportunity to turn anti-PD-1 into super anti-PD-1.
Got it. Completely makes sense. Maybe moving on to the EDGE-Gastric study. So, how do these results fit into your pipeline strategy, and what are you looking for in the EDGE-Gastric study read out in Q4 of 2023?
Right. So EDGE-Gastric, just as a reminder, that is our phase II study in gastric cancer and esophageal adenocarcinoma. It's looking at the same combination and the same, patient population as we're looking at STAR-221, which is our phase III study.
Right.
The reason we ran EDGE-Gastric was some regulatory authorities require some either safety and/or efficacy data to start a study, a phase III study in their region. So that's why we ran those two studies in parallel. But obviously EDGE-Gastric, the other advantage, is we're getting a lot of supportive data along the way. So this first data set that we'll have will be of Dom plus Zim, plus chemo. As Terry was alluding to earlier, it's in about 40 patients. We'll have two scans on every patient. Every patient will be eligible for at least two scans, and it will be ORR data and probably six months PFS data. So, you know, pretty important data sets. We're excited about the data, as Terry was alluding to.
We'll break it out by PD-L1 high versus PD-L1 low, so people will see the breakdown by PD-L1 status. We hope to show that data before the end of the year.
Is there any difference between your compound versus the Roche's one, which could make it more active in PD-1 low patients? Because Roche did not work on my PD-1 low.
No, I mean, I think, you know, we could have an advantage just across the board is in combination with chemo, where we see less AEs.
Okay
Relative to what we've seen yet. So far, we haven't seen a lot of Fc-enabled TIGIT plus PD-1 plus chemo data. I think the one data set that caught people's eye was BeiGene's data set at, I think it was at ASCO, also in gastric cancer. So exact same setting as STAR-221 at EDGE-Gastric, where they did seem to see more toxicity than what you would expect with PD-1 plus chemo. So if you look at Merck's study of Pembro plus chemo, BeiGene saw about two times the incidence of SAEs, relative to what Merck saw with just PD-1 plus chemo. So because we're seeing lower tox and you're not having to take patients off of treatment because you're seeing more tox, that could potentially translate into an efficacy advantage in that.
Got it.
So, Terry, I don't know if you want to add anything?
No, I think, well, I think the one comment that we make is this data set is interesting from a standpoint of this setting, but. Again, if you're looking and thinking bigger picture, you can think of the BeiGene data set as a surrogate. It's an anti-PD-1 plus an Fc-enabled anti-TIGIT. Not saying this in any negative way, but they're vanilla antibodies. They're blocking antibodies. They hit the target the same way. BeiGene's a little higher dose than Merck. We know that Merck, in their co-formulation, selected a 200 milligram dose, even though they explored a 200 and 700 milligram dose. We know clinically, these molecules have been shown, the one thing in humans, they do deplete peripheral T-regs. Ask 100 immunologists, they're gonna tell you that's gonna lead to immune AEs.
You do see more immune AEs, so we think on the efficacy front, backing down may, over time, depending on the setting, have some effects on efficacy. And so I think, this data set, as small as it is, will give another good read on what's being seen in general with two contemporary datasets, ours and BeiGene, in exactly the same setting, with really the only substantial difference being one's with an Fc-silent anti-TIGIT, one's with an Fc-enabled. And I think it bodes well for our doublet more holistically, when you think about the doublet being, becoming one thing, ultimately, and wherever the opportunity is for anti-TIGIT, anti-PD-1.
Got it. Completely makes sense. So, one comment and question we get from investors is that, I think the right—the ARC-10 using pembro as a competitor was the right decision, obviously, because pembro being the standard of care in that setting. One question we get on that is that: how do you prove Zim is equal to pembro here? I mean, in this trial, you won't be able to show that, but, like, what other data set you'll be generating by that time that will help people understand?
Sure. So the first thing is, from an approval and actually adoption standpoint, because we're running in every setting that we're a registrational trial, we're always running versus the global standard of care. So in both ARC-10 and STAR-121, we're running Keytruda and Keytruda chemo. In STAR-121, we'll have the best side-by-side evidence. We're running a Zim chemo arm. It's 1 to 4, so you will see some side-by-side comparative data. But all the data to date across all of the anti-PD-1s, I think, does point to the anti-PD-1s behaving as anti-PD-1s. Whether it's the registrational study that was run with Zim in China, where actually in classical Hodgkin's lymphoma, if you look at those data, numerically, probably better than Keytruda. No one would try to say it is.
If you look at a very contemporary study of Keytruda plus chemo versus the Glaxo anti-PD-1 plus chemo and PERLA, PFS for the Keytruda arm was 6-point-something months, whereas in the approval study for that, KEYNOTE-189, it was nine months change. So the differences you see in anti-PD-1 have to do with cross-trial comparisons. They're actually not just sort of these amorphous things. You can point to differences between the trial, whether it's patient demographics, whether it's how central testing, how long it took to be for patients to get the data back, and whether they had to go on to other therapies, what the standard of care was when studies were done, you know, a decade ago versus now.
So we feel very confident in Zim, and in fact, all the anti-PD-1s looking like anti-PD-1s. But to your specific question, the place where we'll have direct comparative data is in STAR-121. But physicians have told us time and time again, "You beat the standard of care, that's what's gonna get used." And in fact, you see it as we sit here today. We point out, you know, if you look at stage three lung, people use Durva. That's what was approved. If you look at the GI setting, you've got Nivo, Nivo plus chemo. That's what's used. People don't grab Keytruda plus chemo.
Awesome. Maybe moving on to HIF-alpha program. So what kind of data should we expect? What would be next update look like? And yeah.
Why don't you talk about it?
Yeah. So the first data that we'll have will be from the de-escalation. So that's from sort of nine to 12-ish patients. That would be safety, PK/PD. So demonstrating again that, you know, we're getting 3x plus the exposure that Merck gets with belzutifan with AB521. And then we'll also have probably anecdotal efficacy data, so we do have some clear cell RCC patients that are in that dose escalation. It's probably not gonna be a huge number because the dose escalation wasn't a huge number. So the sort of anecdotal efficacy data that we could have is, you know, here's a patient that received four prior therapies. You can see the response or clinical activity that we observed in that patient. We are already enrolling the expansion cohort, so we started that about a month and a half ago.
That is enrolling extremely well. So we're trying to get that enrolled and get data from that expansion cohort out as quickly as possible. So hopefully, that'll be right behind the dose escalation data. And going back to the dose escalation data, the guidance that we've given is by the end of this year or early next year, and then, like I said, hopefully, the expansion cohort data will be right behind that. And that would be 30 patients, initially ORR, and then, you know, eventually, we would have PFS data from that expansion cohort as well.
You know, I'm rarely reductionist, but this is one trial where you're gonna learn so much about the future of AB521 as a drug, which we, we believe it is a drug. And if you think about it, the Merck molecule, it's validated, it's approved, the mechanism is well understood. There's a well-defined PD marker, there's a well-defined safety profile, and it looks like they're leaving efficacy on the table because of that absorption-limited pharmacokinetics. The 120 milligram dose was selected purely because if you go to higher doses, you can't get greater exposure. So what you're gonna see from our study is a very clear, it's this fold, more exposure, it's a more potent molecule. You can see the PD marker ratio of how we can nail that maximally.
It's roughly a third to a quarter of the dose associated with the Merck drug, and that PK, PD, and safety, you know, I think bodes very well for competing directly with the Merck molecule in a very substantial way.
Great. So I'm on to the one question which I ask every company. September 2024, I hope you are here again. What would make you look back and think about the year from now till 2024, and you'll say, "It was a great year for us?
Jen.
Yeah, going back to the beginning of our discussion, you know, we've got these three data events. You know, we obviously know a little bit more than people do now about those data events. We're excited about them, but I think just getting those data events out there, you know, seeing what comes next after those data events, seeing how people react. You know, I think the one that could take people a bit by surprise in a good way is quemli. You know, I mean.
Right.
You know, TIGIT, we know it's active. I mean, that was, you know, what we see with ARC-7, SKYSCRAPER-01-
Right
Et cetera. HIF-2α, we know that's active. But Quemli, you know, is a first-in-class molecule. You know, pancreatic cancer is a huge opportunity, not just commercially, but obviously, the unmet need there is just massive. You know, you haven't had a new treatment that's been approved for pancreatic cancer, for first-line pancreatic cancer, since 2013 with Abraxane. So I think that's the thing that, in a way, we're probably most excited about because that could really be game-changing, in a way that people haven't really appreciated yet. And then we're also excited about what else we could do with CD73. I mean, you know, we think it's an active mechanism. We think that it should work very broadly in settings where chemo works, and so we think there's a lot of opportunity there with CD73, that we're just starting to realize.
Got it. Super helpful. Thank you, Terry and Jen. Really appreciate you joining us today, and I hope you all learned a lot today.
Thank you.