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

Hello, everyone, and a warm welcome to the ASCO 2025 ARC-20 Cas and Cabo Data Call. My name is Emily, and I'll be coordinating your call today. [Operator Instructions] I will now hand over to our host, Pia Eaves, Vice President of Investor Relations, to begin. Please go ahead, Pia.

Good morning. Over the weekend, we issued a press release announcing data from the casdatifan plus cabozantinib combination cohort of our Phase Ib ARC-20 clinical trial in clear cell renal cell carcinoma or ccRCC. The full release, along with today's slides, are available on the Investors section of our website. These data were presented at the ASCO conference on Sunday, June 1, 2025. Before we begin, I'd like to remind you that management will be making forward-looking statements, which are based on the current beliefs and expectations of management and are subject to risks and uncertainties that may cause our actual results to differ. We encourage you to review our SEC filings. We are joined today by our CEO, Terry Rosen; our CMO, Richard Markus; and our COO, Jennifer Jarrett. I'll now turn the call over to Terry.

Thank you very much, Pia, and good morning, everybody. I'd like to start by thanking Dr. Toni Choueiri at Dana-Farber for what we think was another awesome presentation. Yesterday's session was the third oral presentation for casdatifan at a major medical meeting in just the past 7 months. I think this speaks to casdatifan's groundbreaking potential as an RCC therapy and the high level of investigator enthusiasm for the program. Until this presentation, we've only shared data for cas monotherapy in late-line clear cell RCC from our Phase Ib ARC-20 study, and it demonstrated clear improvement across every efficacy measure evaluated relative to historical benchmark data for belzutifan. Today, we're excited to share with you the initial data from our ARC-20 cohort that evaluated cas plus cabozantinib, the most widely used TKI in clear cell RCC. Before we dig into the data, I'd like to make a few points. First off, while early, we believe these data unequivocally show that cas plus cabo should be the best-in-class HIF-2 alpha TKI combination in support of our goal to make this combination the next standard of care in IO-experience of clear cell RCC. Second, today's presentation is timely as we're on the cusp of initiating our first Phase III study for casdatifan, PEAK-1. With intense investigator enthusiasm and support and no directly competing studies today, we expect the PEAK-1 study to enroll extremely fast, and our goal is to get to data as quickly as possible. Third, with our eVOLVE-RCC02 study sponsored by AstraZeneca, we now not only have a very compelling Phase III strategy in the first-line setting, but is also cost and resource efficient for Arcus. Finally, I want to emphasize this again. We own the rights to this molecule. This gives us significant strategic optionality and flexibility. I'd also like to highlight a few things about Arcus on Slide 5. While casdatifan has emerged as the crown jewel of our development program, and we're aggressively pursuing registrational studies in both the first-line and IO-experience settings, we have 2 other late-stage programs. Our first Phase III trial of dom-zim, our anti-TIGIT anti-PD-1 antibody combination is expected to read out next year. We're also rapidly enrolling our Phase III trial of quemli, our CD73 inhibitor, in first-line pancreatic cancer with enrollment completion expected this year. As of the end of last quarter, we had $1 billion in cash and investments, and we're well positioned to advance our pipeline through the PEAK-1 readout. We benefit greatly from strong partnerships with Gilead, Taiho and AstraZeneca. This enables resource and capital-efficient means of advancing our programs as rapidly as possible. So now on to the data. Slide 7 summarizes the key takeaways. First, despite a quite limited median follow-up of only 5 months, we are already seeing robust efficacy with a confirmed ORR of 46%. Perhaps even more important than the response rate is that the responses are very durable even in this early data set. To date, in the cas/cabo cohort, all responses have confirmed and all responders remain on treatment. Additionally, both responders and stable disease patients are clearly benefiting from treatment and will contribute to PFS. This is most visible in the spider plots that we'll share later on this call. Now moving on to safety. Importantly, the AE profile for cas plus cabo is consistent with what would be expected for either agent alone. There are a few points I'd like to call out here. First, we're seeing minimal signs of overlapping toxicity, enabling us to achieve greater than 90% dose intensity for casdatifan and over 85% for cabo. This is pretty remarkable for 2 anticancer drugs administered in combination. Second, consistent with what we saw in the monotherapy cohorts, we observed no adverse effects related to cas that were higher than Grade 3, demonstrating the potential to combine cas not just with TKI, but with other mechanisms. So this is also important in the context of our collaboration with AstraZeneca. Lastly, only 5% of patients discontinued either cas or cabo due to AEs. And this is really key. No patients have discontinued both drugs. As I mentioned, is important because if a patient must continue one drug due to AEs, they could continue to benefit from the other agent. We expect that this will amplify the durability and duration of combination therapy relative to cabo monotherapy. So now turning to Slide 8. There are now 2 robust data sets that demonstrate casdatifan's potential best-in-class profile across all key clinical outcomes. On the right is a reminder of the data we presented in February, which for all 3 independent cohorts showed substantial improvement in ORR, rate of primary progression and median PFS relative to the Phase III benchmark study for belzutifan LITESPARK-005. These findings are further validated by the strength of our initial cas/cabo data, which demonstrate a meaningfully higher confirmed ORR than the 31% seen in the benchmark study for belzutifan plus cabo LITESPARK-003 with a very low rate of primary progression. All of this should translate into a long PFS with cas plus cabo combination. On the next slide, Slide 9, I'd like to remind you of the drug properties that underlie our conviction in casdatifan's superior clinical profile. Unlike belzutifan, which does not achieve meaningfully higher dose exposures above its approved dose of 120 milligrams, cas has linear and dose proportional pharmacokinetics. So this is a driver of all that clinical benefit. The manifestations of this profile are clearly and quantitatively evidenced by the deeper and more durable pharmacodynamic effects on the peripheral biomarker for HIF-2 alpha inhibition. This is the gold standard, EPO suppression at our go-forward dose of 100 milligrams. The superior PK/PD profile underlies our great clinical results relative to what has been reported for belzutifan. On the next slide, Slide 10, we show how unique the HIF-2 alpha field is in oncology because it's just a 2-horse race due to the intrinsic difficulty in creating good drugs against this target. In addition to cas outperforming the benchmark for belzutifan on every efficacy measure evaluated, we have a clearly differentiated development plan relative to that for belzutifan. Specifically, our Phase III trial PEAK-1 is evaluating cas plus cabo versus cabo monotherapy, the gold standard TKI in IO-experience patients. This is differentiated from Merck's approach in this setting where they are evaluating belzutifan plus lenvatinib with the added complexity of having cabo with the control TKI. We expect that physicians may have more difficulty managing lenva's AE profile relative to cabo, which could limit the safety and effectiveness of the combination. In the first-line setting, we're evaluating casdatifan in combination with volrustomig, AstraZeneca's anti-PD-1 anti-CTLA-4 bispecific antibody. This is the first TKI-sparing first-line regimen evaluating the HIF-2 inhibitor. And therefore, it really differs substantially from Merck's first-line strategy where they're combining belzutifan with anti-PD-1 and a TKI. On Slide 11, we have more information on this study in collaboration with AstraZeneca. The eVOLVE-RCC02 study is utilizing a seamless Phase Ib/III design, which will enable AstraZeneca and us to realize our shared goal of assessing the safety and efficacy of this TKI-free combination for RCC patients as efficiently as possible. Importantly, this collaboration enables us to pursue the first-line setting in a cost and resource-efficient manner, particularly given that AstraZeneca is sponsoring and operationalizing the study and expenses are shared. Considering the robust efficacy and favorable safety profile that we've seen with cas, we envision cas could become the backbone therapy for TKI-free regimens in RCC, and this is illustrated on Slide 12. TKIs can be highly effective and in fact, nearly every clear cell RCC patient receives a TKI in their treatment journey. But given their severe side effect profile, there's a lot of enthusiasm among patients and physicians for new therapies that delay the use of TKIs to later lines. This slide summarizes how HIF-2 alpha inhibition, and particularly cas has the potential to shift the future treatment paradigm for RCC, particularly given casdatifan's low rate of primary progression and robust response rate, which is TKI-like. I'd now like to turn the call over to Richard to review the cas plus cabo data that was presented this weekend at ASCO.

Thanks, Terry. Slide 14 shows the design of ARC-20. As you recall, in February at ASCO GU, we presented data for the first 3 cohorts evaluating cas monotherapy in late-line clear cell RCC. Today, we'll be focusing on the cas plus cabo cohort highlighted in dark green. The data cutoff was March 14. And as of this date, there were 42 patients in the safety data set, which included all patients who were enrolled at least 1 month before the data cutoff and received at least 1 dose of cas. This is also the data set provided to the FDA to support our IND filing for the initiation of PEAK-1. The efficacy evaluable population included 24 patients, all of whom were enrolled at least 12 weeks prior to the data cutoff and were, therefore, enrolled early enough to have been eligible for 2 scans. Importantly, we also include any patient who discontinued due to progression prior to the first scan. Baseline characteristics are outlined here on the next slide, Slide 15, I note that ARC-20 is a global study, and this cohort alone includes patients from 15 sites in 3 countries. Only 21% of patients were categorized as favorable IMDC risk with the remainder being intermediate or poor risk. 83% had received 1 prior line of therapy, while the remainder had received 2 or more. It's important to point out that of the patients with 1 prior line of treatment, the vast majority received a combination therapy, typically either anti-PD-1 plus anti-CTLA-4 or an anti-PD-1 plus a TKI, which is consistent with the contemporary treatment paradigm. For the previous therapy, 40% of patients had received prior TKI and IO and 60% had received only IO therapy. Both these patient populations benefited from the cas plus cabo therapy, so the prior treatment regimen does not seem to matter from an efficacy perspective. Slide 16 shows the ORR table for the cas plus cabo cohort. As you can see here that the follow-up time is relatively short with a median follow-up of just 5 months. Despite this, we are already seeing a confirmed ORR of 46%, including one partial response, which is confirmed after the data cutoff. Notably, there has already been a patient who achieved a complete response and only one patient progressed on or before the first scan. As Terry mentioned earlier, impressively, our ORR already exceeds benchmarks for the cabo monotherapy, which ranges from 20% to 40% and most importantly, the 31% ORR observed for belzutifan plus cabo in the Phase II LITESPARK-003 study. And again, this is with limited follow-up in this cohort. In fact, if you look at the first reported data for belzutifan plus cabo from LITESPARK-005, the ORR was only 22%. Slide 17 shows the waterfall plot for the efficacy evaluable population. Nearly every patient experienced some reduction in tumor volume from the cas plus cabo treatment. And as Terry mentioned, all responders and the vast majority of patients remain on treatment as of today. We've also provided a spider plot on Slide 18. This visual is highly compelling and very consistent with the durability of benefit we've seen already with cas monotherapy. As this graph shows, both responders and stable disease patients are benefiting from therapy and similar to what we are observing with cas monotherapy, we expect this benefit will be maintained for long periods of time and tumor reduction for many patients also appears to be deepening with longer time on treatment. So as you can see here, we have a few patients with tumor reduction already approaching the 30% threshold. And as a reminder, when we showed you the spider plot for cas monotherapy, we had multiple stable disease patients still on treatment for more than a year without progression. And this illustrates the durability of the HIF-2 alpha inhibition mechanism and demonstrates how stable disease patients can and will contribute meaningfully to PFS. The swim lane plot is on the next slide. And again, it shows that nearly all patients continue on treatment. This plot also illustrates some delayed responses, those that are occurring beyond the second scan, as is shown in yesterday's oral presentation. Toni Choueiri also highlighted these delayed responses, which are another one of the hallmarks of HIF-2 alpha inhibition treatment. On Slide 20, we have included some interesting data on dose intensity for both the cas plus cabo cohort as well as the 4 cas monotherapy cohorts as of their most recent data cutoffs. As you can see, cas dose intensity is close to 100% for all cohorts across a total of over 100 patients, and the results are highly consistent regardless of whether cas is administered as monotherapy or in combination with cabo. This is important as it means patients are receiving close to full doses of cas mono or both cas and cabo for the duration of their therapy, thereby optimizing and maintaining efficacy. Our investigators have been particularly impressed by these data as one reason that oncology drugs fail event-driven readouts is the inability to administer the most efficacious doses over a prolonged period. Now on to the safety results, beginning on Slide 21. As expected, the most common adverse event for cas was on-target anemia. In most patients, dose reductions or withholdings were not utilized to manage anemia until approximately 3 months or more following a patient's first dose. This means that patients are receiving the full doses of both cas and cabo early on when treatment and the ability to bring the tumor growth under control is most critical. When cas was withheld due to anemia, it was usually just for a short period of time. The next slide describes Grade 3 or higher treatment-related adverse events, of which there were very few. Anemia and hypoxia were the 2 most common Grade 3 adverse events attributable to cas. And most importantly, grades appear similar to what was observed with cas monotherapy. For hypoxia, in only one case did the patient discontinue cas, and this was compounded by a concurrent Grade 4 pneumonia. There were no Grade 4 or higher AEs that were attributable to cas. Discontinuations and dose reductions are highlighted here on Slide 23. And impressively, only 2 patients or 5% discontinued drug due to adverse events and no patients discontinued both drugs. So to summarize the data from this cohort on the next slide, the efficacy with cas plus cabo combinations thus far is robust, including an ORR that exceeds comparative benchmarks for cabo monotherapy and for belzutifan plus cabo. The vast majority of patients are benefiting from therapy and nearly all patients remain on study. The safety profile for the combination appears consistent with what is expected for each agent alone and patients have been able to maintain high dose intensity for both cas and cabo. With very few discontinuations and relatively short dose interruptions, at least one drug was typically onboard almost continuously, again, enabling maximal efficacy to be achieved. We expect that this manageable safety profile will enable patients to drive the full benefit of cas plus cabo. And this will translate into very durable responses and disease control and ultimately to a meaningful PFS that's progression-free survival improvement relative to cabo alone. As shown on Slide 25, with 8 cohorts now ongoing, we expect to continue presenting a steady cadence of data from ARC-20 over the next 18 months. In the fall, we will share more mature results from the 4 cas monotherapy cohorts in late-line clear cell RCC, including ORR and PFS data. Next year, we'll be able to present more mature data from this cas plus cabo cohort as well as initial results from the TKI-free cohorts added to the study this year. As we look ahead to the fall, I'd like to remind you of the PFS results we shared at ASCO GU this past February on Slide 26. These results are very impressive and for the 50-milligram BID and 50-milligram QD cohorts combined, the median PFS was 13 months. Now with more than 12 months of follow-up for the 50-milligram QD cohort, a median PFS was not reached. And as of today, this cohort and the 100 mg QD cohorts have still not reached a median PFS. This bodes very well for the PFS results as these data sets continue to mature. Next, I'd like to review our development plan for casdatifan in more detail. The design for PEAK-1, which is on track to begin imminently is shown on Slide 28. At our earnings call in May, I highlighted the important differentiators of the study relative to Merck's LITESPARK-011, which is being conducted in a similar IO-experience population. Without going into the details again today, we believe our study has a better design that optimizes for both probability of success and speed to market. As Terry mentioned, in the first-line setting, we are pursuing the combination of cas plus volrus in collaboration with AstraZeneca. The eVOLVE-RCC02 study will be sponsored by AstraZeneca and is expected to initiate soon. The study is designed as a seamless Phase Ib/III trial and the details will be available on ClinicalTrials.gov any day now. We could not be more excited about this collaboration and the opportunity to work with AZ on another study. On my last slide, Slide 29, I'd like to highlight the new cohorts added to ARC-20. As of today, over 200 patients have received casdatifan across all cohorts in ARC-20. The 3 new cohorts you see here are designed to evaluate cas as PKI-free regimens in first and second-line clear cell RCC. Results from these cohorts could support the expansion of casdatifan's development program and our greater vision for cas, which is for all ccRCC patients to receive a HIF-2 alpha inhibitor and for casdatifan to be the HIF-2 alpha inhibitor of choice. So with that, I'll turn the call over to Jen to discuss the market opportunity for cas in clear cell renal cell carcinoma.

Thanks, Richard. I'll start with Slide 31, where we show the rapid growth of belzutifan, which Merck markets as volrus, since its approval as monotherapy in clear cell RCC. This is a very encouraging early indicator for the market potential for HIF-2 alpha inhibitors in RCC, particularly since belzutifan is used primarily in the third-line or later setting today. We believe belzutifan's use was primarily in later line due to its high rate of primary progression, which was over 30% in LITESPARK-005. Despite this limitation, last 12 months sales were over $560 million and maybe even more importantly, run rate sales based on the most recent month of scripts is at an impressive $720 million, and Merck has been tolling volrus as a potential multibillion-dollar peak sales drug. With cas' superior profile and its potential to achieve long duration of therapy in earlier line settings, we believe cas greatly exceeds the revenues that you see here. Moving to Slide 32. Here we show the market share for the currently approved treatment regimens for clear cell RCC. This slide highlights that our initial Phase III studies for cas, eVOLVE-RCC02 and PEAK-1 target the largest segment of the clear cell RCC market. If you were to assume that we displace only the current regimens that we are combining with, so anti-PD-1 plus CTLA-4 in the front-line setting and cabo in second-line, we would capture 32% of the front-line market and 29% of the second-line market. This alone would translate into a multibillion-dollar revenue opportunity. On top of this, we believe we can take share from other regimens such as anti-PD-1 plus TKI in the front-line, and I'll talk more about why we believe this is the case on the next few slides. First, on Slide 33, we describe why cas plus cabo is poised to become the new standard of care in IO-experience clear cell RCC. First, cabo is the preferred TKI in clear cell RCC and its use is growing. This is due to its more benign safety profile, particularly less hypertension and its simpler dose titration schedule relative to volrus. Clinicians have repeatedly told us how comfortable they are using cabo and in particular, managing the toxicity of cabo to optimize its clinical benefit. This is the advantage of using cabo over a less well-understood TKI. And today, cas is the only HIF-2 alpha inhibitor being developed with cabo. Additionally, with the 2-drug combination, given the lack of overlapping toxicities, we significantly increased the probability that a patient remains on at least one drug at all times. This is in contrast to TKI monotherapy treatment today where doses are often withheld due to toxicity, potentially preventing patients from receiving maximal clinical benefit. Last, with the delays in the LITESPARK-011 readout and given our use of PFS as a sole primary endpoint, we have significantly closed the gap between the timing of potential readouts and therefore, potential approvals for LITESPARK-011 and PEAK-1. On the next slide, we highlight why cas has the potential to be combined with every standard of care agents in RCC, and that is because of its lack of overlapping toxicities with other mechanisms. Even anti-PD-1 and TKIs have several overlapping toxicities, notably GI-related, hepatotoxicity and rash. In contrast, other than fatigue, cas does not appear to have overlapping toxicities with these other agents. This is why we believe the HIF-2 alpha mechanism, specifically cas, could become the backbone therapy of choice across the RCC treatment regimen. In the first line, as you heard from us today, our initial strategy is to combine with anti-PD-1 plus anti-CTLA-4 using AstraZeneca's bispecific antibody. Clinicians have a strong preference for TKI-free regimens in earlier lines of therapy, which is why ipi/nivo garners over 30% of the market and its share is growing. The key limitation of ipi/nivo in the setting is this approximately 20% rate of primary progression, and we believe that by combining cas with AZ's anti-PD-1 CTLA-4 bispecific, we can bring this number down, further improve PFS and OS and with minimal additional toxicity. On Slide 35, we highlight the market potential for the key segments we are targeting with our first Phase III studies. Together, these 2 segments alone represent a massive $5 billion addressable market. A key driver of the significant market opportunity is the potential for cas combinations to result in extremely long durations of treatment, particularly given cas' relatively favorable safety profile. This is huge for patients to achieve meaningful clinical benefit for many months or even years and without significant additional toxicity. I'll end on Slide 36, where we illustrate how the use of cas should evolve over time, starting with IO-experience patients with PEAK-1 and then to first-line patients with our eVOLVE study and ultimately potentially displacing TKI-based regimen. Taking into account the long durations of therapy expected in the patient population shown here, this obviously translates into a huge market opportunity for cas. I'll now turn the call back to Terry.

Thank you very much, Jen. As you can see on Slide 38, we're extremely excited about the potential for casdatifan as the best-in-class HIF-2 alpha inhibitor. Given the enthusiasm around this mechanism of action in general, for casdatifan in particular, we believe we're well positioned to capitalize on a huge market opportunity in clear cell RCC, which today is dominated by only 2 classes of therapy, IO and TKIs. As we've said before, our vision is that every clear cell RCC patient will receive a HIF-2 alpha inhibitor during their treatment journey and for cas to be that HIF-2 alpha inhibitor of choice. And we'll continue to share more data on cas over the next 18 months, which will support this vision. We have tremendous appreciation for your continued interest in and your support of Arcus. And so we'll now open the call for questions.

[Operator Instructions] Our first question today comes from Yigal Nochomovitz with Citi.

Congrats on the nice data. So this is very interesting. You have a 46% response rate at 5 months. I'm just curious how you're thinking about how that may evolve given, as you pointed out, the majority of patients are still on therapy and the combo with belz/cabo is a much basically fivefold longer follow-up at 31%. Any thoughts on that? Or to ask the question the other way, what was the response rate with belz and cabo at 5 months? I'm assuming it was lower than 31%.

So thank you, Yigal. The first point I would make, we'll see how it plays out. Obviously, -- the -- if you look at the waterfall plot or any of the plots, but I like the waterfall plot because they have all these green arrows under it. Clearly, stable disease patients are benefiting. The vast majority of patients have tumor reduction and still are on study. So you might expect it to go in a positive direction. The onetime point that we can comment on with respect to the belzutifan/casdatifan study is when they presented their earlier data, that was with about 12 months of follow up, again, at work at 5 and change, and they had a 22% response rate at that point and then went on to the 31%. So we'll see how things play out, but things look good as we sit here today.

Okay. And then sort of more conceptually, my question is, this is the first combo data with cas with the TKI. Just curious if this gives you any clues in terms of how you're thinking about what you might see with the combo efficacy. Obviously, I know it's with an IO therapy, but with volru, does it give you any clues as far as what that might look like with that combo?

So I think the only extrapolation that we would make as we've discussed that is the fact that it did emerge that the AE profile of the combination is so A plus B looks like A plus B that we feel pretty confident going into the combination with volrus that we'll see the same again because there's no reason a priority, particularly with these data in hand, I think as we combine with the IO that also doesn't have any overlapping toxicity that we would see anything symbiotic in that front. And I think that's reinforced by really just the intrinsic AEs that you do see from casdatifan are simply those on-target AEs, anemia more frequently, hypoxia, less frequently and nothing else. So I think that bodes well. The other thing that's clearly happening with time that I think will continue to favor us is the physicians are getting so comfortable in this community in managing even those minimal AE and hypoxia. And that's why when you look at that dose intensity and lack of any patients coming off of both therapies, it really, I think, bodes well for other combinations as well as we think about this in a future sense. And we are -- as we talk about, planning to leave no smart stone unturned.

Our next question comes from Peter Lawson with Barclays.

I guess the first one is just on, I guess, safety, just feeding into the prior questions. Are there any safety effects that build over time that could preclude combinations? And then the second question would just be around go-to-market strategy and kind of whether you need a partner to commercialize this?

Peter, this is Richard. I'll take the first question, and I presume Jen will take the second one. But as far as your question on safety over time, not really -- the short answer is not really, meaning we really -- looking at our cas monotherapy data for which we have patients that have gone on longer, there's nothing really that starts to emerge later. So it's really, again, the anemia and some patients with hypoxia and what we're really learning and physicians are getting more and more comfortable with is the anemia factor and how they can actually just treat through it and keep patients on casdatifan some EPO support or a quite short break from cas and then restarting and continuing and often even in that case, the anemia may not come back. So they are getting better and more comfortable with even treating and continuing cas through an anemia. So that's really what we're seeing even in the monotherapy where we have longer experience now.

Yes. And just one other thing to add on that. One of the things that investigators have pointed out is that the toxicities for TKIs tend to show up pretty early. And so when they look at our data, which is obviously still on the early side, that was something that gave them a lot of confidence that you shouldn't see a lot more safety events over time. So we're super excited about what we're seeing from a safety perspective. I think to your question on the market opportunity, I mean, that's kind of like the beauty of RCC. This is definitely a market that we can commercialize a drug in on our own. If you look at the market today, I mean, there's really only 2 other classes of therapies that are used, which are IO therapies are anti-PD-1 and CTLA-4 or TKIs. And so it's actually a much simpler market relative to a lot of other oncology spaces out there such as lung and breast, where it's first line, second line, third line, et cetera, there's now a lot of different options. So we're very, very comfortable that we can commercialize this drug on our own and that we can develop this drug on our own. And that's to evolve. One of the beauties of that study is that AstraZeneca is operationalizing the study, as we talked about for the first time today, it is a Phase I/III study. So the idea is that we enroll the Phase I portion, do a quick look at safety, make sure we're not seeing anything, and then we'll open up the Phase III portion. So you're not having to start a Phase III study from scratch. So it's a really, really efficient way for us to pursue the front-line setting. We also mentioned that there's cost sharing. So yes, so like I said, very, very comfortable about us taking casdatifan forward on our own.

Our next question comes from Daina Graybosch with Leerink Partners.

In the discussion at ASCO yesterday, the topic of OS came up. And with the assertion that as there's more competitive therapies and options in RCC benefit on OS matters more. So you have OS as a secondary endpoint. And I wonder if you could talk about how -- whether you're powered to show any benefit on OS as that secondary or a trend? And what you think of the assertion that you may need to be competitive?

Daina, it's Richard. Thanks for that. That's a question for sure that everyone did hear yesterday, and we're very confident in our PFS as a primary endpoint, and it's quite clear that we can take this all the way with the registration with PFS as the primary endpoint. But as you, I think, know as well, the OS is already the key secondary endpoint. So it's all prespecified. We'll be evaluating OS and have that data also when we evaluate PFS. But the key for us is that the PFS -- the statistical aspect, all the kind of alpha goes to PFS to have the best chance there for the success. But OS will be available for us at that time as well.

Our next question comes from Asthika Goonewardene with Truist Securities.

I want to dig back into some of the questions that Yigal asked first, but add on to that. Can you maybe talk a little bit about what you think the potential contribution is of cabo in that 46%, just because we know cabo has a very strong upfront effect? So I just want to -- and based on what you've seen and how the data evolved, how much of that was driven by cabo and how much of that is in casdatifan? And maybe the related question is, when do you start seeing the real casdatifan effect start to come in and the synergy coming down? And then second, let's go back to the safety here a little bit more because we know with cabo, the 50-milligram dose, you get a lot of upfront toxicities, which you guys have talked about already. But you didn't see that with this -- with that level of toxicities with the data that was presented. Can you maybe hypothesize as to why that is?

So let me start with the first component and talk about how we see the 2 drugs combining and not try to quantitatively say X percent is from this and Y is from that. But particularly, while we're seeing the really strong start on ORR, particularly want to tie it into what one really cares about in the end and what's going to be the approvable endpoint, which is durability and more specifically progression-free survival and ultimately, OS, as Daina mentioned, we certainly hope to have benefit there as well. So I think you can start working backwards since what's most sort of well-defined is cabo alone in that patient population. That's where you have the largest and the largest experience. And so if you were to start with a blank piece of paper, just think what does cabo therapy look like clearly, you're getting a certain response rate that's somewhere in the 20-plus percent, maybe as high as 30% or 40%. And -- but what you see is the durability that's only about 10 months or so PFS. So when you start to think about that and you parse that population out, you can essentially have -- you'll have some that are -- have that quick response to cabo. I would draw the analogy to chemo. So I would sort of draw an analogy to this combination, the chemo plus immuno and this is cabo TKI plus HIF-2 alpha. And so you got that one component that has early response. You have another that kind of has a weaker response and then you have some of those that basically don't respond well at all. And so what I think what you get here that's really positive is you have this orthogonal biology and orthogonal profile. So the HIF-2 inhibition, as we've shown, our molecule can start affecting things early. And you will see that in some cases that you get this amplification. We had a number of early responses. But as Toni pointed out yesterday in his presentation, you're seeing responses all along the spectrum of timing. And I think that starts to give a sense of how the combination is effective. So some of those patients that, let's say, would have responded very well to TKI. I think what you're going to see with those patients is this will enhance the durability when you bring in the duration that HIF-2 brings as a therapy, particularly because it is so well tolerated. Then second, when you think about those patients that are maybe having a, call it, a limpy response to the TKI, that's going to be strengthened by having, again, an orthogonal mechanism. Keep in mind that over 90% of tumors in clear cell RCC have some component of HIF-2 as a driver. And then you're going to have those patients that really aren't responsive very much at all to the TKI and HIF-2 is going to help there. But I think the thing to think about when you really want to synthesize the data out there, if you look at our monotherapy data, and you look at the durability that we're already seeing in a later line population, both the stable disease patients and the responders, and I'll remind you, I think Richard mentioned that in both our 50-milligram once-a-day and 100-milligram once-a-day cohorts, we still haven't reached median PFS. So I think if you project that on to what we're already seeing, the place that the HIF-2 alpha is really going to change things for patients is that durability and the PFS, and that's how we look at those 2 mechanisms combining. I know that was a long answer, but I still remember your second question, which related to the safety aspect of what we've seen to date and does it seem like there's any difference. And I'll let Richard and/or Jen comment on that.

Yes, I can comment on safety-wise. I think the question was the cabo safety we're seeing in particular in this case. So I don't think we're seeing anything quite unique really from the cabo group. It's early in our study. So we'll continue to track all these patients, and we'll track them for quite a long period of time, as Terry said. So I don't think there's anything other than this is a specific set of 42 patients. And what we're seeing is high tolerability of the combination, which, as Terry said, bodes really well for the durability. But I don't think there's anything really special or unique safety-wise that we're seeing out of cabo here.

The one thing that I would add on that, that I just think is sort of an aspect of adjusting how the world looks at things. The one comment I would make is that we're so used to seeing anticancer therapies where the combination, the second agent brings some advantage, but also exacerbates the AEs. And so given the -- again, I'll use the term, the orthogonal biology and the orthogonal clinical manifestations, you really do get advantages from the HIF-2 mechanism without bringing a whole bunch of additional AEs. So you're certainly less likely to see anything that looks worse than you might have seen with the other agent alone. And as Richard said, as time goes along, we'll see if there's any difference or we just kind of merge it baseline cabo.

Yes. The only thing we are probably benefiting from is just clinicians have so much experience with cabo and now with a growing amount of experience with HIF-2 in addition. So that is probably helping as well that they're just getting better and better at managing the toxicities, premedicating, doing whatever. And so that is the big advantage of using a TKI like cabo that clinicians are so experienced with.

Our next question comes from Li Watsek with Cantor Fitzgerald.

This is Dan on for Li. Can you elaborate a little bit on the differentiation in the post-IO setting in ARC-20, where you have 2 cohorts, one is the cas/cabo combo that you presented yesterday and then one is cas monotherapy? And how do you think about the differentiation in the market as well as what are you expecting in a different efficacy or safety profile?

I could comment briefly on what we think, as you said, the latter part, and then Jen can take on the front part of the question there. But -- so we have both in post-IO setting, a combination with PD-1 or zim with cas versus, in this case, cabo. So I think we are pretty clear with what we're seeing now in the cas/cabo, as Terry said, the orthogonal aspect, and I think the same kind of bodes for the zim. So the PD-1 also doesn't have overlapping toxicities with that of cas. So I think they're both very viable combinations for the post-IO setting, especially from a safety aspect.

Yes. The big difference between that newly added cohort is that patients do not have had prior TKI. So these are all patients that either got pembro adjuvant, which is kind of a new patient population to be tapping into or patients that got anti-PD-1 plus CTLA-4 or maybe they just got anti-PD-1 in the front line. But the idea with this cohort is to sort of leapfrog TKI usage because like I said, these are all patients that have not seen prior TKI yet. So we're trying to test the assumptions that we could displace TKIs in earlier lines of setting with this cohort.

I think that's a real important part of the strategy. I think it's a real important part of the differentiation when you think about the rate of primary progression that you see when you look at casdatifan versus belzutifan, that sort of enhances that possibility of going earlier, so flipping the paradigm, which is what physicians and patients would love that you get the TKI late as opposed to the TKI early. And so because we have that low rate of primary progression, it does allow you to move earlier and get into a paradigm shift, if you will, where you don't require the TKI just, again, using that same analogy I was talking about before, where the TKI somewhat behaves like chemo in getting that initial response. And in fact, that was one of the things -- when you think about the combination with volru, it's -- ipi/nivo still has a roughly 20% rate of primary progression. So one of the advantages of combining with casdatifan, and I think that's why it's such a nice combination with AstraZeneca is really, I think, a smart combination is that you get that opportunity to expand on the CTLA-4 PD-1 axis with something that may allow you to go into a broader population with the HIF-2 inhibitor as opposed to thinking about what is generally more toxic to try to combine that with the TKI regimen.

Our next question comes from Emily Bodnar with H.C. Wainwright.

I was curious if you could provide a bit more context on the patient complete response and if there was anything noteworthy about the patient's baseline characteristics? And then also, if you can touch a bit on the prior lines of therapy, so majority of patients had only prior IO. Is that kind of what you're expecting to see for the PEAK-1 trial as well? And I believe you've also brought up potential for patients who had IO just in the adjuvant setting. So is there anything you can kind of extrapolate from this data to what you would expect to see in those patients?

Thanks. I'll try and hit all those. I know the first one was the patient with the complete response. I don't think there was anything particularly unique, not yet certainly identified by us for that patient. We do expect some complete responders, and this one did come pretty early, which was great. But I don't think there's anything unique about that patient's characteristics. Your comment or question about what we're seeing in IO and it was more -- I think you said in the adjuvant. I wondering what we're seeing -- expect in PEAK-1 is probably our -- we do expect this population to be reflected in what we enrolled in PEAK-1. So we do think this is I'll say, an early look at what we might see in PEAK-1 is, I guess, the best way to say it or think about it. I think we had about 20% of the patients with the adjuvant setting IO therapy. And I think that's probably a reasonable estimate of what we would see in PEAK-1 as well.

And 2 other things to call out is probably the most important prognostic factor from a demographic perspective is the IDMC risk. And so if you look at our mix versus pretty much any other study out there in this patient population, so LITESPARK-003 and KEYMAKER-U03B and it looks almost identical to those. And then when we look at the patient population that had just prior IO versus patients that had prior IO plus TKI, the ORRs were pretty similar. We didn't -- the ends are still small, so we didn't want to make too much out of it, but it doesn't look like if you just got prior IO or you got prior TKI, that meant that you were going to do better. So that was definitely very encouraging to see, especially the patients that got prior TKI did just as well as patients that got just prior IO. And then the last thing that I make -- last point that I make that I think is a little bit different about our study population is that my guess is and we tried to look at the LITESPARK-003 patient population KEYMAKER-U03B is that more of our patients got both combination therapy upfront versus getting anti-PD-1 and TKI sequentially. And then we also believe we have more patients that got like second-generation TKIs upfront. So almost all of our patients got either prior lenva or axi versus my guess is if you look at some older studies, they are getting more prior cabozantinib and sunitinib to just more of the first-generation TKI. So those are just some nuances that is pointing out.

Our next question comes from Jason Zemansky with Bank of America.

Congrats on the update. I just did want to follow up on the cabo safety profile. I mean, if you look at the LITESPARK-005, the rates of discontinuation and dose interruptions for cabo alone were about 75%, but it looks like in ARC-20 it was about half of that, which is quite stark. Is the assumption that most of this is due to greater experience with cabo and the addition of cas here? Or is it something else?

So I think that Richard stated it well. And I think as Toni pointed out yesterday and just you kind of heard that discussion by Jen that we think the patient population is very similar. The one thing -- to your point, like if you look at those data, as they sit here today, which, again, I think Richard aptly pointed out, we're early, and we'll see how things evolve. But what we would say is if after the fact you had those numbers look like that and someone you felt there was truly some advantage, the one thing that we would think could come into play, and that goes back to what I was talking about of having a second agent that adds a lot of benefit, but doesn't come with toxicological baggage is that it is known if you talk to physicians that treat patients that from a prognostic standpoint, patients that get tumor reduction early, start to get healthier and stronger and better able to fight their cancer. And the problem is that most of those therapies are poisoning them at the same time as they're helping them. So if you're giving a TKI alone or you're giving chemo alone, you're sort of making the patient sicker at the same time as you're sort of treating the cancer. And in this case, you're getting an enhancement in the efficacy and the patient starts to get better and stronger. And after the fact, if you were to see some advantage in terms of the profile of the other agent from an AE standpoint, it would be likely due to the fact that you're creating a stronger, better patient as you go along. And think of it this way, if you took a healthy young person and gave them a TKI, it won't be a great thing, but they probably tolerated a lot better than your great grandmother. And so if that were to play out, we would rationalize it, not just some random happening, but the fact that your combination agent is bringing benefit and allowing the patient to become stronger and one of those advantages of being stronger is more tolerant. But we're not ready to say that we think there is a difference yet.

Got it. It looked like yesterday in describing PEAK-1, they mentioned that you're going forward with a 60-milligram dose. Any potential that you'd lower that to 40 milligram?

Yes. No, we're starting with 60. We have full dose cabo and full dose cas. We think that it's well tolerated as a combination with that dosing. So that is certainly what we're doing in PEAK-1. And obviously, they can be dose reduced if needed, and they would follow the normal cabo dose reduction algorithm, but we're definitely going to be starting with a full dose of both agents.

Thank you. We have no further questions. And so this concludes our call. We thank you all for your participation. You may now disconnect your lines.