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

Good morning, everyone, and welcome to the ARC-20 Phase 1b Casdatifan Data Call. My name is Emily, and I will be coordinating your call today. [Operator Instructions] I will now turn the call over to our host, Pia Eaves, Vice President of Investor Relations, to begin. Please go ahead.

Good morning. Today, we issued a press release announcing data from our Phase 1b ARC-20 clinical trial of casdatifan or cas and ccRCC. The full release, along with today's slides, are available on the Investors section of our website. These data were presented at the EORTC-NCI-AACR conference earlier today by our lead investigator, Dr. Toni Choueiri of Dana-Farber Cancer Institute. Turning quickly to Slide 2, I'd like to remind you that on this call, management will make forward-looking statements which contain risks and uncertainties. These are outlined in our SEC filings which we encourage you to review. I am joined today by our CEO, Terry Rosen; CMO, Dimitry Nuyten; and CCO, Eric Matthews. We are also joined by Dr. Rana McKay from the University of California, San Diego. As a specialist in renal cancer, she will speak to the current treatment paradigm for RCC and the significant unmet need that remains in this disease. For Q&A, we will be joined by our President, Juan Jaen, and COO, Jennifer Jarrett. With that, I will turn it over to Terry.

Thanks very much, Pia. And thank you all for joining us this morning. First off, I want to say thanks to Dr. Toni Choueiri who earlier today gave an outstanding presentation of the initial clinical data from our ARC-20 study evaluating casdatifan, our potential best-in-class HIF-2α inhibitor and clear cell RCC. Thank you so much, Toni, for the awesome presentation, and more importantly, for the great work that you and your colleagues do on behalf of patients. This large and robust data set includes efficacy data from our first 2 expansion cohorts. These included over 60 patients, all with clear cell RCC. We really couldn't be more excited about these data. They clearly demonstrate cas's potential to play a meaningful role in the treatment of clear cell RCC. So let me get right into the new data and start with Slide 5 of our presentation. This summarizes key elements of cas' differentiated profile from that of belzutifan, the only approved HIF-2α inhibitor. Data from the 100 milligram cohort of ARC-20 showed a substantially reduced rate of primary progressive disease and a higher ORR relative to the results generated for belzutifan in the registrational study LITESPARK-005. While the data for the 50 milligram cohort are even less mature than for the 400 milligram cohort, we believe these data clearly show that this dose, that's half of our planned go-forward dose will also show differentiation relative to the profile of belzutifan. Median PFS has not been reached in either cohort, and you'll see later on this call that the swimlane and spider plot analyses suggest that the median PFS will in fact be higher than that of belzutifan. Note that I said the median has not been reached, not just that the median is immature. In addition, we observed a comparable safety profile, particularly with respect to grade 3 or greater hypoxia and anemia, the two primary on-target toxicities associated with HIF-2α inhibition. It's a very important part of the profile. We're also pursuing a differentiated combination strategy for cas. Last month, we had a successful type B meeting with the FDA, and we're now full steam ahead toward the initiation of PEAK-1. That's our first Phase III trial for cas. The PEAK-1 study will evaluate cas plus cabo versus cabo in patients who previously received anti-PD-1 therapy, a very simple design. We also recently announced the clinical collaboration with AstraZeneca to combine cas with their anti-PD-1 anti-CTLA-4 bispecific antibody volrustomig or Volru. Importantly, Merck is not currently pursuing this particular combination with belz. So by combining cas with Volru, we could have a potential best and first-in-class combination therapy. AstraZeneca will be operationalizing the study, makes it particularly resource efficient way for us to pursue the opportunity. So turning to Slide 6, many of you have asked how can cas achieve an improvement in efficacy measures without a higher incidence of AEs. So I want to comment on the link between HIF-2 biology, on-target side effects and anti-tumor activity. HIF-2α alpha as a transcription factor, is involved in the regulation of literally hundreds of genes, each with its own dose response, sensitivity to HIF-2α inhibition. The easiest way to make this point is to compare the 2 on target AEs that everyone is aware of, anemia and hypoxia. As we all know, robust HIF-2α inhibition leads to some level anemia in the vast majority of patients, while hypoxia occurs in far less frequency. That is because the pathways downstream of HIF-2α inhibition driving these 2 AEs are completely different. So the bottom line is that the pathways downstream of HIF-2α inhibition that drive anti-tumor activity are also different from the pathways that drive the on-target toxicity of anemia and hypoxia. And this is why we believe we can achieve greater anti-tumor activity without a higher incidence of on-target AEs. And frankly, the data now speak for themselves, it's no longer theoretical. Slide 7 summarizes the HIF-2α inhibitor landscape. It's extremely sparse. In fact, we believe it's now just a 2-horse race between us and Merck. Other than belzutifan and cas, only 2 other HIF-2α inhibitors have been reported to be in clinical development in the U.S. And recent data presented suggest that neither molecule will have a competitive profile. This reflects the fact that because HIF-2α is a transcription factor, it's exceptionally difficult to design a good small molecule inhibitor against the target. In contrast, casdatifan has an ideal PK and PD profile and appears to approve efficacy without compromising safety, all with simple once daily dosing regimen. There are multiple anti-PD-1s and TKIs approved and used today for RCC. And we believe that there will be appetite for multiple HIF-2α inhibitors. With our better molecule and corresponding development strategy, we're very well-positioned to compete with the only other meaningful player, belzutifan, and to expand this already multi-billion dollar market. So on Slide 8 we highlight the differentiation of cancer's pharmacokinetic and pharmacodynamic profile underlying its clinical advantages. On the left, the black dotted line represents erythropoietin, and I'm going to call it EPO from now on. EPO suppression for belzutifan at its approved dose of 120 milligrams, and the purple line represents EPO suppression achieved with cas in increasing doses. These data show that at 20 milligrams, that's 1/5 of our go-forward dose, we achieved the same pharmacodynamic activity as measured by EPO suppression as that of belzutifan at its approved dose. Therefore, with 100 milligram dose of cas, 5x the PD equivalent of belzutifan, cas is designed and in fact does hit HIF-2α far harder than the tumor. Second, as you can see on the right, unlike belzutifan, cas does not have dose-limited pharmacokinetics. They're linear, they're dose proportional, all the way up through a 200 milligram dose, the highest dose we've evaluated in our dose escalation study. At steady state, belzutifan does not achieve meaningfully higher drug exposure as the dose increases from 120 milligrams to 200 milligrams. So simply put, increasing the dose is not expected to improve target engagement or efficacy. That's the problem with absorption-limited pharmacokinetics. Slide 9 shows some very important hot off the press pharmacodynamic data, which very tangibly illustrate these advantages of cas that we are -- that we believe are driving its improved efficacy profile relative to that of bel. These are very new data. I think many of you haven't seen these data. They're actually included data from Merck. On the left, you'll see some recently published data showing EPO suppression over time for belzutifan. And by week 13, EPO suppression is minimal. It's almost nonexistent. In contrast, on the right, we show the same analysis for cas, that cas achieves approximately 60% suppression of EPO, very close to a maximal effect over a very sustained period of time. These data reinforce our confidence that cas can have a differentiated efficacy profile over that of belzutifan and represent a very clear demonstration of the PK, PD advantages of cas over bel. These new data are both striking and compelling. See, this might actually be one of the most important slides in the deck. Slide 10 provides the punch line. While we recognize the caveats of cross-trial comparison, data from the 100 milligram expansion cohort, which is our planned go-forward dose of cas demonstrated differentiation on every available key efficacy measure compared to the published data for LITESPARK-005, specifically primary progressive disease rate and ORR. And although median PFS has not been reached, it's clear from the data to date that cas will exceed the 5.6 months medium PFS for bel. Further data maturity will determine by how much. Importantly, this is without any meaningful difference in safety, as you'll see in the last 2 columns. All of this was achieved despite a more advanced patient population and with shorter follow up relative to that of LITESPARK-005 as well as other belzutifan studies. On Slide 11, we show how the response rate could likely continue to improve as the data set matures. Specifically, this slide shows the bridge between our confirmed response rate and unconfirmed response rate. For the 100 milligram cohort, there are 2 pending responses. While not all pending responses will definitely confirm, every confirmed response would increase the confirmed ORR by just over 3%. Also in both cohorts, there are a significant number of patients with stable disease that we always call still in play. Given the IO-like kinetics of this mechanism, we can certainly expect additional conversions to responses with more follow up. And even more importantly, the stable disease patients in both the 50 milligram and 100 milligram cohorts are deriving meaningful clinical benefit, and they're going to contribute to median PFS. In summary, we have already exceeded the bel's confirmed ORR in the 100 milligram cohort. In addition, the 50 milligram cohort, so a second expansion cohort is on track to exceed the bel's benchmarks as well. On top of that, we believe the data suggest there's a pretty clear path to exceeding a 30% confirmed ORR as the data sets mature. Overall, these data demonstrate that cas has the potential to change the standard of care, capture substantial market share and further increase the RCC market by extending treatment duration. I'd now like to turn the call over to our guest, Dr. Rana McKay of the University of California of San Diego. Dr. McKay has significant experience in the treatment of clear cell RCC, in the development of novel agents, including belzutifan. She will also be a member of our steering committee for PEAK-1. We're very lucky to have her today with us, and we're thrilled to have her involved with our program. Rana?

Thank you so much, Terry, for that wonderful overview, and really happy to be here today with you all. Beginning on Slide 13, the treatment landscape for patients with metastatic renal cell carcinoma has been largely evolving over the past several years, and combination immunotherapy regimens are now the frontline treatment standard for the majority of patients with advanced disease, for patients with intermediate and poor risk disease, either IO/IO or IO/TKI are both life prolonging options for these patients. For those individuals who have favorable risk disease, the standard practice is to offer combination therapy with IO/TKI or IO/IO. And in unique situation, TKI monotherapy. IO-based combination therapy has improved survival for patients with advanced disease. However, only a small subset, maybe 10% of patients, derive long-term benefit, and the bulk of individuals, 90% of patients, progress to lethal, refractory disease. In the post first-line setting treatment options largely include TKI monotherapy, the combination of TKI and mTOR inhibition, and most recently, belzutifan, the first HIF-2α inhibitor approved for this disease, introducing an agent with this novel mechanism to action for the first time in RCC. Unfortunately, in the later line setting, treatment options are palliative in nature with the goal of improving survival, delaying progression and maintaining quality of life. TKI can be associated with significant toxicity, including chronic fatigue, diarrhea, rash, mucositis, weight loss and other side effects and unfortunately, long-term, durable benefit to any one of the given regimens is largely lacking. Now on Slide 14, there remains a significant unmet need in the second line setting following IO combination treatment, given that IO resistant RCC is a universally lethal disease. The current standard of care is largely TKI monotherapy with c cabozantinib, based on original data from the Phase 3 METEOR trial, which tested this agent post VEGF TKI, and from the control arm of several Phase 3 trials in the modern era, following receipt of frontline IO agents. The PFS is relatively short, ranging from 7 to 11 months across these studies and treatment resistance is inevitable. All patients develop progression due to either intrinsic or acquired resistance. This unmet need creates a significant opportunity to investigate combination treatment strategies that delay the time to progression, deepen responses and reduce the rate of primary progression. Ideally, we can do this with combinatorial strategies that don't have overlapping toxicities and preserve quality of life. I'm very excited about a strategy that leverages the existing benefit of VEGF TKIs with the addition of a HIF-2α inhibitor that has the potential to decrease resistance, deepen responses, delay progression, while also maintaining quality of life. Turning to the next slide, Slide 15, the approval of belzutifan has provided an important new treatment option for patients who fail on anti-PD-1 and TKI therapies. Belzutifan is a breakthrough therapy for patients with better tolerated safety profile than those of TKIs, specifically no hypertension or other typical TKI-related toxicities such as rash or diarrhea that can have a meaningful impact on a patient's quality of life. Here we highlight the data from the LITESPARK-005, the registrational study leading to the approval of belzutifan in late-line clear cell RCC. These data establish belzutifan monotherapy as a new standard of care in late-line clear cell RCC, but there is clearly opportunity for improvement. First, one of the greatest limitations of belzutifan is its very high rate of primary progressive disease, specifically 33.7% of belzutifan -- for belzutifan compared to 21.5% for everolimus. So 1 in 3 patients in the clinic who received this agent will have no benefit for therapy, and there is no way to identify these patients upfront. This is a major limitation of belzutifan, and many patients are not able to go on to receive additional treatment after they progress on belzutifan. There is a tremendous opportunity to improve on this endpoint. Second, belzutifan is showing an objective response rate of 21.9% after more than 18 months of follow up. And the existing efficacy demonstrated with cas and treatment refractory disease, there is the potential to achieve a higher response rate with casdatifan. Lastly, while belzutifan demonstrated a hazard ratio for PFS of 0.75 and a medium PFS. The medium PFS was only 5.6 months. We expect that by achieving a lower primary progression rate and improved objective response rate casdatifan treatment should result in improved progression-free survival. My last slide, Slide 16, highlights the segments of the first and second line RCC market. As you can see, there's a laundry list of different agents available for physicians to choose per the guidelines. Despite all of these options, about 1/3 of patients are given cabozantinib monotherapy in the second line setting. The cas development plan targets the largest segment of the second line treatment and has the opportunity to take share of other segments as well. Importantly, nearly 80% of patients will not have received cabozantinib in the front line and are therefore eligible to receive cas-less Cabo in later line setting. I want to also touch briefly on the Phase Ib study where AstraZeneca and Arcus are combining an anti-PD-1 CTLA-4 bispecific with casdatifan. Clinicians will embrace this regimen because of the potential to establish another TKI-free option in the frontline setting without substantially compromising quality of life. nivo ipi is a standard of care and preferred regimen in the frontline RCC setting given established long-term durability. But its Achilles heel remains the relatively high rate of primary progression. Currently, 1 in 5 patients will have primary progressive disease. By combining cas with a PD-1 CTLA-4 backbone, this weakness will be addressed without introducing overlapping toxicities that have been observed with other triplet regimens. The combination of CAS with cabo in second line and IO in frontline has ignited enthusiasm in the oncology community, offering potentially potent synergy that overcomes the limitations of current standards while maintaining a manageable safety profile and the opportunity to redefine treatment paradigm and improve patient outcomes in RCC. I'll be here for the question-and-answer session, and we'll be happy to answer any questions then. I'm going to turn the call over to Dimitry.

Thanks, Rana. I'll first review the ARC-20 study design on Slide 18. The dose escalation portion enrolled patients with any advanced solid tumor and was designed to assess safety in PK and PD to identify the optimal dose to take forward. For the dose expansion portion in RCC patients, we started with 100 milligram daily dosing, which is our planned go-forward dose for Phase III, and we have since enrolled both a 50 milligram and 150 milligram cohort to satisfy dose optimization requirements. For the 100 milligram daily dose cohort, this was administered as 50 milligram BID or twice a day because at the time we only had 10 milligram capsules available. We have since transitioned to 25 milligram tablets. So going forward, patients will receive 100 milligram QD or once a day. We have shown that the capsules and tablets are bioequivalent and the exposure for 50 milligrams BID and 100 milligrams QD is similar. This data is in the appendix of our presentation. On this call, we will focus on the data from the first 2 expansion cohorts, the 100 milligram daily dose cohort and the 50 milligram cohort. On Slide 19, we show the patient demographics. And here, you can see that both the 100 milligram and 50 milligram cohorts enrolled heavily pretreated patient population. All patients had received prior treatment with an anti-PD-1 and at least 1 VEGF-R TKI and more specifically 27% in the 100 milligram cohort and 29% in the 50 milligram cohort had received 4 or more prior lines of therapy. As a reminder, in LITESPARK-005, only patients with 1 to 3 prior lines of therapy were allowed. In the subgroup analysis conducted both by Merck and [indiscernible] for their respective studies, the response rates in patients who received more prior lines of therapy were lower. Turning now to the overall response rate for 100 milligram cohort on Slide 20. I'll start with the rate of primary progression, which was 18.8%. This compares very favorably to the 33.7% in LITESPARK-005. The rate of primary progression represents the patients whose disease progresses on or before the first scan. High rates of primary progression will negatively impact the early part of the Kaplan-Meier curve for progression-free survival and ultimately median progression-free survival and other landmark readouts. As Rana mentioned, primary progression was considered to be the largest weakness in the LITESPARK-005 data set. We have observed substantially better results relative to belzutifan for this important efficacy measure in both cohorts. The overall response rate is 34%, which includes 1 patient who responded after the data cutoff. Including this patient, there are 2 patients whose responses are pending confirmation, and there is 1 patient whose response did not confirm due to subsequent progression. The confirmed response rate is 25%. Even with only 11 months of median follow-up, this is numerically higher than what was observed for belzutifan in a similar patient population. As you will see in later slides, there are many patients with stable disease still on treatment. So there's plenty of opportunity for further improvement in the overall response rate with longer follow-up. The next slide, Slide 21, includes the ORR table for the 50 milligram cohort, which has an even shorter follow-up of 8 months. Despite this, we are seeing an overall response rate of 25% and a confirmed response rate of 21% with 1 response pending confirmation. This is already in line with LITESPARK-005. The disease control rates are above 80% for both cohorts, reflecting the low rate of primary progressive disease. As a side note, one of the patients who was assessed as having progressive disease in the 50 milligram cohort remained on study therapy because they were doing clinically well despite the progression seen on the CT scan. And 2 scans or 12 weeks later, they had a documented 33 percent reduction in tumor volume. For formal RECIST criteria, this patient is recorded as having progressive disease as best overall response, not as a responder despite their longer-term benefit. And in our efficacy table, this is captured as a progressive disease patient. Note that there are slight differences between the efficacy evaluable population and the overall population for each cohort with 3 non-evaluable patients in the 50 milligram cohort and 1 in the 100 milligram cohort. 3 out of 4 patients were deemed ineligible for the study shortly after enrollment due to kidney function in 2 patients and hemoglobin in 1 patient. The fourth patient discontinued treatment before the first scan due to an unrelated AE of acute kidney injury. These patients were all on study for a short period of time. The 1 ineligible patient in the 100 milligram cohort discontinued just after the first scan, which showed stable disease with a 29% tumor volume reduction. The other 3 patients discontinued before they could complete their first scan. On Slide 22, we provide some additional context for interpreting our overall response results. We have mentioned that we -- sorry, we have already mentioned the more heavily pretreated patient population and the shorter follow-up for ARC-20 relative to any belzutifan benchmark study. And you can see that very clearly on this slide. Also, regardless of what study you look at, a Phase I or Phase III, the overall response rate for belzutifan has fallen in a very tight range of 19.1% to 21.9% for patients who received both prior anti-PD-1 and a VEGF-R TKI. And both of our 50 and 100 milligram expansion cohorts already meet or exceed these benchmarks. Turning to Slide 23, we show the waterfall plot which demonstrates that the vast majority of patients on casdatifan experienced tumor reduction. We've also highlighted the patients who are still on treatment, of which there are several in each cohort. Slide 24 shows the spider plot which illustrates the very durable disease control achieved with casdatifan. These are great spider plots, particularly for a drug administered as single agent in a late-line population. Duration of response is still extremely immature. In fact, all patients who responded with one exception, maintained their response at time of the data cutoff. Insofar as the probability of future responders note, for example, the patients in the 50 milligram cohort offering slightly above the 30% reduction mark. It's also clear that patients with stable disease can do well with no growth in their target lesions over a prolonged period of time. The registrational endpoint is progression-free survival and whether or not a patient reaches a 30% reduction is not relevant if that patient remains free of disease progression. Median progression-free survival has not been reached in either cohort. To be clear, this means that at the time of the data cutoff, more than 50% of patients in both cohorts had not progressed or their disease had not progressed. But you can see on the Slide 25 -- sorry, on Slide 25 in our swimlanes, you can see -- you can get a good sense for the time on treatment. And you can also see the first patient here is going beyond 1 year of treatment. It is also notable we see very few progressions beyond 26 weeks. So once a patient reaches 26 weeks on treatment, they tend to do very well. We've also included a really interesting patient case on Slide 26. This patient received pembro and lenva with the best overall response of stable disease and her tumor progressed within 6 months. As a reminder, the overall response rate for this regimen is over 70%, 70%, and the median progression-free survival is 23 months. So her outcomes were clearly quite poor. However, after receiving her first dose of casdatifan in November of 2023, a response was noted as early as 6 weeks later at her first scan. That response has been maintained for the last 9 months and was ongoing at the time of the data cutoff. Finally, on Slide 27, I'll briefly speak about the safety and tolerability. We have not seen any dose-limiting toxicities to date in our 20 up to and including the 150 milligram cohort. Evaluation of the 200 milligram cohort is ongoing. Anemia and hypoxia are the 2 on-target AEs that we are monitoring closely. And thus far, the incidence of these AEs has looked very similar to that of belzutifan in LITESPARK-005. To date, we have not seen grade 4 treatment-related AEs or treatment-related deaths, which is pretty remarkable for an anticancer therapy. And now on Slide 28, we'll summarize these data for you. You can see we show a clear improvement on every efficacy measure that has been reported for belzutifan, and we achieved this without meaningful deterioration in safety. This is what gives me significant confidence that we have a drug with an extremely competitive profile and that we are very well-positioned in the HIF-2α field. You heard earlier on the call how Rana highlighted the primary PD rate being an incredibly important point for patients and physicians. And here, we show for 2 cohorts combined that we have achieved a primary PD rate that is about half of what was reported for belzutifan. This is huge. And that brings me to our development program and how exactly we will capitalize on this excellent profile. On to Slide #30, you can see that the data we presented today are just the beginning, and we expect to present a steady cadence of results from ARC-20 over the next 12 to 18 months. Specifically, in the first half of next year, we expect to present median progression-free survival results from the 150 milligram cohorts. We'll also share efficacy results from 2 other monotherapy cohorts and the safety data from cas plus cabo combination cohort in the second half of next year. Followed by early efficacy data from the combination in either late 2025 or early 2026, we believe all of these data further derisk our Phase III study, PEAK-1, which leads me to our late-stage development plan for casdatifan. You can see on Slide 31 we are currently planning to take casdatifan forward into 2 settings. First, following IO treatment and second in IO-naive clear cell RCC patients. We recently had a successful meeting with the FDA for PEAK-1, and we are moving forward as rapidly as possible with the study initiation. In parallel, our clinical collaboration partner, AstraZeneca, is leading efforts on casdatifan plus Volru combination. Both of these concepts have been received extremely favorably by the clinical community, and we are very excited to get going with these trials. The design for PEAK-1 is shown on Slide 32, which is expected to begin in the first half of next year. The design of PEAK-1 is simple which should drive rapid study enrollment. Specifically, we are planning to enroll approximately 700 patients randomized 2:1 to casdatifan plus cabozantinib versus cabozantinib, the leading TKI prescribed for second-line clear cell RCC patients. The study population will include both patients who received an anti-PD-1 as adjuvant therapy and those who received an anti-PD-1 as part of their first-line regimen for metastatic disease with or without a VEGF-R TKI. On Slide 33, I want to highlight why we are excited about PEAK-1. As you know, cas and cabo both have meaningful single-agent activity through independent mechanisms and importantly, with minimal overlapping toxicity. There actually have been 2 studies from Merck showing that HIF-2α inhibition plus a VEGF-R TKI could improve upon TKI monotherapy. Though these studies are relatively small, the median PFS observed in these studies, 13.8 months for cabo plus belz and 11.2 months for lenva plus belz appear to be meaningfully longer than the median reported for TKI monotherapy. I'll end on Slide 34 where we lay out our clinical strategy for casdatifan. To date, casdatifan has been administered to over 150 patients and 56 healthy volunteers. We will continue to expand the development program for cas. As Terry and Rana talked about earlier, we are very excited about our clinical collaboration with AstraZeneca to combine CAS with their potentially best-in-class anti-PD-1 and CTLA-4 bispecific. Additionally, we are planning to initiate additional cohorts in ARC-20 to evaluate casdatifan in certain clear cell RCC populations. We'll talk more about this in the coming months. We are also considering additional tumor types for the development of casdatifan. I'll now turn the call over to Eric to discuss the potential market opportunity.

Thank you, Dimitry. So just quickly here, we've prepared an overview of the market opportunity for cas and RCC on Slide 36. You can see the different lines of therapy and the total market estimates for the U.S. So first, focusing on the IO-naive metastatic segment. We're projecting a market size of 12,200 patients. And just for background on that estimate, we use an annual incidence of 21,000 patients for all RCC histologies, of which 75% are clear cell. And we then take out prior adjuvant treated patients who also had prior PD-1 adjuvant within 12 months, yielding the number you see there for a total addressable market of 12,200 patients in the U.S. Then for the top 7 major markets globally, we estimate this number to be about 22,000 patients. And these newly diagnosed IO-naive metastatic patients typically remain on treatment for about 18 months. So all of this then translates into a global market opportunity of well over $3 billion. As you saw earlier, the anti-CTLA-4 PD-1 share of this market is already substantial, and it's growing in the U.S. and globally, largely due to the potential for incredibly durable responses. And in the U.S., just recently, this regimen was added into the NCCN Guidelines as a preferred regimen for favorable risk patients, expanding its utility. Yet as you can see here, the benchmarks for ORR and PFS with this approach could still be improved dramatically, which is the opportunity for combining with cas. There have been attempts, as you know, to combine TKIs with PD-1, CTLA-4, but these efforts have faced tolerability challenges that have impacted delivery of the regimen, whereas we believe with cas' safety profile, as you've seen, and its potential for driving early sustained responses, cas plus Volru could prove to be a winning combination for this newly diagnosed metastatic RCC population. Next is what we call the post-IO population comprised mostly of second-line patients, but increasingly including first-line patients who have seen prior adjuvant. We estimate this population will be about 11,300 patients in the U.S., almost equivalent to the IO-naive population. Approximately 20,000 patients in total for the G7 major markets. Median treatment duration in this population is currently 7 to 11 months, the former if they receive the TKI in frontline or the latter if they've only received prior immunotherapy. The opportunity for cas-cabo here as the potential best-in-class HIF2-TKI combination is clearly very attractive to patients and investigators, and we're hearing tremendous interest in rapidly bringing this regimen forward with the PEAK-1 study. As you saw earlier, the market of available treatments here is really fragmented with cabo is the most predominantly used TKI consistently across regions. So building upon cabo, we believe cas-cabo will consolidate a market-leading position in this indication, taking share from other TKIs and everolimus, yielding a market opportunity of well over $2 billion a year. So that brings us to the late-line segment, and these are patients who have already cycled through 2 or more lines of therapy, where median PFS is unfortunately only about 5 months. This is where belzutifan is currently approved. Today, Arcus is focused on the earlier post-IO and IO-naive settings and combinations, and we are evaluating a range of additional subpopulations in RCC. So just zooming out now, the overall RCC market is already very large with TKIs generating over $5 billion in annual sales combined. Following bel's U.S. label expansion into RCC in December of last year, there was an immediate and dramatic doubling of U.S. scripts, now generating well over $500 million in annualized run rate. Keep in mind, this inflection was driven by new scripts in the late-line monotherapy setting, where, again, PFS is only about 5 months. By moving differentiated HIF2 combinations into earlier lines of treatment, we have the opportunity to help a larger number of patients with extremely durable responses and manageable safety, enabling patients to benefit from continuing on effective treatment significantly longer. So I'll now turn the call back over to Terry.

So I'll finish by saying thank you all for joining us today. We're obviously thrilled with these data, most importantly, with the opportunity we have to change the standard of care for RCC and make a meaningful difference for patients. I'll now open the call to questions.

[Operator Instructions] Our first question comes from the line of Peter Lawson with Barclays.

I guess first question would just be on kind of thoughts on where you think PFS could eventually land based upon what you've seen for the disease control rate and kind of the reduction in fast progressors and kind of your thoughts on what you think the best correlation is there for PFS, if it's disease control rate or PRs?

Thanks, Peter. I don't want to get ahead of ourselves. You can look at the swimmer lanes yourself as we're out in the 100 milligram cohort at roughly 11 months of follow-up. We still haven't hit the median. We think we're going to be substantially better than that 5.7 months of PFS that we report for belzutifan. And in fact, we plan to report these data sometime as early as possible next year. So we think we're going to be clearly better, but we don't want to throw a number out there yet. You can do your own estimating.

Got you. Okay. And then the time line for completing the Phase III, kind of your estimate there and kind of when you think you can be on the market?

So we haven't shared that yet. We'll give a little bit more color on that, but we expect it to enroll very rapidly. So the enthusiasm for the study has been high. We're moving full speed to get it going by the first half of next year. And we have no more to really say about that, but I'll just offer Jen or Dimitry opportunity if they want to jump in and give any more color on that.

No, thank you, covered it well.

No, thanks, Terry, no...

Go ahead, Dimitry.

Yes, no, I was going to say the same thing. I think most importantly, as I highlighted, study design is very straightforward. There's a lot of enthusiasm. And we -- let's say, we hope to get started the first half of 2025.

Our next question comes from Daina Graybosch with Leerink Partners.

Congratulations on the data. I'm going to do one detailed science and one commercial question, but start with the science. So Terry, on the comparison of the long-term pharmacodynamic data, I guess what gives you confidence that the waning trend in EPO that you showed was observed with belzutifan is also relevant for the downstream HIF-2α pathways that drive hypoxia and the tumor shrinkage?

So what we would comment is more illustrative of the point. I'm going to make a couple of points on that. So the first thing to realize is what we've said all along. Belzutifan at its clinically used dose is basically just approaching maybe not exactly at the maximal effect on EPO suppression. And what you can see is that there's a feedback mechanism, we can get more into that at a later point that clearly makes the HIF-2, at least in the kidney, less responsive to that. So you're essentially seeing almost no effect. Now as we've always said, we're not going to change our story that EPO suppression is not directly related to what's going on in the tumor. But you basically have no margin for error as any type of mechanism that kicks in that might make your molecule less potent on HIF-2 or if there's greater expression in the tumor, et cetera, you're going to see a waning effect. And given that we're at 5x starting relative to that PD effect, we have a big buffer, and that's certainly the explanation why we're not seeing anything falling up. I think it's a big concern if I saw those data that you have that waning effect with belzutifan. And we think that certainly could bode well for the durability endpoints, whether they're PFS and in fact even OS as we start to move beyond what we've already seen with belzutifan. Juan wants to jump in here and say something.

I think -- Dan, a critical observation here is that prior to seeing our data, you could have postulated that the kidneys found a different way of making EPO. So some other regulatory pathways kicked in. When you see the cas is able to continue to suppress it, that tells you that the compensatory mechanism involves some enhanced version of HIF-2α activity, whether it's more protein or it's been modified in some manner, that there's more HIF-2α activity that the standard dose of cas -- of belz is no longer able to suppress whereas cas is able to suppress. Whether the same HIF-2α involving compensation happens in the tumor, the clinical data is starting to speak to that point. But I think that its HIF-2α dependence, it's really, really critical.

Got it. My second question is on zanza. You used to have a combo, I think that Exelixis was running. You didn't talk about that any longer. And I wonder if you can comment on that in light of the collaboration deal between Exelixis and Merck.

So why don't we bring Jen into the conversation? Jen, do you want to handle that question?

Yes. So we did have a clinical collaboration with Exelixis. So you're right, we're combining cas with cabo. That was a study called STELLAR-009. We announced in early September that we were not going to continue with that study. As you know, we had announced a few months before that, that we were prioritizing cas plus cabo, which is the ongoing expansion cohort that we have in ARC-20. We think both are great TKIs, but we just felt like there was more familiarity out there with cabo given that it's the clear standard of care, as you heard from Rana, in the setting that we're pursuing. So we both felt like there were a lot of strategic reasons for us to continue with STELLAR-009 and that we would both move on, that we would focus on ARC-20 and PEAK-1. Exelixis, as you pointed out, did do the clinical collaboration with Merck. There weren't a lot of details in the release other than to say they're going to be doing a Phase Ib. I assume that Phase Ib is going to be proceeding, the 2 Phase III studies that they also referenced in that release. But obviously we don't know details like you guys. But I would say that we continue to feel very, very good about our lead combination of cas plus cabo and love the idea that we're going right on top of the standard of care. And to the point on enrollment, which we're really optimizing for, we believe this is a study that's going to enroll very, very quickly and is a very simple study for investigators to get their hands around.

The next question comes from Jonathan Miller with Evercore ISI.

Congrats on the data. I'd like to -- let's start with a dose response question. I think, Terry, Jen, you've been saying for a while, you don't expect to see major dose response between these 2 dosing arms. When I look at the waterfall charts, it does look to me like I'm seeing deeper responses from the 100 milligram cohort, although the headline ORR is obviously in the same ballpark. Can you talk a little bit about how responses are evolving over time and where you might look for a dose response if there was one to be observed [indiscernible] time?

Yes. Thanks, John. So first off, if you look at the rate of primary progression, they're basically the same. Numerically, the 50 milligram looks a little better. We are seeing kinetics that particularly anecdotally that we hear we might be seeing things a little earlier. We also, though, as we may have talked about publicly already, we've seen patients respond even at 14 months. Keeping in mind the Merck data, they actually saw 60% of their responses occurred within 6 months and then 20% more each of the next 6-month period, so between 6 and 12 months and then another 20% between 12 and 18. I think it's hard to speculate at this point on any differences between those 2 arms. I'll remind everybody that the 50 milligram cohort is really hitting the target heart. It's 2.5 the PD equivalent of the belzutifan dose that obviously changed the standard of care. And so we're 2.5-fold that, and then you've got the 100 that's double that. So I think the place where we might see some difference would be on durability, and that will simply take time to see. So whether we actually see some PFS differences between the 50 and the 100. But we feel that both of those doses are basically on the upper part of a conceptual dose response curve towards the top of the inflection, if not on the flat part. We're -- by and large, I would actually say, I think the 100 milligram dose, you're squeezing all of the water out of the rock of HIF-2α inhibition that you can get. So that's how we handicap it right now.

And John, one other data point I just wanted to point out in the presentation where you did see a bit of a difference between 100 mg and 50 mg was the time to response. So that was quite a bit shorter for 100 mg. We'll see some of these later responses come in if that changes, that difference that you see, but that was something that sort of stood out as being a bit different about the 100 mg dose relative to the 50 mg dose.

Okay. Makes sense. And then maybe as a follow-up, I'd love to ask about the -- how you expect the IO combos to develop here. Obviously, the bispecific Phase I is in Astra's hands for being operationalized. But how do you report on Astra's speed there, their enthusiasm for these combos there and your plans on IO combo regimens potentially beyond Astra's bispecific. You showed, for instance, on your slide, Merck is planning their own PD-1 plus HIF-2 plus TKI triplet combos, et cetera.

So I'll ask Jen to answer that. And Rana, if you want to add anything on to anything that Jen might say, that would be awesome as well.

Sure. Yes. We probably can't say a ton yet, but we have a strategy sort of similar to what you're suggesting that is not directly competitive with a combination that's in development today. So we do like the idea of combining with IO upfront, and I think you'll see us do that. But probably today, I can't say more than that. And I will just say that AZ is very excited about the collaboration. We obviously had a very long relationship with them. Interestingly, the team that we worked with on the study that we're now planning to do with them, the same team that we worked with on [ PAC 8 ], the study that we're doing with them for our TIGIT antibody. So we know them well, and we're really excited to get that study up and running. Rana, maybe you want to add something.

Happy to chime a little bit about the reference to the triplet with Merck, that is building on the backbone of pembrolizumab plus lenvatinib. So it is adding belzutifan to that backbone. And I think the idea of having a TKI sparing option in the frontline setting is dramatic. And we know from data from subsequent multiple studies now that IO post-IO really does not work. And when you give ipi in later-line settings, you are not able to maximize on the durability of that agent. The CR rate is 0. Response rate maybe at best 10% to 15% when that's used later. So I think multiple studies have demonstrated that in the upfront setting, like maximizing that benefit of IO is critical. So I think the linkage to a combination that includes CTLA-4 inhibition in frontline is huge, and sparing the TKI is also an excellent strategy, and that is not being done.

Our next question comes from Li Watsek with Cantor.

Congrats on the data. A couple of questions from me. First, can you maybe just talk about your confidence level for an OS benefit for the [ PK1 ] study given what we have seen from belzutifan LITESPARK-005 study?