All right. Good morning. Welcome everybody to our Casdatifan Investor Day. We will be sharing some exciting new data across first, second, and late-line kidney cancer. Before we begin, I'd like to remind you we'll be making forward-looking statements, and please review our SEC filings where the risk factors are laid out in detail. Today, you'll be hearing from a number of members of our management team, as well as our guest speaker, Dr. Rana McKay, who will be joining us virtually. After Dr. McKay's section, we'll stop for a short Q&A with her, and then we'll take more questions at the end after the full presentation. I'll turn it over to Terry to begin.
Fine. Oh, I'll repeat that. Thanks again for all of you who are joining us, including those of you who are online. We have a lot to share today. It's a pretty exciting day for us. We're obviously here to talk about casdatifan, but before we jump into that, we just wanted to spend two minutes on the background of the company. Even though it looks like a lot of familiar faces, for those who are out there and know less about the company, we just want to give you a drop of background. Arcus was founded in 2015. We've been around for about a decade. The company was founded with both a focus and expertise around small molecule drug discovery. We're a very R&D-centric organization. You can see we had that goal of advancing one to two molecules into the clinic each year.
I think we've done a quite good job of that. We built up quite a portfolio of programs. Casdatifan is sort of right down the middle of the fairway, really reflects the paradigm of Arcus. We have our molecule, quemli, that's in frontline pancreatic cancer. We're actually doing an interim analysis of its registrational trial this quarter. That'll read out. On the other end of the spectrum, we have a portfolio of very early programs that are just into the clinic or earlier. Those are mostly in the immunology area. The reason we point that out is because as casdatifan approaches commercialization, we feel we have a very sustainable pipeline to continue to grow the company with other molecules, both addressing similar opportunities and of similar quality. I'm going to use this slide somewhat as an executive summary as we talk about casdatifan, high level.
You should think about what we're laying out for the day. I think the first thing that I wanted to talk about is our holistic approach here. We recognized some time ago that casdatifan represented a true, unique molecule, dramatically different from belzutifan. In thinking about the opportunities, what we did is we looked across each line of therapy in clear cell RCC at what were the vulnerabilities of the standard of care. I'm going to talk a little bit about that on the next slide. But essentially what we're going to lay out today is the plans that we had, the data that we've generated, and the plans we have going forward.
But suffice it to say, I think when you walk out of this room, that you will believe that casdatifan, hands down, is the best molecule, best drug in clear cell RCC, bar none. We think you will recognize that it does have a $5 billion-$10 billion commercial opportunity, and we have a lot of tailwinds in our development program. Essentially, we are going to talk about first line, second line, and late line. What we are going to share today, it is not just totality of the data. In each of those modules, we think you will walk away thinking we have the best molecule to address each of those. So for those who like the minimalist version where everything can be done in five minutes and feel like you have to be able to say everything in five minutes, this slide has it. So those are real curves.
We will be talking about more specific data as I go along. But what I want to do here is just lay out those vulnerabilities in each line of therapy. So when you think about the front line, the most commonly used regimen is ipi/nivo. Ipi/nivo has about 30% of the front-line market. The reason that ipi/nivo is used number one is because it provides the best overall survival. So it is used 30%. Why is not it used 100%? And its limitations are very clear and very simple. It is its rate of primary progression, which is almost one in five, and then just PFS in general. Landmark PFS for ipi/nivo is 50% at 12 months. So 50% of the patients have progressed.
The clear message, and this is a theme you are going to hear throughout the day, everything we are doing is highly aligned with how physicians want to treat clear cell RCC. So as we are going to change the paradigm, we are essentially going to give an umbrella of a casdatifan-based regimen on top of the regimens that they already like to use. So the idea on ipi/nivo is get that disease control to where it can take that 30% to maybe 50% or more conservatively. The second line is a very interesting one. So the second line has two vulnerabilities, two limitations. So the first one, we have now had LITESPARK-011 done. Merck with belzutifan is zero for three on hitting overall survival. So one thing that we have been told in the second line, if we could achieve an overall survival signal, that would be game-changing.
You would essentially take the whole market. Now, the other thing about the second line is that includes a patient population that probably has the worst prognosis of any population in clear cell RCC. And that is the group of patients that come out of the front line having experienced a second-generation TKI. So I will talk a little bit more about this in detail a little later on, but that is 60% of the patients coming into the second line that you would say miscellaneous.
So two clear, simple opportunities, overall survival, and then also it would be great to address that patient population that is very underserved by the second-generation agent. And what we are doing there is we are going on top of cabo. So just like ipi/nivo is the most commonly used regimen in the front line, cabo is the most commonly used regimen in the second line.
It's 40%, and in terms of monotherapies, it's actually 80%. It's very favored by the clinical community. Finally, the late line. You generally look at this late line as almost like a salvage therapy population, a new lease on life, extraordinary overall survival. A picture's worth 10,000 of my words, and I won't give you 10,000 words. Everything that sort of describes casdatifan's advantages over belzutifan can be described as coming from this property of the molecule. We have a better quality molecule, and the first thing that we recognized where you could really see a dramatic difference is when you look at the effect of casdatifan on the biomarker for HIF-2 alpha inhibition in a human, and that's the ability to suppress erythropoietin. Not only does casdatifan more deeply suppress that production, but you can see it's incredibly durable.
As far as we know, it goes on in perpetuity. Whereas the V-shaped curve, which was published by Merck, what you see is somewhere between one week and 13 weeks, you lose essentially all effect on that biomarker. The thing I'd like you to keep in mind is that whether you're taking those three 40 mg tablets, your effect on erythropoietin production beyond that 13 weeks looks very minimal. I think that's particularly important when you start to think about biologies that are interacting when you're thinking about combination therapies. The left-hand side will show more specific data, but the left-hand side represents what we'd shown with our late-line patient population about a year ago. Basically, much better data compared to belzutifan, about half the rate of primary progression, 50% higher ORR, a two-plus fold improvement in PFS.
What we're going to show today is a very strong OS signal in that same population. The key point here, as I mentioned before, these two molecules are completely different. This is not an incremental difference. What we'll show you today as you see our data is that belzutifan doesn't define what HIF-2 alpha inhibition does. It defines what belzutifan does, and essentially, casdatifan squeezed all the water out of the stone of what you can get out of the true promise of HIF-2 alpha inhibition. I think when this target first came into recognition, there was a lot of hope. There was a lot of sort of promise as to what robust inhibition could deliver for patients. I think what you're going to see today is how that manifests itself when the curve looks like the bottom curve.
We've looked to be in the lead in the science of HIF-2 biology and studying casdatifan. This highlights a recent Nature paper, and I just want to point out two key aspects of what we showed in this Nature publication. The first one is we were able to correlate efficacy with pharmacodynamic biomarkers. Basically, what we showed was that you could correlate ORR or PFS with that ability to suppress erythropoietin production. When you think about the previous slide, given that correlation, you can see how important the difference between those two curves is. The second thing that we showed is that patients that had higher HIF-2 alpha activity actually did better. Why is that important? HIF-2 alpha activity is one of the most negative prognostics you can have in RCC or cancer for that matter.
What that tells you is it's not only correlative, but it's causative, given that we're having the best effect in the patients that have the worst prognostic. This is another very important aspect of durable, robust HIF-2 alpha inhibition. This is the spider plot for the 120 patient monotherapy cohort that we've been talking about. I'd like this to get imprinted in your head because I'm going to show three spider plots today, and they're just at various stages. I don't really love the metaphor if you've seen this movie, but this movie tells you how the other spider plots are going to play out with time. What you can see, and this is remarkable when you look at this mechanism. Keep in mind, this is a late-line patient population.
Third plus line of therapy and 25% or more that are fourth plus line of therapy. If you look at that x-axis, you're out past 30 months. What you can see is not only do you get early disease control, rate of primary progression, ORR, but these patients just go on and on. Whether you're a yellow stable disease or whether you're a green or blue responder, all of these patients are getting tremendous benefit. If you look up there, 25% of the patients are still on therapy after two years. Basically what you're talking about, this drives obviously patient life benefit, but also it drives hazard ratio for both PFS and OS. Very critical to see that basically once a patient is on control, they essentially go on and on.
I would emphasize, you can see this is out past 30 months, you're going to see an amazing OS signal. When people first started to see this even a year ago, they suggested, "Hey, maybe you're going to see a strong OS." This just calls out two particular patients to amplify the last point. What you can see from these two patients is here are two patients that were stable disease in out past two years, they actually became confirmed responders. Again, it's a very, very durable mechanism of action, and it's probably no more benefit to these patients, but it just illustrates the point that you can even see deepening of the tumor response out past two years. I'm not going to say much about the slide. You're going to see it way too many times.
Both Rana and Richard are going to comment on this. As I was talking about, when we thought about our strategy, it's a very integrated strategy. These patients, some of them are going to be on therapy for over 10 years. The sequencing of all of the different opportunities is a very critical aspect to how these patients are managed. As I said, the way we thought about approaching this was to ensure that we're going to change the paradigm for treating clear cell RCC, but we're going to do it in a way that's highly aligned with how physicians want to treat clear cell RCC. The brass ring therapy, again, that we're building on. The reason I say it's brass ring, that's the ipi/nivo. Frontline, it's the largest patient population. It's going to give the greatest durability.
When I showed that spider plot, that also drives commercial opportunity. There's no competition in the frontline. As most of you know, belzutifan failed in the frontline, so our only competition in the frontline is to improve upon ipi/nivo. The second line, as I mentioned, we're going on top of cabo, and late line, we're going on top of tivo. These are three regimens that are all favored already by physicians. But essentially what we're doing is putting an umbrella over them. We're going to have a casdatifan enhanced opportunity for them to use those therapies. I'm going to switch now to, I have about a half a dozen data slides. I'm basically going to steal all the thunder from Juan and Richard, and then they'll put a little bit more meat on the bones. Juan will talk more about the translational aspects of this.
Richard will give you a few more details about the clinical aspects, and also he'll tell you a lot more about the development plan. A year ago, I think it was basically in this room, this is the data set that we were focused on. If you look on the left, that was last year. On the right is the Merck data set. Now in the middle, we have similar maturity to that of the Merck data set. I'll just run through this quickly. This is, as I talked about before, you can see the confirmed ORR. It's about 50% better than what you see with belzutifan. The PFS is about 2x that. Very importantly, you see that progressive disease that's about 50%. That's hugely enabling.
That's why we're able to go into the frontline with, again, what physicians wanted to see was a TKI-sparing regimen, so building on top of ipi/nivo, where you don't have the baggage of the TKI. So you've got the opportunity to give patients years of life before they ever see that TKI. With that high rate of primary progression, that's why Merck always needed to combine with a TKI to get that early disease control. While these data, as always, this is a theme for this target, and particularly for casdatifan, the data always get better with time. That was one of the first points that I want to illustrate when I showed you that spider plot. Similarly, you can see bits of improvement, but the real sort of home run now on this is the overall survival.
You could see the opportunity inherent in the spider plot, but basically I'd emphasize that yellow oval, we're seeing a median OS of three years at this point. When we saw these data, it was mind-blowing. If you look what belzutifan was, belzutifan showed a five point six month median PFS and basically did not improve upon the OS of everolimus. Here you can see, if you look at the confidence interval, when we come back and look at this a year from now, it's going to be even better. Look at all the tick marks of patients still on. So three year overall survival. Importantly, this is really important, Juan is going to show you data where, as I mentioned in the Nature paper where we showed that PFS and ORR were correlative with pharmacodynamic markers, we're going to show the same thing for overall survival.
These are real, and they are highly likely to read through to other lines of therapy. In fact, I am going to show you in the second line, we are already seeing a strong OS signal. To put this number in context, in second line, cabo monotherapy has an OS of about 20 something months, low 20s. So almost the same as belzutifan late line. We are already doing better. The other thing I will just tell you here, when I show you curves for the second line, cas/cabo, we have already put them on top of this OS in the late line monotherapy PFS, and those second-line data are looking better. You can expect that the second-line data, as they continue to mature, are going to mature similarly, and I will show you the spider plot that points to that. Now let us turn to the second line.
The first thing I want to do is briefly address the patient population coming into the second line. As I mentioned, 60% of the patients that come into the second line have received the second-generation TKI, and those patients have the worst prognosis of any patient group in this particular setting. Now, for second, put casdatifan aside, and I just want to illustrate what Merck just recently published in The Lancet publication describing the patient population in LITESPARK-011. What you can see is there is a couple of unusual things about that. First of all, they have zero cabo patients, obviously. They have zero lenvatinib patients as well. Then they have about 25% or more patients that received a first-generation TKI, like sunitinib. That is very rare, particularly in the U.S., that is almost not used anymore.
The point we just want to make here is to illustrate that this patient population looks nothing like the real-world patient population that is delivering 60% of either axi, cabo, or lenva, priorly treated patients into second line. Why is that important? It comes back to what I mentioned about those patients who are underserved a little earlier. If you look at that upper blue box, lenva is the most notorious in that context, and basically real-world data that were published this year illustrates if you got lenvapembro in the front line and then you get cabo monotherapy, you are basically going to have 5% ORR, a median PFS of four to five months and change, and most devastatingly, an eight months overall survival.
Keeping that patient population in mind, that is illustrative of differences in the patient population between LITESPARK-011 and the ARC20 cas/cabo cohort, as well as our PEAK-1 registrational trial, which is going to include plenty of those lenvatinib patients. Now, we actually felt for a number of reasons, Juan will share them, that we had an opportunity to completely change the trajectory for those patients that had those second-generation TKIs, and I will get to that in a few slides. What this actually shows now is the waterfall plot and the spider plot for those patients that received cas/cabo in ARC20. What I want to emphasize, not only the depth of response, the number of patients still on study, but that spider plot should look really familiar to you.
When you get casdatifan, if you get past that second scan, you are going to be very well-served by the drug. You are going to be on it for a long time, whether you are a responder, whether you are a stable disease patient, you just get incredible durability. As I emphasize, all of these patients are contributing to the hazard ratio for both PFS and OS. As Bob speaks to you, and he will be pretty conservative on our estimates, they are also contributing to the commercial opportunity. You can see, again, we have something that a year from now is only going to get better, and all of that is contributing to the implied hazard ratio, the shape of that Kaplan-Meier curve. Now this actually compares cas/cabo to belz/lenva from LITESPARK-011, then you see the control arm of cabo from LITESPARK-011.
Despite those patient population differences that we talked about, we are overlapping on our PFS curve, but what you can see is there is already a dramatic separation on overall survival. What you can see is that the curves are separating at six months. To just give this some context, if you look at the two lower curves, those represent the belz versus lenva versus cabo from LITESPARK-011, and they actually did start to see some separation about 16- 18 months out. At this time point, when they had the same maturity data that we have now, belz one they had a hazard ratio on OS of 0.9. It actually improved over the ensuing year to their final analysis to about 0.85. Interestingly, they almost hit on OS.
Given what you are seeing here, this is sort of the first piece of evidence to suggest, in addition to what we have shown for the molecule as a single agent, casdatifan cabo is going to hit on overall survival. I will show you more on this topic. Now, this emphasizes the point that we talked about those TKI pretreated patients. This is another one that probably had us just staring at the data. The first thing is, on the left, what we have basically shown are the PFS curves for patients that did receive a TKI in the prior treatment and those patients that did not receive a TKI in the prior treatment. What you can see is they are overlapping. Now on the right-hand side, we have shown swimmer plots for the subpopulations. Those reds are all lenva patients.
The yellows are other TKI patients, and the blues are those patients that received IO that did not receive a TKI. The remarkable thing about those lenva patients, recall the data that I said, those patients had four to five months expected median PFS, eight months expected median OS. If you look at that right-hand oval on the reds, that is out past 600 days, 20 months. So those patients that prior to casdatifan, really in a devastatingly poor situation, now you have got basically looking just the same. If you look down those swimmer plots, what you basically see is all patients have equal opportunity when you come onto casdatifan. The second and probably most important aspect of this is those ovals that we have highlighted.
All of those patients are going to continue to contribute to the PFS curve as well as then that PFS curve feeds into the OS curve, and this is just going to be a dramatic hazard ratio as you think about how this therapy will mature. This slide is meant to make the point of how our registrational trial is going to look. What we have here is cas/cabo is the teal-colored curve. What we've shown here are the corresponding PFS and OS Kaplan-Meier curves from every registrational trial that's involved cabo, including LITESPARK-011. The important takeaway message from this is PEAK-1 control arm will be the first that ever includes meaningful levels of any second-generation TKI other than AXI. AXI's probably the most gentle. Basically what's absolutely the case is that the control arm for PEAK-1, it's going to be a worse population.
It's going to perform worse than any of these cabo curves. You can already see how strong we're looking. Again, I would emphasize, if you look at the OS curve, this is going to hit on OS. Like I said before, when we look at this a year from now, those curves are only going to be more separate. You've got a lot of patients that are contributing to this curves. Now I'm going to show you what's. I think that last slide emphasized where you had a patient population that has the highest medical need. This is the brass ring patient population where the commercial opportunity is huge. In the last six months, as we've been talking to people, everyone's been wondering, what are we going to be able to show that tells them that we're going to be successful in the front line?
I think if anything's been a home run, these data are going to be a grand slammer. What this shows is, again, same theme. We've shown you a waterfall plot. You see all those asterisks, all those patients are still on. Keep in mind, this matures. This doesn't have a TKI in it. This is basically casdatifan plus anti-PD-1 without the ipi. ipi/nivo is 12 weeks of ipi/nivo, and then it's nivo for the rest of the time. This is just the cas anti-PD-1 component. The first thing you can see is waterfall. Like we said, in January, we didn't show up, but we said it looked amazing. It only looks more amazing now. There will be more responders. We've had two more confirmed responders probably in the last month or two.
You can see all of those patients with asterisks still on. Secondarily, look at a spider plot. Basically, when you think about the spider plots I've shown you, they're nice in terms of predictive ability because we're going from patient population that's most advanced, single agent towards a healthier patient population and in a therapy that includes an important combination. Again, a pretty amazing spider plot. If you look, you do not see 45% angles shooting up. What does that tell us? Well, this is nice. If I was only going to have one slide today, it would be this slide. What this shows you is basically how does casdatifan anti-PD-1 compare to ipi/nivo? The middle curve on the right is from CheckMate 214. Its control purpose, actually, and it's summarized in the far right column of my table.
It's the lower curve on the Kaplan-Meier curves. That's pembro in the same patient population. What you can see is pembro gives you just the, what does anti-PD-1 alone do? You can see 30% rate of primary progression, 30% and change confirmed ORR, and then it has a median PFS of about seven to eight months. But the real takeaway is in the left-hand side of the table. First off, remember, with ipi/nivo, you get about one in five patients are primary progressors. Now, we had shared, we saw a 7% rate of primary progression, two out of 30. We're already at a 50%. A lot of people during the course of the last six months said they'd love to see a five in the ORR. So for the ORR lovers, we're at a 50% confirmed ORR. We think that will continue to get better.
Most importantly, the landmark data. We knew we were going to have 11 and a half months of follow-up with cas anti-PD-1. We were excited about that because we know landmark PFS for ipi/nivo is basically 50%. When we plant our PFS curve on top of CheckMate 214, you can see yet another set of curves where you can drive a truck between those two curves. And in fact, when you look at our 50% landmark PFS, it's 69% compared to that 50%. To put that 69% in context, it's a small print under the graph. It should be in big print. That's essentially, it's better than what you get with anti-PD-1 plus TKI in the front line. So whether it's axi or cabo, and it's essentially the same as what you get with lenva, pembro. This builds on a theme that we've seen.
We've said this to you a number of times. Every single time we've looked at casdatifan as monotherapy, it performs in terms of disease control and everything thereafter, like a TKI, but without the baggage. So what you're seeing here is a PFS curve that would be just as competitive with anti-PD-1 plus TKI, but without that TKI baggage. The other yellow oval that is important to highlight. So you've wondered, people have asked, we've been running a cohort of 30 patients for our safety data set of the triplet. This is ipi anti-PD-1 cas. Richard's going to show you the safety data on that, but I'm going to give you the spoiler alert. It looks fine. As a matter of fact, that safety data set already fulfills.
We had our discussion with the FDA and EMA in terms of the safety data set that we need to start our registrational study at the end of this year, and we already have the data in hand that meets those specifications. What you also see, though, is out of the 25 patients that have made it to first scan, one primary progressor. So very consistent picture. We're delivering on both of what were those key things that investigators, not our narrative, what investigators wanted to see, basically dramatic improvement in PFS, and most importantly, dramatic improvement on the rate of primary progression. A very consistent feature of casdatifan. It performs just like a TKI, but it doesn't come with the baggage of TKI. There's two baggage with the TKI. The first is the quality of life.
The second is, and you are going to hear this particularly when Juan speaks, TKI drives HIF-2 alpha activity. That is why those TKI pre-treated patients do so poorly in the second line. Sparing that patient a TKI is both a quality of life feature and a future feature in terms of driving their efficacy going forward. Let me just elevate for a second here and summarize what I have shared. We have looked at the front line, we have looked at the second line, we have looked at the late line. We have contextualized our data limitations in each of those lines. I think we have shown unequivocally that in each line, we dramatically address those limitations. Our second line study, we expect to be fully enrolled by the end of this year.
Our frontline study, we will be starting it the end of this year, and we are starting the work to enable that late line phase III study combining with tivo. I think you will see some more meat on the bones, but I think when you put this all together, you see that not only do we in fact have the best drug for treating clear cell RCC, but it generally will be a $5 billion-$10 billion drug. The next thing I would like to do is turn things over to Rana McKay. Rana is a phenomenal clinician. She has worked with casdatifan, she has worked with belzutifan. As you can see, she is a leader in everything in her field. It says she is a rising star. I say she is a star. She is a superstar.
The second thing I would really point out, in this world where you all the time, it is such a misnomer, you hear KOL. Rana actually has both. Most of them are just Os. She is emphasis on the K and the L. We really appreciate having Rana here to share some perspectives on the current treatment paradigm and how she would expect that to evolve. Welcome, Rana.
Thank you so much for that really kind introduction, and it is really such a pleasure to be here with you all this morning. Really showcasing the evolution of the data around casdatifan and the tremendous opportunities in renal cell carcinoma with this agent that we are all super enthused about, really for our patients to ensure that they live longer and live better, because that is really what is driving everything that we are doing here with the program. Let us go to the next slide. Just to showcase the patient journey, I think Terry alluded to this as we talked about the different landscapes. The evolution of kidney cancer treatments have really changed with the introduction of IO therapy in the front line and in the front line setting.
Now we largely have two key options, an IOIO option in the context of nivolumab plus ipilimumab, of which 30%-40% of patients will receive that combination. It is actually the most commonly utilized combination if we think of things at the regimen level or the combination of PD-1 TKI, which is the remaining share of individuals. But PD-1 therapy, anti-PD-1 therapy is the backbone of front line treatment for patients with modern day RCC. For the individuals that progress onto the second line, which is actually the bulk of individuals go on to develop the disease progression. There's tremendous opportunity in the front line setting. We are not curing the bulk of patients with metastatic disease, and many go on to develop resistance, which is a huge vulnerability.
The bulk of the options for patients that are IO experienced now includes either TKI monotherapy or recently TKI HIF-2 alpha inhibitor, which is really proof of concept to the strategy behind the P1 program and opportunities in the second line space. There are many individuals that go on to develop disease progression, that receive third plus treatment options, of which we're basically cycling TKIs with TKI monotherapy. HIF-2 alpha monotherapy can also be introduced in that context, and we are seeing that there is growing utilization of belzutifan, which actually is a good thing as we think about the opportunities for casdatifan and MOA in developing a best in class HIF-2 alpha inhibitor. Next slide. Now when we break it down into those individuals that are actually receiving TKI based treatments, which we think there's tremendous opportunity here.
When we look at the anti-PD-1 TKI regimens in the front line space, there's almost an even split between lenvatinib, nivolumab, cabozantinib or an anti-PD-1 combined with axi, the bulk of which of that is pem axi. In the second line setting, again, we see TKI monotherapy as the predominant backbone. I think what's striking here is the astounding utilization of cabozantinib in this second line setting. It is the go-to workhorse in the second line. People know it, people go to it, people use it, clinicians and practitioners alike. When we look at the later line setting, tivozanib is an agent that can be utilized. Cabozantinib could also be utilized in that context later on or other TKIs.
I think what Terry highlighted really so beautifully in his talk is that across every single line here, there's also an opportunity for integration of casdatifan, targeting the vulnerable opportunities that are present in existing modern-day treatments. Next slide. We've chatted a lot about the frontline setting and what nivo/ipi brings to the table and what it doesn't bring to the table. As a regimen, it is one of the most widely utilized regimens for frontline RCC, particularly in the academic setting. The tantalizing piece about the use of nivo/ipi is that in a small subset of patients, looking around approximately one in 10 patients, if you will, there can be long, durable remissions that can be observed with nivo/ipi frontline. That is an opportunity that we want to continue to expand about.
Unfortunately, one in five patients develop primary disease progression and have no response to treatment, and that is the number from clinical trials. When we look at actually real-world data sets, that number is likely much higher. There is a huge opportunity. There is no biomarker that exists to try to identify who these individuals are. This is where casdatifan can have a dramatic role with potentially reducing the high rate of primary PD and expanding the PFS in the frontline setting in the context that also spares a TKI. This is something that clinicians desire, patients desire, and is a huge opportunity.
From the early data that we are seeing from ARC-20, even just with the cas/zim combination, we are already seeing progress being made in that space with the introduction of casdatifan in the frontline setting with decreasing primary PD rate and expanding the PFS rate. Obviously, the data will continue to evolve, but this is incredibly exciting, and we are incredibly excited about the triplet in this combination. Next slide. As we think about second-line opportunities, and I think this is the workhorse right now with the PEAK-1 study, cas/cabo really has the potential to improve PFS and potentially OS and provide dramatic benefit in TKI pretreated populations. What is striking about the data around cabozantinib is in the post-IO data sets from LITESPARK-0101, from CONTACT-03. The performance of cabozantinib has been pretty consistent, right around 10 and a half months for a median PFS in that context.
When we also integrate all patients who have been exposed to a prior TKI, we do see reduction of the activity of cabozantinib. As is highlighted, the key here is cabo point where we see a drop off in the PFS for those TKI-exposed individuals. I think the opportunity here really lies in that the cas/cabo could improve upon this PFS. Cabo is the most commonly utilized, most commonly prescribed TKI in the second line setting. Anchoring to that agent that clinicians know well and are utilizing very well, I think is a really strong play. There is significant comfort with utilizing this drug, managing toxicities, but then also an opportunity to dramatically improve upon the benefit and potentially build with expanding the response rate, expanding the PFS, and also the overall survival in this context. Next slide.
When we think about the late line setting, where there is a tremendous opportunity as well, with every line of therapy, there are patients that drop off and are not able to receive a subsequent line of treatment. I think there is a tremendous opportunity here as well. When we look at the drugs that are utilized, belzutifan monotherapy, tivozanib, other TKIs that patients may cycle through, the rates of primary progression are high, and almost a third of patients who receive belzutifan greater than that have primary PD in the later line setting. Those are tough odds to stomach when you are facing potentially your last and final line of treatment potentially in RCC is that 1/3 likelihood that the treatment is not going to respond at all. We are sitting on a PFS that is less than six months median with both tivozanib and belzutifan.
There is tremendous opportunity to build upon this. The bar is quite low. Casdatifan, we view it as a best-in-class agent from the data that Terry has shown regarding the pharmacodynamic properties of the agent. I think there is tremendous opportunity with the combination in the later line setting. Next slide. This is that slide that I think Terry had alluded to that myself and Richard are going to really talk through. There is a lot of opportunity to color the entire landscape of RCC treatments from frontline, second line, and third line with casdatifan combination treatment options that leverage the vulnerable opportunities of the already existing standard-of-care agents in this context. There are opportunities for a TKI-free regimen with PEAK-20 of cas nivo/ipi that I think is going to potentially be a game-changer in this context, lowering primary PD and expanding PFS.
I love Terry's phrase of the benefit of a TKI without the baggage of a TKI. Then there are also opportunities to combine with a TKI. There may be people that have rapidly progressive disease and need that dual inhibition where we look at PEAK-21 of cas-axi-PD-1 versus axi-PD-1. Then also thinking about novel combinations, as we have seen many of the bispecifics entering into the testing landscape across the board. There are multiple agents and I think tremendous opportunity there as well. In the second-line setting, the most robust data coming from the ARC-20 combination and providing rationale for dual complementary targeting of the HIF/VEGF pathway with the combination of casdatifan and cabozantinib there. That is going to be a tremendous opportunity there in the second line.
Then later line, thinking about PEAK-30, as I said, the bar is low with tivo monotherapy in the later line setting and looking at a combina, [inaudible]
Okay, thank you. This is Daina Graybosch from Leerink Partners. I have a question about front line because we are going to go into ESMO and see the LITESPARK-012 data set. In that abstract, my read, similar to what we saw with COSMIC-313, is reduced treatment exposure of all three drugs in the triplet may have contributed to the lack of benefit there. I wonder if you agree, and as we see Arcus move into front line, if there is anything you can do in these trials to better keep patients on all three of the triplet therapies to avoid the negative outcomes we saw in LITESPARK-012.
Yeah, no, thank you so much for that question. It is incredibly timely. I think it is very different when you have agents that have overlapping toxicity. I think this is what was the Achilles heel of COSMIC-313, also LITESPARK-012. I think when you run into issues with hepatitis, you are not sure if it is IO-mediated, you are not sure if it is TKI-mediated. Patients end up stopping both agents because you are trying to tease it out. COSMIC-313 was also a blinded study, and so you did not know what patients were on and just stopped everything. I think that this is very different when you have a side effect profile that is very distinct and unique from that that you experience with the IO. So the IO-mediated itises that you see are. We are not seeing hepatitis with casdatifan, pneumonitis, colitis, diarrhea.
These things that we know happen with TKIs do not really happen with casdatifan. I think that is a real beauty of the agent. The other thing that I will highlight is when thinking about a TKI-sparing regimen, the backbone of len/pem is a high bar actually in the frontline setting. I think that is also another reason why the LITESPARK-012 study failed. Because I think there are more vulnerabilities with nivo/ipi that can be leveraged with combination strategies when we are looking at the frontline.
Li Watsek from Cantor. Dr. McKay, I have two quick questions. First, can you talk about your TKI sequencing decision, especially around lenva using frontline versus second line in light of the LITESPARK-011? Then for belzutifan, how would you choose between the mono versus lenva plus belzutifan in second line?
Yeah. Very good question. I think in my context, I tend to be a dramatic user of nivo/ipi. I want my patients, if they are able to experience a long, durable remission, if they have the opportunity to experience that. For those individuals that have really rapidly progressive disease and really need a TKI, they are either in visceral crisis, have high volume of lung, liver, what have you, metastases, and really need a TKI, cannot afford them to have that one in five primary PD rate. I am predominantly using lenvatinib pembrolizumab in the frontline setting. When I do use a TKI in the frontline, it tends to be lenvatinib. We have seen really high response rates with that combination, low primary PD. So for the person who needs it, that is the combination that I go to. Cabozantinib is my go-to second-line agent.
I think that by using either nivo/ipi or len/pem frontline, I think it opens up the opportunity for cabozantinib second line. I think the data around monotherapy use versus the combination of belzutifan plus lenvatinib, I think a lot of it is TKI toxicity. What's the fitness of the patient? That regimen was just recently approved about a month or so ago. I think it's just being kind of considered in practice and trying to identify who is the right individual for that. Generally, my TKI sequencing is lenvatinib frontline, cabozantinib second line, and then depending on the fitness of the patient, belzutifan, and the underlying kind of distribution of metastases, even profiling selection of either belzutifan or tivozanib in later line.
Great. Thank you so much. Jon Miller from Evercore ISI. You just mentioned that you're an enthusiastic user of ipi/nivo in the front line, but it remains a minority regimen. In your opinion, what does it take to drive use in the first line off of TKIs? Is there a realistic possibility of expanding TKI-free regimen use with HIF-2?
Absolutely. I think if you decrease the primary PD rate, you will increase utilization of nivolumab. People don't want to get burned. They want to start somebody on a regimen, and they want them to do well at their first scan assessment and not have to be faced with making a treatment decision or having the patient have primary PD. Primary PD is not a good thing, and it can taint an individual's experience with utilizing that agent. I think if we can decrease the primary PD rate with a well-tolerated agent with non-overlapping toxicity with the backbone of nivo/ipi, that's going to be a huge win and is going to drive up utilization. As individuals are utilizing belzutifan in late line, it's a not great monotherapy. Again, that primary PD experience is going to taint utilization.
I think to integrate it in the front line in a TKI-free sparing strategy will be a win.
Tara?
Hi, Tara Bancroft from TD Cowen. I am wondering if I could build upon one of the previous questions of when making prescribing choices, would you base your choice more on hazard ratio or absolute numbers? Which numbers actually matter most to you? In those considerations, when you are comparing data sets, how much would you consider baseline differences in the different patient populations that are enrolled? How important are those nuances to you in prescribing?
To me, they are very important because I am very granular, very detail-oriented. I think, thinking about broadly how a community oncologist may interpret the data, I think there is a lot of education. I think it is all the above. It is not just the hazard ratio, because the hazard ratio is reflective of, well, what is the performance of your control arm? If your control arm is underperforming or over-performing, that is basically going to be an interpretation of the hazard ratio. Yes, absolute numbers matter, but it is a combination of the two. I think the absolute numbers can be, you have to think about the patient population when you look at that. If there is more favorable risk patients that have low volume disease that do not have an intact primary, your objective response rate is going to be automatically higher. We see that.
The likelihood of somebody achieving an objective response that has an intact or a complete response with an intact primary is incredibly low when they still have their primary intact. These are all the things that I think matter. All these trials are different. They enroll different populations. I think the other thing, too, is we also have to think about the timing of when the trials enrolled. CheckMate 214 was the first foray of challenging the standard of care of sunitinib frontline with nivo/ipi being utilized in a large phase III in RCC. There is a learning curve that happens with actually being able to familiarize yourself with how to treat AEs, how to manage AEs. If you actually look at contemporaneously these trials that we are enrolling later on, people learned how to use the medications and how to use combinations and how to mitigate tox.
We learned how to adjust and modify the criteria and management algorithms. I think that these are all really important things when we are considering interpretation of the data.
Perfect.
Cool. Thank you. Dr. McKay, thank you for your time this morning. Charles Zhu from LifeSci Capital. I have one question here. Based on your experience as well as any available clinical literature out there, how would you describe patient outcomes segmented by whether or not they are able to receive all four doses of CTLA-4 during the induction phase relative to those who receive three or fewer cycles? Thank you.
Yeah, very good question. Actually, not to say we do not know the right number, the question to ask is why are the patients not able to receive those regimens? Is it truly driven by an IRAE that we think could potentially be suggestive of an enhanced effect to treatment? I do not know whether patients received one to two. I think the dichotomy is in the one to two versus the three to four. It all depends on what is the reason that the treatment was stopped. Was it for an IRAE? How was that IRAE treated? With steroids or other agents that could potentially blunt the activity of the immunotherapy.
This is why it is actually really critical to not have overlapping tox, because if you are stopping the, you have hepatitis that comes on from a TKI that has a long half-life, and you wait a couple of days and still have high LFTs, and nobody feels comfortable with that. Then you are layering in steroids or tacrolimus and other things like that are basically blunting the immune response. That is not a good thing.
Hi. Salim Syed from Mizuho. Thanks for the color, Dr. McKay. One from us on the second-line data granularities here. We are getting some questions just on the 14 and a half months median PFS that was presented today. I was wondering if you could put that into context. I think some people were expecting a slightly higher number for that cas/cabo second line, especially when you look at the late line 13 months median PFS. I am wondering if you could put this into context for us, especially as it relates to the OS data that was presented today, where we are seeing separation earlier than what we saw in LITESPARK-011 and over LITESPARK-011 bel's lenva. How do you put this into context for your overall thinking here?
And then second to that, could you comment on just your thoughts around HIF-2 alpha having a greater contribution on the back part of the curve versus a TKI?
Yeah, very good question. I think regarding the 14 months, first off, I will say this is a very respectable and very encouraging PFS that we are seeing greater than a year in the second-line setting in a very heavily pretreated patient population with multiple prior exposures. I am certainly not discouraged by that. And when we think about the bar for approval and the bar for utilization, as we stated from cabozantinib is sitting in that 8-10 month range, depending on degree of TKI exposure. I think that is actually quite encouraging. I think the latter part of your question around integration of HIF-2 alpha, I think we are seeing post-VEGF progression.
There is upregulation of the pathway, and casdatifan really has the tremendous opportunity to overcome that expression, and pathway upregulation probably much better than we are seeing from our pharmacodynamic studies than belzutifan quite honestly dramatically suppresses EPO and blunts the pathway. And so I think the complementary aspects of VEGF and HIF in this context, and specifically with CAS, is going to be a really good thing in the second line.
Hi, Dr. McKay. This is Jane from Goldman Sachs, and I am asking this question on behalf of Rich. My question is, how do you think CAS will translate in a TKI-containing regimen first line, given the first-line data here is without TKI? Is there anything you see from the second-line CAS plus cabo that could be read through? Basically, is there any concern for using a first line in a TKI-containing regimen with axi based on TKI combo data in second line? Is there anything you see in the first line presented today? Thank you.
Yeah, very good question. I think the thinking here is in every way that a clinician could think about utilizing a modern-day regimen, there is an opportunity with casdatifan being added. That's really the take-home message. While we haven't shown you today the combination of a frontline regimen that includes casdatifan with a TKI, with a checkpoint inhibitor, I think that is the strategy. The strategy is targeting in every way that a clinician may think about, okay, I'm going to use nivo/ipi for this. Well, I'm going to add casdatifan to it. I'm going to use IO TKI for this. Well, there's an opportunity to potentially test casdatifan in addition to that. I'm going to use cabo second line, or I'm going to use a TKI second line. Okay, add casdatifan to that. I'm going to use TKI monotherapy later line. I'm going to add casdatifan to that.
The data is pretty dramatic and robust that across the board with an agent like casdatifan who works pretty well or beautifully as a HIF-2 alpha inhibitor and is likely best-in-class agent of combining.
We'll take one last question.
Hi. This is Supawat Thongthip from Truist Securities. Dr. McKay, just as a clarification to your earlier comment. So with belzutifan and lenvatinib now approved, if cas and cabo were also approved, what would drive your choice between the two? Would it be primarily the profile, the PFS and OS profile we've seen today, or the ease of integrating cas plus cabo, given that cabo is already a backbone? Secondly, what do you think of PEAK-1's probability of success, having seen the results that we've seen today? Thank you.
Yeah, very good question. I actually think that it's a little bit the latter and for the fact that cabozantinib is already being widely utilized as the go-to second-line agent. METEOR really set the bar high. It was actually one of the first studies ever in kidney cancer on its first analysis to improve all three objectives of objective response rate, PFS, and OS, statistically significant improvement across all three of those parameters. The utilization of cabozantinib in the later line setting is anchored in high-level phase III data. There is no data for lenvatinib monotherapy. The best data that we have for lenvatinib utilization later line is from 150 phase II trial of lenvatinib everolimus. Lenvatinib, in people's experience and in the data, has a lot of toxicities that's associated with it.
I think to anchor with an agent that has a really high bar with regards to the rigor of data that led to its utilization and also comfort and ease of use is going to really be an excellent thing. I think that the data here really foreshadow what we hope to see in P1. I am incredibly optimistic about this study being positive from the data that we've shared with you thus far and being integrated into the landscape.
Thank you, Dr. McKay, for your time today. We'll let you get on with the rest of your day. I know there were some more questions in the queue, but the management team will take questions at the end. I really appreciate you joining us today and we'll let you get to it.
Thank you.
The next section, Juan will talk about the translational science.
Thanks, Rana. Good morning, and thank you for your attention. I am going to take about 10 minutes. My job today is to walk you through a couple of key points. The first one is I am going to link the biology, the PD changes associated with casdatifan treatment with every relevant clinical endpoint in our combined 121 patients late-line monotherapy cohorts. I am going to link the PD changes with primary progression. I am going to link the changes when I am introducing a new PD marker today. I am going to link both of them with objective response rate with PFS and with OS. The reason that I think the monotherapy cohorts are the ideal place to do this is because we have one single agent. We do not have confusing multiple mechanisms of action, potentially pushing some of these PD markers in opposite directions.
HIF-2 alpha regulates probably hundreds of genes, many of which are relevant for clear cell RCC biology. EPO does not really play a particularly important pathological role, but it has the advantage that it is something that we can measure easily in the periphery, and it gives us a really good integrated sense for how much HIF-2 alpha activity a particular patient is experiencing and the changes associated with HIF-2 alpha inhibition.
I will remind you that at the approved dose of 120 mg, belz reaches a maximal level of inhibition, the median has been described more than once, around 64%. Terry already alluded to the published data whereby the EPO levels in serum revert back to baseline after about three months. In the case of casdatifan, we can reach that level of 60 some percent reduction in EPO at a dose of 20 mg daily as we continue to go up in dose.
The figure to the left represents an integrated assessment of maximal level of EPO reduction as a function of daily dose of casdatifan. By the time we get to 50 mg daily, we pretty much reach the maximal level of inhibition, and it does not get in a meaningful way any higher as we have gone all the way up to 200 mg in the dose escalation. Not only do we achieve in general, a higher level of EPO reduction with a median of 84%, 85%, that PD effect is sustained for the duration of treatment, as Terry reminded you, that shown here for the 100 mg cohort, but pretty much the same for any of the other dose levels. This is a graphical illustration of where each one of the 121 patients is represented by one bar.
What we've done is illustrate each patient's maximal reduction in serum EPO levels and color-coded to illustrate best overall response. What you can see very clearly is that there. Well, maybe you can't see it on these screens, but if you turn around, you can very easily see that there are way more blues and greens towards the right than towards the left. Patients with better objective responses tend to be associated with deeper EPO reductions. That dashed line running horizontally represents the median of 84%. You've seen versions of a figure shown here to the left, the PFS comparison. This is simply updated with the latest clinical data cut. There's no attempt to do any kind of patient selection here. All we've done is to split the population down the middle using the median of 84%.
You can very clearly see that those patients that experience greater than the median in terms of maximal EPO reduction tend to have a longer PFS than those in the lower than the median have. The representation to the right, still immature because many of the patients are still alive, fortunately. But it shows you a similar segregation when you look at the OS curves. Patients with deeper EPO reductions tend to live longer. Now, this is, I think, remarkable that this PD effect is first noted a couple of weeks into treatment and sustained, but nevertheless, it starts to be detectable right at the beginning of treatment. It's reading through to a clinical outcome two, maybe three years later. Now, if you're not convinced of that relationship, I'm going to introduce PAI-1 as another marker of systemic HIF-2 alpha activity.
In this case, we could spend an hour talking about the potential pathological role of this protein. It may be more than just a PD marker, but for today's update, we'll just use it. It's just another independent measure of systemic HIF-2 alpha activity. Illustrated here and going back, combining both the EPO and PAI-1 relationship between the extent of reduction of these markers and the best overall response. And you can see here that similarly to what we've shown previously for EPO, shown here now for PAI-1, is the patients with an objective response tend in general to have deeper PAI-1 changes than those with stable disease or those with primary progression. In the case of PAI-1, the average or the median level of reduction is 45%.
What that tells you is that, and it's independent of casdatifan dose, what that's telling you is that when it comes to systemic circulating PAI-1, about half of it is transcriptionally regulated by HIF-2 alpha. The other 50% of that protein is obviously regulated by other pathways. When we do the same median split and look at the deeper versus less deep reduction in PAI-1 levels, you see something very similar to what I showed you a second ago. Patients that experience the deeper reduction tend to have longer PFS and longer OS than those in the lower half. Again, not an attempt to select patients, just an attempt to link biology, the efficacy or the activity of casdatifan with the clinical outcomes. Now here, this is, I think, one of my favorite slides.
What we've done here is said, "Let's now make an attempt to identify a natural breaking point. Let's identify the 15% or so percent of the patients that have a lower level of reduction on either biomarker or both, and let's separate them." When we do this, a couple of things happen. First of all, you see the majority, that darker upper curve, that contains about 80% or so percent of the total population, does pretty darn well. You also notice that it barely has any primary progressions left in it. If you look on the other hand to the three curves towards the lower part of the figure, those represent patients that are in the worst 15 or so percentile of either biomarker or both. You see what we've done is essentially taken out all of the primary progressors out of the 121 patient pool.
What this is telling us is that those patients roughly, and this ties in with some of the work that we published in the Nature paper, where we define high HIF-2 alpha activity in different ways. We did it using immunohistochemistry, the amount of protein. We did it looking at the transcriptional signature of HIF-2 alpha. We always come back to the same place, where roughly 15% - 20% of tumors that don't appear to have a lot of HIF-2 alpha activity, and they really don't care that much about casdatifan intervention. The other 80% of the population that has meaningful levels of HIF-2 alpha tend to do pretty darn well. Before I move on to the second part of my update, which is introducing the concept of adaptive HIF-2 alpha activity, I'd just like to recap what I hopefully have convinced you, given the limitations of time.
There's a link between the extent of HIF-2 alpha inhibition that we can achieve with casdatifan, as measured by two completely independent circulating PD markers. There's a relationship between the rate of primary progression, objective response, PFS, and OS. So the numbers, the curves that Terry showed and that Richard's going to come back to, are clearly reflective of biology and the strength of HIF-2 alpha inhibition that cas is able to reach. Now, shown here, I'll remind you that the figure to the right is taken from a Merck poster. It's breaking down. In this case, we could have done something similar with the LITESPARK-005 data, except that the LITESPARK-013, the phase II data, was broken down more granularly. What you can see here, the three curves represent the number of prior TKI regimens that those patients that were treated with belz had experienced.
You can see that there's a very clear degradation of belz performance as the function of prior TKI treatment and as a function of the number of TKI regimens. If you look at our monotherapy data shown in the left of the slide, you see that there's absolutely no daylight between patients that received one prior TKI, two prior TKIs, or three prior TKIs. We think this is a reflection of both the biology that sets in as a result of the treatment with a TKI, but also the strength of HIF-2 alpha inhibition associated with cas. We're working on a paper that should be submitted pretty shortly here, that pretty much integrates a whole bunch of data that leads us to propose that anti-angiogenic treatment results in an increase. I won't call it upregulation because we don't think it's necessarily transcriptional.
There's just an increase of HIF-2 alpha activity in tumors that have been treated with anti-angiogenic agents. The evidence comes in two forms. In the tumor, we see increased protein stabilization. We've developed probably the world's most sensitive and specific immunohistochemistry assay for HIF-2 alpha, and HIF-1 alpha, for that matter. We see increased levels of HIF-2 alpha protein. We see increased transcriptional signature of HIF-2 alpha in patients that have received prior TKI treatment. In the periphery, we see increased levels of EPO and PAI-1 in patients that have received TKI within approximately about six months. There's sort of a window of time after which that prior TKI exposure is no longer clearly detectable in the periphery. This is my last slide, just trying to introduce the concept. It's not really all that complicated.
We all know that clear cell RCC is a tumor that comes into existence under the pretense of being hypoxic when in reality it's not. That translates into a tumor that's probably one of the best vascularized solid tumors that there is. The first time they see a blood supply drying agent, the TKI, they do relatively well. You dry up that blood supply, tumor shrinks. There's no doubt about it. The problem is what happens after that true hypoxia sets in. Evolutionarily, the tumor and the non-tumor cells are pre-programmed to respond, among other things, by relying on HIF-2 alpha activity for survival. That probably means that outside the tumor, there's an upregulation of HIF-2 alpha, but also the tumor may undergo selective pressure to identify clones that are perhaps more dependent on HIF-2 alpha, better able to thrive through a condition of true hypoxia.
Now, the model explains why after a pretty good response on a first round of TKI treatment, you really don't do very well on second and third TKIs. Whatever benefit you're getting from the second or third line of TKI may have just as much to do with a non-VEGF inhibitory activity of that TKI, probably not getting much from an anti-angiogenic effect after the first round. Secondly, you've now set in motion something that if you're a borderline HIF-2 alpha inhibitor is not particularly good. You now have a bigger foe to fight against. There's more HIF-2 alpha activity. How well you do with your HIF-2 alpha inhibitor probably depends on how strongly that HIF-2 alpha inhibitor is able to fight against that increase, that adaptive HIF-2 alpha activity.
We think this explains why belz is so dependent or performance is so degraded by prior TKI treatment while cas appears to be unaffected by that. I don't want to leave you with the impression that cas works better in prior TKI treated patients. It's simply it does well against people that have never seen a TKI, and the data supports that it performs just as well in patients that have seen multiple rounds of TKI treatment. With that, I'll turn it over to Richard.
Great.
Okay.
Yeah. I was told I should try and be brief. We've been here for a long time, so I'll try and move fast, but there's a few key points I'd like to make. There we go. You've seen this before. It's the development plan. It's really important to understand, and Dr. McKay, I think, highlighted that when you see a molecule with the impact of casdatifan, it really needs to be developed and available to all patients with clear cell RCC. I mean all patients in the front line and subsequent lines of therapy. As we put together the development plan, it is to add casdatifan to the preferred existing regimens across each line of therapy.
In the front line, the key here is to be TKI free, and we know that the physicians, and good for the patients, they can now live 5 - 10 years, and hopefully longer survival with adding casdatifan. So they have to decide what to use in the front line, also thinking about what they'll use in subsequent lines of therapy. Being TKI free with long-term survival, that's the ipi/nivo regimen. The only reason it's not used by far more is the liability of that primary progression, as Dr. McKay said, where one in five patients actually show progression at the first scan, which is very disheartening for the patient and the treating physician. So the key here is to meaningfully reduce that rate of primary progression, which would drive the ipi/nivo plus cas use far higher even than it currently is.
Importantly, there are going to be patients that are highly symptomatic, so they do need a TKI up front for early inhibition there. That we have selected axitinib and PD-1 as the partner for casdatifan. Axitinib was deliberately selected because it, I'll say, is the kinder, gentler TKI from a tolerability perspective for the patient. That's because it's a much more selective anti-VEGF TKI, so better tolerated. That theme will continue. Importantly, we also have three different partnerships for additional potential frontline treatment of patients. That's our partnerships with BMS, Pfizer, and Summit, all testing their VEGF PD-(L)1xVEGF bispecific with casdatifan in frontline patients. So those will be additional frontline opportunities. Quickly moving to the second line. Everyone's familiar with PEAK-1, our cas plus cabo versus cabo. I think very high probability of success.
The key here is we selected cabo, again, deliberately because it's by far the preferred choice in TKIs in the second-line setting. So very easy for the prescribing physicians who already use cabo predominantly in that space to just add casdatifan to the current algorithm. Lastly, look at the later line setting. There we're combining cas with tivo. That's again, a very selective, very well-tolerated TKI, and we think that combination will take over all of the monotherapy use in the third line and be the dominant regimen in the late line as well. Quickly, this is our ARC20 study. It's our phase II with multiple expansion cohorts that generates all the data that we've talked about today. This is for data in the monotherapy as well as many different combination therapies, again, across first line, second line, and late line.
We have over 300 patients that have been on casdatifan and over 200 patients that I'll discuss and describe today in the data sets we're sharing. Quickly, I'll talk about the late line setting with our, as you've seen, unprecedented overall survival. You've seen this before, so just to highlight the keys here is we actually, at the 24 month or two year landmark, have 2/3 of the patients still alive and surviving. I think what's more amazing is, as Terry said, that three year landmark here is about 50% for the median still alive at three years. This is late line. I'll remind you, this is generally a median of fourth-line use, meaning the patients had a median of three prior lines coming into this cohort. Not sure why it won't advance. Safety, we've shared the safety before. It generally looks like the safety profile of belzutifan.
We've described it before, but now we have much longer follow-up, and the safety profile has not changed despite having the much longer follow-up now. So late line setting, again, unfortunately for patients, there are three different monotherapies that are used or available. None of them have a PFS of even six months, and none have shown an overall survival benefit. We think this is a very real opportunity to add cas to tivo and to take over the entire third-line regimen, and we can meaningfully improve both PFS and OS for these patients. We strongly believe this because of these data here. Here we overlaid our cas monotherapy. Again, basically a median of fourth line, and that's compared to TiNIVO-2 in the gray. Those are patients that received tivo monotherapy in the second or third line. That trial enrolled both second and third-line patients.
Not hard to see the very clear difference in PFS on the left, but also very clear difference in OS on the right. These are the monotherapies. Now adding tivozanib and cas, compared to tivo, should only further separate the curves. We are enrolling now an arm in ARC20 that is a randomized arm, casdatifan plus tivozanib verse tivozanib, and this arm is enrolling all patients that have had prior HIF-2 or prior belzutifan experience. Our phase III trial will enroll all comers. That'll be patients that have had or have not had prior HIF-2 experience. Quickly moving to the second line, cas plus cabo. We think this is really positioned to be the regimen of choice, particularly because we've started with cabo, which is the TKI of choice.
You've seen this from Terry, the PFS on the left and OS on the right, where we meaningfully beat all prior benchmarks for cabo. That's why we have such strong expectations and high probability of success since PFS is the primary endpoint of the trial. But also think we can very meaningfully beat OS on the right. Keep in mind, LITESPARK-011 had a hazard ratio of 0.85, so they almost had significance. We only have to be a little bit better than what they were in order to achieve a significance, and that would be a complete game changer. From a safety perspective, so far the experience really is looking like cabo's safety profile, plus the anemia and hypoxia associated with the potent HIF-2 alpha inhibitor. The trial you've seen before, this is now enrolling.
It's enrolled very well. In fact, we're going to complete enrollment in the next few months. It enrolled really to expectation. It's cas plus cabo versus cabo in a two to one randomization. Importantly, all the statistical power, our alpha, is on PFS as a sole primary endpoint. The other reason that we can probably win on overall survival is we then recycle the alpha, so the statistical test of OS has all the statistical power recycled from PFS to OS, so we don't have to split our alpha. We have to be slightly better than 0.85, and we have more power or full power because all the alpha will be allocated to OS.
Move to the frontline where we've seen now the unprecedented early data, and you saw this, and I'll just highlight again that the main reason ipi/nivo is not the regimen for every patient is the risk or concern of primary progression. You can see we have two cohorts now in ARC20, cas plus zim, so without ipi with 11 and a half months median follow-up. We have cas plus zim plus ipi that recently fully enrolled, and that was for a safety set for the FDA and EMA. I think Terry said we have all the data now and it's packaged up ready to go to the agencies to start our phase III trial. You can see now we have two separate cohorts, both showing single-digit primary progression. That's going to be the meaningful element there that will really drive adoption.
I think enrollment in this trial is going to go very fast. You can also see the confirmed response rate of 50%. That's quite high, particularly compared to ipi/nivo. On the right, you've already seen how the separation of curves there are with the primary endpoint is progression-free survival. That's where we are to win on this trial. Of course, ipi/nivo already has the advantage on overall survival compared to all the other regimens. Importantly, and there was a comment or question earlier about the safety, this could be, I think, clearly the first triplet that shows high tolerability for the patients. It really does look like ipi/nivo, that's the immune AE profile, plus the hypoxia and anemia associated with HIF-2 alpha inhibition.
We don't have the overlapping toxicity, and as Dr. McKay said, predominantly that's the GI toxicities and liver hepatitis, colitis, pneumonitis, et cetera. None of that is associated with HIF-2 inhibition. We've had dose intensity-wise, exactly 78% of the patients that have been on the trial long enough to get through four cycles, meaning more than 12 weeks on the study. 50% of them did receive all four cycles of ipi along with the zim, and then an additional 28% got three of the four cycles. Quite high utilization to get, as Dr. McKay said, and that's, I think, how we all think about when we prescribe ipi/nivo. The goal is to get them through at least two and ideally three or four cycles of ipi. At that point, all the patients are off ipi and they'll just be on the PD-1 plus cas.
And as I just showed you, our cas plus zim, we have gone out now 11 and a half months of median follow-up, so we are very confident and comfortable with that safety profile. Otherwise, the immune-mediated events that we are seeing are very consistent with ipi/nivo, both in frequency, severity, but also timing, and that is the important thing is when do these happen? And they have been happening very consistently with what we see generally with ipi/nivo. So as expected, cas doesn't change the immune-mediated events. So the frontline trial is cas plus ipi/nivo versus ipi/nivo. It is an all-comers from an IMDC perspective. And another important point, the PFS is the sole primary, so all the statistical alpha is on PFS. And as you can imagine, if we meaningfully reduce primary progression, that instantly elevates the PFS curve by that magnitude, and that will carry forward.
So high probability of success on hitting on PFS. And then after the first four cycles of ipi as depicted on the bottom for the continuation of the trial, which can go years, it is really a test of cas plus nivo versus nivo. So with that, I will hand it off to Bob.
Thank you, Richard. So we have had a chance to share a lot of data today, and we think that a clear picture has been painted, and we believe that is the best therapeutic for the treatment of clear cell RCC in terms of casdatifan. When you look at each of the key therapeutic attributes, we think that picture becomes super obvious. So let us start off with disease control. And Dr. McKay mentioned the importance of disease control and giving a patient an opportunity to succeed in each line of therapy. In our late line monotherapy, our disease control is on par or better than a TKI monotherapy in any line of therapy. That is really been the gold standard when you think about disease control historically.
And I think we showed you a lot of spider plots today across different lines of therapy, and you saw that sort of consistent theme around disease control, which we think is quite impressive for casdatifan. Moving to survival, both OS and PFS, we have showed really exciting OS data, obviously in the late line, and it is emerging very nicely as well in the second line, alongside very compelling PFS in every setting.
Importantly, we have talked about the tail of the Kaplan-Meier curves here, and we think that that is going to continue to show through as all of our data matures over the coming years. Finally, tolerability is important. Dr. McKay talked about the ability to combine casdatifan without overlapping toxicities across treatments, and we think that is very unique, having a treatment like casdatifan that doesn't have that overlapping toxicity. So Terry and Richard have already reviewed this slide.
I just want to make sure that everybody's left with the impression that we are laser-focused on the first two commercial opportunities for casdatifan. First of all, PEAK-1 is poised for success. That's a $2 billion opportunity. Then perhaps even more importantly, we have a clear line to success in the frontline with the ipi/nivo combination, where we have a $4+ billion commercial opportunity. The plan obviously lays out additional opportunities both in the frontline and the late line as well. Let me talk about how that development plan then translates into the commercial opportunity with a little bit more specificity. Starting with the second line, and this is important, I think Dr. McKay underscored this, we're building on the physician-preferred regimen in the second line.
In fact, when we spoke to 50 physicians and asked them the question, would they utilize the newly approved LITESPARK-011 regimen or cabozantinib monotherapy, 50% of those physicians told us they were going to stick with cabozantinib monotherapy. That's because of their preference for cabozantinib. We think that's really important in terms of how we built that regimen in the second line. In the frontline, this is the largest commercial opportunity with no HIF-2 alpha competition. In PEAK-20, as Richard highlighted, we're combining with the regimen and Dr. McKay made this point, physicians reach for to provide patients with the best opportunity for long-term survival. The weakness here obviously is disease control, where 50% of patients do not stay on therapy beyond a year.
We've shown you data to support our belief that casdatifan can address that key weakness in the frontline and that it can be the dominant regimen in the frontline, which is a $4 billion opportunity. In addition to PEAK-1 and PEAK-20, we're also exploring opportunities or regimens that address the VEGF pathway in the frontline, AXI and the PD-(L)1xVEGF bispecifics and a compelling combination with dominant later line share in tivo. I think one of the themes that continues to emerge in our data across lines of therapy is the incredible durability of casdatifan, and we think that that introduces a huge upside to the overall commercial opportunity and duration of therapy, and we think that story will continue to play out over the next few years. With that, I'm going to turn it back to Terry to wrap up and then we can take your questions.
Thanks, Bob. Thank you all again for your attention and we look forward to the Q&A. I just want to close. I'm not going to go through these in detail. You can look at them. But we have a lot of milestones coming over the next few years and I'll just bucket them. We're going to continue. As you've seen, we've done medical conference after medical conference after medical conference. We'll be continuing to do that. We do that not only to share the data with you, but to generate the sort of investigator enthusiasm we work, I think with amongst the best investigators in this field, and we feel it's important to get all of our data out there. So what you're going to see is the data that we shared today.
Obviously, as they continue to mature, we'll be sharing them on a fairly ongoing basis and we'll continue to publish as Juan was mentioning. I think then the next probably most important aspect, Richard talked about fully enrolling PEAK-1, that's coming shortly, starting PEAK-20. That's going to enroll gangbusters and as Richard suggested, not only is it going to enroll quickly, but it's going to read out quickly given that primary endpoint of PFS. Then you should expect to see a continuous flow of information around these other studies that we're doing. We barely touched on it, but obviously everyone was looking at all the various ESMO abstracts. You saw the exciting data with the anti-PD-1 VEGF bispecific. We think that could end up being an important basically version of TKI anti-PD-1 without the baggage of the TKI.
We're well positioned to have the one reagent that has the monopoly, and I think there's a large number of those bispecifics and we're working with three of them, and we'll be executing the study that's in collaboration with Summit. We're confident in how that will get done and enroll. I think the next part that's other separate bucket is our studies with tivo, our studies with axi, our studies with the bispecifics, and you'll see a continual flow of information. I just would finish by summarizing, bring you back, elevate a second. I think our data show that not only across all lines, but in each line, we have the opportunity to completely win. The other thing that should not be lost on you, and this is something that Bob, why I would say everything he tox about is the conservative end of things.
This is a therapy. When you think about HIF-2 alpha drives this cancer in all of these patients that basically we're seeing efficacy. So whether it's 80%-90% of these patients. It also may even be induced in those patients where the TKI is bringing some benefit and they were either lower HIF-2 alpha activity or no HIF-2 alpha activity. So extraordinarily important in the setting and the other thing to keep in mind, there is zero evidence of any meaningful resistance development to HIF-2 alpha inhibition. So we believe that you will be able to sequence casdatifan with the other agents for the duration of their therapy. So when we talk about backbone, a lot of this discussion is sequencing this, sequencing of that.
We actually believe all the evidence, and we'll be generating more evidence ourself, that you will be able to have a backbone therapy of casdatifan, much like you see Bev in colorectal cancer, where patients will benefit from casdatifan throughout their 5- 10, hopefully 10+ years on treatment. I'm going to stop there. Thank you all again. Thank the presenters. Thank Rana in her absence. I think Rana did a fantastic job of both laying out the current treatment paradigm, how it may evolve, and would rather have her answering the questions from you guys then. Also, appreciate that, and we'll be happy to take some questions now.
All right.
Will someone speak up?
We have a very large queue of questions.
Yeah. One person raise their hand, everyone will jump in.
Thanks very much, Terry. This is Yigal Nochomovitz at Citi. I am wondering if you showed the data for OS for frontline broken out by TKI versus no TKI in the prior therapy. I know you showed the PFS broken out. I assume it was consistent on OS.
So on the frontline?
Yes. And second line.
Oh, okay.
Second line with prior TKI versus no prior TKI. Did you show the OS broken out by those?
Yeah, we need much longer follow-up to
Okay
to track OS, really.
But we are already seeing the same type of separation. But to Richard's point, they're simply not mature, as you might imagine.
Yeah.
It's very early.
Okay. I'm just curious, with regard to the spider plots on HIF-2 alpha, maybe Juan can answer this, I gather the results would be similar based on the baseline HIF-2 alpha, as well as the HIF-2 alpha expression. Did you look at that as well?
You want to comment on that one?
Can you repeat the question?
Just I was curious if the spider plot conclusions regarding HIF-2 alpha expression were consistent when you looked at the baseline HIF-2 alpha data as well.
Yeah. The thing with baseline HIF-2 alpha assessment is that we don't have a sample for each one of those 121 patients.
Yeah.
So when you start to get away from the PD marker, you're interpreting 40 samples as opposed to 121. But yeah, generally, and again, we published this, when you look at baseline HIF-2 alpha protein levels or transcriptional signature, evidence that HIF-2 alpha is firing, the correlations, the relationships that I described to you in the PD markers all hold up.
Okay. And then just the last one is, obviously, you have amazing data with the single-digit primary progression, but is there any rationale to do a biomarker strategy to even lower that so that you just don't treat the people that have extremely low HIF-2 alpha driving the tumor, and that's not really necessary?
I can take that. I think short answer is no, because even those with low still have benefit from cas. It's just those with higher HIF-2 or higher EPO, we can cover it with casdatifan. But even in the less deep suppression, that's, say, stronger, we've had stronger responses than belz would have. So I think it still treats everybody.
You should just grab that from Yigal. You're a big guy.
Terence Flynn, Morgan Stanley. I guess the first one is just for Richard. You mentioned in PEAK-1 that OS was powered for 0.85. Just wondering if that's similar for PEAK-20. I know primary is PFS, but just can you tell us anything about the OS stats? And then Bob, you mentioned a doc survey, 50% stay on cabo in light of the recent approval, but did you ask about future cas and what uptake would look like of those that are staying on cabo versus the Lenvima Welireg regimen? Thank you.
Yeah, sure. So let me, I guess, clarify a little bit. So I said LITESPARK-011 had a 0.85, so we just have to be a little bit better to have a meaningful impact, both statistically and clinical impact there. And we are well-powered because we'll have all our statistical power will go from PFS to OS. So while in LITESPARK-011, they split their alpha accordingly, so I think we have both ways to win there on or both ways to improve and win on OS, which I think is the game changer, right?
I think that's, for those of you who didn't catch that, I think that's a really important point about the stat plan in terms of not splitting the alpha and basically getting 100% of the alpha to go on to OS. Not only do we look like we're having a better OS signal, but I think we're well positioned statistically.
Yeah, the way our survey was structured wasn't-- Hopefully it's picking up. Yeah, there we go. The way our survey was structured--
Even less mature than the aggregate data, but it is trending very nicely.
I think it might be in the deck.
Yeah.
It's in the back of the deck. On the surface, it may look better. We've tried to emphasize that we believe truth is best represented by the full 120 patients, but the 100, if anything, it looks better than the other.
Yeah, okay. The next one, the two situations, obviously, there's the question as to whether PEAK-1's going to hit, and then obviously there's a commercial question.
Yes
Around whether that regimen is going to be competitive because people like cabo. I think we know the competitive one. On the trial itself, can you just talk a little bit about the evolution of TKI use and cabozantinib and how that might play out? I think you guys assumed 10 months roughly or so as a comparator, but just maybe give us a little bit more color. Also tie that into how it seems that, obviously, with whatever's happening with EPO suppression, that you have longer, your responses get better over time. Your PFS gets better over time.
Yeah, I think the trial design assumed, I'd say, even a relatively conservative estimate for cabo's performance. So we expect it to perform well by all prior benchmarks, and we're still very well powered to have a win over that. I think reality, and as we've seen going forward, is we expect the cabo arm to perform less well than we assumed from the models, if you will, because of all the prior lenva and second-generation TKIs for the patients coming into our trial. From a trial design perspective, we were pretty conservative and assumed cabo would perform based on prior benchmarks, on the higher end of the prior benchmarks. So I think that's a very high probability success that we would win on PFS from that.
Li Watsek from Cantor. I have two questions. First, for the PEAK-1 trial, do you guys have any sense in terms of percentage of patients that might have prior lenva exposure or prior next-gen TKI exposure? Any reason to believe the number may look different relative to ARC20?
Yeah, I think the number would be similar to ARC20 because they're enrolling near term, let's say, to each other. The only thing we exclude by design, of course, is cabo. I think there's going to be very little first-gen TKI. There'll be some because it's a large global study, so there's still some first-gen use, but the TKI part will be predominantly axi and lenva. My guess is they'll be somewhat split between the two, depending on what part of the world. Lenva's not everywhere in the world compared to axi. I think axi will still probably be more than lenva by trend from that standpoint as a global trial. I still think that the IO use and TKI use will be maybe roughly equivalent at the end of the day from a split.
But we're still enrolling it, so I can't say definitively, but we're getting close to enrollment.
Okay, great. My second question is around the front line, just from a mechanistic point of view. Can you talk a little bit about your conviction level for the IO/IO combination versus TKI regimen-based combination in light of LITESPARK-012? I wonder if you can comment on the biological interplay between IO versus TKI with cas. You share some pretty interesting data of the impact of TKI on the HIF-2 activity. I wonder if you have to look at IO's impact on HIF-2.
I'll speak at least to the first part, and then I'll speak a little second on that one. We see the opportunity for casdatifan in much the way Rana talked about it. Given the orthogonal profiles of the AEs, we see it as being quite well matched to also go with an axi type of treatment. The reason we picked axi is for the reason that when Richard described it as kinder, gentler. We see that our therapy is dominated by the HIF-2 alpha activity, and perhaps there's a spark, almost like with chemo, that the TKI would bring to that. We want a TKI that does no harm.
Axi's selective, it's well-used, it's time-tested, it's a gentler TKI, and we feel like all the benefit that you get with cas, we know you basically can even get benefit with monotherapy, casdatifan, that it will enhance the TKI. The other aspect of this, and this is an important part. When we talked about the OS signal, the OS benefit is likely coming from both directions when you combine with the TKI. Then I'm going to come back to LITESPARK-012 in this context. If you think about it, those patients that come into the second line, for example, and I'll apply it to the first line. The patients that come into the second line, those that had a TKI have had their HIF-2 alpha activity dramatically increased by the prior TKI treatment.
What you saw, and we didn't overstate it, but basically you bring those patients back as if they hadn't been treated with the TKI. Now, the secondary piece is that while they're on the TKI plus the casdatifan, same thing's going on that would've been going on the frontline. The TKI is causing a push for that tumor that's already addicted to HIF-2 alpha to become even more dependent on HIF-2 alpha. But having casdatifan that's there for the entire ride, those patients then move to their next, when they progress on that other TKI, is they go on to their next line of therapy, they're in a much better position than if you think about a situation where you didn't have a TKI, or if you had belzutifan. Think about lenva plus belzutifan, then think about that biomarker curve.
As you're tending to lose your effect on HIF-2 alpha with time, now those patients are starting to look like naked lenva treated patients. As they go into the later line, they're basically, that's so you're affecting the OS in both directions. Now, coming to the frontline, the reason, and we've been stating this for probably six months, when we heard about the LITESPARK-012 failure, is we attribute that to nothing to do with biology. We attribute it to the shorter effect of belzutifan. If you think about it, that was a perfect storm for not having a strong hazard ratio. You have the toxicity that's obvious even before we see the data, just from the abstract.
It is dramatic when you think about you had a 37% rate of discontinuation from, already known for lenvapembro, that was reproduced in this study, and then the combination drove it up to 50%. That is the first thing. The second thing is, for 24 months now, let us say if you took a median PFS, you are driving increased HIF-2 alpha activity, and then you have got a molecule that is falling off in its HIF-2 alpha activity, however you want to define that, somewhere fairly early on. All those things conspire to be that it is more the molecule than the biology. We think the biology is going to be just great.
For the reasons Rana suggested, we have very strong advocacy from the physicians that even if you drive with cas ipi/nivo to greater than 50%, certain physicians are still going to want to reach, particularly with fast-growing bulky tumor for TKI, and they should have a casdatifan inclusive regimen. Otherwise, I do not think there is anything biologically related to the immunotherapy alone, but Juan can comment what is known on that.
Yeah. We have baseline information on over 200 patients, not just the monotherapy ones. We can interrogate baseline characteristics looking back at no prior therapy or IO only. I can clearly state that no therapy does not look any different than IO only in terms of how much HIF-2 alpha is there at baseline. Now, asking the question, your other question was in reverse, does HIF-2 alpha inhibition impact the response of the tumor to an anti-PD-1 or an IO regimen? Aside from the clinical data that you just saw, we have a whole bunch of additional measures of cytokine changes occurring that have been described in the context of pembro or nivo. We will be showing at triple meeting that we see similar things in our combination of cas with zim.
We also have a whole bunch of preclinical data leading to the conclusion that HIF-2 alpha inhibition should not impair the responsiveness of the tumor to an IO regimen. In both directions, I think they are fairly independent. There is some data out there from some of our collaborators showing that the best predictor of a rapid response in a frontline patient treated with IO only is mRNA levels of EPO, presumably HIF-2 alpha indicating, which ties in very nicely with the fact that we sort of seem to take the head off those primary progressors. Presumably HIF-2 alpha may be a key driver of non-responsiveness to IO only.
Yeah. Jane asking for Richard Law from Goldman Sachs. If you only compare the second line cas/cabo PFS curve for patients without those seven prior lenva patients, then compare those to LITESPARK-011's belzutifan curve in a more comparable population, how would the PFS curve look compared to belz plus cabo? Sorry. Yes, belz plus lenva. A follow-up question is that in PEAK-1, since you allow lenva use, how much worse do you believe the cabo control will look relative to the cabo arm in the LITESPARK-011?
You want to take all of that?
Sure. Yeah. I think the key really is with what we see.
Here
What we see is with our drug, with patients who had lenva in a prior, or they had axi in a prior, we just don't know about cabo in a prior, by definition, basically look the same. Cas is able to overcome whatever I'll call it damage the prior lenva's doing with regards to that enhanced HIF-2 activity. I think if conceptually you removed that seven patients from the TKI prior profile, then we're predominantly going to be in axi profile. But the axi profile looks very consistent with or without the lenva subjects. I think that's going to be very different than, like you said, the cabo control arm. We know that cabo following lenva does very poorly on PFS and sadly on OS, probably because of that HIF-2 driver that lenva really stimulates.
The cabo control arm, I'd say, would perform on the lower end of any prior benchmark for those patients that had prior lenva by comparison, even though, like I said before, we assumed cabo would perform on the higher end of the prior benchmarks and we'd still win.
I believe that actually in our deck, I think we show the seven lenva patients broken out. Juan told me it's silly to put it as a Kaplan-Meier curve. We wanted to illustrate the point and
That's on slide 84.
Secondarily, to your question, just in general, we pulled the seven out to look at the curve and without those seven, not surprisingly, it looked essentially the same. Those lenva patients performed quite well. Jon?
Charles.
Thanks. Charles Zhu from LifeSci Capital again. I had another question on your frontline triplet safety. On slide 62, you highlighted, for example, you had a few cases of hepatitis, colitis, so on and so forth. I wanted to ask, what's your true denominator here? Are all 31 patients in the ITT truly safety evaluable? How should we think about when IRAEs tend to occur, especially as patients receive subsequent cycles of ipi? Thank you.
Yeah, it's a good call-out because, as I pointed out, maybe 18 out of the 30 had made it past the four cycles. The safety data includes all of them, so even if they were in their second cycle and they had an event, they were captured there. Because the average or median time to these immune-mediated events were in that 10- 12 week frame, so it's very typical.
That's why I have some that follow the third course of ipi, and that's why if they got three and not four, they probably had an immune-mediated event, so they didn't get the fourth. Other immune events happened after the fourth or fifth cycle later on. Those are, I think, very typical time frames for an ipi/nivo immune-mediated hepatitis colitis picture.
You're very patient in a public setting. I didn't know that. I saw you're not used to that.
More patient in public. Hi, Jon Miller from Evercore again. I have a question about the first line as well, specifically on the doublet arm, which shows really nice-looking PFS, but you have a lot of favorable risk patients in that population. CheckMate 214 didn't have it in that benchmark. We know favorable risk patients do a lot better on TKI, maybe not as much or have any difference at all for ipi/nivo. How does risk status impact cas efficacy, and how do you expect that to play out in the first-line trial?
Yeah. Well, we broke them out. There's a slide somewhere in there that we broke out the tumor responses by favorable versus intermediate/poor, and to me, they basically look the same. cas and zim performed the same. If you want to be, I'll call it a skeptic and carve out the favorable and just focus on the intermediate and poor risk, cas plus zim still looks better than the comparison group from CheckMate at their intermediate and poor, because you can break out their data as well. I think breaking it out by favorable comparing to favorable or intermediate poor comparing to intermediate poor, we look better on either of those fronts. Then, of course, in the phase III study, we'll stratify by favorable, intermediate, or poor.
It's an all-comer, but we'll make sure that the risk stratum are equal in the phase III trial to make sure that they're balanced for that comparison.
Strategically, obviously, as you know, ipilimumab-nivolumab ended up just logistics, timing everything, and intent is used primarily in the intermediate poor because of the great PFS you get in favorable risk with the TKI. But the intent here obviously is to have it sort of be an all-comers population. Just for the record, in that data set, the one complete response that we've seen to date is in the poor.
Maybe I'll just quickly add, I know the approval originally was just intermediate and poor and not favorable, but the broad use today is across the spectrum, and that's probably why as we enroll this cohort and enrolled fast, we have the distribution because the patients and physicians do like to use ipilimumab-nivolumab regardless of the IMDC risk, and that's a global, Europe the same. There's a lot of use across all risk groups, despite the initial interpretation, which has evolved over time to appreciate its use across every group.
It was interesting. It was subtle, but you remember when the question came up to Rana about her use of TKI and she emphasized she's almost like 100% ipi/nivo, and a lot of the academic centers more and more go in that direction where they basically want to get that OS. So obviously bringing cas in and lowering that rate of primary progression would be a big help.
But there's a disconnect between what they're doing, what she's doing in the academic setting, and what the broader community
is using, where they're very hesitant to give up on IO, my sense is especially in favorable risk patient.
That's why we need data to turn it around. Exactly.
Hi, I'm Priyanka from JP Morgan. It was mentioned that in frontline there was no overlapping tox with the zim/ipi combo. As we're thinking about the registrational data set, and given that these patients will be in the combination for a longer time, how does the anemia and hypoxia-related side effects, how do you think those could drive physician prescribing choices for casdatifan in frontline versus second line?
Yeah. I think the, I'll call it beauty of this combination as a triplet is it's barely a triplet, if you will. Right? Meaning the first four cycles or up to the first four cycles is all three, and then after that it's really CAS plus PD-1, or in this case nivo. We're very comfortable with that profile. Like I said, in our cas-zim group, we now have nearly a year of median follow-up, so we have a lot of experience there. The anemia and hypoxia, we see that occurring regardless of line of therapy. It's clearly HIF-2 driven. I'll say that I'll expect all patients to have anemia to some degree when they're on casdatifan, especially in a prolonged use. So we tell all the investigators, physicians, just expect it, look for it, treat it fast using ESA if your institution allows an ESA, erythropoietin-stimulating agent.
Otherwise, if they get anemia, depends on the degree and if they're symptomatic. If they're symptomatic, they'll get a transfusion, but they stay on CAS. That's the key is we haven't seen discontinuations of casdatifan due to the anemia. Hypoxia occurs similar rates as belzutifan. Again, that's considered on target, but again, that's regardless of the combination partner, meaning zim, a PD-1 versus cabo, from that standpoint. And hypoxia is obviously not a desirable event, but the higher grade hypoxia is still quite low in occurrence, and that'll drive patients to take a holiday or lower the dose of CAS, but they generally are staying on CAS even with events of hypoxia.
That's the AE that they'll end up lowering down to 50 mg maybe, but the key is you saw in the PD data, even at 50 mg of CAS, that's a more potent HIF-2 inhibitor than the commercial dose, 120 mg in belzutifan. So I'd say even though they dropped it down, they're still on conceptually full strength inhibition of HIF-2.
Last couple questions.
Hi, I'm Supawat Thongthip from Truist. Thanks for the rationale for why cas plus cabo could extend OS despite somewhat similar PFS. I'm just curious if you have post-progression data, subsequent therapy post-progression data in ARC-20 to support the hypothesis. Related to that questions, now belz and lens have been approved. Just curious if its potential use in progressor in the PEAK-1 study might alter or influence the OS outcome in PEAK-1. Thank you.
Sorry, can you restate that last part?
Presumably that belz and lens now approved
and I was wondering if it could be used, for example, in the cabo control arm in PEAK-1 and may change how the OS hazard ratio might look.
Yeah. So the cabo arm in PEAK-1 is exclusively cabo, right? So there will be no len, bel patient coming in.
Post-progression.
There is no prior HIF-2 alpha exposure allowed.
Yeah
in that study.
Right. So that's exclusion. You're saying what'll happen after PEAK-1 progression? After progression on PEAK-1, a patient could go on belz, for example, if they were on the cabo arm, if they didn't have the treatment arm. That'll be up to the physicians as to what to give. Now it's a global study, and belzutifan isn't even that really available outside the U.S. for these patients from an access and reimbursement perspective. I would expect the majority in either arm really to go on to an alternative TKI probably, or they can go on len/eve if they haven't had len before. I don't expect much len/bel anywhere in PEAK-1, and it'll be len/eve, tivo, or in the U.S. they could get belz if they were on the cabo arm.
Last question.
Hi. Excuse me. Daina Graybosch from Leerink Partners. I'm wondering if BMS is contributing to funding PEAK-20 and whether you allow patients to receive subcutaneous nivolumab, particularly when you go down to the doublet with cas.
Yeah, I can comment on the financial arrangements first. We haven't disclosed the financial arrangements, but I would just comment that, and we said this previously, that if you look at the totality of our collaborations, with BMS, with Pfizer, and with Summit, the aggregate of those is very financially attractive to Arcus. In the aggregate, we're spending less than we would've spent on one arm for ourselves.
I was asking about PEAK-20.
Oh, sorry. For PEAK-20. Yeah. No. And we haven't disclosed anything there yet.
Okay.
But the trial, just to be clear, if anything there that would just relate to some sort of supply. So there's no collaborative aspect of PEAK-20 with
Yeah
BMS. If we did something, it would involve just supply.
Yeah. We are planning on IV use, not sub-Q use.
Yeah.
We have in mind it, but just for consistency in a global trial, which is very important that everyone knows how to use it. We are picking centers that obviously have a lot of experience with ipi/nivo, but predominantly that experience is IV, so we are sticking with what they know.
Okay. Thank you everybody for joining us. We will end the webcast now, and we are happy to take questions one-on-one afterwards.
I will thank everybody, too. We really appreciate the engagement, not just today, but on an ongoing basis. So thank you very much.