Okay, great, thanks. Good morning and welcome to day two of the 43rd Annual J.P. Morgan Healthcare Conference. I'm Eric Joseph, Senior Biotech Analyst with the firm. One of our first presenting companies this morning is Relay Therapeutics, and presenting on behalf of the company is CEO Sanjiv Patel. There's a Q&A after the presentation. For those who have questions, we'll bring a mic over to you. And for folks tuning in via the webcast, also feel free to submit questions via the portal there. So with that, Sanjiv.
Thank you.
Thank you to Eric and thank you to J.P. Morgan for the kind invitation to speak at this year's conference. But most importantly, thank you to those of you on the West Coast attending at 7:30 A.M. It's very much appreciated. So I'll be making a series of forward-looking statements today. I ask you to check our website for all the full disclosures. And so with all that said, let's begin. Nine years ago, we set out to use our research platform, which we termed the Dynamo Platform, which is a combination of computational and experimental tools to make novel medicines against targets in both precision oncology and rare disease. And whichever metric you want to measure us on, we've been very productive. We've nominated eight development candidates, we've received four INDs, we've generated two clinical data sets that we believe show true proof of concept for these novel assets.
We're very excited about what lies ahead of us in research. We've got four unnamed programs behind all the programs that we've talked through, we will talk through this morning. 2025 is a year for us to turn the page and really focus a large part of our efforts on clinical execution in very large commercial opportunities. The biggest of those, and where we'll spend most of our time today, is in breast cancer, where we'll start our first phase 3 pivotal trial with RLY-2608, the first mutant selective PI3K alpha inhibitor. Behind that, even though it's such a large commercial opportunity, we're not just a breast cancer company. We'll start three additional clinical programs this year. One of those will be in oncology targeting NRAS-driven solid tumors, and two will be in rare diseases targeting both vascular malformations and Fabry disease.
So with these four large clinical programs, you can see why 2025 for us is all about execution in the clinic. And so let's just linger for a second on why breast cancer is such a large opportunity for us and where all of our focus really will go in 2025. The first reason is it's such a large opportunity. There are 140,000 patients in the U.S. alone that have PI3K alpha mutated-driven breast cancers. The second reason is we've gone a long way towards validating this opportunity in the clinic. And just before the vacation period, we were in San Antonio and we shared a very robust data set. It was early, but it showed 9.2 months' worth of PFS in hormone receptor positive, HER2 negative, second line and second line plus breast cancer patients. And that's against a comparable standard of care of five and a half months.
So we have gone a long way to show that we can create a step change in efficacy for this very large group of patients. The third reason why we're excited is we have a very clear pathway now in the clinic to generating clinical data that will get us an approval in a very large slice of a very large pie. And we have a clinical execution machine that has been shown over the last five years to be able to execute very successfully across trials across the world and has already executed its first pivotal trial with our FGFR2 program. And finally, we have a very extensive balance sheet that would allow us to run more and more novel combinations and take more and more slices of this very large pie.
Hopefully that gives you the logic for why we're so excited about our opportunity in breast cancer and where the majority of our focus will be in 2025 and beyond. So let's start talking through the four clinical programs. First one is PI3K alpha driven. PI3K alpha mutations are one of the largest precision medicine opportunities that's left out there unserved. There are over 500,000 PI3K alpha driven mutated patients in the U.S. alone, and that's spread across breast cancer, vascular malformations, and solid tumors outside of breast cancer. Unfortunately, they are targeted with non-selective inhibitors, and these are challenging to be taken by patients because they are plagued with off-target toxicities such as diarrhea, rash, and most importantly, hyperglycemia. So unfortunately, those patients cannot have the dose intensity that they require, and that leads to a lack of efficacy in these patients.
And so several years ago, we set out with our Dynamo Platform to address this by creating a mutant selective inhibitor that would allow us to dial out these toxicities, allow these patients to stay on therapy for longer, and allow us to get the efficacy that these patients really need. And we believe we've done exactly that with RLY2608. It's the first mutant selective, isoform selective, allosteric inhibitor of PI3K alpha. And we're very pleased over three years ago now to bring it into the clinic. The first place we're going to deploy RLY2608 is breast cancer. And the reason why is it's such a large opportunity, as we said. Over 140,000 patients in the US alone have PI3K alpha mutated-driven breast cancer. And we see already approvals in both first and second line with the non-selective inhibitors.
And so we have a clear pathway towards getting this approved and making RLY2608 the combination partner of choice across all lines of breast cancer therapy. And it's a very large market, and so we'll start our first pivotal trial this year, and that should give us a very large slice of this market. Now, it's very confusing, breast cancer, because you see all these kind of dynamic nature of agents. We see agents appearing against the cell cycle where you see CDK2 and CDK4 selective agents appear, and there's obviously a lot of agents now that are oral targeting the hormonal space. And so we just want to simplify you for this whole complexity. So let's be clear. In precision medicine, there's a very simple paradigm, which is target the mutation. So there are 140,000 patients with PI3K alpha driven mutations. And this mutation in the end is foundational.
It appears early and is a clear driver of the disease, and so to simplify this for you, we believe that PI3K alpha driven inhibitors will be the foundation of all of these treatments. As these additional agents appear and add efficacy, they're purely additive to RLY2608, and that's why we believe it's such a big commercial opportunity. PI3K alpha driven mutations will be treated by PI3K alpha inhibitors, and the best and first one we believe, with the data that we've shown, is RLY2608, so let's get into what is making us so excited. It's the data, and so the ReDiscover trial started just over three years ago and has three arms to it: a monotherapy arm, a doublet arm, and a triplet arm, and so just before the vacation period in San Antonio, we showed a very clear, robust data set from the doublet arm.
So, these are in hormone receptor positive, HER2-negative breast cancer patients, second-line and second-line plus. And so what you see here is some of the cross-trial comparisons, which are difficult to do given the slightly different patient populations of the approved agents and the agents in development. And what you see in these agents is numbers such as five and a half months' worth of PFS, CBR rates in the 40s, and response rates in the teens. What we showed in 64 patients with nine and a half months' worth of follow-up is a very clear step change in efficacy. We see 9.2 months' worth of PFS. We see 67% clinical benefit rate, and we see a confirmed objective response rate of 39%. These are numbers that we've never seen in the class before, and it's going to step change efficacy.
Now, if you dive into the details and start looking at second line only patients to give you a like-for-like comparison, this 9.2 months' worth of PFS moves to 11.4 months. The response rate is now 40% as a confirmed objective response rate. So if we believe that we go head to head with any of these therapies that are either approved or in development, we feel very confident in our agent. We also know there are a range of sponsors that are running trials that are specifically targeting the kinase domain mutations of PI3K alpha. And again, here we showed a cut that shows 11.4 months' worth of PFS across all lines of therapy, second line and second line plus. And what we believe to be a remarkable 67% confirmed objective response rate.
Again, in San Antonio, we saw some of the other sponsors report their data in this space, and it wasn't on the same page. We believe RLY2608 does exactly what we hoped it would do. It dials out wild-type toxicities, allows patients to have greater dose intensity, and that translates into greater efficacy. You can see that in the tolerability profile, which is very clearly, we believe again, early, but shows a step change. Very low rates of grade three AEs and very specifically, very low rates of grade three hyperglycemia, which is the toxicity that oncologists really do not want to manage. If we take a step back here, in heavily pretreated patients with hormone receptor positive, HER2 negative breast cancer, we see a step change in tolerability, and that translates into a step change in efficacy.
Our 11.4 months' worth of PFS translates very nicely against the standard of care of five and a half months. So this gives us the confidence to start a phase three pivotal trial this year. That phase three pivotal trial will be in post-CDK4/6 pretreated patients, and it will be RLY2608 plus fulvestrant up against the emerging standard of care, Capivasertib, which has had a remarkable commercial launch in its first year. And it just shows you the unmet need that sits in this market. And so this trial will start this year. We'll provide further details on that once we are on the other side of an FDA interaction with written minutes and a clear clinical plan trial in hand. And we look forward to sprinting towards getting this started this year. So with all of that said, let's just take a step back.
This is a very large market commercially and in a very large pie. This first trial will be in post-CDK4/6 patients. So it will take all of the second line and second line plus market. And this is where the real unmet need is. These patients have benefit today from the current standard of care of five and a half months. And so if you increase that to the numbers that you've seen today, you can see that in this very large market, 13,000 patients, if you can extend that, you end up with a very large commercial opportunity. But given the fact that we're in post-CDK4/6 patients and post-CDK4/6 is now being used in the adjuvant setting, we'll also capture a sliver of the earlier lines of therapy with this trial. And so we take a very large slice of this market with this first pivotal trial.
But that's not where we're going to stop. We continue to try and build to take further and further slices of this pie. And that's why you see us run novel triplets. And so we're in dose escalation both in RLY2608 plus fulvestrant plus ribociclib, current standard of care, and 2608 plus fulvestrant plus atirmociclib, Pfizer's selective CDK4, which we believe will be the emerging standard of care. And so this positions us well to take additional slices, and we'll continue to add novel combinations to our armory. And so if you just take a step back, you see the excitement for RLY2608. This is clearly a very large opportunity. This is clearly a standard of care that needs to be disrupted. The data is early that we've shown in San Antonio, but it's robust.
64 patients, nine and a half months' worth of follow-up, 9.2 months' worth of PFS. We have a very clear clinical plan to take a very large slice of this market, and we set ourselves up to take all of the other slices over time. So that's our breast cancer focus. Now, RLY-2608 can be used outside of breast cancer because, as we said, PI3K alpha driven mutations exist outside of breast cancer in both other solid tumors, but also there's 170,000 patients with vascular malformation driven PI3K alpha mutations. Now, this is an umbrella term that's PI3K alpha mutated driven that leads to overgrowths in both lymphatic and venous vasculature and appears in various different phenotypes. But they share one thing in common, which is they are poorly served with therapies today.
They are unfortunately treated by surgery, local bandages, sclerotherapy, and 25%-40% of them end up with systemic therapies. And again, these systemic therapies have one thing in common. They're ineffective. They have tolerability challenges. Alpelisib is approved, for example, in a sliver of this market, and we know it's plagued with the tolerability challenges of diarrhea, rash, and most importantly, hyperglycemia. So unfortunately, these patients do not stay on therapy, and unfortunately, these patients do not get the efficacy that they need. And we're very excited about the profile of RLY2608. As we've seen from the robust data sets that we've shared, it has a very safe, clearly tolerable profile. And remember, these patients are diagnosed early in life and can be on therapy for decades.
And so having a very tolerable profile is important, and we hope and we believe these patients can stay on therapy for much longer and get the efficacy that they deserve. And so again, in this very large market, 170,000 patients, we will deploy RLY2608 for the first time this quarter and hopefully get this very large patient population treatment that they deserve. So if we move to our third clinical program that we'll start this year, it's another oncology program, and we have a program targeting NRAS-driven solid tumors. There are 25,000 patients that have NRAS-driven solid tumors in the U.S. alone. They're concentrated in tumor types such as melanoma, non-small cell lung cancer, and colorectal. And again, there are no specific agents that selectively target NRAS.
So today, the therapies really target agents and nodes in the MAPK pathway, RAF, MEK, and also you see the emergence of the Pan-RAS agents. Again, all of these non-selective agents are plagued with tolerability challenges. And again, you hear the familiar story. They have tolerability challenges, leads to lower dose intensity, leads to lower efficacy. And so again, we set out with our research platform a few years ago to create the first exquisitely selective NRAS agent. And the hope was that we could dial out the toxicity challenges and provide a step change in efficacy. The preclinical data that we have shows exactly that. And here's a range of different tumor PDX models in the different tumor types and different NRAS mutations. And what you see is very rapid and deep regression, and you can see the standard of care on there as well.
And so we believe we've achieved what we wanted to achieve. In exploratory animal toxicology studies, we can achieve very high levels of exposure. In the study that we have just completed, it's tenfold above the efficacious dose, very tolerable. And so again, we've achieved what we hoped we would achieve with our research platform, exquisitely selective inhibition of NRAS that leads to deep and rapid regression in these PDX models and that is very tolerable. And so we will be excited to have this in clinical trials later this year. The final clinical program that we'll start this year, and you can see why we're so excited about focusing on clinical execution given the opportunities that sit in front of us, is in rare disease and Fabry disease.
Fabry disease is a genetic disease, lysosomal storage disorder, that is caused by a very broad range of mutations in the gene that codes for alpha-gal, which is an enzyme that traffics substrates in and out of the lysosome, and when it becomes disordered, it's ineffective, and it leads to a substrate buildup that then leads to all the complications of Fabry disease. About 8,000 patients in the US alone, and the current therapies sit with enzyme replacement therapy for alpha-gal, and then there's a small molecule chaperone marketed as Galafold that is doing commercially very well. Peak sales are going to be close to $750 million. It's a significant market that has been created here. The challenge is, unfortunately, alpha-gal has some very significant limitations. It's an inhibitory in nature.
Now, it's doing exactly the opposite of what it sets out to do, which is to activate alpha-gal. It only is amenable to 40% of the patients with Fabry disease, and the other 60% are termed non-amenable, and the fact that it's inhibitory means you can't combine it with the standard of care enzyme replacement therapy, so we set out a couple of years ago to make a non-inhibitory small molecule chaperone, and the data that you see here shows that we've done exactly that. We've created an allosteric inhibitor of Fabry disease that is non-inhibitory, so it leads to greater activation of alpha-gal, hopefully greater benefit to these patients, has a much broader coverage of the mutations outside of just the 40% that migalastat, Galafold covers, and given its non-inhibitory nature, it's combinable with the standard of care ERT.
We believe this will be a step change in efficacy and in coverage for Fabry patients and look forward to having this in the clinic later this year. If we start to summarize, you can see why 2025 for us is a year of turning the page and focusing on clinical execution. We have three new starts in vascular malformations. Remember, 170,000 patients in the US alone that are today unserved. NRAS-driven solid tumors, 25,000 patients today that are currently served with non-selective inhibitors. And then Fabry disease, 8,000 patients today, many of them unfortunately are termed non-amenable. And so we'll start all of these three clinical programs this year. But the giant in all of this is hormone receptor positive, HER2-negative breast cancer. And we will start our first pivotal trial in this very large market this year with RLY-2608.
The data that we have is very compelling. 9.2 months' worth of PFS, 11.4 months' worth of PFS in second line only patients. In kinase domain, 67% confirmed objective response rate. These are numbers that we've never seen in the class before. And so we are going to sprint towards starting this pivotal trial this year, and we believe we have the team and the execution capacity to do it. And we have a very extensive balance sheet that will allow us to continue to take larger and larger slices of this pie. So I look forward to updating you across the year as we execute on all of these priorities, and with that, I'll hand it back over to Eric Joseph for Q&A. Thank you. Well, great. Yes, we have time for questions.
So maybe just picking up on the addressable market served by the pivotal study that's expected to get underway later this year. It seems as though it's not perhaps a strict sort of second-line opportunity, right? There could be prior CDK4/6 experience could extend to those receiving treatment in the adjuvant setting. So maybe just talk about sort of sizing, I guess, of the opportunity that could be conserved by the patients comprising, making up, the study population.
As you know, it's a very dynamic space. As we've talked about, CDK4/6 inhibitors are being utilized in different indications. We believe up to 25% of patients in the adjuvant setting may receive over time a CDK4/6. And so this traditional kind of term of first line, second line may become rapidly antiquated. And what we are looking at here today is a population that we will go after, which is post-CDK4/6 treatment. And so traditionally, those 13,000 patients in the term traditional second line are going to be applicable to our trial. But increasingly, as you see CDK4/6 come earlier and earlier, they'll be applicable too. So the exact numbers will evolve over time as the use of CDK4/6s evolve. But I think all we can say is it's a step change in market opportunity for us.
In that second line market, there's already 13,000 patients, and there's only five and a half months' worth of PFS. So even in that market, if you could extend it to the numbers that we've talked about, you end up with a very significant commercial opportunity. And then you tack on the additional kind of first line patients that become applicable to this, and you end up with a giant commercial opportunity that sits in front of us.
Is there data or sort of the experience of capivasertib in the post-CDK4/6 adjuvant setting? I don't know if that's kind of part of their existing label. Is that sort of perhaps preclude being able to accrue that particular population given that agent being the comparator arm?
You want to take a note?
Yeah. So the Capitello-291 trial did allow patients who had progressed rapidly through adjuvant therapy, but at the time that trial was run, there was very little usage of CDK4/6 inhibitors in the adjuvant setting, and that trial did not require prior CDK4/6 inhibitor therapy. So we've not seen a subgroup analysis of the performance of capivasertib in patients who progressed on adjuvant CDK4/6 inhibitor therapy.
Okay. But the one thing we do know is even in that study, in CDK4/6 naive patients, which would be an even healthier set of patients to see how CAPI performs, and in PIK3CA mutated patients that were CDK4/6 naive, you saw ostensibly no difference in the median PFS.
Okay. And I guess, is there a particular, let's say, quota or cap of prior adjuvant CDK4/6 patients that you'd want constituting the study?
You want to take it?
Yeah. I mean, I think at this point, the trial that we're proposing would have the ability for patients who had seen an adjuvant CDK4/6 inhibitor to come on therapy, but we're not necessarily proposing at this point any pre-specified cap on how many patients or a minimum for how many patients that would need to represent. Okay. All right. So just wondering a little bit, so it sort of could enable an initial approval and an indication that doesn't align perfectly, but sort of runs adjacent to the indication where Roche's Itovebi is currently approved. Maybe just kind of make the contrast between the addressable population perhaps served by that agent and relative to where you hope to be with 2608.
You want to take it?
Yeah. I think the interesting thing is when you look at the traditional definitions of front line and second line, you have to keep in mind in the front line setting, it is clearly bifurcated in terms of the standard of care regimens for those patients that are deemed endocrine sensitive, which is about 60% of the front line population, and those patients that are deemed endocrine resistant. That's about 40% of the front line population. So the absolute numbers of patients that are addressable are actually almost twice as many in the traditional second line definition because those patients come back together in the second line. And so Alpelisib label sits in the endocrine resistant population in the front line and is even further narrowed by the strict metabolic criteria in which they filtered for that trial.
At the largest, the best size of the market that Alpelisib could address is about 40%, but when you take into consideration the metabolic restrictions, it really drops down to about 15%-20% of the Western population. There's not a lot of overlap between the way we are designing our phase 3 and the patients we would aim to address and the current label and metabolic restrictions that would exist based off their entry criteria for their phase 3 trial. Right. That trial was run exclusively in CDK4/6 naive patients, including in the adjuvant setting. Maybe just kind of picking up on the point about metabolic fitness, where are you planning to kind of set or metabolically fit or metabolic eligibility in your post-CDK4/6 study, particularly around sort of the A1C as a threshold for entry?
You want to take it on?
Yeah. So consistent with what we've done in the ReDiscover trial, we've enrolled patients on study who have hemoglobin A1C up to 7% and who have fasting plasma glucose up to 140. So what that's translated into in our ReDiscover study is a population that's actually quite reflective of a Western population with about a third of our patients who have either a BMI over 30 and/or an A1C over 5.7. So patients certainly at risk of developing diabetes and developing hyperglycemia with a PI3K alpha inhibitor. And I think we'd be consistent with that going forward into a pivotal trial.
Allowing a patient population, again, very reflective of a Western patient population, allowing patients on study with metabolic syndrome, prediabetes, obesity, and given the data we've seen so far and the very limited hyperglycemia we've seen so far, I think with a representative patient population, we'll be able to continue to maintain a differentiated profile with regard to hyperglycemia.
Yeah. I think just to comment on that, the AE profile that we've shown in the ReDiscover trial has about a third of patients that have high BMI or at the upper end of the hemoglobin A1C. So again, it shows how impressive the tolerability profile is. I think we will exclude active type 2 diabetics from the study. And it's not that we don't think that the RLY2608 will be beneficial to those patients. We can run that in additional kind of studies on the side in parallel. I think our view is in the end, these trials are going to have cross-trial comparisons made to them. And so let's try and get those comparisons to be as accurate as possible.
You made a point that patients with a PI3K alpha mutation will be treated with a PI3K alpha inhibitor. Will they also be treated with fulvestrant or endocrine? I guess the real question is, is there optionality for any PI3K inhibitor monotherapy in the setting that you foresee?
You want to take it on?
Yeah. No, I mean, I think breast cancer in the modern era is definitely a disease that's treated in combination, right? So I think what Sanjiv was highlighting is the fact that the backbone of therapy here, a commonality across therapy, will be targeting PI3K alpha. It's an early mutation. It arises. It drives the disease. We also know estrogen receptor drives the disease in these patients, and there's interaction between estrogen receptor and PI3K alpha. So we anticipate that you'll have something targeting the estrogen receptor as part of that regimen as well. There's not a path forward, at least in any patient population, that represents a significant commercial opportunity for monotherapy development in breast cancer. I think monotherapy would be looking at very late line therapy in patients who have completely lost any estrogen dependence. So there's clearly going to be an anti-estrogen on board as well.
And then the question is, as you look across either the cell cycle dependent mechanisms or emerging new mechanisms, there could be the opportunity in certain patient contexts to add in a third agent. Obviously, we're starting with CDK4/6 and CDK4, but we anticipate over time there'd be the opportunity to look at additional mechanisms given different tumor contexts as well. But as we've shown with our data to date, with our proposed clinical plan, going in with this combination of something targeting PI3K, again, a driver in the tumor, and something targeting the estrogen receptor, also a driver in patients who have positive disease, can lead to very meaningful clinical benefit in these patients and really warrants that being the focus for moving forward in development.
Yeah. And I think the point we were trying to make is obviously it's a dynamic space, both in the kind of selective CDK4 space and in all the agents now that are targeting the receptor. What we have is the first leading PI3K alpha mutant selective agent. And so we believe we'll get that in the clinic as quick as possible, get it to patients as quick as possible, and that will become the foundation of all of these different combinations. And we look forward to all of the data that emerges in the other spaces. But as it does, it will be purely additive to the benefit that RLY2608 can deliver.
For what we still consider the true first line metastatic breast cancer setting here, currently managed by CDK4/6 inhibitors, how are you thinking about pursuing that perhaps with a triplet regimen? I mean, that's a regimen that's currently being evaluated. I guess what do you ultimately need to see there to feel confident about pursuing that opportunity?
I think it's exactly what we're doing, which is you see us doing dose escalation with the current standard of care, Ribociclib. And then we're obviously also doing it, what we believe will be the future standard of care, selective CDK4 inhibitors. And obviously, with our partners, Pfizer, we've started that dose escalation. And so I think that's our focus at the moment, is to try and create the right set of data then to go forward and then make the decision around which is the right triplet to run in the first line.
The right set of data, is it mostly safety or there's an efficacy component to that as well? And to the extent that's the case, I guess maybe just put a little more context around that.
You want to take it on?
Yeah. I mean, it's clearly initially focusing on safety and focusing on combinability. I think on the efficacy side, there'll be data that we're generating within the context of our own trial, but we'll also be seeing data that's generated in external data sets as well. And along with the question of what the appropriate CDK inhibitor would be to use in the setting, we're going to see oral SERDs emerging in this space as well. And I think you can anticipate you'll see us starting combinations with additional estrogen targeting agents as well over time. So we'll be looking at our own data. We'll be looking at safety. We'll be looking at tolerability. We'll be looking at early signals of efficacy. We'll be looking at the external landscape. We'll be looking at what the apparent best-in-class future standards of care will be.
We'll integrate all of that into a comprehensive plan for advancing 2608 into other settings in HR positive breast cancer as well.
But again, to reiterate, we'll be very thoughtful about this issue, is why you see us generate the data. This is a very well-served market in the first line. Multiple years' worth of therapy currently is the standard of care. And so these trials are dynamic in nature in that you see all of the different agents moving. And as you know, it is fragmented, endocrine sensitive, endocrine resistant. And so that's why you see our focus on. We're focusing on 100% of patients that are post-CDK4/6. It's a giant patient population, 13,000. They're currently underserved, five and a half months' worth of PFS. If you can increase that to the levels that we have been able to show, 9.2 months, 11.4 months, and then you get the additional sliver of these earlier patients that are post-CDK4/6 in the adjuvant setting, you have a giant commercial opportunity.
So that's why you see us focus there. And then obviously, we'll get to these earlier lines as things settle.
Maybe just shifting gears to the vascular malformations program here, very significant commercial opportunity and scope there. There's a product currently approved for a small subset within that subset of diseases. Maybe just comment a little bit on sort of whether, biologically speaking, whether there is any difference in the sensitivity of those various subtypes to a PI3 kinase alpha inhibitor, sort of how you assess that non-clinically, and then the scope of subtypes that you plan to initially evaluate in the phase one trial.
You want to take it on?
Yeah. So there's not an extensive clinical database that we're able to look to to answer the question clinically about the relative sensitivity of the different subtypes of vascular malformations to PI3K alpha inhibition. But what we do know is the same mutations that drive cancer also drive these PI3K alpha-driven vascular anomalies. And what we're able to do in our preclinical models is we're able to build models where we can take, again, these same mutations where we've shown in cancer, we get robust inhibition of these. We get clearance of CT DNA, other clear pharmacodynamic evidence that we are potently inhibiting these mutations. And we can go into preclinical models of vascular malformations with these same mutations that exist in the clinic and show really pronounced efficacy of 2608 in those vascular malformations models.
I think given the initial clinical proof of concept that we've seen with Alpelisib and PROSE, we have a high degree of confidence in the biology and the linkage of PI3K alpha to driving this disease and the ability to have a disease-modifying effect with 2608 as we go into these patients. As we go into our initial clinical trial, we initially will be doing a randomized dose assessment across three different dose levels, all of which are biologically active. We anticipate that there'll be primarily patients with a PROSE subtype who are enrolled in this initial part of the trial. That's, again, where there's the existing data set from Alpelisib and the ability to benchmark to an external standard.
We also will have the trial open to other subtypes as well and anticipate we could see lymphatic malformation patients where 80% of lymphatic malformation patients have a PI3K alpha-driven lesion. So we anticipate we could see those patients represented as well. Any age restrictions in the initial phase one trial? I mean, you have a fair amount of pediatric presentation here. So, yeah.
Yeah. So the initial cohorts will be focused on patients 12 years and older. We'll be using flat dosing based on everything we've done in cancer patients to date. We then will have the opportunity, as we get experience in this patient population, the protocol anticipates being able to open dose finding in younger populations as well, down to infants.
Okay. Maybe just wrapping up with a question on the RAS inhibitor program, certainly getting a little more presence here in the presentation today, and just picking up on the point about sort of just the qualities of your compound and selectivity profile, just to have a clearer understanding, is it a mutant selective approach that you're targeting here? I guess what's the selectivity for mutant NRAS over wild type, and is sparing wild type important for widening, having an effectively wide therapeutic window?
Yeah. So this is a place where initially we were informed by what you see in mouse genetics. So you can knock out NRAS in a mouse and not see any meaningful phenotype. And that's because you get compensation by HRAS and KRAS. So I think what the experience has been, if you knock out all three RAS family members, you see both a significant phenotype in mouse studies. And then in the clinic, we've seen high rates of rash in other AEs with pan-RAS inhibition. But our hypothesis going in is if you just are able to inhibit NRAS, you can get compensation in normal tissues by HRAS and KRAS and shouldn't see any significant toxicity.
And consistent with that, what we've now shown with the molecule that we're advancing towards the clinic is we can go into NRAS-driven tumor models and get very rapid and deep tumor regressions across different models and different genotypes. And at more than 10x the concentration of drug that's necessary to get that deep anti-tumor response, we can go into toxicology species and essentially see no AEs whatsoever.
You're not seeing any toxic findings with this, despite the fact that we've been able to show that you're seeing meaningful inhibition of NRAS across normal tissues, which again is consistent with what we see in the mouse that other RAS family members, if you have the selectivity, are able to compensate for inhibition of NRAS, which gives us confidence as we move into the clinic that this has the potential to be a highly efficacious mechanism in NRAS-mutated tumors while sparing many of the toxicities that are seen with inhibiting either downstream nodes in the pathway or inhibiting all RAS family members.
Okay. Great. Well, I think we'll leave it there for time. So thanks for.