All right, good morning, everyone. I'm Stephen Willey, one of the senior biotech analysts here at Stifel, and glad to have with us for the next session, CEO and President of OnKure Therapeutics, Nick Saccomanno. We're going to just have a kind of an informal discussion here. There is a dashboard for listeners to populate questions with, so we'll try to get those asked and appropriately answered if any do show up. Nick, thanks a lot for your time today. Maybe before we get into Q&A, you can just kind of give us a quick overview of the OnKure story.
OnKure founded some time ago, pivoted to precision oncology, and put our focus on PI3Kα mutant selective molecules. We also populated the company with a number of expatriates from the ARRAY organization. Over the last couple of years, we've been building a portfolio of mutant selective inhibitors, where the furthest advanced is OKI-219, which is currently in expansion studies in the clinic.
The ARRAY ancestral tree of medicinal chemistry is extensive.
It's deep roots. People that are trained with no regrets at ARRAY have found themselves into a number of organizations, not only in Boulder, because this is sort of like this proliferation of individuals, but a lot of them have gone back to the coast and also find themselves in biotechs and major pharma there. It really was a great training ground. Thanks to ARRAY and the other companies that have provided some attention to Boulder to allow us to have this really kind of now very healthy biotech ecosystem.
Yeah. I do not want to date myself, but I think we go back to ARRAY with Bob and Kevin, and a lot of those molecules ended up seeing the light of day.
I've known Kevin most of my life. If anybody wants to do the math, they can.
All right. You mentioned you're developing novel small molecule therapeutics against PI3Kα. Maybe you can just broadly start off by speaking to the relevancy of this protein in the context of malignancy and maybe the spectrum of known mutations that are responsible for oncogenic signaling.
Yeah. If you just look at the epidemiology right now and you look at the well-known activated oncogenes, it's the most frequently mutated protein found across cancers, and particularly solid tumors. It finds its home quite prominently in breast cancer, but lung, endometrial, head and neck, prostate, and others also find themselves sort of graced with this oncogenic menace, which finds itself and usually is occasioned with other mutations or other aberrations in the cell that drive the cancer. PI3Kα has been a drug target, I would call, for three generations in the sense that first it was, can we make an inhibitor that's selective just for the protein, and then the subtypes of proteins, alpha, beta, gamma, and so forth.
More recently, it was recognized that it would be very important to have mutant selective inhibitors, mostly because PI3Kα is found in pretty much every cell in the body. It integrates a lot of different signaling. If you are not selective for mutations, you are going to start changing normal physiology into something which looks a bit more like side effects. Just to set the stage, as we understand it right now, there are a lot of hotspot mutations that lead to activation of PI3Kα in the context of cancers. We also see them in the context of other non-malignant syndromes. Certainly, we could talk about that. I think the area is beginning to pay more attention to that. The predominance, so the nomenclature, which I am sure you have heard from some of my colleagues in the space, is the kinase domain and the helical domain.
When you look at those two, they usually account for a good, depending on what cancer you're in, they account for the majority of all mutations in some cancers. The kinase domain is more prevalent, and in others, the helical domain is. When you think about any particular area, you really need to be thinking about molecules that are very selective for the mutation that you're targeting versus wild. The type of selectivity you're going to need is really, as we're finding now through experimentation, which pretty much started back with the pioneering work of Novartis, you're going to need something 8-10, 15x selective, the mutant you're targeting over wild type. Because if you don't have that, then you're going to start inhibiting wild type, and then there's some well-known side effects, which are dose-limiting, efficacy-limiting, and really quite horrible for anybody, especially patients.
That's a nice context. Selectivity is very important here. Adequate, tolerated, target inhibition, mutant selective, and certainly, it's what we believe is going to rule today in this particular space.
You talked about the need for kind of a threshold selectivity gradient, the need to spare wild type. Can you maybe talk about those things just in the context of the drug development progress that's been made on the target to date and really what's been kind of rate limiting in terms of this first class of inhibitors that we've seen to come to market? Is it really just the wild type sparing functionality of the drug itself?
As many here will know, drug discovery programs run into terrible of all sorts. There's going to be on-target toxicity and off-target toxicity. This is really just, it's just like mathematical pharmacology, right? If your selectivity is this much, and you're inhibiting wild type more than, let's say, I don't know, 30%-40%, then you're going to start provoking pharmacology, which is a characteristic of wild type inhibition. What we see, I think the ones that are most clearly seen and that we look for are hyperglycemia increases and insulin metabolic perturbations, which come by inhibition of wild type. It's really just, if my selectivity is so much, as I start dosing up, I am slamming the mutant target, but then I really have to worry about to what extent am I inhibiting wild type.
We recognize that certainly, if you're at EC50 of wild type, you're into a territory where you're likely to see side effects.
Okay. I guess the flavors of inhibition that we've seen from kind of the class of compounds that are out there, it's either active site or it's allosteric. Do you think it's only really that selectivity gradient, that desired wild type sparing is only really achievable with an allosteric molecule, or do you think that there's a way that a drug can be designed via active site inhibition that can also accomplish that?
Yeah. I mean, you have to let history be your guide here. We've seen a number of very, very good organizations, right? Not like sort of like neophyte organizations trying to take the orthosteric site and make it highly selective in a mutant selective fashion. It really hasn't gotten you anything. That at least was probably the high watermark of molecules that have moved forward, and they're seeing like 4-5x selectivity. I mean, the reason why is that the activating mutations in PI3Kα are distal from the orthosteric site. We do see mutations pop up in the active or orthosteric site as a function of drug treatment on target resistance.
The mutations, the kinase domain, 1047, helical domain, 542, 545, that's amino acid positions, those are far enough away from the active site that the active site really isn't sensing the mutations at a distance, right? The way the protein works doesn't really necessitate a mutation in the active site like so many other kinases. The two allosteric sites, which people have been working on, is the site which is proximal to the kinase domain mutations, that's 1047, 1049. There is the site where companies, really in Scorpion, have focused. That's a separate allosteric site, which in most instances is wild type in nature, but still to a very great extent senses that it's part of a mutant protein. Molecules that will bind there can, in fact, show selectivity immune over wild type.
Allosteric sites, the importance of them is oftentimes defined by how closely they're linked to the function of the protein and how different they are in a disease context versus healthy, sort of giving you a sense of how easy it's going to be to make a molecule that could sense that difference of the allosteric site mutant versus wild type.
Okay. Is there a shared activation profile via these mutants whereby, I mean, I know that there's obviously companies that are pursuing pan-mutant inhibitors, right? Do you get the sense that some of these mutants are activating the protein in the same exact way and so that there is kind of a cryptic allosteric site that you can target that can give you coverage against all of them?
Yeah. I think you've asked a very sophisticated question in essence.
I would have taken the under on that, by the way.
Oh, you win here. The fact is that there is a protein, let's call it a companion protein called p85, which acts as sort of the brake and the accelerator of PI3Kα. In whichever way it binds, it can actually modulate the enzymatic activity of the kinase. That we find, this is something that OnKure has been studying, and I'm sure some of my colleagues in the space have been, is that the allosteric sites are tied to the function of p85. You can imagine, I have a mutation, it changes the protein, and if I want to activate the protein, what does it do? It's going to change the nature of the way p85 is inhibiting PI3Kα in the p110 space.
I think you're exactly right in pointing out that the allosteric sites that we've been finding are connected to a more deep-seated change in the way the protein folds and works. They can all funnel back to sort of a pseudo-conserved mechanism of inhibition. We found that. In the allosteric site we're studying, and in the context of the way we're studying it as well as the others, we recognize that there's only so many tricks this protein can do, and a lot of them seem to be leading back to p85. I do want to point out that in our portfolio, we do have 219, which is 80-100 fold selective. We do have a pan-mutant selective program, which aims for that 10x or greater selectivity across all mutants.
We recognize that it is tapping into a sort of a conserved mechanism of inhibition of the enzyme, which uses the allosteric site to modulate activity at a distal point.
Okay. Maybe we can talk about 219 a little bit. Maybe you can just set the stage by kind of talking about what your target product profile was for this drug when you launched the discovery campaign initially?
Yeah. We based, I mean, a lot of this was in the literature some time ago where there was an allosteric site which had been found by some researchers. We reckoned that this probably was a site which was worth drug. It was defined by a molecule called PIC108. We picked that up recognizing that this was something that could be morphed into a drug and move forward. As we were looking at sort of the epidemiology, we recognized that the H1047R mutant was highly prevalent in breast cancer and a standalone population large enough to warrant developing a drug all the way to the marketplace. Of course, there is a broader opportunity if you think about inhibiting all the mutants with a single molecule, but that is something that we said we will save for another day. In fact, we are working on that now.
OKI-219 was pointed right at this one site with the intent of making a no-regrets compound with the highest selectivity possible so we could slam the target as hard as we wanted and never worry about on-target toxicity, that is to say, toxicities that come by inhibiting wild type. We aimed for that specifically. We also recognized that we needed to pay attention to DDIs because this drug was going to be used in combination. We needed to make sure that this drug was well tolerated for high selectivity as well. We also wanted a molecule that was brain penetrant because we felt that at least in some context, clinical context, we would want to be treating intracranial disease.
With some very high bar characteristics and a real focus on the absolute highest selectivity we could get, we made OKI-219 and we put it into the clinic. Not surprisingly, we're happy to see that our exposure, our pharmacodynamics predict that we will get great target inhibition and that we are not to this point and are very unlikely to see any of the dose-limiting on-target wild-type inhibition-based toxicities.
Okay. Yeah. I mean, we can talk a little bit about some of that preliminary single-agent dose escalation data that you presented. It was at San Antonio, I believe, back in December. I guess it looked like you were achieving near continuous coverage with the 900 mg BID dose. I guess based upon my unsophisticated eye, there did not appear to be much of a shift in some of the PK parameters between 600 and 900. I think you are dose escalating up to 1,200 and 1,500. I guess, why continue to push dose? Is there a DLT that you observed in your preclinical tox models that you would expect to become rate limiting here at some point as you push dose?
Yeah. Let me take those questions and go back. There was nothing in the preclinical toxicology or in the IND enabling toxicology that would lead us to believe that we could not keep pressing up to complete target inhibition. The pharmacokinetic data we presented were very early. Given that the world was surmising that the molecule we were making was going to suffer from some of the same problems that eventually led to the demise of LOXO-73, we chose to publish a very early data set. You will note that it is four patients per dose group. In a sort of a crude sense, it is dose proportional. If you look at day one, day one exposures, it is dose proportional with pretty good hold on variability. As you add additional patients to those levels, and as we go up, we fully expect to see dose proportional exposure.
What looked like sort of a capping of the exposure is really very unlikely. We are hoping to be able to present those data second half of this year.
Okay. I'm guessing this is obviously a pretty challenging patient population in which to show any kind of efficacy, right? I think these patients were not only heavily pretreated, but I think you were requiring a 1047 mutation to be present at baseline, which meant that presumably most of these patients had already seen a prior PI3K inhibitor.
We allowed for prior PI3K, including molecules that come from our competitors' companies as well as molecules that have been approved. We had no problem with that. We wanted to get those patients on. You're right in pointing out that monotherapy studies in a phase one setting are really very challenging. They're not intended to say single-agent activity, certainly at the lower doses and for all comers. We wanted to really make a decision on PK and tolerability in the beginnings of efficacy. The data that we presented at three and 600 really were lower doses. You see stable disease in our swimmer spots. We hope to be able to present more mature monotherapy activity as we go to six, 900, potentially 1200. We expect those data to come out the back half of this year.
We did look for a 1047 mutation coming in, but we also allowed patients with higher levels of HbA1c, patients that were in fact prediabetic to come on the study. Again, because we did not anticipate that there was going to be any hyperglycemia because of wild-type inhibition. We also allowed for patients that had stable brain mets, again, sort of consistent with our desire to have a brain penetrant molecule. The efficacy data from October is really very early. As far as that data set, for me, it shows that we have got PK, we have got tolerability, we are showing the beginnings of pharmacodynamics, we showed some ctDNA. At that point, we can make the decision, is this molecule worth developing?
In my experience, just quite a number of compounds, it was very clear to us that this molecule certainly was worth developing and that development plan we had laid out for it was perfectly appropriate.
Okay. Maybe you can just talk about what you view as the next development steps for this program? I know that you've talked about wanting to run a combo study with fulvestrant, I think, as perhaps a next step.
Yeah. We have started dosing at six and 900 BID with fulvestrant as the background. We will enroll patients with HER2 positive, excuse me, HER2 negative breast cancer. Of course, we will have the same alternate inclusion criteria. We are enrolling that study right now and hope to have interim data the second half of this year. We are also guiding towards or thinking certainly about triplet studies in the hormone receptor positive domain, a CDK4/6 or CDK4 selective molecules. There are certainly some very interesting ones, plus fulvestrant or letrozole. That would be an important study because those would begin to, depending on how you run them, represent frontline studies. We hope to get there quickly. We have also presented some data on the utility of OKI-219 in the context of other cancers.
Perhaps most pointedly, we've presented some data regarding the ability in preclinical models to be able to control HER2 positive, positive HER2 positive breast cancer. That's a study that we've communicated that we have an interest in running, which OKI-219 plus trastuzumab or trastuzumab, whatever form we use, plus tucatinib would be a very interesting study of a smaller population, but certainly a population of great need.
Should we expect to get some of the fulvestrant combo data with the next cut of the dose escalation data later this year?
Yes. In the second half of this year, we will provide an update on the progress in the doublet, which again, I think most people will agree that as far as a test bed for these molecules, a clear, more relevant clinical population in a regimen which could represent standard of care, that's going to be an important measure of efficacy. Monotherapy studies, I mean, I think everybody, again, would agree. They're limited in their ability to tell you how good your molecule is going to be with regard to durability and combineability, which are going to be very important features of this molecule. I think the fulvestrant study is going to be a much better measure of the value of these compounds. Again, most of the industry recognizes that and are moving to those types of studies to prove the value of these molecules.
How do you think about the current competitive landscape in the mutant selective space, right? There has obviously been a lot of things that have been happening. Relay is obviously advancing their pan-mutant inhibitor. LOXO decided to shelve their 1047 selective drug, ended up turning around and buying the Scorpion asset. It just seems like there are a lot of moving parts right now in terms of the space. Just what's your overall perspective of kind of what's going on right now?
Yeah. I mean, I think there's some really interesting, I think the Relay molecule is a really interesting molecule, RLY-2608, worth developing, the Scorpion molecule worth developing. And of course, I'm going to say OKI-219 is worth developing. The questions that are going to be important to answer are going to be regarding is there adequate selectivity to cover the 1047 mutant? Is there adequate selectivity to cover the helical domain mutants? I think everybody is being conservative and recognizing that many of the molecules we're looking at right now are going to have a higher likelihood of addressing 1047. The selectivities we've seen thus far make the 542, 545 or helical domain mutants maybe a little bit more of a stretch. We'll see. We'll see how that plays out.
Importantly, tolerability and combineability in multiple lines, not only in breast cancer, but in other cancers, is going to be really important, right? Whether it's Scorpion, now Lilly, or Relay, or whether OnKure, combineability, tolerability, and the selectivity, which is high enough, right, to be able to address one or both sets of major mutants, are going to be the questions that are going to have to be asked and answered. I think the clinical trials are ongoing or starting right now that will give us a clear view of that probably in the early 2026 timeframe. That's how I see it. Again, everybody recognizes the very specific problem statement of making the right molecules in which specific mutational settings for PI3Kα. Now we have to test bed them in the clinic.
We also have to see what are the next molecules that are coming along. Again, I referenced our own pan-mutant selective program, which will be nominated compound shortly and getting into development, hopefully the end of 2025, beginning of 2026. Again, with sort of that higher bar selectivity, which might provide you greater confidence of going into all the major mutants as opposed to restricting yourself to just one mutational class over another.
I know you are also developing, you have a med chem effort against the helical domain mutations. How does the selection of a DC for the pan kind of philosophically align with the med chem effort that you have on the helical domain mutations?
If the helical domain program is really designed to give us kind of like a 219 for the helical domain, right? 219, kinase, and then the other program further back. We do believe we understand something about how the helical domain rearranges to be able to affect inhibition of the protein. However, I would not necessarily rule out there being a single allosteric site that can give you sort of that very, very high selectivity, like 25-fold selective all mutants over wild-type. What we want is a molecule which is going to have sort of that no regrets, high, high selectivity in the context of the helical domain. There are some cancers. If you look outside of breast cancer, there are many solid tumors where the predominant, overwhelmingly predominant mutant is the helical domain.
There could be a huge opportunity there to make a great medicine because the same problems or the same problem statements of selectivity, combineability, tolerability are all going to be waiting for us there as well. These are really rather large patient populations as well.
Okay. Maybe just in the last 30 seconds here, you can just kind of give us an update on where cash runway currently stands and that runway allows you to execute on?
For our stated strategy, our cash runway takes us to the fourth quarter of 2026.
Okay. Very good. Nick, enjoyed the conversation. Thank you.
Yeah, it was great. Thanks for having us today.
Be in touch.
You bet. Take care, guys.
Take care.