Welcome to this next fireside chat with OnKure Therapeutics. With us today, we have Nick Saccomanno, President and CEO. Nick, welcome. Thanks for joining us.
Thanks for having me.
Maybe for those that are less familiar with the story, could you just give us a high-level overview of the company's area and focus today?
Right. Based in Boulder, Colorado, we were formed about a decade ago and did not start working on PI3K alpha until about three years ago, where not only discussion in the community, as well as some scientific papers we had noticed, led us to believe that there was a real opportunity to create an interesting new set of drugs that would be highly mutant selective. Since that time, we have built a portfolio of molecules, one currently in the clinic that we will talk about, OKI-219, and a number of other molecules which exploit the target in a variety of ways.
Great. OKI-219, as you mentioned, is your most advanced. For our candidate, it targets the PI3K alpha H1047R mutation. Yeah, tell us about the compound and perhaps how it's differentiated from other PI3K inhibitors.
Sure. As context, H1047R represents about two-thirds of the mutations of the patient populations that is found with mutated PI3K alpha. The other third, or maybe 40%, are represented by mutations in a separate domain. H1047R is the kinase domain, and the other mutations are found in something called the helical domain. The primary differentiation point amongst all molecules in this area is selectivity for mutant or specific mutants over wild type. The reason for that is that inhibiting wild type PI3K alpha, which is found in every cell in the body, leads to a constellation of predictable, understandable side effects, which are really unacceptable in the context of any therapy, and most especially cancers. OKI-219 binds to a reasonably well-understood allosteric site. It derives greater than 80-fold selectivity mutant over wild type.
As such, we could dose it to a level where we can sufficiently inhibit the target in a cancer cell without inhibiting to any extent the wild type target found in all cells. It is this attribute that will allow molecules with this level of selectivity to be useful in cancer and other conditions, because, again, you could dose to a level with no side effects, and those side effects certainly will not exacerbate the types of side effects you've seen from the drugs that you might combine with. Many other drugs have moved into development. Several drugs have been approved, but none of them really seem to have sort of a high bar selectivity that allows them to avoid all the tolerability issues associated with wild type inhibition.
If you compare OKI-219 to some of the other mutant selective inhibitors, how does the wild type selectivity, I guess, compare to those?
Just in going, the molecules that are approved are alpelisib, no selectivity. Inavolisib, four-fold, mutant over wild type. Really, in Scorpion, have molecules 2608 and 478 that are really interesting molecules, but their selectivity is different on the two different domains, but really never gets much above 8x selectivity across any of the mutants. Those interesting molecules certainly should be developed, but when you have eight-fold selectivity, as you start titrating up and inhibiting the mutant protein, a significant amount of wild type is inhibited, and you could see the predictable side effects. It's like it's almost hallmark PI3K alpha inhibition, that is to say, rash, dermatitis, but most indicatively, the hyperglycemia and increases in insulin, which are the limiting features of therapy.
OKI-219 never gets to any sufficient wild type inhibition, so we do not see any increases in insulin or any dose-limiting or dose-interrupting types of side effects.
What's the selectivity over wild type?
Greater than 80x. I think there's a certain sophistication you have to have in looking at these selectivities, given that they're based on sort of a deep literature that was created for the most part by Novartis, but also you have to compare apples to apples because numbers really matter here. There is a real difference between a molecule that has 4x selectivity and 15x or 10x selectivity, simply because you can keep patients on therapy.
Gotcha. Then what about selectivity relative to different isoforms? How important is that in your opinion?
You know, it's very important, but for the most part, there's been four generations of PI3K inhibitors. I would think that the last two have actually gotten away from any level of inhibition of beta and gamma. Alpha selectivity is guaranteed. You don't go forward without it, especially if you want to get into sort of the breast cancer area, as well as some other areas that we're becoming interested in. That's left behind, like the first two generations, like starting with Wortmannin like 60 years ago. Right now, most molecules don't even get into the game without being fully alpha selective.
Okay. You do also have a pan-PI3K mutant inhibitor. How does that compare to the?
We have a pan-PI3K alpha inhibitor that from a single allosteric binding site that we've been working at, that binding site being an allosteric site, found the protein we can detect and derive mutant selectivity for any of the hotspot mutations. OKI-219 only inhibits the H1047R mutant, but our pan-mutant selective with greater than 10x selectivity across all mutants can inhibit any of the hotspot mutations found in cancers. I point out this 10x is an important number. It's not something that we made up. It's something which falls directly out of the math. That is to say, if you require EC80 or EC90, 10x selectivity makes the wild type inhibition 30% or less. You can inhibit PI3K alpha 30% or less, and that's tolerated. Anything above that leads to dose-limiting side effects.
Gotcha. Okay. And then OKI-219 has been in a clinical study for some time, in a phase one study. Just remind us of what you've seen so far and where you are with your study.
We're in three separate clinical studies right now at different levels of completion. Monotherapy, it's a basket study. A doublet in mostly breast cancer patients with fulvestrant, a SERD, oral SERD. Most recently, we've started enrolling a triplet study, two triplet studies. One in the front line, ER-positive breast cancer in combination with ribociclib plus fulvestrant, and a separate study in HER2-positive breast cancer in combination with trastuzumab, which is the generic version of trastuzumab, and tucatinib, which is a small molecule HER2 inhibitor. In fact, trastuzumab and tucatinib represent standard of care at this time in the second line. Our hope was, and it always has been, as a small company with an interesting asset, to move as quickly as we could, being as capitally efficient as we could, from first dose, first patient to these front line studies.
Because we feel that given our selectivity and our tolerability, that we were going to differentiate ourselves from other molecules in development or on the market by moving into the front lines in a well-tolerated, combinable regimen that could represent an early line of therapy. We moved monotherapy, ran the study, a science experiment, right? We ran the study, got to the doublets, have those ongoing, and are now enrolling triplet studies that would represent the data that would give us the conviction, as well as the design to move into registrational studies in early lines.
Right. What were the learnings from the distinguished study so far? The monotherapy?
We haven't reported, and we will be reporting data in the first quarter of 2026. What we learned from each study was how to run the next study. You put in monotherapy in there, you understand what patients to include, what patients not to exclude, what dose levels to go in. The monotherapy study, and of course, it also helped us with a conversation with the healthcare authorities because you have to get studies approved. The monotherapy helped us design the dose levels, as well as the inclusion/exclusion criteria for the doublet. The doublet study provided us with a conviction in the design to move in these triplet studies. We will present the efficacy data from all these studies in a way that can be compared to our competition in the first quarter of 2026.
Are these dose escalation cohorts, or did you already identify an RFPT?
In the monotherapy study, from PK and tolerability, we were able to discern what the likely expansion doses would be. We moved into the doublet study. We recognized that 600, 900, and 1200 B.I.D were going to be the doses we would focus on, and that provided us the information to start recruiting the triplet study, saying one study begets the next. Of course, because we wanted to move from first patient, first dose to front line triplets within two years, we managed to do that, and we picked the doses as we went along. What the final expansion dose is going to be for the triplets is yet to be determined. I mean, that's an important decision you make after you have all the data, but we feel that we're probably closing in on what that dose level would be.
Gotcha. I believe OKI-219 also has CNS penetration pre-clinically. How important of a differentiator is that versus some of the other molecules in development?
Yeah, when thinking about brain penetration, you know you're going to have to, there's always, generally speaking, unless you have a CNS agent that's been designed that way from the beginning or it's very low dose, is that you're going to need to put a lot more drug in the periphery in order to get adequate levels centrally. Thankfully, OKI-219 has a ratio of brain to plasma, as we call it Kp,uu, which is 0.6, which is really very high. It's one of the highest that I've seen for kinase inhibitors. Especially if you're going into certain cancers, and especially HER2-positive breast cancer, I mean, there's some other cancers like lung and endometrial where CNS disease is important. Having a Kp,uu to be able to get 60% of the exposure in brain that you would get in periphery really is very important in that.
When we were thinking about moving forward with earlier lines of therapy, going into the HER2 setting and putting ourselves with a very brain-penetrating compound on the back of tucatinib, which actually the group that I have discovered tucatinib when they were at Array, also a brain-penetrating drug, then it made a lot of sense that this could be important in the context of PI3K alpha-driven breast cancer, HER2-positive, where CNS mets were noted. It is important. It will not be important. It is less important for ER-positive breast cancer simply because of the occurrence of CNS disease there. With respect to HER2 and other cancers outside of breast, it is very important.
What percent of patients have CNS mets in HER2-positive and negative breast cancer?
In HER2-positive, it's, oh gosh, I wish I knew. I think it's 60% progressive with central disease. In HER2-positive, it's much, much lower. It's much lower for reasons that aren't extremely well understood.
Gotcha. I think the HER2- positive cohort is somewhat unique relative to your peers. I may be wrong, but my impression is that others are not necessarily going into the HER2- positive breast cancer space. What is the frequency of the mutation there?
The frequency of the mutation is one third of what you see in positive breast cancer. You would see it in, I would guess, 5% of the cancers, but it's still a significant population and a significant opportunity given the length of therapy.
Right. It sounds like the important update, Nick, the most important update, the most mature data is on the fulvestrant doublet combination. Is that fair?
Yeah, I think it's a coin flip between that and the ribociclib triplets.
Right. And so.
Because we're very unlikely to go back and run a registrational study as a doublet and then fight to differentiate ourselves in the second or third line. We're much more likely to find real value in differentiation by competing in the front line with the triplet in HER2-positive breast cancer.
Right. There has been, I think you're right, the second line market is getting fairly crowded. There has been data recently from a few molecules. I guess, what is the bar, you think, in terms of efficacy in the setting for your cohort? How do you think about that?
Yeah, so in trying to just make it very simple in thinking about this, and given that our differentiation point is going to be tolerability in combination, then when you think about the bar to get over, you would look at, in a similar population, you would look at matching efficacy, which is going to be overall response, PFS, like mostly PFS, and beat on tolerability. If you match efficacy and beat on tolerability, you usually win because you keep people on drugs, drug intensity stays up, missed doses stays down. That is when you win. If we want to compete for the front line, you have got to compete us to win at the INAVO study, inavolisib, palbociclib plus fulvestrant, which right now in that line is really the only regimen that is approved. That is the bar you are going to need to be.
Makes sense.
Right? Seven months, it's going to be the PFS. Actually, even OS is put out there. The one way to get around inavolisib, fulvestrant, and palbociclib in the front line is that the patient population that was studied and approved for was somewhat contracted to patients with lower HbA1c levels, as well as how long was duration of therapy on prior line. There is an opportunity to actually either go head to head against them or to develop around them in a population for which that regimen is not yet approved.
Right. Are you limiting HbA1c levels in any way?
No, we're not. We don't. That's always a topic of conversation with the agency. They know that these drugs, if not exquisitely selective, are going to lead to those types of increases, and they're going to want if you put a patient on drug and then you're treating with metformin, you have to take them off a drug, that's not a responsible study. Thus far, we've had no issue with delimiting our HbA1c, and we'll continue to do that.
Right. Will the data, I mean, I know both the first two cohorts have completed enrolling at this point, I think the doublet and the triplet. Will there be enough follow-up to get visibility on PFS as well next year?
There'll be the beginnings. That is one of the reasons why we held off in reporting, because we want to be able to tell the full story about OKI-219 and make an informed decision about moving forward. With the number of patients we've had on, most, if not all, the patients will have seen six months of follow-up. We should be able to say something about CBR in the beginnings of a PFS, depending on what kind of centering of the data we've had. Certainly, we'll be able to start moving towards a projection of what PFS would be and get it directionally again. In an effort to compare ourselves to the competition or to the other drugs, overall response rate is not as important as PFS. Combinability and tolerability are probably more important than that.
We'll have all that information and the beginnings of durability, which is a longer conversation. Durability in the context of PI3K alpha is you need to start studying why are patients progressing on these therapies. You need to start doing that now. We're beginning to understand under drug pressure, how does PI3K alpha mutate to render the drugs that are being used inactive? We intend to fold that information into the kind of clinical studies we're doing.
Gotcha. For the triplet, yeah, maybe just help us set expectations for this data disclosure. How many patients are going to be valuable in that cohort? Are these all first line patients, or is it a mixture? What is the comp? Are we looking at Monalisa II, for example?
The HER2-positive population, that's a smaller population, will recruit more slowly. Let me answer it in another way, because we really haven't provided guidance on what the first cohort will look like. With 15 patients, three cycles, we should be able to get a sense of tolerability, combinability, and the beginnings of efficacy, if not certainly things like ctDNA. The positive study, given the population, should recruit rather rapidly. There's a lot of interest in that. In fact, one of the motivations to move this study forward was the interest on the part of clinicians to have a regimen that included a PI3K alpha plus ribociclib, because no other PI3K alphas thus far have been truly successfully combined with ribociclib, which is emerging as sort of a preferred CDK4/6 backbone in positive breast cancer.
Okay. Yeah, just maybe going back to the fulvestrant doublet, other companies have excluded sort of PTEN AKT mutation patients, for example, when they look at the data. Did you include or exclude those types of patients?
We excluded patients with a known deleterious PTEN mutation. We did not exclude patients with prior PI3K alpha therapy. We wanted to try to understand that. Although many of the studies which have been run or are being run have a very low incidence of prior PI3K alpha. Not only in predicting and understanding response, not only is the number of prior lines going to be important, everybody knows that five versus two lines, and you're going to impact the overall response as well as PFS, but also what prior lines, right? Because you treat a cancer cell with one drug, it has one set of resistances. You treat it with another drug, it's something else. You have to be very thoughtful about, especially in these studies, these earlier studies, where you have to accept patients with multiple lines of therapy.
You have to remain mindful of exactly what those patients have seen, when they've seen it, and how often they've seen it.
Right. Have you disclosed, I mean, in terms of just enrollment criteria in the doublet, do you expect patients to be 100% post CDK4/6, or what about?
The majority of patients will be post front line. They will be generally estrogen therapy, endocrine therapy resistant. They will have seen a CDK4/6 mostly. They will likely have seen letrozole or something like that.
More likely cert naïve, presumably. And then what about chemo?
Not in the triplet studies, but in the other studies, you will have seen a significant amount of chemo. Chemo is interesting. One has to think about, will patients be less likely to respond or more likely to respond to a PI3K inhibitor if they've seen chemo and no PI? More attention needs to be paid to that, especially if you're introducing the drug into later lines of therapy. What has the patient seen and how many prior lines have they seen?
Maybe just to summarize, in an ideal scenario, I guess, what would you want to see from the doublet and perhaps the triplet study to advance? The next step would be your phase threes, I'll say.
Really easy for me to remember, match the competition on efficacy and beat them on tolerability.
Gotcha.
If it's in the front line, it's the INAVO120 study. If it's in the second line, it's going to be anywhere that alpelisib and abemaciclib relay, even inavolisib and capivasertib, wherever they enable, all sort of fall out the same place. If you beat them on tolerability, you beat them on, then it's just going to be a preferred regimen in any of those lines.
Right. Okay, that makes sense. Yeah, what about your pan-mutant inhibitor? What are your plans there?
We've identified the molecules that we're likely to take into development. We tend to run a number of IND enabling studies prior to DC, just we move the line so that we reduce the risk. Those molecules are going to have sort of that high bar, 10x and above selectivity across all the mutants. Keep the projected dose down so that DDIs are not a risk. These are going to be combination therapy. We'll be reporting on those in the first quarter with the hope of having INDs in 2026.
Great. Maybe just going back, sorry, one more on OKI-219. When you say better tolerability, is this mainly on hyperglycemia, rash, or are there any other particular side effects that you think are critical to improve upon?
We have not seen any side effects sort of crop up that would indicate some kind of off-target toxicity. That is great. It is. We have seen no indications of treatment-related on-target adverse events, but those are the ones that certainly are in tenor of.
Right. So in terms of just improving upon the competition now, it's mainly on hyperglycemia and then GI?
Stomatitis and rash.
Gotcha.
Of course, you only find this when you do the studies, the other risk associated with exacerbating toxicities of combination partners, which you're only going to understand as you do those studies.
Right. Some peer companies have had some DDIs with CDK4/6 inhibitors. Is that something that can be predicted or expected?
It is something that can be predicted. Certainly, you can understand the metabolic fate of your molecule and see whether, in fact, there's going to be any risk to either being a perpetrator or a victim with respect to your own. We've done some preclinical studies to minimize it. Of course, if your dose is low, your dose gets, the less likely there's going to be a DDI. Removing DDIs when thinking about 4/6 inhibitor is really important, given that any fluctuations in that exposure is going to lead to fluctuations and a lack of predictability of those side effects, which docs also have to manage. You just can't put a drug in there that's going to make abemaciclib do this, because that's going to make the treating clinicians nuts.
It sounds like you didn't see that preclinically so far.
We did not see that preclinically. There are ways to predict it and minimize that risk. More people should be doing that.
Sounds good. All right. Thanks. I really appreciate the time. And then we'll have the big data disclosure next year. Looking forward to that.
Thank you.
Appreciate it. Thanks.