Good afternoon, everyone. Welcome to the Targeted Oncology Panel. I'm Tara Bancroft, joined by my colleagues, Yaron Werber, Eva Privitera, and, sorry. Troy. Very, very sorry. It's a long day. Okay, so on this panel, we have Mike Andriole. Andriole? Andriole. Andriole, the CEO of Chimerix, Sam Kintz, the CEO of Enliven, Mark Manfredi, the CEO of Ikena, Ben Zeskind, the Co-Founder and CEO of Immuneering, and Clay Siegall, the President and CEO of Immunome. So I figured we could start it out with some topical thematic questions and then move into company-specific questions. So I guess to start, we can maybe start on this end and work down. But what do you think are the biggest unmet needs in targeted oncology at the current moment? Whether, you know, whether it's technology-based, regulatory or commercial considerations, things like that. Yeah, there's a lot of ways to go with that question. Yeah. I assume it was directed to me. Sure. Yeah. You know, from a regulatory perspective in Targeted Oncolog y, part of the promise is you can get an early signal on activity, early signal on response rate, and help guide the next wave of development. Particularly in oncology, we've seen the regulatory winds of change move from sort of one end of the spectrum to the other in terms of acceptance of single-arm data for accelerated approval. And so from a regulatory perspective, we've seen exceptions to that phenomenon play out over the last couple of years. But there's more uncertainty, I think, on the regulatory accelerated approval path for Targeted Oncology, maybe than there's been in the last 10 or 15 years. And there's part of me that's a little agnostic as to what direction it heads in. If we're gonna go completely to randomized controlled studies and nest an early endpoint in that, or continue to look in certain situations at single-arm data for accelerated approval. But more clarity on that, I think, from the agency would help the field immensely. So I'll take it maybe from the biology perspective on unmet need. You know, there's been tremendous advancements in targeting pathways and targets that have been undruggable in the past. So if you look at the RAS space, there's been incredible progress on, you know, building off G12C as a target to now drugging the other mutations like G12D and others. But I think there's an unmet need in maximizing the potential of those. So you see the G12C there, the durability is quite short, and then you go beyond the initial indication of non-small cell lung cancer to, like, colorectal, where you have to add different agents. So there's just a lot more biology to figure out how to maximize some of these major drivers, and so I think you're gonna see, you know, other... Just take the RAS space for a sec, second. You're gonna see other, you know, progression of targeting G12V and other mutations, but then you're gonna see how to add other therapeutics onto that and other modalities. So now going from small molecule targeting to degradation of, you know, some of these drivers, I think there's a lot of unmet medical need. If you look at, like, just the EGFR space, where, you know, there was EGFR approval many, many years ago, but they're still figuring out how to make the durability greater, greater response rate. And I think that's gonna still happen in some of these newer areas like the RAS space. There's a tremendous amount of biology to, to figure out. I think that there are some tumor types that we need to be better at, and I'd love to see everyone working on targeted therapy to do some more with pancreatic cancer and AML, and diseases that don't have a lot of good therapies, and do things with good therapeutic windows. I look at this field, and, I'm probably older than everyone here, so, I've been involved in six drugs that are on the market now and, in targeted therapy. And there's, what, about 25 or 30 really good Targeted oncology drugs, and there needs to be about 125 or so or more. And I view it that maybe that's 20 years from now, or whatever, there'll be 125, and we won't use cytotoxics because we won't need to. And we'll, you know, the oncologist won't care if you have colon cancer, lung cancer, leukemia. They'll wanna know what's, you know, unique about your tissue that went awry, and they'll feed that information into ChatGPT 8.0, and in their office in the little screen. And then they'll tell them what three targeted drugs to combine that'll give you a 92% five-year disease-free survival with very little toxicity. And all the patients will be happier because they're not getting a lot of chemo or problematic drugs. And, you know, we I hope that every company here can contribute to that. We're not competitors. We're competitors versus cancer, and we all need to make drugs, and there needs to be another 100, and to get to where ChatGPT has enough information to tell us what to do. I'll try to incrementally build on what these other panelists have said. I mean, a couple of themes that I'm hearing that I think resonate with my feeling about where we are in targeted oncology. So the promise certainly is that there are sub-patient populations that respond, are exquisitely sensitive to these targeted mechanisms, to the extent where monotherapy activity can meaningfully extend overall survival, and in the extreme cases, turn cancer from an acute condition to a chronic disease. So that's the promise of targeted precision oncology, and certainly what brought me into this space. Some of the challenges we're hearing today is it's just not enough, all right, for enough people, and for multiple tumor types. There are different drivers of the biology, and so you're gonna have to hit it in multiple different ways. Combinations are challenging because cancer therapeutics do not have wide safety margins or therapeutic windows. So getting to the point where we can do what Clay mentioned and rationally combine therapies requires wider safety margins. So there continues to be room, not just in discovering new targets, which will be important, and we need to do so in these tumor types that have been underserved and historically challenging to develop drugs in. But we need to be able to make better therapeutics with wider safety margins. And that gets back to one of the first challenges that was mentioned is: well, how do you do that in today's regulatory environment, right? If you have to start at the end, the biology has already changed. But if ultimately we need to get to the beginning in combinations, you need safer drugs with wider safety margins to get there. And so we do need to put more thought into how do we get to the relevant setting for what's gonna be important in a future context of the standard of care, which for many tumor types, will be combinations. And so it's sort of an integration of, you know, innovative regulatory strategies to test, you know, the hypothesis of making better drugs earlier, so that we can then combine them through novel combinations for these patients that are underserved from a biological unmet need standpoint. I certainly agree with all the previous comments. I guess I would phrase it as broader activity and better tolerability, right? You know, I think, you know, clearly, targeted therapies like the G12C inhibitors are a huge breakthrough in a pathway that was previously considered undruggable. But you're kind of blocking one lane of the MAP kinase superhighway, right? And it, you know, it slows down traffic, but the tumors can just mutate again and switch lanes and get around it. So I think to develop, you know, maybe a little less targeted, right, a little broader therapies that can block all the lanes of the highway of a particular pathway and make it a little harder for cancer to get around it, you know, I think can help with some of these challenges. And then better tolerability. Look, I think, you know, the philosophy shouldn't be no pain, no gain, right? You know, I think really, there's an opportunity to think carefully about what it is that causes toxicity in a lot of cancer medicines and really fundamentally improve on that. And, you know, I think the FDA's Project Optimus is a great step in kind of pushing people to think in that direction. Or you don't, you don't have to dose escalate to an MTD, where you, you make someone really sick and then back off. You know, you can choose a, a dose in a more, a more rational and more, more molecularly driven way. And, you know, I think that, that combination of better tolerability and broader activity will really, really get to some of these goals and enable us to, to help a lot more patients. And so maybe moving on to the second question, and we'll start maybe from you. Sure. How predictive are preclinical model systems in assessing safety and efficacy of targeted oncology candidates, and how does that vary across different tumors and different targets? Yeah, and I think they vary, and I think it depends how carefully you're thinking about the model. So, you know, let's talk first about in vitro models, right? You know, flat cells in a petri dish in fetal bovine serum is not a great model for a tumor, right? That, you know, there's not many patients that have flat tumors or, you know, a cow, calf serum in their veins, right? So, you know, there's humanized 3D tumor growth assays that do a better job of that. You know, so there, I think there's innovation there that can help make in vitro models more predictive. And then on the in vivo side, you know, I think we're fortunate in oncology that we have reasonably well-predictive models, but you really have to understand what you're modeling, right? You know, each model really represents certain lesions at a certain stage in the disease, right? So to really understand and predict a clinical context, I think it's important to really take that into account the heterogeneity among lesions, right? The changes in cancer as you get to earlier versus later lines of therapy, and make sure that the models are really taking all that into account. But I think with that kind of thinking, you know, you can actually create almost a mosaic or a collage of models that help align with the activity you're trying to predict. Sam? Yeah. I think as far as you know, cancer goes in oncology, that the precision oncology, the targeted therapy, preclinical efficacy translates pretty well. I think that's been one of the strengths of the area is you can pretty much line up free drug exposures that are required to elicit whatever the desired response or you think threshold response needs to be, and those more or less line up with you know, what you will see in humans. The challenge I think has been more so on the safety side. For known established targets where at least the on-target liability in humans has been established, the safety models are pretty good. But as we know, they don't. That's where things get quite a bit different, particularly when you're trying to model a therapeutic window in rodents. Because, you know, mice tolerate chemotherapy pretty well because they're treated in a clean environment. It's not usually the chemotherapy side effects that result in issues in the clinic, it's the secondary things that happen, infection and all of that, that leads to, you know, limited therapeutic window, and that's really difficult to model in preclinical models. But I think the efficacy side is pretty good. The safety, as long as you're comparing free drug exposures in the relevant species, higher species, give you a sense for at least a margin to evaluate. I think where you get into trouble is when you require higher exposures in these rodent models that may not capture some of the toxicities associated with the target, in the higher preclinical species. And unfortunately, a lot of programs fail due to that. And you see a lot of attrition there. But really, there's no perfect safety model until you get to the clinic. So I'd sum it up as efficacy is pretty good. Safety, yeah, you need to test it in the clinic. I think that a lot of the tumor models that are used are from cell lines on the back of a rodent, and they're clonal, and they have the same display on every cell, and human tumors are heterogeneous, and so it is not the same thing. So a lot of people get thrown off with activity in a clonal cell line on the back of a rodent and think it's gonna work the same way, so it's not. So, I think we need to continue to make good models that are more representative of a human tumor. Yeah, I agree with that. I think there's... It highly depends on the target and the disease you're trying to replicate. You know, there are some model systems with certain targets that are highly representative in terms of efficacy, the plasma concentration necessary to humans. And then, there are models that can't be modeled preclinically, and people try. I think the IO field is a perfect example where that translation's been horrible. And so you really need to know, like, what is the model system you're gonna use, and what is it gonna tell you? And so if, you know, if that target has been drugged before in the clinic, you know that translation. You have a better sense, at least, of that translation. The hardest thing is when you're going after a brand-new target. You have no idea if that target's gonna be replicated in that model system, and that's when you're flying a little bit blind. So it can't be the only. It's necessary but not sufficient of why you want to go forward with that target. So I think you have to be wise there. And then, the same thing for toxicities. I think there's, you know, follow-on molecules that are second generation, third generation of targets that you know that translation from a preclinical setting to a clinical setting, and those are the easiest to say, "Okay, at least I can predict better my safety margin." And again, I think when you're in a new setting with a new target, particularly if you have two species and one species has a toxin and it and the other one doesn't, it's where you're a little bit more flying blind on the safety side. I agree with all the comments. The only thing I would add is, you know, we have a lot of examples where programs have died in vivo, and we're just not comfortable moving forward. I think there are times when you have to exercise judgment, recognizing the heterogeneity of some of these models and recognizing that human cells behave differently. And thinking back to the last company I was with, we had in-licensed an asset based almost entirely on the clinical data as opposed to the non-clinical data. And it turned out to be a successful project and has become standard of care in its disease. Yet, when we in-licensed it and went back and ran it through our own models, had reached the conclusion that if it were up to us and we only had the models we had at the time, we probably would have moved it into the clinic, which was an interesting conclusion. So I think it's important to exercise judgment on the limitation of the models you're using and are there paths forward, because nothing can replace, you know, our actual human clinical data. FDA, I'm gonna move to a regulatory question. So FDA is changing things, you know, faster or has been beginning to move away or changing the bar for accelerated approvals. You know, both in terms of the confirmatory data needed and potential needed randomized data right away, and also the tumor-agnostic approach is getting revised, right? It's in the process of getting revised. It's still not 100% clear what that pathway is gonna look like, potentially partially because the therapeutics are not, they're not agnostically tumor-agnostic, right? They have different activity in different tumors. So maybe, Mark, we'll start with you. At what point do you start having visibility from FDA as to what they want? Yeah, I think, you know, the first thing to look at is what data you have, right? You get a sense of, you know, what the indication is, how your compound's performing, and if it's in a very defined population, then it's an easier conversation than it's tumor-agnostic, because then you need more data to go across different tumor types. But I think, you know, if you have that data and you have a potential, you know, path forward with, you know, with a single-arm trial, then you have enough to have that, at least have that initial conversation. But I think, having the conversation earlier where you don't have data across different tumor types where the genetics are playing, and we're in a situation where we are going after mesothelioma that have NF2 mutations because it's a, you know, the tumor type that has the most, percentage-wise of the mutations, but it does go across many different tumor types. But that's the first place for us to focus and, you know, collect data to then go to, to FDA to have this discussion, as opposed to going across different tumor types. I don't think the FDA has changed that much. I think that every now and then, Pazdur comes out and says, "No more single-arm approvals. You have to do this, and you have to do this." And he says it for the masses so that everyone knows he's there, and but you bring great drugs with great data, and the FDA moves, and they approve a single-arm phase II, or they do whatever. So I think it's important for the FDA to now and then level set things, and they do tweaks here and there and add some projects and programs. But at the end of the day, if you have a drug that's the fourth in the same class, and it doesn't look that different from the other one, it's gonna take a long time, and if you have a drug with an 88% response rate and in a class where there's nothing there, they will move so fast. I mean, I've had approvals that have taken months and months and months, and I had one approval for 11 days. So, and Pazdur told me it was the fastest one I did for now, and I don't know if there's a faster one. So I, I think they still do whatever they want based on the data. I mean, I completely agree with that. I admittedly have more limited experience, but they're very data oriented, right? So you make an argument based on historical data. That data is getting because of the attention, I think, that single arm approvals got. I think there are better data sets out there to refer to and to benchmark your own programs against, 'cause that's also of huge value. You don't want to keep investing in drugs where you're not, you know, meeting those internal benchmarks. But they respond to data and scientific and rational arguments, and so I think just, you know, using all of the additional data sets that have been developed out there in terms of the natural history of the disease, standard of care, how that's changed. And data can come from anywhere, and then, of course, your own data. It needs to kinda meet that bar to have a discussion and to engage, and then multiple paths can open up. But there isn't a one-size-fits-all. I think one area where the agency has been particularly innovative, and it's helping to shape the data that comes to it, is Project FrontRunner, right? So this is an effort where they're saying, you know, "Go out and take your agents into studies in earlier lines of therapy," you know, whether it's as a monotherapy or as an add-on with an existing standard of care. And I think that's new, right? That's different from the standard paradigm, and it's you know, it's trying to shape the data that comes to them. And you know, I think it's to the benefit of patients. You know, in our phase II-A, we have two arms that are first line in combination with standard of care, and then we have three monotherapy arms. I think that's, you know, that just creates more options for patients. Those first-line combo arm, I mean, that's a great, that's a great option for patients. So, you know, I think this, this idea of moving drugs into the front line earlier in their assessment, which FDA is promoting through Project FrontRunner, is a, you know, a positive innovation. And maybe on the commercial side, I think we've seen in the past a couple of these Targeted Oncology launches have maybe progressed a little bit more slowly than investors would have liked. And so I guess, how should investors think about the launches of these agents? You know, are they being too optimistic, or are there challenges that you can't foresee? Are there commercial challenges that you might run into later on after the launch, that you might not foresee in the development stage? And maybe I'll open it up to the entire panel for whoever feels like they have relevant commentary. All his drugs were big, though, so he cannot- No, that's not true. He overperformed. That's not true. I launched a cervical cancer drug that was really small. Tivdak. And, I think sometimes you have a paradigm where the doctors are so set in their ways, you have to change it, and other times it's a disease where doctors are more open. I'll give you an example. Hodgkin lymphoma. They were treating Hodgkin lymphoma for 40 years with four chemotherapies, well, three and a steroid, ABVD. And you come out with a new drug, and the drug is way better than the cytotoxics. And the doctors, so, some doctors, you know, if you're at Dana-Farber, they're changing it right away. And I remember we had, with Adcetris, we had, like, zero sales in the state of Utah. I remember I got on a plane, went to Utah, and sat with my salesperson, and we went to the major University of Utah, you know, cancer center and a couple of the key community centers. And we said, "You know, why aren't you using this drug? It's better for patients. Look at our data, look at this and that." And they said: "Well, I don't know." And they said: "We'll have to look in how we treat it." And some of these community centers and big ones, they have, like, it's an antique. They do what they do. And they just didn't even know, and they say, "Oh, we've seen it," whatever, and you have salespeople talking to them, but they just, they're stuck with it... and then, it takes a long time. And so Wall Street says: "Oh, well, you know, geez, this is much better. You should have this much sales." It's not always that way. And then there's other times where it's just the disease is a group of doctors that just, I mean, they're very innovative, and it depends. You know, in Hodgkin lymphoma, the standard of care hadn't changed for 40 years, so they were so entrenched, and some of these docs were treating patients. They don't treat a ton. Hodgkin lymphoma, it's a rare disease, so they're seeing, you know, four, a year, and it's just, they don't even know there's a change, even though it's so much better, and they don't even know. In some diseases, they're on top of it. I, I think it's a hard thing to do for Wall Street, but it catches up, and if Adcetris didn't sell and Wall Street said, "Well, it's gonna be a $100 million drug," until last year, it was $1.7 billion. So it, but that took a while to get there and took six more labels. So I, I don't. You know, it's really hard to do that. It depends on the disease, the doctors. If there's a lot of innovation in the disease, the doctors are used to looking for it, and then they'll take it, the new drug, up faster, or if there's a massive unmet need. I think in Targeted Oncology, to build on that, you know, it starts with incidents and making sure you've got your arms around that, but then testing rates matter, right? You would think in 2024, you know, every tumor in the world would be reflexively tested at diagnosis, and we all know that's clearly not happening routinely in every tumor type. And every mutation, when it is, is not being necessarily screened for. So understanding that and then understanding the competitive landscape, there's particularly in targeted oncology, there are a lot of fast followers, right? Targeting the same mutation, trying to build a better product profile, but the number of patients stays relatively finite. You also see some overlapping mutations and alterations in some of these tumors. You're competing not only in your target of choice and in your class of drugs, but other potentially target agents for that too. So, building a share model in that environment, I think is more a little more complicated than we think, depending on the competitive landscape, and I think can lead to underperforming some expectations. Okay, well, we're gonna shift to company-specific questions now. I can start with Ikena. So can you tell us how and why you chose to target the Hippo pathway, and then what types of indications that you're enrolling? So we became interested in the pathway, as well as just the industry, for two reasons. One, there's genetic alterations in the pathway, which causes driver, you know, activation and tumorigenesis. So if you look at NF2 loss of function, I mentioned earlier, mutations in mesothelioma, but it goes across many different types of tumors, like non-small cell lung cancer, et cetera. And there's also fusions in the pathway that cause activation, as well as amplification. So a lot of genetic evidence that it's a driver in certain indications. And then second to that, there's a role in therapeutic resistance, and when you put pressure on a pathway, meaning add an inhibitor to EGFR or RAS pathway, you get activation of this pathway, which circumvents the inhibition of the other pathway, so it's a compensatory activation. So if you look at that market, it goes to, you know, again, EGFR inhibitors, RAS inhibitors, MEK inhibitors, other RTKs. So it's a really vast amount of, you know, clinical space beyond the monotherapy opportunity. And so initially, as with any dose escalation and then exploring expansions, we're focused on monotherapy first, so we did dose escalation, looked at some a very ultra-rare population called EHE. It's a soft tissue sarcoma, looking at clinical benefit. We disclosed that last year and now looking at mesothelioma for disclosure in the second half of this year with additional EHE data. And that will show that, you know, the pathway's safe, number one, and then two, for clinical activity as a monotherapy. From there, going on to the combination with EGFR, osimertinib in non-small cell lung cancers. Okay, great. I'm gonna pass it over to you now, Clay. So you obviously experienced a tremendous amount of success at Seagen. So what made you want to go back to work at Immunome? And, like, can you tell me what the goals of the company are? What's the DNA of Immunome? Well, I looked into buying a tiki bar in the Caribbean. That would've been fun, but I would have missed you, Roan. Thank you. I think the last time I saw you was in Napa. Yes, that's true. Yeah, and we were not drinking wine. No, I like making cancer drugs, and I like the benefit of it, and there's, you know, there's nothing like going into a cancer center and having someone sitting in a La-Z-Boy, and sitting, talking to them while they're getting infused with your drug, and they're shaking cold, and they have a blanket on them, and they have, like, an audiobook there, and they're so excited to talk to you about your drug and learn about it. And whether it's in a clinical trial where they're explaining to you that they wanna do it to help other people that are coming after them, or they're getting commercial drug, and they're looking forward to seeing their, you know, nephew play a softball game. I mean, it's nothing like it, and when I was 19, my dad was diagnosed with cancer, and he died a few years later, and I, my mission has always been to make cancer drugs. So I, you know, I don't know. You know, the question is like, why are you doing this? Because I like it. Yes, you're right. I don't need the money. But that money is only one part of the equation. And why Immunome, you asked? Yeah. Yeah, what are the goals that you're trying to achieve here? It's making great therapies for cancer patients, and we're focused on targeted therapy. We're not gonna do gene therapy with CRISPR, which is great. We're not gonna do cell therapy, which you need a different infrastructure. So we're gonna do targeted therapy of antibody-based, small molecule-based, and radio, radionuclide. And we're gonna do high science, and we're gonna try to not work on what we have, but work on what's best that we have. And we already have a big pipeline, and we're doing work inside in our labs, and the playbook calls for inside the labs and to in-license an exciting thing. That was the playbook at Seagen, and always had four, five, six, seven molecules running into the clinic. That's the plan because despite all the great science we do, and despite all the great funding everyone does, it's still hard, it's still hard as hell to treat human cancer. And, and, you need, you know, a lot of, shots on goal, and I hate that phrase, but you need a lot to... and great science and good luck to make this work and make an impact on cancer in a big way. And I'm not talking about incremental drugs. Make real advances, you need a lot of work and effort, so it's fun. Let's move on to... You're developing a short-acting MEK inhibitor, and that's based on your informatics. So you're trying to change the paradigm, which has been kind of moving toward slightly better tolerated, long-acting MEKs, and you're looking with a very different signature to impact RAF and RAS, both up and downstream, the RAS pathway. And you're gonna release phase I data this month. Can you talk about what we're gonna see and how that builds on what you've shown at AACR last year? Sure, absolutely. So yeah, I think you summarized it well. Right, we're using a Deep Cyclic Inhibition mechanism. And really the goals are to, you know, improve significantly on the tolerability that you've seen for prior MAP kinase pathway inhibitors, and have broad activity, right? You know, universal RAS was the enrollment criteria for our phase I. And, you know, as you said, we're, you know, we plan to share the top-line results this month in March. And I think it's important to appreciate in the phase I, our endpoints were tolerability, candidate RP2D, those were the primary endpoints, and PK, that was the secondary endpoint. So, you know, I think you can expect robust updates on each of those based on the kind of design of the trial, the patients selected, and, you know, all that. And then we've also committed to sharing, even though they weren't endpoints of the study, PD data, circulating tumor DNA data, and initial clinical activity data. And I think you can expect that that'll help to kinda inform and understand the selection of the candidate RP2D. It'll really help, you know, help folks understand why we decided to expand the plans for the phase II-A. So a few months ago, we announced we were increasing the number of arms in the phase II-A. We're adding additional sites and investigators, and of course, those changes require, you know, submitting a protocol amendment to the agency. You know, you need each of the investigators to believe it's worth their time, worth their patients' time. So I think the data will help folks understand, understand better that decision, and then really how we're doing on this, this goal of universal RAS, right? I think the ctDNA in particular will be, you know, is, is a valuable way to look at, look at how we're doing on that front. And the concept with the short acting, you want to talk about the cyclic inhibition? Because MEK inhibitors as single agents historically haven't shown much in terms of response rates, and they've really been combos. But with you, you're really looking to modulate the pathway in a different way. That's right. Yeah, and, you know, look, I think traditionally, historically, MEK inhibitors have been chronic, right? They've shut down the pathway 24/7. And it turns out we have the MAP kinase pathway for a reason other than for cancer to hijack it, right? So if you just shut down the pathway, you're causing a lot of harm to healthy cells that also rely on this pathway. So the idea with Deep Cyclic Inhibition is that while healthy cells and malignant cells both use the pathway, they use it differently. Right, the cancer cells are always on. They're addicted to this sustained high level of MAP kinase signaling. The healthy cells are more easygoing, right? They can tolerate more interruptions, lower levels of signaling. So the idea with Deep Cyclic Inhibition is we hit the cells with this very high free fraction Cmax, much higher than you could achieve with a chronic inhibitor, to really strangle the cells, the malignant cells in the MAP kinase signaling. But then with a short half-life, it's about two hours in humans, we have a complete release, a near zero drug trough. And we believe that, you know, that essentially makes every day a drug holiday, right? Every day, the healthy cells get the MAP kinase signaling back in time to, we believe, be less affected. So that's really the concept of Deep Cyclic Inhibition. And you're right, it goes against kind of the conventional wisdom in target oncology, which is to make drugs with a longer half-lives, but our, you know, our data is what led us to it in the first place. You know, I think our preclinical data supported it, and we look forward to sharing our phase I data in March. Moving over to Sam, your lead asset, ELVN-001, is a TKI being developed in CML. There's currently six approved TKIs in CML. What's the remaining unmet need, and how is ELVN-001 designed to address it? Yeah. So, I think our lead program represents sort of the opposite end of the spectrum here, where we're just building on what's known about the effectiveness of TKIs in CML. And for those of you that haven't been paying attention since, it's been a long time since much innovation has happened in CML. Imatinib, which is Gleevec, was, I think, the first approved, you know, precision TKI. There were some multi-kinase inhibitors that have been approved prior to that. On the commercial question, actually, Novartis got a lot of flack in moving forward the first BCR-ABL inhibitor, 'cause no one thought there'd be a commercial market there, and it went on to be the largest TKI in terms of commercial market over time. Today, if you looked at the branded premium price prevalent patient population, it's probably a $14 billion end market opportunity. Of course, now they're generics, and since the approval of imatinib in 2001, overall survival in this patient population has not been reached. It's been so effective that the newly diagnosed CML patient has the same expected overall survival as the age-matched general population. So really, CML almost isn't a cancer anymore. It's a chronic condition, and actually probably more similar to HIV/AIDS, in that patients will live decades, but they require daily TKI therapy. And that actually is what's led to what we view as the new unmet need. This is a chronic condition. A quarter of patients in the prevalent patient population have exhausted at least two prior options, and there are limited treatment, you know, options available for them. So there have been decades of innovation in this space, and I think it's an example of where we can make these drugs better to help improve patient outcomes in terms of their treatment goals today, which really are quality of life and convenience. So if you're a patient with CML, taking drug for the next 30 years of your life, well, you don't want to feel like you're on cancer therapy every single day. And that's where market share has really shifted, especially with Novartis, which has been a leader in this space for the last 20 years, with their new drug, Scemblix, asciminib, which was recently approved in third line plus CML, but will be approved very shortly in frontline CML, so available for all these patients. What's most amazing about that drug is it's proved that through more selective BCR-ABL inhibition, you can improve not only efficacy, but also safety and tolerability for these patients. But we know the vast majority of these patients won't stay on this drug forever. They will develop resistance and require TKI therapy with different mechanisms. So really building on the insights that me and my team have had for now 15 years, first working on this target, actually not for cancer, but for neurodegenerative diseases like Parkinson's disease, you know, coming in with a different mechanism for drugging that same target is highly likely to be active in that setting for patients that cycle off Scemblix, this new treatment option, and can capture that instant share of patients coming off there. So this is an example of, you know, we expect 50%-70% of patients to be on generic TKIs and doing relatively well, but it's a large enough prevalent patient population, and there is still a significant unmet need for these patients for which the other TKIs aren't working. There really isn't any commercial risk if you meet that differentiated profile. There are six approved drugs, all of them make over $500 million, including Scemblix, which was just recently approved in the late line setting, and within a year and a half has already reached, you know, a $500 million annualized revenue run rate. So it's an example of you can get a lot of mileage commercially out of these small patient populations if you have transformational benefit, and if you take an acute cancer and turn it into a chronic disease, then even these small patient populations can deliver huge value, and obviously, there's benefit to go with it. So moving to Mike, I know you all at Chimerix have done a lot of work in the glioma space, and I want to ask about the phase III ACTION study with the dordaviprone in H3K27M glioma. But first, maybe you can just remind us of some of the efficacy data that you've seen from dordaviprone in H3K27M glioma patients and how that's compared to the standard of care there. And then separately, maybe why you think... or maybe if you can just outline for us the reasons why you think the phase III study is a good chance to succeed. Yeah, it's a good, good question, Troy. I think it was Clay who noted at the start of the panel the importance of more targeted agents to areas of very high unmet need, pancreatic certainly being one of those, high-grade glioma certainly being one of those, one of the highest unmet needs, I think, remaining in oncology. We're targeting a certain mutation in high-grade glioma and brain cancer. It's a H3K27M mutation. It's a mouthful, but it's an automatic grade four by WHO criteria, negative, a prognostic for survival in an already really, really difficult tumor to treat. Nothing is indicated specifically for that, and that chemotherapy temozolomide really doesn't work in this MGMT unmethylated type of mutation in tumor. And so as a consequence, in a recurrent setting, n atural disease history would be low single digits in terms of response rate expectations, maybe even down to zero, depending on how you want to define that. Maybe on OS at recurrence is about five months in the literature and supported by our own natural disease history study. So a particularly difficult situation for these patients with this diagnosis. Our phase II data set is predicated on the first 50 patients or single-arm data that met a homogeneous definition that was outlined with FDA on assessing response by a relatively new measure for assessment criteria called RANO for high-grade glioma, and had a 20% response rate by that measure, and a 30% response rate by a similar measure called RANO-LGG. The durability of those responses really was what struck us in an unexpected way. So eight months of onset of response and just over 11 months durability of response. So a statistician won't allow me to say PFS, but you could think of it as PFS as a year and a half in that responding cohort of subjects. And so really unprecedented in terms of comparison to the natural disease history in a really difficult treatment setting. We're predicated on that, initiated a phase III study. We're enrolling a 450-patient randomized study globally in this patient population. We're on track for interim OS survival in 2025, and final OS expected in 2026. We've moved from sort of the recurrent setting to the frontline setting in that study, which we expect, without going into too much detail, a higher treatment response, in that setting. We've also enriched the population, by, eliminating performance status below 70, where we really didn't see any responses in the, in the, phase II experience. And then, only the one time I've said thank you to the FDA, on Project Optimus, certainly an interest from, from the agency in exploring more dose optimization in this study. Optimus came about sort of in the middle of this clinical program, and so we've added a second treatment arm. Initial conversations were maybe we'll look at a less intense dosing schedule, but where we actually ended up, considering we haven't actually hit an MTD, and given the safety profile of the program, we essentially have a double dose in a separate treatment arm. So getting twice the frequency of the same dose as a separate treatment arm in that study. So several reasons why we think there are a number of tailwinds to the probability of that, and we're looking forward to interim data next year. Okay, so Hi, Mark. So what are you thinking of doing in mesothelioma this year? You mentioned it before, but if you could go into a little bit more details just about the goals in that indication for this year, and if you are going broadly into genetically defined and other mesothelioma and perhaps even what kind of improvements can we expect to see in that population? Yeah. So we have two goals for monotherapy for signal. One is EHE, which we talked about, which is the first data we disclosed in the fall, where we saw improvement of, you know, clinical outcomes. We saw tumor shrinkage. We saw clinical benefit even during the dose escalation. And just to stay on that for a second, we'll continue to enroll patients in EHE. Just a reminder, those are 100% of those patients have a genetic alteration, so they're easily defined at the clinic, and there's really nothing out there for them. There's no approved therapies, and you know, we've gotten a lot of interest from the community. For mesothelioma, it is a population that, from the genetics, for the Hippo pathway, has some diversity, so there's NF2 dysregulation, so there's loss of function, monoallelic, biallelic. There's loss of function mutations. There's epigenetic silencing of the gene, all of which does the same thing, which either dysregulates the protein or eliminates the protein and activates the pathway. And then there's the other mutations that are there. So we're going to be looking at all of that, plus patients that are wild type. So we'll be and we are screening patients that have this loss of function, but also taking all comers, and we'll be deconvoluting the activity from the RP2D for the second half of this year. You know, we think that, if you look at the totality of the genetic alterations, that 70% is really the actionable population. And so that's what we'll be looking for, for looking at the correlation of activity and those genetic markers. And so we'll be looking obviously for continued safety. I didn't mention this, but the pathway has a liability of the kidney, where you have proteinuria as a marker of toxicity. The initial clinical data that came out from a competitor in April of last year showed clinical activity in mesothelioma, but dose limitation of this kidney signal. Our initial data has shown that we don't see this kidney signal being a dose limitation. So continued safety from that side, as well as the anti-tumor activity in the mesothelioma population. Okay, Clay, back to you. So you have announced several different agreements, acquisitions recently, even just since January. So I guess, you know, you could choose to talk about any, but I think the AL102 asset is, is interesting. So I wanted to ask what a ttracted you to that asset, and when we might get phase III data? Because I know you just announced the phase III enrollment completion, and also if you would like to debate Yaron desmoid tumors. I, I will lose the debate with Yaron. Desmoid tumors are not very life-threatening, but they're really painful and difficult. There's about 1,650 new patients a year. They're the U.S. numbers. About 30,000 prevalence, about 6,000-7,000 of them roll into therapy, and they get radiation or TKIs or something that's hard to tolerate. None of it's been approved ever, and then last year, nirogacestat was approved in November, which is phenomenal. They're the drug originally developed by Pfizer, for neurologic disease. It's used twice-a-day pills at 150 milligram per dose, so 300 milligram per day. I think what the company developing it did was great. The first time there was ever a drug for desmoid tumors, and it's a real groundbreaker. What I was interested in, in AL102 from Ayala, I used to call it Ayala, but they corrected me and said, Ayala. This drug was originally developed by Bristol Myers Squibb, and it was developed for cancer, and it's not used 300 milligrams per day, it's 1.2 mg per day. So it's one grouping of pills instead of two groupings of pills, and so it's a little simpler, and it's more potent. The potency is important, at a similar, roughly similar toxicity profile. You have quicker responses, you have deeper responses, you have faster pain relief, you have quite a lot more anti-tumor activity, tumor volume reduction, T2, a measure of cellularity to look at the normalcy of a, of a lesion. So what we saw in there was just that this was a best-in-class drug that Ayala had really no... They were chronically underfunded and didn't have the wherewithal to do CMC, and so it was a really great drug that was withering on the vine with the wrong company. But if I could turn the clock back a number of years ago, when I was at Seagen, I found a drug that's now called Tukysa. It's available all over the globe to treat breast cancer, and especially breast cancer that's metastasized to brain. That was a drug that was being developed within a tiny company called Cascadian, and that company was chronically underfunded, didn't have money for CMC, didn't have money to really develop it. When I bought that company for Seagen, you know, the comments were, "You're crazy." You know, it's an important drug that helps women with brain mets, secondary to breast cancer, and has a survival advantage. So I saw a lot of the same things here, which was a fantastic piece of chemistry, a great medical product. With Tukysa, there was already two HER2 tyrosine kinases on the market, but both of them were problems because they both caused a lot of GI toxicity, and they bound to not only HER2, but they bound HER1. In this case, it's just a very different. It's not like a second generation or me too, it's a very different chemical agent with 250 times the potency and works quicker and works better. So just trying to make a difference in the life of patients. So that's why I went ahead and rescued it, if you will, from a small, underfunded company, and we're gonna do the best we can for cancer patients and bring it to the market. Ben, let me go back to you. I think you teed this point earlier in your remarks. So you're already—I mean, the hint that you gave is you're already moving compound now into a mono in first line, but—and you're doing combos as well, but you're gonna do a PDAC first line, I believe, first and second line, and also first line in, I'm sorry, first line in lung cancer, in a RAS specific lung cancer. And as you said, you're gonna have to do combo arms. But the concept of moving monotherapy into first line, what led physicians to adopt this protocol in their practice? Yeah, it's a great question. And again, you know, it's the five arms, right? So the two frontline combos in panc, monotherapy in melanoma, monotherapy in lung, and then the monotherapy in pancreatic. And I think you're asking specifically about the monotherapy arm i n pancreatic. And actually we- I was just trying to get ahead of the data. Yeah. Try to figure out how you're able to go into front line as mono? All right. Well, here, actually, let me just walk you through the data. I'm just kidding. But, you know, I think, that was actually a suggestion from our investigators. So we weren't originally planning to do that 'cause it's first line, and you have a standard of care that you either have to add on to or let go first. But what our investigators pointed out is that the frontline standard of care chemotherapies for pancreatic cancer are so toxic and so poorly tolerated that there's actually patients that just don't wanna do it. They just don't want to go through that for whatever. And Brett Hall, our Chief Scientific Officer, who's sitting in the audience, he, you know, he was involved in these conversations with the investigators, and they said. You know, it's not gonna be necessarily a ton of patients, but we have these patients that want a non-chemo frontline option in pancreatic cancer, and there isn't one for them. Can you at least make that a possibility for your trial? So, you know, there'll be an informed consent, of course, and but I think, given the feedback we received from multiple investigators, we're, you know, we're happy to have that as part of the trial. Are you complaining that he has a single agent for frontline? No. For the childhood. Very intrigued. I mean, that's pretty darn good. Thank you. Moving on to Sam, can you talk about the design of ELVN-001 phase I, and give us an update on enrollment on potential timelines for initial data? Yeah. So we, we've guided toward this year for initial data, which will be our phase I-A data, which includes significant backfill in the, you know, dose exposure cohorts that had evidence of activity. So as a reminder, this is CML, late line CML. There are lots of available therapies and as I noted, these patients, the vast majority of patients don't ever develop progressive disease, so they always have other options. That being said, what we're looking for in our phase I-A data set, are data that are clearly superior to the other active site TKIs. The placement, the positioning, it really is gonna be in this post-Scemblix, so post-allosteric inhibitor context, and we believe even in late line data, because the other active site TKIs are known to have almost no meaningful response rate in these patients. We should be able to show both meaningful clinical efficacy in these late line patients, including patients that have had prior Scemblix, but also the directly comparable safety margins, so target coverage, tolerability, and safety, compared to second gen TKIs. Then, Mike, maybe just the last few minutes on the panel, can you just provide a little bit extra color on the next steps for the ON C206 program after the dose escalation work completes late this year? Do you think you would go into a pivotal controlled study with this agent, or do you think you'd do maybe like a smaller phase II study, similar to what you did with ON C201 first? Yeah. So, for the audience, ONC206 is a second generation compound to our lead, targets the same molecular pathways, but is 10 times more potent. We think has the potential to unlock sort of potential outside of CNS tumors. And so we're in a phase I/II, phase I dose escalation studies now. That's progressing well. We expect to have probably recommended phase II dose in the second half of this year. In terms of how we're thinking about development strategy for that compound, the work is ongoing, and so we'll likely update the market at the end of the year on that strategy. There are sort of two leading scenarios. One is, just as you described it Troy, the scenario where you leverage the clinical experience of the parent compound, not just in the setting that it's in, but also in other settings where it's been, and leverage the phase I experience of 206 to create a quick to market pathway, potentially, or quick to data pathway in a randomized study that maybe with an adaptive design, you could expand if you saw the signal you wanted to see. The other scenario is we're doing a lot of work on unique genetic signatures for that compound, pre-clinically. And if we went down that path, I can imagine a scenario where a smaller kind of single arm phase II, but we'll see where the data takes us. Those decisions haven't been made yet. First things first, we need to get to a recommended phase II dose, make sure we've got a safe dose and schedule to proceed, and we'll update the market at the end of the year. Okay. All right. I mean, I think with just a couple of minutes left, that we can probably leave it there. So thanks, everyone, for being here on our panel and for you all being here with us.