One of the biotech analysts here at Citi. This is the targeted oncology panel, always gonna be a very exciting topic. We have three uh excellent CEOs hereand from different companies around the country. From Shattuck Labs, Taylor Schreiber, CEO, welcome. Avanish Vellanki, Rain Therapeutics CEO, welcome. And Bill Lundberg, CEO of Merus. Welcome to all of you. There is a fourth person, Jacob Chacko. Hopefully he'll show up soon from work. Maybe not everybody is super familiar with what you're doing, you know, let's not go into a long discussion, but just maybe a 2-3-minute quick overview of your pipeline. What are your key assets? What are the key data points that we should be looking forward to in the next uh year or so? You wanna start? Go first. Um Taylor Schreiber, CEO of Shattuck Labs. What Shattuck is developing is a series of dual-sided Fc fusion proteins in immuno-oncology. We have two compounds that are currently in phase I development. Our lead compound, we call 172154. This is a molecule that simultaneously blocks CD47 on macrophages and activates CD40 on antigen-presenting cells. This is a um next-generation-type CD47 inhibitor. Uh we've just finished enrollment in the monotherapy portion of our dose escalation trial in ovarian cancer patients and have moved on to multiple combinations in ovarian cancer, one with liposomal doxorubicin, two with nivolumab later in the year. We're enrolling in AML and MDS trial right now as well. Those trials will all provide an interim read in the first half of next year. Second clinical program is SL-279252. This is a molecule that's a dual PD-L1 inhibitor and OX40 agonist at the end of a dose escalation trial uh right now as well. We'll leave it at that for now. Well, I appreciate the opportunity to call and thanks for the opportunity to share the story. Rain Therapeutics is a late clinical stage biotech, focused on our lead program, which is an oral small molecule inhibitor, MDM2, for the purpose of reactivation of p53, the guardian of the genome. MDM2 is a key regulator, many of you may know the story, regulator of p53. MDM2 levels can become enhanced through a variety of causes or mechanisms. This can enhance the removal of p53, this all-important protective anti-cancer transcription factor, and lead to an MDM2 different tumor type. MDM2, in addition, as a strategy, is nothing new. Big Pharma spent a lot of time and a lot of money on this mechanism, because ultimately it could target half of all cancers, half of all type p53. Unfortunately, what I think was not anticipated is some very specific, on target cytopenias that prevented the development of this category of drugs, thrombocytopenia, neutropenia, anemia. Therefore, the next generation of MDM2 inhibitors, version 2.0 or version 3.0, has really started to address recovery of the bone marrow and dosing strategy to overcome this problem. We're talking about dosing more like chemotherapy rather than as a TKI. We're not trying to create sustained target coverage above the IC50 or IC90 anymore. We're trying to hit the tumor hard above the Cmax, maximize Cmax, and then back off, allow the bone marrow to recover. Rain has identified a novel dosing schedule, which we believe runs to pretty exciting tolerability data. Our clinical data has shown uh rates of thrombocytopenia, neutropenia, anemia that seem to be far improved from historical rates in the MDM2 inhibitor landscape. Uh it led to a progression-free survival that is triple to quadruple that of historical rates of the standard of care and it also showed in the prior clinical study that patients with enhanced level of MDM2 gene copies had tumor shrinkage, tumor responses. That led us to kick off a phase III global randomized registration enabling study called the MANTRA study. We kicked off enrollment last summer. We completed enrollment well ahead of schedule this past summer. We have data, I think in the first half of next year. We also kicked off the MANTRA-2 study, which is an MDM2 amplified tumor-agnostic basket study late last year. We expect to have interim data from the first 10 patients for end of this year. We also announced plans to start the MANTRA-3 study in Merkel cell carcinoma and the MANTRA-4 study, which is our second tumor-agnostic basket study in patients that have lost a gene called CDKN2A, not CDK4/6, that's not CDK2. CDKN2A is novel biology, and both those latter two studies will start before the end of this year. $100, a little over $100 million in cash to provide our runway into the middle of 2024. That allows us to complete enrollment for all four studies. In conclusion, I think we remain very hopeful that this dosing strategy helps reinvigorate this MDM2 inhibitor debate. Hopefully, there's an avenue to address uh those half of cancer patients with wild-type p53. Thanks, Gal. Appreciate the opportunity, the support from your team and the opportunity to present here today and for the meetings today at the conference. Merus is a company focused on developing bispecific and multispecific antibodies for cancer based on a technology called the common light chain. This allows us to make essentially monoclonal antibodies, except they bind to multiple different things. Technology platform has enabled us to develop multiple clinical stage assets. Zeno, our most advanced, is headed towards registration filing, hopefully for approval. Our next molecule, Peto, has shown early but very intriguing data in head and neck cancer. Our next program, MCLA-129, targets both EGFR and c-Met as a bispecific. We'll be reporting out initial clinical data on that program in the second half of this year. Our last program, MCLA-145, is a bispecific T-cell engager. We also have a strong balance sheet with $396.8 million in cash that will take us well through a number of value inflection points for this program and to continue to bring the platform forward. There you go. Thanks for having us. I'm sorry that I snuck in late here at the end. Came just in time. My name is Jacob Chacko. I'm the CEO of ORIC Pharmaceuticals. ORIC stands for Overcoming Resistance in Cancer. That, in a nutshell, is the mission of the company. Our pipeline has been assembled through both internal discovery efforts and capabilities, but also paired up with some opportunistic business development over the years. The unifying theme across the pipeline is precision oncology, which I think is why I was invited to be part of the panel today. Um in general, we tend to focus on certain cancer types, lung, prostate, and breast are highest on the list. The team that we've assembled at ORIC, in, well, gross oversimplification, but on the clinical side, has really hailed from Ignyta, which was where I was just before this. Obviously, at Ignyta, we developed a brain penetrant targeted therapy, and that it explains in part our focus on lung cancer. In addition to that, we have a research team, research leadership that hails from Genentech with lots of expertise in breast. Then we were founded ultimately by Charles Sawyers and Rich Heyman, with expertise in prostate cancer. That explains sort of the uh thematic areas of focus from uh a tumor type perspective. We have three different programs that are in the clinic right now. They all entered clinical studies this year. They're in dose escalation right now. There is a small molecule orally available inhibitor of CD73, where we are testing in multiple myeloma as a single agent. After that, there's an EGFR Exon 20 program that is brain penetrant, that we're testing in lung cancer, and then a PRC2 inhibitor that we're testing in prostate cancer. In general, we try to, or at least we pride ourselves on trying to find either a biological or a chemistry angle of expertise or differentiation that we're bringing to bear. For example, in the case of CD73, we're the only company that is testing CD73 as a single agent and the only one doing that in multiple myeloma. In case of EGFR Exon 20, for example, we found a brain penetrant molecule that we then licensed. That's obviously a space that is underserved today because the two approved agents and all the agents, most of the agents coming after it do not have brain penetrance. We'd like to try to find an angle like that to explore. Those three programs, like I mentioned, are in dose escalation studies now with first publicly available data sets in the first half of next year. Finished Q2 with $240 million of cash on the balance sheet. That provides us cash runway comfortably into the second half of 2024 and beyond, a number of value inflection points. Okay, great. We'll come back to more of the company specifics a little later in the discussion, but maybe just to get the broader discussion going in terms of target selection, how do you all think about picking targets? What are the criteria that you use to decide which targets are most interesting to go after and maybe which ones are not as interesting? What are the biggest challenges in developing uh targeted or precision oncology therapies? You wanna give it a That's a perfect question. I'm sort of an oddball on this panel 'cause we're the IO company. You know, in our case, we have because these molecules are multifunctional, where one side goes after an immune checkpoint, our thesis has been that we will have the ability to outperform a benchmark monospecific checkpoint inhibitory antibody. Um and so in our lead programs, one going after PD-L1, second going after CD47, these are, you know, for better or worse, the only two checkpoints that in our view have um some clinical validation today. CTLA-4 as well, we can't target that with a fusion protein like this. Um and so one side goes after a clinically validated target, and our platform is specifically engineered to also activate tumor necrosis factor receptors, things like CD40, 4-1BB, OX40. And these are a class of primary targets that cannot be maximally drugged with any antibody-based scaffold. That's what led us to building these two dual-sided fusion proteins. Um so in our case, it's really been predicated upon the desire to activate a TNF receptor and unlock that family of costimulatory receptors, and de-risking those first efforts by going after validated checkpoints. I'm gonna pass it off to- I'll jump in. I'll just and just to be brief, I think, a lot of the Chinese targets have been pretty well picked over, right? There's a reality, but there's probably a lot of room for improvement. There's an Exon 20 program here, and certainly with the commercial EGFR Exon 20 program is around 40% response rate. Without brain penetration, there is a differentiation. There's probably ways to pick targets, and you can find validated targets, but perhaps they fell short of expectations. We're seeing a lot of missed expectations in G12C space. So there may be some opportunities from interesting Med Chem to improve points there. In terms of novel targets, you'd probably want something that's available on NGS, identify these biomarkers, uh but it's getting harder. I'll say that the challenge for Merus is a little different because our starting point is that targeting multiple different proteins is the ability that we have with our platform. We really drive the question to what's the scientific hypothesis that we could potentially attack with a multi-specific or bispecific approach. An example is MCLA-129. This antibody that binds to c-Met and EGFR. Where c-Met is a major resistance pathway, feel like I'm in the RAS space now, major resistance pathway to EGFR therapies, and more recently, EGFR has been identified as a major resistance pathway to c-Met therapies. The idea of having a bispecific antibody or therapeutic that targets both simultaneously gives us the potential, and has been recently validated by J&J's Amivantamab, or Rybrevant, to really address more effective therapy in lung cancer. Yeah, I think the panelists have covered it pretty well, you all. What I'd add is just, you know, again, harkening back to my short intro comments, for us, for good targets, we're looking for an angle, and so is that a biological angle where we have a particular insight that others maybe haven't found yet? Is it a chemistry angle, you know, for example, to come up with something that's brain penetrant? Whereas, you know, as Avanish talked about in the Exon 20 space, so many are not brain penetrant. Is there a clinical angle? So is there a patient selection strategy where we can deal with a smaller subset of patients and therefore try to stack the deck in our favor in terms of the ultimate trial? Um but I think Avanish's point is a good one, which is one of the filters used to be, is it, is there white space there? Is it not competitive? That filter is a little bit out the window these days because I think it's nearly impossible to find something that, a target that hasn't already been found by somebody. When you have identified a target that's novel, that's new, that's unencumbered, where there's not a lot of competition, how much preclinical work do you need to do these days to feel good, to feel confident that you have a good shot going into phase I? Has that changed over the years in terms of what you need to have in terms of preclinical POC to feel good that you're ready for a clinical study? Or maybe a few comments on that. Sure. I mean, I can start. I think it has changed. I think the bar is raised tremendously. I think in our case, for example, with CD73, that's an area where everybody knows the adenosine pathway. It's been talked about for years. Everyone knows the three nodes you can go after, and it's been almost just accepted conventional wisdom that for CD73, you're gonna combine it with IO and that's what everyone is doing, big combo studies with an IO agent in solid tumors. We asked ourselves, you know, "Do we wanna be the twelfth first player doing that or find a different angle?" In this case, we got connected through one of our investors to Ken Anderson at Dana-Farber, and he had a proprietary view that he had developed that he hadn't published on yet about CD73 in multiple myeloma. We worked with him for a year developing the preclinical package with our CD73 inhibitors in his ex vivo assays. Of course, the validation of someone who had seen as much as he had seen in that space helped us get comfort with it. I'd say to your point, Yuval, anything less than that, we probably wouldn't have felt comfortable moving it forward in the current environment. I just think that the current environment is too discerning and it's probably not a good use of resources to push something forward without the level of evidence that you see for that. Anyone else wanna add to that? Yeah. I'll add, I think I have a slightly different take. The preclinical data is often necessary but not sufficient, right? I mean, you need to have sufficient preclinical data that you validate your hypothesis. Predictive value of preclinical data for saying, "Boy, this is definitely gonna work in the clinic," is exceptionally challenging, even more so in areas like immuno-oncology. Outside of an antecedent drug saying, "This pathway, you block it, you can really get a response," like, you know, MCLA-129 or Rybrevant or a genetically defined pathway where you have a higher chance of success, it's really difficult to have confidence short of the first, you know, 20, 30 patients in the clinic. Yeah. I completely agree with Bill on all of those points. We view preclinical data as a way to establish your level of confidence in the mechanisms you're approaching, and to a certain degree, the potential benefit, when you run studies against, you know, whatever the appropriate benchmark monotherapy and combination controls might be. You simply cannot extrapolate that data to expected responses in humans at all. I think I'm violently agreeing with both of you guys in the sense that I would argue that three years ago, you didn't have to have even the level of preclinical evidence that we have today. That was the point I was making is that I think 3, 4, 5 years ago, people would just push things into the clinic with a couple of preclinical experiments, and today that's absolutely not enough. That's right. I guess there's increasingly more discomfort with the predictive capacity of preclinical data, right? I think we know better now internally at Rain. Our CSO constantly mentions that one cell line is one patient. You don't run a trial based on one patient's level of confidence. You need multiple level confirmatory models to predict or derive from multiple patients to get that conviction. Yeah, I think we need to be more robust than ever. Let's switch gears. Let's talk a little bit about regulatory landscape for targeted therapies. In terms of the accelerated approval pathway versus more of the survival outcome endpoints, just talk about whether you think there's an evolution at the FDA in terms of moving away from some of the surrogate endpoints towards more of the outcome-based endpoints for oncology. Or if you disagree with that, then that's fine too. Just curious. Can I jump in and be a little provocative? Go ahead. For the sake of having interesting conversations. In terms of what the FDA is going to do, I don't think surrogate endpoints are going to go away. Okay. I think there's something to be said for in an era when there's more drugs in clinical development today than ever before, more drugs of one specific mechanism in development, at the clinical stage than ever before. We talked about this before. How do you truly show a survival benefit when any patient on a control arm is going to ultimately get that mechanism? In subsequent clinical trials, it's going to be exposed to it. How do you actually have conviction that overall survival is going to be more predictive, equivalently predicted today? I would say overall survival is going to be tougher to get conviction around today more than ever. But to answer your question, I don't. I'll just check in on whether the FDA is going to believe in OS more than. Yeah, I'm happy to provide our thoughts on this. Classically, the FDA will approve drug based on clinical benefit, which is how a patient feels, functions, and survives. The accelerated approval notion is a surrogate endpoint like response rate with some mention of durability that is predictive of or correlative of a true clinical benefit endpoint. The noise from the FDA has really been around, let's not use an accelerated pathway to approve the tenth or eleventh or twelfth PD-1, PD-L1 drug. Let's really invest in innovation. I think as long as we're bringing forward innovative opportunities, whether it's, you know, in resistance mechanisms or, therapies in, genetically defined subset cancer like NRG1 fusions or other opportunities, the FDA continues to be really open to that and interested in bringing forward innovation. They just don't want to waste their time or the global resources around the 10th, 11th, 12th PD-1, PD-L1, is my sense. I think there's a recent example of what happened with Belinostat and MDS, where you know there was an exciting phase II composite CR, CRi rate that then ultimately failed on OS. That certainly has affected the way that FDA was viewing Magrolimab in a similar indication, which was originally hoping for an accelerated approval based on some composite CR, CRi endpoint. You know in that case I think there's some degree of confidence that if you look at just the true CRs and not the CRs plus the CRis there may be greater predictive value with OS. I think the potential is out there that that is potential surrogate endpoint. The strength of the data in terms of establishing that predictive value of OS isn't yet there. I think that's a fair position for them to take. Another topic, didn't put it in the questions, but I think it's worth going over, is Project Optimus. The idea that we're going to move away or the FDA wants to move away from MTD, maximum tolerated dose, and move to, I think they said, minimally efficacious dose or something along those lines. How is that factoring into your thought process regarding your trial designs and/or is it, Cat? Because it's relatively new guidance. Thoughts? Take that off. For future development, absolutely. For ongoing development and for future studies, certainly. You know, it's an interesting period of limbo that we're in right now, given when the conversations around Optimus started. I think from here on out, everyone's gonna certainly respect the opinions of the agency in an end of phase II study to look at any sort of optimal dose that could be relevant. You know, I guess the other side of it, just to make an argument, is something else we talked about internally is physicians have for a very long time learned how to optimize therapy for patients. That's been standard clinical practice. I mean, the doctors here on the panel. Certainly it does make actual sense to optimize, identify the minimally effective dose, getting away from MTD. Physicians have obviously always gone down to identify and personalize therapy for patients for quite some time. You're absolutely right. I mean, I've seen it with the KRAS. Most of the oncologists don't start at Lumakras' labeled dose. They start 25% lower just as a matter of habit. Yeah. Yeah. All right. Yeah, no, I totally agree. I'd say that Project Optimus has not changed anything about what we do internally, because both at ORIC and at the prior company where much of our clinical team came from, we were looking at PKs and exposures and PD for target engagement and various, you know, trade-offs. You know, Dr. Bill in the audience was one of our main PIs at Ignyta in looking at the trade-offs of efficacy and safety and trying to pick the right dose, not just push it to MTD. Nothing changes about how we do it internally. I think what does change, absolutely, I agree with Avanish, is that there's a lot of ambiguity about trying to guess what the FDA is thinking. From our perspective of how much data is enough to prove to them that you've picked the right, you know, the right one of two doses. 'Cause the reality is, you know, you can always gather more data at two different doses and try to figure it out. How you know, what's the equation for the trade-off of safety versus efficacy? I mean, it's not as linear or as clear as I think FDA's Project Optimus documents make it sound. I'm not even sure they necessarily know what they're looking for at this point. I think that's the part of the guessing game of how much is enough that we're kind of going through internally right now. One thing we may see, though, is the ask- We as an industry, not just, you know, Mirati as a company, is the ask from the FDA to actually explore a different dose with a greater number of patients in the clinical setting, rather than defining the recommended patient dose based on simply dose escalation plus a lot of modeling. Yeah. You may see the ask for more clinical data. Yeah. You know, within IO, many of these agents never find an MTD, and so it's not a matter of knowing what's toxic, stepping back a little bit, and then solving for, you know, the tail of the PK curve. And they're active across a dose range, and there might be incremental safety benefits of exploring a slightly lower dose than when you know you're sort of 2x saturation of your target. There's also in the dosing, the dose level itself is always what gets the most attention, but there's also the dosing schedule. You know, you look at a patient that has progressed on Pembrolizumab, and they still have fully saturating therapeutic levels of Pembrolizumab in their blood 6-9 months after their last dose, right? Do you really need to give a drug like that once every three weeks or once every four weeks? Probably not. Yeah, I know that's a great question for the Mirati and Amgen people. That's another discussion. One more kind of broad-based question, companion diagnostics. I think the relevance of that might be a bit different for all four of you, but what do you think? What do you think about companion diagnostics in terms of your development strategy and designing those for your trials and for commercial? We've got two different strategies to identify the right genetic population. I think it runs the gamut of what they can be. The first population is a population where there is no companion diagnostic. We don't expect it to require a companion diagnostic because all patients have the genetic profile we're looking for. Liposarcoma, our lead indication, you know, 100% of patients are MDM2 amplified wild-type p53. Find the tumor type that where you have the near ubiquitous gene profile you're looking for. The second trial, of course, is selecting MDM2 amplification across a range of tumors and critically important for us. We're a bit on both sides of it, of having to, if we see success in that tumor-agnostic strategy, we're gonna need to ultimately develop a companion diagnostic. We set up the relationships so we can jump on that if the signals support it. Our lead drug Zeno is for patients with NRG1 fusion cancers, and that's a molecular change in the cell. The patient population is defined not by histologic subtype, but by this molecular change. There is a companion diagnostic that's required or needed to identify that patient population. Look, it's the way our field is going. We're moving away from the old pathologist looks at a slide and said, "Oh, this is a breast cancer, or this is a lung cancer," and towards these molecularly defined cancers that often require companion diagnostics. There are currently around 50 drugs in oncology alone that are used with or require a companion diagnostic. Our view is just that it's critical that patients get comprehensive testing up front with the entire panel that will address the potential therapies that patients could need. I'll also add that many of the older DNA technologies don't always capture many of the newer changes, the newer molecular changes that need to be identified for certain therapies. The NRG1 fusion cancers are best captured by RNA-based assays. I think the field both is and needs to evolve towards having available comprehensive molecular testing for drugs like MDM2 inhibitors and others. Okay. In the last 10 minutes, I promised I'd try to ask some company-specific questions too. Why don't we do that? Maybe Jacob, starting with you. On ORIC-114, that's your EGFR/HER2 exon 20 inhibitor. That's very interesting because from what I understand, it's got a really, really clean kinome tree, which apparently sets it apart from a lot of the competitors in that space. Tell us a little bit more about the design of that molecule and how you got it to have such a clean profile. Yeah. To be clear, we at ORIC didn't get it to have the clean profile. It was our partners at Voronoi that developed it originally. What caught our attention was they had obviously prioritized selectivity and potency, but more than either of those characteristics actually, what was needed in the space was brain penetrance. That was what really caught our attention. We looked at the data that Voronoi generated, brought the molecule in under MTA into ORIC, and again, validated all of that ourselves. In the Exon 20 space in particular, I think, people sort of get wrapped around the axle on biochemical and cellular assays and trying to mix and match what I call the Frankenstein ratios of different kinds of assays to sort of make the preclinical data say what they want in terms of selectivity windows. Ultimately, what we got comfortable with beyond all that was honestly, you know, to the earlier point that the panel's made, as close as you can get was on in vivo models in mice. Can you see regressions and can you do it with relatively little tox? Can you see evidence of shrinking brain lesions in mice, and can you do that with very little tox? Those are the things that got us comfortable that this molecule was actually more differentiated than others that were in the space today. On the brain penetrance angle, and this speaks to something I mentioned earlier, I mean, brain penetrance is now, I think, widely accepted in the lung cancer oncology community as you've got to have it in your molecule for it to be a differentiated molecule. We were just at a lung oncologist conference two months ago, where that's all they talked about with respect to KRAS, ROS, ALK, all the major targets. I'd say 5, 6 years ago, that wasn't the case. We were at Ignyta, I think, trying to make the case for brain penetrance with the molecule that we were developing at the time. There was concerns at that time about the potential toxicities associated with brain penetrance. I think it's all shifted quite a bit, Ugur. That was the key angle for us in seeing the validation of the brain penetrance that got us excited about it. What is the next data point for that program? Yeah. The next data point, I sound like a broken record for each of the three programs 'cause they all initiated studies first half of this year. They are all in three plus three standard dose escalation for the remainder of this year. We anticipate that by first half next year, we'll have the recommended phase II dose for all three of the programs, and first half next year is when we'll have the first publicly available data for all three programs. Bill, turning to you, there's a lot of interest from my clients on your EGFR c-MET, MCLA-129. I think just yesterday, you announced you're gonna have the data, the initial phase I at the SITC meeting, soon, later in the fall. People are making comparisons to the Janssen asset and Amivantamab. Tell us a little bit more about how your molecule differentiates in terms of protein engineering from the Janssen one. You know, to the extent that you can comment, I know you can't give us details, but what should we expect from the phase I update for later in the year from the SITC meeting? We're excited about sharing data on MCLA-129, the dose escalation portion up to the recommended phase II dose at the SITC meeting later this year. 129, just like Rybrevant or Amivantamab, is a fully human IgG1 antibody. It binds to two different targets, c-Met and EGFR, and they are quite similar. They both block ligand binding, and they have generally similar affinities, with the exception of our molecule has more similar affinities, which is a characteristic you want to generally have in your bispecific antibodies. The development of our antibody was through empirical screening through hundreds or even thousands of molecules to find the right one that behaved optimally in a wide range of biological assays. Sort of empirically defined in that way to discover the right molecule to bring into the clinic. One other characteristic that's important is that our molecule is fully ADCC enhanced, which may give it even greater potency or the potential for greater potency in the clinic, so we're excited about sharing that. In terms of the data at the SITC meeting, we will be reading out the dose escalation portion. Again, also a standard 3-by-3 dose escalation trial up to and including the recommended phase II dose, the safety profile to date, the PK/PD characteristics and justification for phase II dose, and then any early evidence of activity or any dose of response that we have through that data set. We're looking forward to sharing that. Avanish, for you have a critical value inflection point coming, I think, in the first half of next year for the phase II MANTRA trial in liposarcoma. Give us a sense as to what you see as the hurdle for success there in terms of PFS, which I believe is the primary endpoint. I also understand that trial enrolled ahead of schedule. Give us a flavor for the reasons for enrolling ahead of schedule. Last, you know, what's the read-through from a positive MANTRA to your other programs for Milademetan? Sure. The first question, hurdle for success in liposarcoma, the prior clinical study that got us excited and jazzed up about the opportunity for the phase III was a 7.5-month median PFS in these MDM2 amplified subset of liposarcoma, 7.5 months. Standard care is around two months. All right? Tripling and quadrupling. We designed a study assuming it has a ratio of 0.5, a doubling of PFS, although it's overpowered, so we can certainly get a stat sig less than a doubling. We think that there's a lot of room for error in terms of success there. We assumed control arm PFS of three months. We assumed experimental arm PFS of six months, which is actually being more conservative than the actual data variability shown. I think there's a lot of room for potential positive upside from the phase III, but the hurdle for success, anything perhaps above 4.5 months may be clinically meaningful to clinicians. In terms of the read-through to other studies, we're pursuing a strategy that looks at MDM2 dependence broadly. You know, our lead indication, liposarcoma, all MDM2-amplified and wild-type p53. Second indication, we're selecting for MDM2 amplification at a high level of gene copy number, all p53. Our third is an MDM2-overexpressed indication Merkel cell, and fourth is a novel biomarker strategy. We're looking at all MDM2-dependent tumors. We would expect that if MDM2 enhancement amplification overexpression is oncogenic in one, there could be reason to believe that it's gonna be oncogenic in multiple tumor types. We hope to show that we can, through a tolerable dose regimen, impact that and restore p53. Great. In our final few minutes, in terms of the CD47 landscape, obviously we've seen some changes there, some exits, but your molecule is different. It's a CD47, CD40, so there's some special engineering there. Can you talk to us about what the differentiation is? Then you have an update in the first half of next year, I believe, for three of your programs. What do you see as the bar for efficacy for those studies? Sure. Yeah. This is the first differentiated CD47 inhibitor from a mechanistic standpoint in the clinic. The CD47 axis is broadly referred to as the don't eat me axis because when a tumor cell expresses CD47, it essentially wards off macrophages. The mono-specific antibodies will block that signal and enable macrophages to gobble up tumor cells if they are appropriately decorated with a pro-phagocytic signal. That's all it does. There's nothing about those drugs that tell the macrophages what to then do with the tumor cell fragments that they have consumed. All of the antitumor activity of the axis actually relies upon not just macrophage consumption of tumor cells, but then macrophage-mediated display of those tumor antigens to a T cell. CD40 is the most abundantly expressed costimulatory molecule on macrophages and other antigen-presenting cells. By stimulating that axis on the very same macrophage that is consuming a tumor cell, you dramatically amplify the ability of that macrophage to display antigens and at least acknowledging all the comments I made previously in appropriately controlled preclinical studies that translate into real benefit. We're hopeful that will be the case. In terms of what we will share next year, we will share the balance of the data from the monotherapy dose escalation trial in platinum-resistant ovarian cancer patients. Recently noted that we dose escalated through the top dose of 10 mg per kg, which was a really important dose for us, knowing that all prior CD40 agonists haven't been able to dose escalate beyond 0.3 mg per kg without running into dose-limiting liver tox and cytokine release syndrome. I think we can maximally drug the pathway. We'll also include the initial data from the combination in platinum-resistant ovarian cancer patients with liposomal doxorubicin. It's a chemotherapy that stimulates immunogenic cell death in ovarian cancer cells and has a very low monotherapy signal of 10%. Demonstrating the contribution of components with SL-172154 we don't think will be an enormous lift. We're looking for about a 30% overall response rate in that group. In AML and MDS, we'll share data, also from the initial, monotherapy run-ins, as well as data Azacitidine combinations. In the dose escalation part of that trial, it'll be primarily relapsed refractory patients. In that patient subset, any responses are actually a positive signal. If you saw something anywhere close to 20%, that would be a very positive signal. We're really looking forward to that in the first half of the year. Perfect timing. Thank you all very, very much. Much appreciated, and good luck.