All right. Thanks everyone for joining for the next session of TD Cowen's Oncology Innovation Summit. I'm Marc Frahm from the biotech team here at TD Cowen. We're really pleased to have joining us for the next session, the CMO from Monte Rosa, Filip Janku, where we will be talking about the full range of their pipeline, which does go a bit beyond oncology. Given this is an oncology-focused event, we will spend probably a bit more time than is often done on the oncology assets. We'll lead off with those. I have a whole list of questions that I and my team have come up with to go through. We do want to make this interactive. For the investors on the line, if you have your own questions, certainly feel free to submit them either through the text box at the bottom of your portal or you can email me directly at marc.frahm@tdsecurities.com, and I will layer those questions into the list. With that, maybe to start off with, Filip, do you want to give a high-level overview that level-sets people on Monte Rosa's approach and what you view as the key value-creating milestones over the next year or so across the pipeline? Yeah, that sounds great, Marc, and thanks for having me. Monte Rosa Therapeutics is focused on development of the molecular glue degraders. We are a pipeline company, so this is all we do, focusing on these therapies, which are small molecules, oral. We focus on the targets which are either undruggable or which are hard to drug. Our pipeline kind of spans over inflammation and immunity and oncology as well. If I start with oncology, since this is the topic of this meeting, we have a GSPT1 program, MRT-2359, which is the oral degrader of GSPT1. That was our first program, which went to the clinic several years ago, which at the moment is expecting the transition to the initiation of the phase II in the castration-resistant prostate cancer with AR mutation after a quite interesting signal, which we have seen in the expansion cohort of our early phase trial. We also expect another event in the oncology space by the end of this year, which is in the second half of this year. We are guiding towards the IND for cyclin E1 molecular glue degrader, which is a part of our cyclin E1 CDK2 program. In addition to that, as I mentioned, we have quite a lot going on in the inflammation and immunity. Our VAV1 degrader, which is the first in class, MRT-6160, that's been licensed to Novartis, and the expectation is multiple phase II trial initiation this year as it's depicted in our guidance. Also we have a NEK7 degrader, which targets the NLRP3 inflammasome and can be the attractive option for these disorders. We have a molecule in the clinic, which is called MRT-8102. We had some data from the proof of concept CRP study already shared several months ago in January. We also have a next-generation program for a different molecule in the same space, hence NEK7 degrader. We obviously have multiple targets, which we are looking at, which we haven't disclosed yet, which we look at ourselves or through our partnerships with Roche and Novartis, which focus on I&I, oncology, and genetic and neurological diseases. All right. Well, thanks for that review, Filip. We'll start with GSPT1, since it's the oncology space where we do have some clinical data. What do you view as the key questions that 2359 in that phase I data set that we already have has already answered, versus prior GSPT1 programs? What do you think we can learn in this next update, particularly in that AR mutant population? I start with your first question, which, how does it kind of fit into the prior efforts which were done in the GSPT1 space. There haven't been that many, but there were a few. Most prominent one was probably from BMS, and that program, but not only that one, but focused on hematological malignancies. In that case, actually, it was the leukemia. Our approach was very different, right? Our GSPT1 degrader was primarily developed for solid tumors, meant to be this targeted therapy with the precision medicine or the targeted therapy angle. What we have seen in the expansion cohort in the phase II expansion of the early phase trial, which is the data which we lastly presented at ASCO GU, was quite encouraging signal of activity in heavily pretreated metastatic castration-resistant prostate cancer, especially in patients with the underlying AR mutations. Among the 15 evaluable patients which we presented, we identified five such patients, either by the genomic testing done by the site or by the genomic testing, which we have done as a part of the trial in the biopsies and the plasma. In these five patients with AR mutations, we found out that all of them had actually PSA responses, and two of these responses were PSA 90. That was clearly a very encouraging signal. I'm going to go to the next question, which you have, what we expect from the next step. The next step is the signal-confirming study. It's a phase II study, which is actually quite thoughtfully designed as a first and pre-planned step to confirm that signal, which we have seen in that expansion cohort. We will be focusing on enrolling the AR mutated patients only. Through the two stage design, we are hoping in this study of up to 25 patients to confirm that signal, and that will then guide the next step of development. Okay. For AR mutant patients, what is the standard of care for these types of patients that are enrolling that have maybe already seen AR inhibitors and maybe a few other therapies? At the moment, it's really none because there are no therapies with FDA approval which would be specifically targeting AR mutants. One thing which we can say that too many therapies, especially if they have a hormone component, the AR mutant patients have actually worse response, because they are inferiorly responsive or non-responsive to AR targeting agents. There is some experience with agents which were tested to some extent in early phases in AR mutant population. I would probably mention AR degraders or maybe some novel CYP11A1 inhibitors. That kind of set a little bit in expectations what kind of a signal you have to see. I think we are fairing really well. I think in heavily pretreated population, the PSA response rate above 40% is probably meaningful in this population. Above 50%, definitely quite exciting. At the moment, obviously it's a limited data set, five out of five patients, but we are running at 100% of the PSA response rate. Even if we got a little bit lucky and maybe the real number is a little bit lower, we would be, I believe, still fairing really well to what would be the benchmark data from AR degrader, so CYP 11A1. That's at the PSA 50 level? That's the PSA response, yes. Combining PSA 50 and PSA 90s. RECIST responses, the data are a little bit more murky. There are some response rates which were reported, which to me look maybe for at least some of these agents, which I mentioned as a benchmark, which kind of vary from numbers which are, let's say, around 30% or less, up to maybe like 40%. In general, the RECIST response rate runs kind of like maybe 10%, 15% below your PSA response rate. Yeah. Okay. Just in terms of the population that you're enrolling, how focused are you on getting RECIST evaluable patients? What percentage of patients do you think in this cohort are likely to actually be RECIST evaluable? That's a great question. We do focus on, in the early phase I or phase II expansion of the early phase study, we actually enrolled only RECIST evaluable patients. There were actually very good reasons for that, because we wanted to make sure that at the end of the day, the RECIST responses are important. Obviously, the PSA responses are helpful, but the RECIST responses are important, and you can actually believe that what you see is really real. That's why for this effort, in the data which we put out at ASCO GU, these were all RECIST measurable patients. I should mention that in these five patients with AR mutations, the two also actually showed RECIST responses, and the remaining three who are either on therapy, two of them are actually still on therapy, who didn't have a RECIST response at the time of presentation. We have actually seen at least some noticeable tumor shrinkage as presented in the ASCO GU data. When it comes to the subsequent signal confirming study, I think that we kind of cleared that hurdle a little bit, that we showed that it can shrink the tumor, not just modulate the PSA. In the signal confirming study in AR mutant population, we will not mandate for all the patients to have RECIST measurable disease. However, there is an expectation that there will be significant proportion of patients which will have RECIST measurable disease. It's kind of hard to figure out what that number exactly will be, but I think somewhere hovering around or below 50% is probably a reasonable expectation from the literature. I believe the phase II is designed to enroll up to 25 patients. How many of those patients do you think you'll have when you update data later this year? What are the criteria around deciding whether you actually go all the way to 25 patients? We haven't guided towards when we might have data from that phase II study. What we guided to is that in the Q3, we anticipate the study initiation. That's right. Yeah. Yeah. I would expect that the enrollment would go quickly. I would just kind of maybe point that the enrollment went fairly quickly in our expansion of the phase I study, even though it was a more early phase study, which included a lot of sites with more like a phase I expertise and not necessarily prostate focus. Obviously, the phase II study will be done by the centers which are specialized in the prostate cancer. We expect that the enrollment will go quickly. We haven't guided to the data disclosure, but it will be updated in the months to come. Okay. The next milestone is the initiation of the study. Yeah. Are there kind of interim checkpoints before you get to that full 25 or do you really just need to go all the way to the 25 to figure out whether the signals would come up? There are. As I mentioned before, it's a two stage design, you have to pass. The study will continue to enroll, if there is a signal at the level which would suggest that it's not meeting the expectation, the study would stop prematurely and wouldn't go all the way to 25 patients. I don't think it's very likely scenario, given the amount of activity which we had in the expansion cohort of our early phase study in the population, which was probably much less favorable compared to what you would get in the specialized prostate centers. I haven't necessarily mentioned that, it was a population which was really heavily pre-treated. Nearly 60% of patients had actually prior Pluvicto. Everybody had a measurable disease, that also kind of enriches you for the liver metastasis and the things which are clearly prognostically adverse. We had 27% of patients with liver metastasis in early phase, prior docetaxel in more than 80%, prior enzalutamide or similar in nearly 80% of patients. It was a population which was actually pretty beat up. Despite of that, we've seen this clear signal. I would hope that, unless we got exceptionally lucky, that the study should meet the efficacy criteria and progress to 35. Okay. If ultimately that signal is confirmed, what do the next steps look like? Is that something that Monte Rosa can do itself or often we see in prostate cancer that partners are needed to be brought in for the big late-stage trials. I'll start with the later point. I think what differentiates this program a little bit is that it would be developed as a targeted therapy in the molecularly defined population. I think these are typically areas, not in prostate cancer, because in prostate cancer, the tradition of targeted therapy hasn't been perhaps advancing as broadly as, let's say, in lung cancer. If there are some examples of success, it's often in the targeted, personalized therapy. I definitely think that we have a capability of developing it. The phase II study will guide the next stages of, or next approach to the development. We do have ideas and models how the development should look like from kind of like maybe faster routes to the label versus maybe more traditional routes to label, but a lot of that also actually depends on the ongoing discussions with the regulators. Great. Do you think you'll have that kind of lockdown when you are able to unveil data or is that likely to come another handful of months after the data? I think if you see the activity and the duration of activity, the levels which would be comparable to what we have seen in the phase II expansion, when I say comparable, I don't necessarily mean 100%, but if it's wherever between that, like a 50%, which I put for PSA response rate as an exciting benchmark. What we have seen, I think after 25 patients, in addition to the five which we already had, that would likely signal that this is actually something real, and you can plan for the plan which would be designed to take you in the gradual steps to the finish line. Okay. Obviously, 25 patients will not do it, but I think that will give you the confidence that you can get there. Yeah. Okay. Obviously, the very early stages of the phase I had a pretty broad enrollment criteria in terms of tumor types and histologies. Then, with this little bit of a signal seen in AR mutant, you've kind of focused there right now for the full proof concept. If that is confirmed in phase II, is the future of 2359 as a prostate drug for AR mutant patients only or are there other tumor types and histologies that you would then look to kind of more extensively explore once that proof of concept's in hand? The prostate AR mutation is definitely the most promising signal, but it's not the only level of activity which we have seen throughout the expansion cohort in prostate cancer. We have seen some very durable treatment effects, even in patients who didn't meet that characteristics. We also done quite a lot of work trying to understand the biology, why the AR mutation is a signal for GSPT1 degradation to be effective specifically in that population. We don't necessarily think that it's the AR mutation just by itself, but we think it's more the AR mutation is more like a surrogate of a scenario where the tumor is still significantly dependent on AR signaling. With that in mind, that would certainly offer other avenues for development, especially in earlier phases of prostate cancer treatment when this AR dependency is more pronounced than in patients who are otherwise pretty beat up and come for the phase I study, in which the AR mutation might be the good differentiator of that AR dependency. We have seen that overlap between the activity of the AR signature in the tumor biopsies and activity in pretty striking fashion even in tumor biopsies. That's what would be the logical step. Obviously, we do have the activity in patients who were heavily pre-treated, including Pluvicto. In fact, out of five AR mutant patients, four of them had prior Pluvicto. That's also pretty encouraging and looking at potential possibilities with radioligand, combining therapies can be another potential avenue of development, which can be built on this AR mutant population. Okay. Maybe we'll move to the next oncology program, which is your work, both against CDK2 and CCNE1, for the same underlying hypothesis. Just maybe, how does the approach to degrading either of the targets, but most prominently CCNE1, differ from or promise a different outcome than what has been seen with some of the prior CDK2 inhibitors that have gone into the clinic? There are certain elements which I think would be mutually true for degraders of both CDK2 and cyclin E1 compared to, say, CDK2 inhibitors. There are some differences which I think can separate within the glues CDK2 degrader from the cyclin E1 degrader. I'm going to start with the first one. One thing which CDK2 program seems to be universally running into is that these are clearly drugs which have to be combined. They have to be combined with either CDK4 or CDK 4/6 inhibitors and the hormone therapy. There is one thing which, even with all the advances in the chemistry, there is still this underlying fact that the CDK2 inhibitors bind to this ATP binding pocket, which is conserved across the entire family of the CDKs. If you have that always results in some limits of when it comes to selectivity, which ultimately do translate into adverse events. We do see it universally across most of these programs, which we have seen at least some data from. CDK2 degrader, molecular glue degrader, as well as cyclin E1 molecular glue degrader, for that matter, don't have that issue because they bind through completely different mechanisms, through the protein-to-protein interaction, via the forming the ternary complex with Cereblon and the target protein and glue. That's one thing which separates them clearly. The second thing is that when it comes to how we separate them internally, one from the other, because there is certainly some overlap which could have been seen. You can see an d we view cyclin E1 degrader in some way as a drug to be developed, maybe as a single agent in more personalized targeted therapy fashion in malignancies with cyclin E1 amplification. I think the biggest representative of that group would be ovarian cancer with a straightforward path, potentially to the finish line. When it comes to the CDK2 degrader, we've seen this one as a more kind of a breast drug, which would be combined with, again, CDK4 or CDK4/6 and the hormone therapy and developed in breast cancer. It doesn't necessarily mean that each of them cannot do the other as well, but we can obviously see this one as a logical distinction, and certainly looking at the science and biology behind it's kind of supportive of the distinction as well. Okay. I completely get that the improved selectivity in just degrading CDK2, not other members of the family, you avoid some of the toxicity that have been seen with the other members of the family. I guess that also means the promise of that is that you degrade, that you inhibit CDK2 more potently, and that could have efficacy benefits. What gives you confidence that that will still be safe and won't reveal additional CDK2 toxicities that are only evident when you go from 50% to 75% or 75% to 90% inhibition of CDK2? I think the game comes down to selectivity because arguably most of the AEs which are observed in that field are probably not really from CDK or well, there is no really cyclin E1, but are actually from the off-target effect. If you are selectively inhibiting CDK2, the AE potential should be really limited. Okay. You mentioned CCNE1 may have more single-agent potential. Do you plan on focusing the phase I on that opportunity right from the beginning or is that just more something to focus on as you get into formal expansion cohorts, phase II work? There is definitely a potential, and this is the way we are thinking about it, to focus on going after targeted population as soon as we can. There is one thing which we learned from our glues from doing it in the clinic for the last number of years, is that you often have a relatively broad pharmacodynamic index because due to the catalytic mechanism of action, you can achieve the expected level or the desired level of your target degradation at multiple doses. I think the way we are thinking about it, we would likely get into the targeted population very early. Okay. Particularly for that CCNE1 heavily amplified population in ovarian cancer, what do you view as a meaningful response rate durability that would justify continued single-agent development? These patients will be platinum or platinum failed. That's a patient population which can really hope for short-term responses with the chemotherapy agents, if anything at all. In the phase I, the responses north of 30%, 35% would definitely be interesting. Obviously, the higher you get, the more excited you get. The duration of responses in 6+ months, that would be definitely meaningful in that phase I population. That would be definitely better with anything else these patients can hope for, if they even can hope for anything. Okay. Then maybe in the last few minutes that we do have, maybe we'll move to NEK7 outside of oncology. As you mentioned, you did present some CRP degradation data and the effects of that on CRP earlier this year. There's also been a handful of similar stage updates from NLRP3 inhibitors, which are in the same pathway. Just what do you think the key learnings are that go across datasets, but also, are there any signs of differentiation yet between NEK7 and NLRP3 targeting that you think are coming out of those datasets? I think the CRP science is actually quite well understood. I think it's now probably well accepted that the more you inhibit CRP and the more patients you get below CRP levels of two or maybe even one, the better off you are compared in terms of cardiovascular risk. I think this is universally accepted. I think our 8102 definitely showed that in terms of CRP reduction, we have a potential which is comparable to IL-6 antibodies and definitely best in class in NEK7 NLRP3 space. What differentiates us from NLRP3 inhibitors, because there have been some others which have data which were on the CRP level, showing the similar ballpark of effects, some of them show effects which is a little bit worse, is that, again, the huge advantage of the NEK7 is the catalytic effect of the molecular glue, which allows you to get to the comparable levels of the CRP inhibition or reduction with the dose levels, which are, relatively speaking, low compared to, let's say, NLRP3 inhibitors. If you look at the NLRP3 inhibitors programs which recently released the data on CRP, often these doses are at the higher spectrum of the doses relative to what was tested in the multiple ascending doses in phase I, while our CRP data which were released were actually in the lower quintile or quartile of what was our maximum and perfectly safe dose level in the MAD. That's actually one thing which is an important differentiation that obviously results in multiple practicalities. We really keep saying it over and over, that's something which we are heavily focused on. We are focusing on the glues which are mono-selective. For inhibitors, especially if they are given at a much harder concentration, achieving that level of mono-selectivity and eliminating the risk of targets, it's somewhat harder and more difficult. Okay. Are there any signs in those datasets that you see that they are inhibiting other things that are causing issues or that it's just more of a, that's a risk when all of these programs go to much larger trials and start finding the 1% AE rates and the half a percent AE rates? There have definitely been reports of some AEs in NLRP3 field in general. When it comes to the earlier program in which we have relatively limited top-line results, it's a little bit harder to comment on. Going again back to our own program, what we put out, we were incredibly pleased with the safety profile which we have seen. We also have, and actually released that publicly, our preclinical GLP data, including the three month study, which shows no findings. We feel really good about that. Also the fact that we certainly can, the safety supports that we can use the higher doses, but it looks like that we absolutely don't have to, and maybe we can use the fraction of the doses, which would be a small fraction of the doses, which would be relative to doses which we explored in the multiple ascending doses study. Okay. You are doing this expansion cohort in ASCVD risk factor patients- That's correct, yeah. That's opening that. Just what will you learn in that small cohort beyond what we were already able to learn in the phase I? It's actually not that small. What we were initially planning was a one dose level with corresponding placebo, and a substantial part of it we released in our January disclosure. Because the data were actually looking so good, we decided to add two additional dose levels with a corresponding number of placebo. It's now actually a much larger study, which otherwise the kind of setup is the same. It's a 20 days dosing with the CRP as a main efficacy readout. What we are hoping to get out of it, I already kind of alluded to. We haven't guided what the doses were for these two new dose levels. I already guided the dose level of 40 mg did what we expected to do. It's probably no secret that we will be focusing on the lower dose levels, not for any safety reasons, but essentially having lower doses has many practical advantages, even for your later stages of development when you need to scale up everything. What we otherwise expect, we expect confirmation of the CRP signal, which we have already presented. We don't really expect any surprise there. What we also expect from that we will inevitably have to do a phase II study anyway, the CRP study, which will have at least three months administration. That would be the requirement to make the product phase III-ready. What we actually hope for, since our CRP effect is rapid and sustained, which is another differentiator from a lot of NLRP3 IL-1 data. If you look at them, often they see a little bit of a rebound after the initial maximum drop at the first week. We actually think that this data can help us to also understand some part of the dose exploration and do perhaps the upcoming phase II more efficiently. Okay. Over maybe late summer into the fall, we're expecting to get ZEUS data from Novo with their IL-6 antibody, I guess, which is in a somewhat overlapping pathway. Just how will you interpret that data? What does good data look like for Monte Rosa? Is there a bad data set from that for Monte Rosa? I will start with that the degrading NEK7 is very different approach than inhibiting IL-6 with ziltivekimab, and there are some important biological differences. Obviously, inhibiting IL-6 with the antibody is very effective to reduce the CRP, since IL-6 is directly involved in that CRP production and release. On the other hand, what the anti-cytokine therapies, and it's not only applicable to IL-6, it also applies to IL-1s for that matter. They don't affect some other upstream effects, which results in significant pro-inflammatory stimuli, such as pyroptosis. I wouldn't equal these two therapies together. Obviously, we wish ZEUS results go well because I think it will be good for the field, and it will further energize the field, which is already quite enthusiastic about these anti-inflammatory approaches. What a good data would look at, I suppose, meeting your efficacy endpoints and not having any safety issue. That would essentially mean that ziltivekimab has a path to label. On the other hand, I just pointed out that the mechanism of action is very distinct, so I wouldn't necessarily make the read-through in every possible aspect to our program. If you inhibit cytokines, whether it's IL-6 or IL-1, these antibodies do it very effectively. That also means that you might be getting rid of cytokines, which are not necessarily sourced through the NLRP3 axis, and which might be needed for the response against pathogens. That's something which NEK7 degrader clearly wouldn't have because you effectively would act just on everything which is NLRP3 driven. These are kind of important distinction, which is important to keep in mind. I don't know what the bad data, if it's bad data, what that bad data exactly would be, whether it would be like efficacy okay, safety not. I think there are an important distinction at all these levels. But definitely, g ood data would energize the field, no doubt about it. Okay. I know we're running over, let me squeeze in one last one on NEK7, which is, in addition to the cardio inflammation that we've been talking about, you've also disclosed plans to start trials in gout in HS. Why are those the most attractive kind of indications for that first wave of phase II trials versus a lot of other areas of inflammation that you could build a case for NEK7? I'll be quick a little bit. I'll start with gout. One of the things which we've been looking for at gout for quite some time when thinking about this program, and one thing which we really like about it is that if you look at in which biotechs and in which fields they were really successful, it often happens for biotechs in a space where you can innovate, when you can be the innovator rather than a follower of other people, let alone the big pharma. We definitely see that space for innovation in gout. The biology is matching perfectly with our mechanism of action. I think it's one of the most supportive biologies with the high level of NLRP3 activity. That's what we like about gout. We think that this can be developed very efficiently and really bring a change to the field, which I think it desperately needs change. When it comes to the HS, again, something which we haven't spoken publicly, but have been looking at for quite some time, and definitely the conviction in terms of when it comes to the IL-1 targeting therapies have been increasing over the last couple of years, starting with lutikizumab from AbbVie. Obviously, the LOTUS data from Avalo also helped to further solidify the relevance of the axis in HS. We definitely believe it's a relevant approach for HS. I think the crucial will be the execution to make sure that your efforts are not hampered by any way by some surprising placebo effects. We think with the right knowledge, right team, right KOL engagements, that we actually can deliver on that. Perfect. All right, with that, unfortunately, we're going to have to cut it off. Thanks a lot, Filip, for joining, as well as the investors on the line. Thanks a lot. Thanks for having me.