Okay, great. Thanks everybody for joining us here on day one of Piper Sandler's annual healthcare conference. I'm Joseph Catanzaro, one of the biotech analysts at Piper. It's my pleasure to welcome, for this session, Biomea Fusion. Joining us is their COO, Ramses Erdtmann, and their Chief Development Officer, Steve Morris. Ramses, Steve, thanks so much for joining us. Maybe before we kick off into some Q&A, Ramses, I could give you a minute or two. You could give a bit of a quick intro on Biomea, what you guys have been up to, what we have to look forward to. Okay. Thank you very much for having us. It's very nice to be here. And Joseph, you were at the IPO in 2021, so you know the story really well. The company has evolved, obviously, since then. We were actually originally, as you will remember, an oncology-focused company. And we found that the target that we were looking at is a protein called Menin, which is a scaffold protein that was a little bit underserviced by the industry and academia. There were a few papers you could find, but as we got deeper involved into Menin and found a way to selectively inhibit it with a covalent binding compound, we saw the safety profile was actually very well tolerated. It was just relatively benign. We embarked upon a journey of trying to address Menin in diabetes, which has never been done before. It was only done in animals, and so we went from animals to human cells, and we showed the same effect there that you could control diabetes. As you inhibit Menin, the body makes beta cells, and these beta cells assume a function of producing insulin, and that insulin downregulates the sugar in the body, and when we saw that in animals, we knew it could be done. We saw it in cells. We were really excited to see the proliferative effect. We then went to humans, and we went first at a, the regular steps, so you do single ascending dose, multiple ascending dose studies. We finished those. We showed that data at ADA, and we could see that there is a subset of patients that are doing really well, and then others were doing okay, but we're still trying to find the dose. So we couldn't really tell, is it the dose? Am I at the right dose yet? And so we went through the dosing exercise, and we have now published a couple almost like a week or two ago. We published data that was telling the full story, at least from an early stage, that diabetes, as I think everybody who's in the field will agree, is a very, very heterogeneous disease. Patients come and have diabetes that are overweight and obese overweight, and patients like you and me who are just not really overweight, and they show up with diabetes. And so it becomes more of a precision exercise to identify what is the right treatment for each of these patients. The ones that are more like the phenotype of being slender, and there are about 50, in America it's about 55% of the patients are fitting more into the insulin-deficient phenotype, which are more slender, demanding for more beta cells, right, to be produced, versus the heavy one, which are more insulin resistant. We have now shown that if you inhibit Menin in both of them, you have a profound effect. Yet in the ones that are insulin deficient, you have about two and a half times more the effect. This is small n's, we know. For the readout that is coming up, we finished a phase two study. We are now showing in this larger patient pool of over 200 patients, we categorize them by this phenotypic approach where you say, "These guys are more insulin deficient. These guys are more insulin resistant. We categorize them, and in these groups, we can now show, based on BMI and other characteristics, how are you doing. And that is important because it allows us to go to the FDA and say, "Look, I can tell you which is my best responder, and I can go into late-stage development. Great. So you mentioned the COVALENT-111 readout that we'll get 200-plus patients in a couple of weeks. I maybe want to start off by going back to the data we've seen thus far and what you think sort of we know about icovamenib, then its mechanism of action and what it's doing in the context of type 2 diabetes, and maybe what we don't know that maybe COVALENT-111 may help inform. Yeah, that's—it's an open question, which is very helpful because it sort of is an important question. What we do know is we know the safety profile from about 400 patients. We've done oncology work with the drug. We've done diabetes work with the drug. We've done healthy volunteers. And we know it has a very, very good safety profile. So because we can see it in all these different disease settings, what is related, what is unrelated. So we do know the safety profile. The FDA just finished a review, as you guys may have known, where they put us on hold. They said, "Hey, wait a second. What are you guys doing there? And you're 267 patients deep into diabetes, and we see these liver transaminases going up. And why is that?" And so they put us on hold. We went through all patient narratives, and they basically said it wasn't a terrible exercise to go through. It was actually good because it validates the quality of the data. It validates the safety profile. And what they came out with is, "Continue the study. Don't just start at 200. Start at 100 and then go to 200." So we know starting at 200 causes transaminase rises that are transient for the most part, but that are somewhat of a phenomenon that we want to avoid because it's a grade that gives concern to the FDA. And we now have to figure out why that is. It could be on target because of the reduction of sugar that has to go somewhere. And so what we do know is we don't know that mechanism. We are thinking we have some ideas about it, and we're doing the background work on it. Okay. But what we do know is if you start at 100, it's just, it's a very safe compound. One. Two, what we do know is we have a dose response from 100 to 200 to 400 to 50. We understand roughly the PK profile and the response it gives me at these various dose levels, yet with a small end. What we do know is we're hitting Menin because we've seen the data that we showed at ASH last year, Menin, the Menin gene is downregulated through the work of our compound. So Menin is reduced. We do know roughly the extent of the reduction. It's about two-thirds. So we're not downregulating Menin to zero, which would be a problem. We know that as well. Okay. What do we not know, which isn't just as important? We do not know, and this is exactly why we're doing this study. We want to understand who is the target patient, and we want to understand in this target patient what is the right dosing that I should apply. And those are the two main things we're going to get from the readout. We will completely identify the target patient. Not completely, but we will get a very good understanding. And we will also understand, is it eight weeks? Is it 12 weeks? Is it 100? Is it 200? And that will be the outcome of this study. That's what we'll learn. So, I want to go to that point of the target population. So at ATTD Asia, a week or so ago, you had presented some of the subgroup analysis that you did within the context of the dose range and data set. Maybe speak a little bit to how you're defining these patients in terms of insulin deficient, insulin resistant. What are you looking at in baseline features, and the proportion of the patient population? Those breakout. Because the response of just like in pregnancy, where the body as a natural response grows the beta cell pool to overcome the resistance of pregnancy to provide sufficient insulin to the mother and the growing fetus, just we have the same compensatory effect in diabetes where an obese patient characterized. These are the five characteristics that we didn't identify, but that academia has sort of identified by 2016, 2017. You can see the first papers coming out. There've now been many, and thousands of citations were done to one specific paper. It's the Ahlqvist paper from Sweden. There's a study that was done in 2017, and they use criteria that everybody else has agreed on. It's age at onset, it's BMI, it's HbA1c, which HbA1c is a measure of sugar. BMI is obviously giving you characteristics of resistance. And then you have HOMA-IR, measurement of in insulin resistance, and HOMA-Beta, measure of how your beta cells are productive. And so with these five criteria, they cluster patients. And if you, you know, want to simplify it, you go by BMI because high BMI, insulin resistant, low BMI, more or less insulin. And this is nothing set in stone here, but this is kind of where you can get 90% of the patients clustered. And these studies are then done on Asian population. If you, for instance, go to the presentation that Juan gave at ATTD Asia, which Juan is our CMO, and he had a professor from Hong Kong there, and she provided background on the heterogeneity of the Asian population. And they, for instance, have obesity starting at BMI of 25, right? Which is kind of funny, right? For us, that's not even overweight. But for them, there's a different phenotype. And they also are trying to go and the whole effort is based on the idea of, "Can I give a precision medicine to a diabetic patient? Because should I get a GLP-1? Should I get an SGLT2?" And by the way, there are 60 approved agents, and all they do is try to figure out who is who. None of these agents address the underlying cause, which is what we're doing, the depleted pool of beta cells. They have not gotten to that yet. And so we believe 219 could be sort of a backbone to these therapies. But we will show you in the readout that's coming up how are we performing in these various subsets. Yeah. So I want to hone in on that a little bit in terms of the, the upcoming readout, 200-plus patients. Maybe just talk through how what your expectations are in terms of how you're going to present the data, right? There's three dosing cohorts. There's these patient subgroups that we're speaking to, how you think the, the sort of package is going to come together. Yeah, so we're looking at three dosing cohorts, one eight weeks, 12 weeks at 200 and 100 milligrams, so we're trying to understand so far we've looked at four weeks of dosing and then 22 weeks of nothing. And we really made sure these patients do not change anything. They just stay on background therapy. And then we looked at how is this patient doing off therapy, our therapy, while still maintaining whatever else he was doing before. It was a failing patient. They had HbA1c's above seven, so they were failing their background therapy. And we looked at them at week 26. Now what we're doing is we're saying, "Okay, we need to understand what does eight week do, what does 12 week do, and what does a dose of 100 versus 200 do for us." You will see in the readout that's coming up within the next couple of weeks you will see phenotypic these four clusters. We will give you top line on those with HbA1c, and we will give you a very detailed review of the safety profile. So that argument is finished for all of us, or that we actually know what is the safety profile in a large pool of patients. Then the rest of the data we will try to give to a larger conference and get it peer reviewed. So you get a top line from us before the end of the year. So in terms of the subgroups, is the strategy going to be to use the same definition that academia has used or use BMI as a sort of? We will do both so that you can see it's similar, but we will for sure follow what is the standard. And there's a computerized model you can get online, and you just plug in the numbers and there it goes so that it's clear that we're not trying to maneuver the data, so that it's just based on whatever academia is thinking. And then we apply something that would be used probably more in an FDA trial where who does HOMA-IR, right, in a practitioner's office down the street? They probably would use BMI, which is much simpler. And we will do that stratification as well. Do you think you, I guess I want to get at the level of confidence you'll have, you have that this data set will tell you what the right dose and what the right schedule is, right? If you're looking again at like subgroups, right? So then you're, you're going to have within a subgroup patients that are across different doses and durations, whether there's enough of a sample size to sort of inform like, "Hey, this is the dose and schedule that we, that we take forward here. I think there's just, there's never enough, right? And you can always address the N as a critical criteria and say, "Hey, you know, your N is insufficient because," and take that as another aspect, 75% of those patients. So there's three groups, 72 each. 75% of those 72 are frontline. And then you have patients that are more disease progressed. So you have those subsets in there. I would say the N is insufficient, of course, but it will be directional for us to say as a company, "Yes, we will go, let's say with a dosing cohort. We start at 100, go to 200. It's a 12-week cycle. Done." And we will see differences between 4 weeks and 12, that's for sure. And we can say, "Did what our preclinical data seems to indicate that with longer dosing duration, you get more mass? “Does that matter?” We will know that and we will know, is 100 versus 200 better, so I think we will know as a company, “Yes, 200 is better. Yes, 12 weeks is better,” and that's what we're going to do going forward. I think yes. What do you want to see within that sort of, you know, that subset of patients that you expect to be responders, you know, their week 26 placebo-adjusted A1c number? Like, where do you think the bar is to say this, this is a viable and meaningful therapy in the context of other therapies in type 2 diabetes? So number one that I heard from, because we talked to pharma as well, and pharma has a viewpoint, and that's an important viewpoint because they sell drugs. One of the aspects we heard when we first started this exercise that we are non-chronic. Non-chronic definitely is something that is novel in this field because all other agents out there, look at every one of them, they're chronic. The second aspect of difference is oral versus injectable, where oral, not all agents are oral. Actually, the fewer agents are oral. Half of the patients reject an injection, and that's by statistics. So now you're saying, "Okay, so you're oral, you're non-chronic. What are you, what are you competing with?" You're competing with nothing because nobody addresses the depleted pool of beta cells. You could argue in a prevention study, you're preventing diabetes, which we're not reading out on because we have disease patients in these studies, right? But from a viewpoint of prevention, would I not want to take an agent that replenishes the pool just to prevent the disease? Of course I would. So any benefit that shows or is indicative of a preventing aspect that the pool is growing under this drug influence that has occurred for a moment in time, and then the pool has grown, and you can, we know this pool is active and more active because we see the HbA1c going down, could that not be sufficient as a validating MOA? Yes. One. Two, I also want to know, or it's also validating for us to know that in this subset of insulin-depleted patients, we work better because that shows that the MOA is working. Those resistant patients shouldn't work better, actually. So there's these two aspects that are not answering your question, but they're for us good to know that yes, in a larger patient set, this was not a lucky strike that we had in the updosing, which we did and which we published on, but this is now in a larger patient set. We can prove the MOA is working, and the MOA is giving us a larger function of the beta cells. One. Too, now I'm competing in a relapse refractory setting, and I would like to compete, or we would like to compete as a company in the sort of Metformin failure that happens within the first four years of your disease indication. Metformin fails. It gives you a nice benefit, and then it stops. And typically the second agent comes in, and that second agent is an SGLT2 inhibitor. What do they do? The orals do 0.5%, 0.6%, 0.7% as a chronic therapy. So you can, if you get above 0.5%, you're meaningful. If you're below 0.5%, people will probably, "Are you really meaningful?" So I want to break that hurdle for sure with the readout in a large patient set. Do I want to hit the 1%? Of course I want to. Do I want to get, you know, I want to get to the ATTD data if I can. But I think if the MOA is validated and we can show that we, that we're doing what we're supposed to do, we may need to dose longer. We may need to dose again. All these things we'll find out over time, but the validation of the MOA is for me critical, and that probably happens at 05, but that's an opinion. I mean, people can have different opinions on this. Yeah. Fair, fair enough. I want to go back to safety. It sounds like, I mean, your level of confidence in the safety profile is pretty high. And I know you guys have been very clear that within COVALENT-111, COVALENT-112, that safety has been looked at. The FDA has okayed it. There shouldn't be anything in there. But I want to speak about like moving forward, like what you will do to minimize that risk that it doesn't rear its head again. Yeah. And that was really cool. And I have to say the experience that we had through going through this hold and going through all the reviews as a young company is obviously an unfortunate event, but it gives you a lot of strength coming out of it because you've gone through it. You've had interactions with your counterparts at the agency, which were actually quite good. I mean, I use the word pleasant because I really enjoyed these conversations. You learn a lot, and you bring in your best people. We had hepatologists, top of their game and from various institutions in the U.S. meeting up with their hepatologists. We agreed that the schedule we had is a good schedule, and we added here and there another review, and we added a review, a group that sort of assesses neutrally what we're doing so that, and we proposed that it was because we felt, "What can I do to improve the understanding, right?" It's not just us interpreting data, but others as well. We provided all that background, and with that, schedule of reviews that we've beefed up and the outside group, the FDA and we felt comfortable going into a larger patient setting, and they agreed to continue dosing with that. Just that alone, plus the understanding, which is kind of crucial that we don't go into a new patient with 200 or 400, but we start at 100 and then updose. You can see a lot of these drugs get titrated. They start low and then go higher. In our case, it's a little bit similar. There is an effect in the body, and this effect needs to be absorbed. So we start with 100, and then we go to 200. Currently, it's at eight weeks. Whether that changes is open, whether we can go to four or something. So then maybe, looking forward beyond COVALENT-111 in the readout, I know a lot is contingent on the data and what it tells you about the drug and the profile and the patients who are benefiting. But what do you envision as the potential next steps for the program thereafter? Where do you go with it, with type 2 diabetes? Once we know the patient's subtype or the characteristics or the phenotype of our patient that we'd like, and they understand the dosing, which is kind of what this readout will bring, we want to go into late stage development and talk to the agency and say, "Look, here's the data," and the phase two meeting, let's sit down and figure out what could be the next step for us. My eye is always on the mountain, and the mountain is get the first label. Why is that important? Because from there on, as you are marching towards it, you have a path towards, a finite outcome. You can always investigate more on the sides, and you should, and you should understand everything, but you don't want to get lost in that. When we had the experience at Pharmacyclics, and I don't want to draw on this because my role was not clinical operations, but the experience, what we learned at Pharmacyclics was once you have identified the working path and your path forward, you would bring in investigators, and these investigators were actually introducing all these potential avenues where your drug could be used. At one point, we had literally over 50 concurrent ISTs ongoing. I'm not saying that we're going to do it at Biomea, but it gave so much knowledge that we didn't have at the time, and it helped us really, you know, identify seven labels at the end. We will, we will march down this narrow path where we're going, which is not quite narrow because it's about over 50% of the patients, but we're marching down a path that is remarkable where we know this is a high responder, this is my target patient, and that's the goal. Then from there on, we will do all the other work. Great. So maybe in these last couple of minutes, we could talk about COVALENT-112, which will also have some data type 1 diabetes. Maybe first, any color on sort of sequencing of COVALENT-111 versus 112, and then maybe the, you know, any read-through that experience in type 2 provides to type 1 and just overall thoughts there in type 1. So, 112, at the time of the clinical hold, we hadn't quite finished the enrollment. We had finished much of it, but we hadn't quite finished the dosing. So it came unfortunate that today we have 20 patients. We will have not 12 weeks of dosing, but with 20 patients, about eight weeks of dosing. And we have less than a handful of patients that finished at 12 weeks. So the n is too small. We have to continue enrollment in that study, and we're going through the motions for that right now so that we will get to a better understanding with full completion of dosing what the effect is in type 1. In type 1, just as a background, you will go, if you go to the Joslin Diabetes Center there, they do beta cell work, and they identify their medalists, those people who have had type 1 diabetes for 50 years. They still see when they do a mixed meal tolerance test, there is still insulin production. Where is that insulin production coming from if your body has attacked all these beta cells? Well, it doesn't attack all of them. Some are not flagged for attack. Those are the cells we hope to proliferate, and we will learn that, and we will learn probably with additional patients how, with what dosing duration and with what drug concentration, because we've tried 100 and 200 here, and we've tried 0 to 3 years and 3 to 15 years, right? We need to understand, is it the early onset that gives us better responses, right? Can I take him, can I take somebody who has been, who has had type 1 diabetes for a longer period of time? That body of data is unfortunately not complete, and we will make it complete, obviously, next year, but for now, that readout will not be solid. Okay. Maybe in the last minute, you guys recently disclosed the oral GLP-1 program that you're working on. Maybe just speak to how you think that sort of fits in with your efforts in Menin and where those two programs maybe can kind of synergize together? Yeah. The data was, so in 2017, a professor at UPenn showed that Menin is repressing GLP-1 expression. When you read that, you think, "Wow, that's exciting." What happens if I take out Menin? Would I have more GLP-1 expression? Could a GLP-1 agonist work better under a Menin inhibitor primed beta cell? What we found is exactly that with the data that came out in October. We were able to improve in cells the insulin secretion of these cells in that environment. And then we said, "Wow, now that's exciting because that could be, we could be a backbone therapy with icovamenib to all the GLP-1s." What the problem of GLP-1 is, if you talk to our CMO, who's done thousands of patients in his center before, he, he says persistence is problem number one, keeping patients on drug. And if you go into, literature, you'll see in some cases it's seven out of ten patients who drop off in surveys, even when you pay for the drug on a GLP-1 therapy. So if we could bring in icovamenib as a backbone and allow the GLP-1 to be more expressed and therefore have more of an effect, could I use less drug? That would then lead to higher persistence. If that were the case, that'd be wonderful for all the patients who are on a GLP-1. There's a combination study that we're planning currently. It's called COVALENT-211. That's in the making. To answer your question, 650 is our GLP-1 receptor agonist that we developed in-house. We declared as a next generation, we wanted to have our own. We wanted to be combinable with icovamenib, and we wanted to address those patients that were not immediately targeted, the insulin resistant patients with a drug. We started this program about 18 months ago, and we want to be differentiated from the current GLP-1s. The idea is, increased protein binding, less PK variability, better safety profile, more therapeutic window. Those are sort of the intent of the 650 program. Okay. Perfect. With that, we're out of time. So, Ramses, Steve, want to thank you for your time. Looking forward to the data. Good. See you in a couple of weeks. And thanks everybody for joining. Take care. Thank you. Yeah. Thank you.