Our virtual conference. I'm Matt Biegler. I'm an analyst here at Oppenheimer. I'm with one of my covered companies, which is Biomea Fusion. Here to tell us more about the story is CEO Mick Hitchcock and Ramses Erdtmann, the COO. Guys, off to you.
Yeah, thanks a lot, Matt. Appreciate the opportunity to discuss this today. Ramses, can you flip on to the next slide, please?
Yep.
As usual, we have a standard legal disclaimer. We won't spend any time on that. Let's move on to a quick snapshot of Biomea Fusion and what we're doing. Here are the two development candidates that we are developing. Icovamenib is a small molecule, oral for diabetes. We recently derived some data from COVALENT-111, which was a study in diabetic patients, and are now moving forward into two clinical trials. The first one in insulin-deficient type 2, and the second one in the ones who are GLP-1 inadequately controlled. We also have a GLP-1 receptor agonist, BMF-650. It's being developed for weight loss. It's, again, it's oral and a small molecule, and we'll tell you a little bit about what's going on with that, with our GLP-1.
sorry, GLP-131 Phase I study. Next, please. As many of you know, diabetes is a big problem. About 1 in 4 dollars in the U.S. is currently being spent on diabetes. While there are many drugs, and 60+ is generally the number that people use, they obviously only work for a certain period of time, and then, you know, things return to the disease state, and people progress in their disease. 80% of the people with diabetes will die from the disease. Ultimately, many years of life are lost, and you can see the numbers there in terms of 38 million people in the U.S., 35 with type 2 diabetes, and 7 million are, in fact, the guys who are insulin-deficient that we're targeting. Next, please.
One of the good results that we got out of our one-on-one study, when we looked at these particular patients, the severe insulin-deficient diabetes patients, we saw that there was a pretty good reduction in A1C over time, and it's significantly improved compared with placebo over the same period of time. An important feature from this graph is to look at the left-hand green bar, and that is the treatment period. What we are doing here is we're giving patients three months of oral Icovamenib therapy, and then the therapy stops. You can see that the A1C keeps going down, not just for another three months, but another nine months post having delivered the drug. We're not sure yet how long that goes down for.
We don't have any further follow-up on these patients, we'll be looking for that in future clinical trials. The important thing is the mechanism. Why is it that it's still going down at three months, after three months? It's because we are producing insulin, the insulin increase is what's responsible for the A1C decrease. If we move into the next slide, our drug is a menin inhibitor, what it does is menin suppresses beta cell proliferation and function. When you inhibit menin, it's like you're taking off the biological break. We can do this with a short 12-week dosing period, because we have then created more cells or more insulin-producing capacity, that doesn't go away when you stop dosing.
You can have long-lasting effects from a short course of therapy. Next. What we're doing here is actually we are recapitulating a physiological process, which is seen in various states, in both things like pregnancy and lactation. You get suppression of menin, and that increases the beta cell pool, and that leads to increased insulin outlet output, which is why gestational diabetes does not occur in every pregnant woman. It's also something that occurs when people get overweight, the fact that Icovamenib is working through this natural mechanism, we think, makes it a pharmacological validated process. Next.
We also noticed with Icovamenib that when you put it onto islets, and these are isolated islets, you can see the experimental design here, it actually increases insulin production, and it does it by upregulating the GLP receptors. Having more GLP-1 receptors obviously improves the efficacy of the GLP-1 hormone that is floating around in the body. This enables you to have an effect on the GLP-1 activity as well. Next slide. We looked in our 111 study at patients who came into the study on a GLP-1, and in fact, they were maintained on the GLP-1, even though they were failing, i.e., they came in at an A1C above 7.5.
What we show here is that we get a similar type of effect as we did in the insulin deficient patients. The A1C goes down over the dosing period, but then it keeps going down, even though we're not dosing Icovamenib for that nine months post the last dose. You can see again, what we've shown here is the patients on the GLP-1s getting Icovamenib versus the ones on GLP-1s are getting a placebo. Next. Here again, the same thing. The rationale for this A1C decrease is increased insulin secretion, and the increased insulin secretion is measured by this particular paradigm called the C-peptide. C-peptide is a piece that's clipped off of insulin and stays in the blood for a longer period of time. It's a good measure of how much endogenous insulin is being produced. Next.
To expand the data that we have in these two particular patient populations, we're moving forward now with two studies. This is the first one called COVALENT-211. You can see here we're essentially repeating the experiment, but this time with a defined population. Here, what we're looking at is adult participants with type 2 diabetes who have already been treated with 1-3 anti-diabetic medications and are still not achieving their A1C target. We're taking patients who come in at either somewhere between 7.5 and 10.5 A1C, i.e., not controlled, and with a somewhat lower BMI, 32 or less.
What they do is they come in, and they get on Icovamenib or placebo, but they maintain their background therapy, then that way we can see what the effect of Icovamenib is over time. We have a six month A1C, which is the primary endpoint, and a 12-month, which is a secondary. The next slide shows you the 212 study, which is the other set of patients. These are the ones who are on a GLP-1, who are not achieving the target, and they come in, they stay on the GLP-1. Again, they have a higher A1C, 7.5-9.5, i.e., not controlled. Controlled is considered seven or below. These are the patients with the higher BMIs, 25 to 40.
Again, they stay on their background therapy so that we can see what the Icovamenib effect is. On the next slide, we jump to the GLP-1 receptor agonist, which is called BMF-650. There's what we've tried to do here is to improve the PK profile so that we get a more consistent response in patients who need this for a reduction of weight. We've shown a generally favorable safety profile, no LFT abnormalities in the animal studies, and we have designed this as an oral drug with a simplified dosing regimen. Let's move to the next slide. Here you see the monkey data. What we have here is five monkeys in each group that we've dosed for four weeks.
This is dosed orally, and you can see there's a reduction in food intake. What we're showing is a reduction in weight. This is very similar, as you can see, to the Carmot compound that was being developed by Roche. What we see here is at the two different doses, 10 and 30 mgs per kilo, we are seeing about a 12% or 15% reduction in weight over the 28 days study. Next. We've moved this now into a Phase I study. On the left-hand side is the single ascending dose. We have completed that part of the study and have finished off those data.
We're now in the multiple ascending dose, and we'll be working through these four dose groups to see how much weight loss we get. We are doing this not in normal volunteers per se, but in normal volunteers that are obese or overweight, and this way, we can see over the four to six week period that we'll be dosing them, we'll be able to see a weight reduction. Just a summary slide. These are the studies that we're doing moving forward, and you can see on the right-hand side here, the key catalysts will be getting primary endpoint data in 4Q 2026 for the 211 and 212 studies.
Whereas, in the bottom there with BMF-650, we'll be, by the middle of the year, getting data on weight loss out of our BMF-650 study. That's all I have.
Great. Thank, thanks, Mick. Thanks for the, thanks for the rundown here. I'll kind of take it into, like, a fireside chat format now.
Yep.
Starting with Icovamenib, could you just talk about how you dialed into this insulin-deficient patient population, and why you think that that's the best place for it? Why has the data supported that idea? Then I think the biggest question you probably get from investors is: how do you identify a patient who is insulin deficient versus some of the other phenotypes that we see
Mm
out there?
Yeah, very good questions. The initial sort of focus was we were trying to understand where the drug might work best, and we had seen these characterizations of four different types of patients from this group in Europe, the lead author of which is somebody called Ahlquist. They developed ways of separating diabetes patients into severe insulin-deficient, mild obese diabetes, and also severe insulin-resistant patients, and there's another group as well. It was a complicated algorithm that looked at a bunch of different things like glucose tolerance and, you know, insulin levels, as well as BMI and A1C. While you can do that, it's complicated, it's not a straightforward thing that would normally be done in a doctor's office.
We used those algorithms, because we were doing all those measurements in the clinical trials. We used those algorithms to initially define the four populations that we wanted to look at, and we found that the severe insulin-deficient patients had a better outcome than the other three groups with respect to this drug. Now, in discussions with a number of advisors and also with the FDA, we came to the conclusion that it would not be good to move forward with that particular algorithm, because the average doctor doesn't want to do this type of
Yeah
you know, analysis.
Yeah.
Not all those tests are available to them. We came down to sort of looking at the data and said, "What are the characteristics in this data set that really define the patients where the drug works?" We came up with these three parameters that I showed you for the, for the, you know, the characteristics that will allow enrollment into the trial, and that's a BMI between 7.5 and 10.5. Sorry, not a BMI, A1C between 7.5 and 10.5, a BMI less than 32, and 1 - 3 prior medications that they were failing. When we look back at our database, those were capturing the patients where we had a good outcome...
Mm
We feel that's the right patient population to move forward with, and it will be, from a commercial point of view, a lot easier to justify and to enable doctors to identify who gets
Yeah
the drug.
What do you think a potential label for this drug looks like in the future? It's essentially, number one, they must have failed some other type of
Yeah
diabetes medication, probably like a metformin.
Yeah.
Number two, they have to have a BMI less than 32, I guess an A1C within that bounds.
Yeah, that would be certainly for the patients who will be followed in the 211 trial.
Mm.
We're hoping that the 212 will give us a separate label, i.e., patients who are failing a GLP-1 receptor agonist.
For those that don't or aren't that familiar with A1C metrics, 7.5 to 10.5, that puts you as, like, modestly diabetic, or is it severely diabetic?
7.5 might be considered modest. I mean, people are considered controlled when they get down to seven, although if they're at 7.5, most doctors would treat them because a normal A1C would be more like, you know, 6 or 5.5. 7.5 is not too far out of range, but 10.5 would be considered, "Geez, you should have been treating this guy for some time.
Yeah. Got it. That makes sense. It's really uncontrolled diabetes that have failed at least one prior and has a BMI less than 32. I'm just trying to
Yeah
make sense of, like, the easiest way to think about labeling here. 'Cause, you know, sometimes you run the risk of over fitting the curve, right? When you look at
Yeah, for sure.
Phase I, Phase II data set, and then go back. I feel like these bounds seem pretty reasonable here, because you're really just trying to capture the patient population that did best, and as you mentioned, you don't have that, a fancy algorithm tool that is easy for these types of doctors to implement in their clinic.
Exactly
I bother.
This should be a lot easier.
Maybe Matt, one point to add is every third patient ends up on insulin in diabetes, just overall. If you look at the 38 million people in America that have diabetes, you have over 10 million that are insulin-dependent
Mm
so roughly, right? Why do they end up on insulin? Because the prior therapies are failing. We believe those patients can be "saved," quote, unquote, from being a lifelong dependent on insulin, because we would provide them years of controlled restoration of the beta cell pool. We're not trying to fight over the patient that shows up for the first time and says to the physician, "I don't know what's happening. My mouth is dry.
Right.
I have all these symptoms.
Yes.
That's for standard of care. We take them when they fail standard of care, and there's nothing else. Just to make the point, 80% of those that have diabetes die of a diabetes-related disease, and the loss of life is 12 to 14 years on average.
Mm.
The healthcare burden is huge, particularly of those that fail. If you're controlled on metformin, you don't need me, and I don't need to help you. If you're uncontrolled on metformin, that's where we come in.
Interesting.
Yeah.
It's a very, very big market, regardless of kind of slicing.
Yeah
Dicing it by insulin deficient versus not.
Yep. What do.
We want to show the FDA, look, there is a real need. They all know it, but we're not trying to fight over what you have controlled. We're fighting for those that are uncontrolled.
Yeah.
Right.
That makes sense. What do you think is a clinically meaningful bar here in terms of A1C improvement, assuming that is going to be the primary endpoint of the pivotal trial? Is it 1%? Is it 1.5? Is it less?
I think from a registrational point of view, if it's 0.5 at 6 months, and it improves to over to 12 months, like we've shown in these studies, I think that what would probably be good enough for registration. I don't know how much use it would get at that level, but if it's one at six months and more than one at 12 months, then I think it'll be a blow your socks off type of regimen.
Matt, the registration, when you look at all approved agents, they all come in. Every agent today is chronic. That's number one. That is being used in diabetes. We would be a non-chronic agent, and they said, "Look, if you're a non-chronic agent, the hurdle hasn't been defined for us." If you look in all the approved agents that were chronic, 0.5 is the minimum hurdle that we saw or found. That was for the DPP-4 inhibitor. We think that's a minimum hurdle for whatever the endpoints are, primary or secondary, whatever they want us to achieve. You've got to give the drug a little bit of time to mature, meaning the cells that are initially created through the inhibition of menin, they take time to come fully online and be fully functional.
That's why we see the curve improving over time. We believe the primary endpoint is six months, secondary is, as it is in our study, 52 weeks, and at that level, at 52, 0.5 is what I personally think, and it's an opinion, obviously, 0.5 is my hurdle.
Yeah.
Right? If I hit 0.05, I'm in the ballpark of approvability.
Does this include, like, GLP-1/GIPs now, the more novel weight loss treatments? These are obviously very effective in diabetes. Do they not work as well in an insulin-deficient population?
They don't work really well. If you talk to Juan or people that are, like, treating these patients every day, that's what they tell you.
Mm.
We see them when they come in the door. They're typically not obese, you know the loss of weight is not going to give you the benefit of insulin now being at a high level. You lose the weight, you reduce resistance, perfect.
Yeah.
That mechanism you don't have working in your favor, so you're dealing with a depleted pool. What do I do with this pool? You can stress it more, but if you stress it more, it doesn't keep yielding, meaning you just don't have enough.
Makes sense.
Yeah.
What are the next steps for the program? Are you just kind of waiting to get meeting minutes from the FDA for buy-in on the design of the trial, and then you can kick it off?
No, we're already moving down that pathway. We've, we filed these studies with the FDA several months ago, and we're now in the throes of identifying sites, getting patients up and running. We'll be screening patients shortly.
Great. How long do you think this thing will take to read out? I mean, if a primary endpoint is at six months, you know, how many patients are you going to have to enroll?
Well, the patient population in the study is fairly small. It's 60. There's 40 patients on Icovamenib and 20 on placebo. We believe we can enroll that in probably 4 months, we think that we'll get the primary endpoint by the end of the year.
Oh, wow! Okay. Is this actually intended to be a registrational trial, or is this trial just kind of intended to support the insulin-deficient patient population?
It's not registrational, but what we had sort of proposed and what we had been discussing internally is the idea that if we could get good results from this study as designed, all we would need to do is expand it into a bigger population for phase III. Clearly, 40 patients treated won't get you a phase III registration, but it should recapitulate the data that we have in a more defined way.
Yeah
We're defining the patients ahead of time now and not retrospectively.
Yeah, that makes sense to me, certainly should spur more investor interest as you get those data and generate them.
Yeah
in the relative near term versus, let's say, trying to enroll, like, a 300 patient pivotal trial that's going to read out 2.5, three years from now. I like that strategy.
Yeah.
Yeah, makes sense.
We've got to get data, and we've got to, you know, get investor interest to help us move the program forward.
Matt, the GLP-1s, you shouldn't forget, right? Well, because we're harping on the insulin-deficient patients. We saw, and these are obese patients, if you're an obese patient on a GLP-1, and your GLP-1 is no longer controlling your A1C, meaning your A1C is rising, we had those patients in this Arkoma study also, and we looked at them because we knew we're raising GLP-1 receptor expression and GLP-1 expression in the body by inhibiting menin. We knew there's mechanistically the benefit, and so when we looked at the 11 that we enrolled, and it's a small n, I agree, that's why we're doing the phase II study now.
Yeah. Yeah.
They all did phenomenal.
Yeah.
The only thing you need is you're failing in GLP-1.
Yeah.
That market is huge.
Yeah
as you probably know.
Yeah.
the failing of standard of care is sort of where we will focus on, and GLP-1
Mm
failure is this other study that we're doing.
Yeah
that as well has simple enrollment criteria, and those patients, we will also prevent from going on insulin.
As you know, there's a lot of people on GLP-1s, and so there's also a lot of people failing GLP-1s.
Yeah, totally. Let's talk a little bit, in the time we have left remaining, about 650, which is your GLP-1 agonist, oral drug. I mean, strategically, I understand the value of having this asset, 'cause I do think at some point maybe you could even pair it with Icovamenib.
Mm-hmm.
It definitely gives you optionality on trying to tackle the broader diabetic population here, if you include kind of the weight
Yeah
those that, not just the insulin-deficient component, but the other maybe more weight-driven components. I think the questions that I do get a lot are just, you know, has the field kind of moved on to more fancy GLP-1, GIP?
Mm-hmm
T he tri-agonists now, GLP-1, GIP, glucagon. You know, where do you see the need for another GLP-1 agonist oral?
Well, as you know, there's a lot of interest and a lot of excitement and a lot of activity, but there's not a lot that's actually made it through to registration yet. That's one thing. We feel we're behind, but we're not that far behind.
Yeah.
Part of the issue comes down to, you know, what do all these other pieces do? One thing, for example, you know, you think about something like Tirzepatide. It has a great weight loss sort of outcome. The question is, do you need all those three activities? I mean, I'm a little stunned from a pharmacologic point of view that they talk about the three activities when they all have different EC50s. It's not clear that at any point in time, that all three of those things are actually in play. Having a GLP-1 receptor agonist, we know GLP-1 receptor agonist from what we've seen over the last several years, they work very well.
What we're just trying to do is to get away from injection and get away from
Yeah
significant time spent on the uptitration. We think with our drug having a more uniform pharmacokinetic profile, we should be able to uptitrate it quicker and therefore get to weight loss earlier.
Maybe one key aspect, and that's the real world data. We see all the fanciness, at the end, seven out of 10 patients in real world surveys, up to seven out of 10, and it depends on what survey you look at, drop off in year one, even with insurance coverage. Why is that? We think it's exactly what we're trying to solve. Can a patient stay on drug, have the benefit over time? Does he want that? Is the body accepting the drug well enough? If you make it patient-friendlier and can allow a patient to sustain years on drug, I think the next wave comes to: How can I keep a patient on drug? That's where we want to fit in.
Yeah. Makes sense.
If you hit the weight loss target at a reasonable range, right, and we all know the ranges, nobody needs 25%. You know
Yeah
that feels almost weird, right?
Yeah. Yeah.
If you hit, you know, the marker that everybody else is hitting with a GLP-1, we know it's gonna be the backbone
Yeah
and we're going in that direction, but keeping them on after year two and year three, that's tricky.
Agreed.
Right.
Makes sense. Always fun talking to you guys. Look forward to the progresses here.
Yeah, thanks a lot.
Thank you.
Great. Thank you, operator.
Thank you.
Thanks, everyone, for joining.
Bye-bye.
Yep. Bye-bye.