Edgewise Therapeutics, Inc. (EWTX) Earnings Call Transcript & Summary

Okay. We'll continue with the next session, which is Edgewise Therapeutics. Good afternoon, everyone. I'm Paul Choi, and I cover biotechnology here at the firm. It's my pleasure to welcome to my immediate left, Kevin, CEO; and Behrad, COO. What we'll do is kick it off -- let Kevin kick it off with some maybe high-level overview of the company, sort of where the lead programs are and then we'll get into Q&A.

Yes. So thank you all, and thank you for joining us today. Edgewise Therapeutics is really a muscle platform company, developing targeted agents for genetically-driven muscle diseases of high unmet medical need. Really founded by Alan Russell working in the muscular dystrophy space. That drug ultimately has come to fruition. Much of what we designed -- this is really starting from scratch. This target has never been worked on. Went through the -- identified the lead molecules, went through the preclinical work, went through clinical work, which validated our preclinical approach. And now the drug is in Phase II and Phase III studies in Becker muscular dystrophy and in Duchenne muscular dystrophy. We had a very exciting readout in December of GRAND CANYON, which was the first controlled study in Becker muscular dystrophy, where we hit our primary endpoint of decreasing muscle damage biomarkers. And also, we're quite close on hitting the functional endpoint of North Star in that Phase II study. Underpowered for the functional endpoints but a 40-patient study but a strong safety profile, hit its biomarkers in a statistically significant manner and had a nice trend for function. We announced in the first quarter that we had completed enrollment of the Phase III confirmatory study in Becker. And so that is an 18-month study. We overenrolled the study up to 175 patients, greater than 96% power to hit a 1.6-point change in North Star. And we'll be reporting data in June, this, in the next couple of weeks on our interactions with the FDA, their evaluation of the GRAND CANYON package and some additional data from our open-label MESA study. We also, in parallel with that program, are working in the Duchenne space. So Duchenne is obviously a space with a very high unmet medical need. There has been some results of some agents in the past year or 2. Unfortunately, this continues to be an unmet medical need in this space. We'll report on our Phase II data, open label in Duchenne, roughly 80 patients with endpoints over a 6- to 18-month period at multiple different doses. The goal here is to show that we've been able to see a change relative to propensity-matched patients of about 2 points. It also -- hopefully, we can convince you that we've identified a dose for a Phase III study, perhaps a patient population. And then we'll be discussing the functional endpoints and the primary endpoint for a Phase III design, which will ultimately be to go to the FDA and the EMA for some conversations on what's appropriate for that part of the discussion in Duchenne. On the HCM side, so as you remember, we, again, defined a novel target for the treatment of hypertrophic cardiomyopathy that came out of our skeletal muscle program. That drug, we showed really positive data in normal healthy volunteers in single-dose data in September. What's highly unusual and significant about this program is that the drug shows that has no concentration relationship with the change in ejection fraction. So as you know, the competitor drugs in this space are the cardiac myosin inhibitors, so changes in ejection fraction, which is associated with efficacy. We've been able to decouple that efficacy from the change of ejection fraction so that provides differentiation for the product. We've also shown an outside sized result on certain feel and function measures, which are part of the primary endpoint for cardiovascular drugs called things like the KCCQ score and New York Heart Association changes. And really great importance is that we're the first drug to show very rapid changes in a measure called e-prime, which is the diastolic relaxation of the ventricle. That's predicted based on our mechanism and preclinical models. We think that translates not only to benefit in the HCM space but also potentially benefit in heart failure in general and HFpEF in particular. And then as 1 more milestone, sometime in the next probably a month or 2, we'll be announcing the entry of -- into Phase I of a second-generation cardiac molecule that will be targeting HFpEF as its primary indication. So a lot going on this year. Just to give you one more flavor of the rest of the year, we are currently dosing patients for 28 days at 25 milligrams in the Part B and C of CIRRUS. We've also completed the protocol for Part D, which is a 12-week study in patients with HCM, both obstructive and nonobstructive. That protocol now is with the FDA and also through the IRBs. So we anticipate in the next month or 2, we'll be dosing patients in Part D with a readout by the end of the year.

Okay. You've definitely framed out a busy remainder of 2025, Kevin. But maybe sticking with your last point on the 7500 program in HCM. I think maybe just to help people understand the program a little better, can you maybe walk us through, first, the MOA and then maybe just remind us what you showed with your 28-day data recently to help set the stage for your Part D data later this year?

Yes. So this drug was designed to eliminate this relationship between the concentration of the drug and the ejection fraction change because the ejection fraction lowering of the CMIs below 50 drives a REMS and a black box. You want to...

From a risk of heart failure?

From a risk of heart failure. So we want to eliminate that particular aspect. And so how did we do that? We identified a mechanism where we optimize the drugs in preclinical models and animal models to eliminate that liability. We also wanted to have a greater effect on the diastolic activity. So we are more potent at low calcium levels biochemically, so we have a greater effect on the relaxation of the ventricle when it's filling with blood before it pumps out. And then the third critical differentiation of the mechanism is we're a partial inhibitor. So we cannot maximally inhibit contraction in the heart tissue. And that combination of being a partial inhibitor having greater effects on the diastolic portion leads to the profile we've seen in the clinic. The 28-day data came out in April. We showed dramatic changes in things like KCCQ and New York Heart Association changes over a 28-day period. We had scores up to 24-point improvement of 5 points is considered clinically meaningful. We also saw deep changes in NT-proBNP, which is a measure of diastolic function. We had deep changes in gradient in the obstructive HCM patients. So we're very pleased with the profile. And the drug is moving forward and I think will be very successful.

Right. So to continue on the last part, you said there now you're continuing with dose escalation at this point. But you're also talking about -- thinking about working through starting off Part D of your study, which will include both obstructive and nonobstructive patients. And so maybe, I guess as you think about the landscape at this stage, how are you thinking about what is considered, I guess, clinically meaningful for Part D of your study? And how does that potentially differ for your obstructive patients versus your nonobstructive patients?

Yes. It's interesting the profile we've seen so far. I think what's -- one of the goals is to minimize things like the number of echoes you need, and to demonstrate that in a broader group of patients, you continue not to see the ejection fraction changes. So that's a goal of the Part D. The second goal of the Part D is that we had a signal of atrial fibrillation in some of the patients. And there's a couple of things that we've done to change that. So now the CMIs, if you look at the Phase II studies, there was a relatively high rate of atrial fibrillation. And patients with HCM in general as a background have high rates of atrial fibrillation. So what we've done is thought through very carefully. First, we need to dose patients who are truly HCM patients. Second, we need to look at patients who do not have a high risk for atrial fibrillation. And so this is important so that we really are following now the guidelines that the CMI set up in their Phase IIIs as opposed to trying to treat all the patients. That's important for the safety profile as well as the efficacy profile of the drug. So I think what we'll do in the Part D is demonstrate really 0 effect on atrial fibrillation and show exactly the same benefits on feel and function, proBNP, gradient.

Right. It seems to me that at least some of the patients who experienced AF events had predispositions to it and various risk factors. And so...

The other thing you have to factor is -- so think about the treatment setting, right? We went to largely the same sites where the CMIs have been doing their Phase IIIs and where you see most of the commercial, which is a center of excellence. So the patients that we were treating were maybe skewed towards, if you look at the baseline demographics, to being more severe. If you look at their baseline gradients, if you look at baseline rates of hypertension, if you look at where the LVEF was, in aggregate, those patients were a little bit sicker. And so that doesn't mean that you can't treat those patients, right? That means that you might want to take a different approach to a more severe group of HCM patients. And you probably don't want to do it in an open label, you want to do it in a controlled setting where your placebo helps you out. And I think that's the lesson learned here was that those patients were not eligible patients for CMIs. A lot of the sick ones were not eligible for CMI. And a physician looked at the patient and said, "I need a therapeutic. Edgewise's profile, particularly on LVEF is really, really attractive." And so I think that's one of the things that we've learned is like you -- those patients are in dire need for a therapy, and we'll continue to study them but probably not in an open label.

Right. And then maybe just returning to thinking about what is considered a good outcome for your obstructive patients versus nonobstructive patients. Any sort of ballparks or just sort of how you're thinking about it?

Yes, what we noticed, which is really important is that the patients -- we think of getting people to normal as being a positive thing, normal meaning ejection fraction of 60. When you come from 70 or 50, we seem to converge on normal. We saw that when we normalize biomarkers like NT-proBNP, we are able to drive the New York Heart Association measure down to asymptomatic from Class II or Class III, which is really exciting. And then with the KCCQ scores, we saw patients go from the 60s and 70s up to 90-plus, which is essentially normal. So in both populations, we saw the ability to take a patient from a disease state to essentially a normal state. And what we heard from many of the patients that their feel, how they felt about the drug was they felt spectacular when they're on the drug. So we want to try to replicate that. Now from question of obstructive versus nonobstructive, I think we all saw the data that the ODYSSEY study with mavacamten did not hit their primary endpoint of KCCQ and peak VO2. We don't have all the data. We don't know why that occurred...

Or the magnitude of the miss.

Or the magnitude of the miss and we'll probably get that data later on in the late summer. But that's -- it really provides an opportunity for Edgewise in that space. The Cytokinetics drug, aficamten is also being studied in the nonobstructive area. And I think that data have been reported, I think first quarter of '26, that will probably come out. So obviously, we're looking very closely at that. But I view that as a really clear opportunity because if you look at our data on ejection fraction, we're absolutely 0 effect. In fact, the ejection fraction when you look at, we started patients who are 52, 54, 56, instead of going down like the CMIs have seen with the nonobstructive patients, our ejection fraction actually went up, which is really highly differentiated in the nonobstructive space. So I think we're very excited about the opportunity in nonobstructive. Again, in obstructive, seeing these KCCQ scores north of 20 is highly differentiated and almost double what we observed with the CMIs. So we're excited in both populations. But certainly, the hope would be that we might be the first agent to have an approval in nonobstructive.

I think where you're going as well, Paul, particularly on the differentiation, in obstructive is like what is the TPP ultimately going to be, right? And I think we're trying to shift from a paradigm where today, the class of molecules were going to -- they have a pre-prescribed REMS that is set up to monitor for safety. The REMS is there to look for patients with dropping LVEF. If we can shift that practice where the decision on performing an echo to look at whether it's gradient or LVEF is at the discretion of the physician in order to optimize efficacy. That's a fundamental switch to really unlock the value, which is not really sitting at the centers of excellence exclusively. It's at a group of physicians that we call Community Plus that are the community cardiologists that have these patients but are hesitant to push them on to a center of excellence right now and can't handle the burden of putting them on a CMI because of the echo titration required. So I think it's a fundamental shift in the thinking. If you think about it, if you were a CMI company, you would want to avoid performing echoes. The fact is they have to do it because of the safety liability. Nonobstructive is a good example. Why would you need to do an echo in the nonobstructive patients when there's no gradient to relieve, right? So I think they've been forced to have to do that. And we're trying to say, imagine you treated obstructive patients like a nonobstructive patient where you don't require a safety echo. That's a very different proposition.

Great. Maybe just continuing on that thread as you think about 7500 potentially coming onto the market in a few years. You talked about the REMS being a key point of differentiation and 7500 potentially being the first nonobstructive drug as well. But just in terms of market development and market identification, I guess, with theoretically 2 CMIs being out on the market ahead of you, where do you see sort of the key areas that you would focus on in terms of commercialization or focusing in terms of your marketing and launch efforts down the road, a little bit further down the road?

Yes, I think as Behrad has said, I think the opportunity is really in the more of the community practices. And one of the impediments is that these community docs, they have their own -- is running their own business for all intents and purposes. They need to have a highly qualified echo reader. They're responsible for that echo result and monitoring that patient and they need to put in place the -- essentially the infrastructure to monitor those patients, to do the paperwork and to make sure those patients in a relatively rigorous way are monitored across the continuum forever for all intents and purposes, and especially when they're getting to the right dose, which is echo-intensive. So the question is for those groups of doctors, it's easier for them to actually send that patient off to a center of excellence. But the center of excellence have a saturation point as well. They can't continually add echo facilities or people. So even though there are some groups that may have 1,000 HCM patients, they can only put 100 of those HCM patients on a CMI because of the requirements. So now we'll see what the REMS turns out to be for aficamten, but we don't know how it is but we know they will have REMS at some point.

The other thing to think about is like the nonobstructive is really interesting because if you look at the dynamics, and Cytokinetics have talked about this as well, and we've seen the same thing from our market research is as a proportion, the nonobstructive patients are constituting a larger proportion of diagnosed HCM patients. So the general thought has always been it's 1/3 of the market is nonobstructive, 2/3 is obstructive. We're seeing a bit of a shift. It's getting closer to 50-50, not quite there. And so if you have a drug that can treat the nonobstructive population, because remember, we have a unique benefit which is driven by this diastolic effect. So we're a lusitropic agent. And so I think that aspect plays out really positively in the nonobstructive. And you saw the data, right? The 50-milligram, our lowest dose, we saw 16-point changes in KCCQ, which was almost double what you've seen with the CMIs. So I think that, to me, is like one of the most interesting ways to capitalize on the market is to go out through the nonobstructive. We already know what we do in obstructive. You saw the data. With almost a single dose, you're driving a pretty dramatic effect on gradient, on feel and function, on NYHA, and on old diastolic parameters, whether it's NT-proBNP or you're looking at e-prime and e over e-prime and other end disease.

So AHA guidelines don't say treat to a gradient reduction. They say treat to...

Symptomatic relief.

Symptomatic relief, correct. So that's what we're really shooting for.

Great. Maybe 1 more on HCM before we talk about sevasemten, which is one of the interesting developments recently in the space has been the Cytokinetics MAPLE study, which hit -- we don't know the details yet, but it hit on the primary endpoint and beat standard of care beta blockers in the head-to-head study. So my first question there is do you think this will drive broader adoption of CMIs in the frontline setting, whether it's aficamten or Bristol's CAMZYOS? And second, given that a competitor in the space has successfully run a head-to-head study, do you feel like this is necessary for your 7500 program for utilization in the frontline setting?

Yes. Beta blockers can be effective in some patients for obstructive HCM. So I think in the obstructive space, I think it's a bit more of a challenge to say that I won't go through a beta blocker first before I move to a CMI. This probably would help, but then you're back to, do I need to do extensive echoes to get people to their maintenance and to their target exposure? So how much of an effect that could be is an open question, I think. I view the nonobstructive opportunity as being even greater for a MAPLE-like study because I think it's kind of well-conceived today by the physicians that the beta blockers don't work at all in nonobstructive. It's just kind of you try it, I hope it works. And there is no standard of care in the nonobstructive. So in fact, I think a MAPLE-like structure in nonobstructive might be a more feasible and commercially viable strategy to be the first drug for treatment of nonobstructive as opposed to trying to push it into the obstructive population. But I think we have to do a lot more work on that space to see if that's the appropriate study.

For 7500?

For 7500, yes.

Great. I want to shift gears to sevasemten and maybe talk a little bit about the DMD space and -- which is one of the most notable market developments this year in the DMD space has been unfortunately a patient death on Sarepta's ELEVIDYS. And so I'm just sort of curious what has the downstream impact been on Edgewise in terms of either patient-inbound interest or clinician-inbound interest in your DMD program. Can you maybe comment on that a little bit?

Yes. I don't know if I've seen a lot of changes from that. There is the rate in which they can get patients lined up for the Sarepta gene therapy, and that hasn't changed, I think. I don't know if there's been a lot of change in the ambulatory patients. I think within the nonambulatory patients, I think there's been a bit of a pause in the thought process because there's just simply less data there. But from my standpoint, people still believe there is some benefit in the gene therapies. Now there has been some data published both last year at MDA on the 2-year ELEVIDYS data and then this year, a 3-year ELEVIDYS data. And there was a fairly substantial decline in the North Star from 2 years to 3 years. So that's starting to become more well known. And the question is, what is the durability of the response and how does that affect the market uptake? And does it affect us or does it affect the enthusiasm for getting some of the second-generation gene therapies? Now from our standpoint, we're a bit agnostic to that, so we have studied the patients who have previously been given a gene therapy. So 2 years out, these are both Sarepta patients, SOLA patients, Pfizer patients. And so we do have an understanding of what the rate of decline is out by 2, 3, 4 years. We'll be reporting on that data later on this month. Also, I think -- I don't think we've seen a dramatic change in retention of patients. I think some patients have gone off the gene therapy. But I don't think it's like a flood to the gene therapy programs.

Do you feel like you'll be in a position in the near term to talk possibly about your Phase III design and just kind of what a post-gene DMD population looks like? Is it ambulatory, nonambulatory? Is it limited maybe to a certain age range of patients or certain North Star baselines? Maybe just at a high level, how are you guys thinking?

At a high level, I think there's good evidence to say that if you can develop a protocol and demonstrate benefit in any population of Duchenne patients, you'll -- the label would essentially read like, for instance, all ambulatory patients, right? So you've seen that with givinostat. You've seen that with the Sarepta drug. So I think that's true. So then it's identifying the right patients. The question for us becomes, is there an age group? Is there a functional baseline that you would use? Is there a dose that you would use? And we should be able to provide that update of, do we have the right parameters to be able to develop that Phase III? Now one of the key aspects of this is what is the primary endpoint. We're quite inclined and we're quite interested in a measure called the SV95, which is a wearable device. That endpoint is approved as a primary endpoint in Europe. It's under evaluation in the U.S., but we don't know how long that will take. So that's 1 -- we think it's a highly quantitative endpoint that's highly objective because...

For extreme populations, yes?

That's right. But of course, we have the North Star, which is the standard and we have things like Time to Rise and 4-stair climb. So we will discuss and disclose multiple functional endpoints. The goal here is that do we have a window in which we can work? And what would be the primary endpoint? Which is probably going to end up being a discussion with both the FDA and the EMA of what's appropriate for this population and talk about the trial design.

And the balance of like how many gene therapy-naive versus gene therapy-experienced patients to include in the study, right? Because what's clear for us and just looking at the process is there's still a tremendous unmet need post-gene therapy as early as 2 years post getting dosed. So if you're seeing a rate of decline that's aligned with natural history based on the data that Sarepta put out at MDA, that tells you that, yes, you provided an initial bump but now they continue to decline at that -- at the same rate as predicted by natural history. So you need something for those individuals.

You bring up an interesting point, which is that these patients post whatever initial therapy they may be getting will still probably have a chronic need for treatment here. So I guess as you think about the landscape of rare and orphan diseases and drug pricing in those categories for what may be a potentially chronic treatment here, what does, I guess, your work or market data work sort of suggest...

Well, you look at risdiplam on top of ZOLGENSMA. We've seen other molecules. SPINRAZA gets used on top of gene therapy. We've even seen in the U.S. some patients who received gene therapy but then have gone on to receive the HDAC inhibitor, givinostat, right? So that flexibility is there. And I think the interesting dynamic for Edgewise is, from a cadence standpoint, we would be launching in Becker sometime in Q1 2028 if we go through the full approval based on GRAND CANYON. We would be selling our price in Becker. And as you know, there's nothing approved in Becker. It's completely a blank slate for us. And so the question is, can we take that value and that price associated with that value and translate it to the Duchenne population? Based on our market research, there's flexibility because, again, it's not -- you're not talking about 30,000 patients. You're talking about a very discrete number of patients, about 12,000 to 12,500 ambulatory Duchenne boys, maybe 15,000 in Europe. So you're not having a tremendous impact on overall budget, even with the advent of the gene therapy at 3 million-plus.

Right. I think the other interesting aspect is for years, people are saying that I'd like to turn my Duchenne patients into Becker patients. Now we have a treatment for Becker. What does that mean to the market now people think about the drug? Because if we can demonstrate effects in Becker, does that translate over to some view of, if I provide a gene therapy to a Duchenne, that's now formally a Becker? They have expression of a microdystrophin. And how does that play through with the regulatory agencies because it will take a while to really understand the natural history of the gene therapy-treated patients to run a trial and there's really no drug for them to take. So what do you do with them?

Right. Behrad, earlier, you just mentioned GRAND CANYON potentially forming the basis for a future Becker's approval. But can you maybe help us bridge the gap between your CANYON data, what the agency has said and how has that informed your thinking on where GRAND CANYON might fit into the regulatory package?

Yes. So as you know, we had some really interesting data from CANYON, right? We hit our primary endpoint, which was CK. We hit TNNI2, which is our on-target biomarker. And we had a 1.1-point difference in North Star, which didn't quite hit stat sig, but what was really interesting about the CANYON data is the first study, placebo-controlled study in Becker patients, and actually, the placebo behaved exactly as predicted by natural history. Very seldom do you see that play out. So I think you look at that data package and you look at the data that we generated in ARCH, which was the open-label study, and obviously, we've captured some data in MESA, which is our open-label study. So we put together a pretty interesting package that we put in front of the FDA. And the question is, hey, is this data the basis for some sort of accelerated path to an approval? Because we've talked about this before. Generally, the FDA hasn't been open to CK as an accelerated endpoint because it's so variable and noisy. We even saw that in our CANYON data in the placebo arm, right?

It's tough to measure.

It's tough to measure. So I think the question is like what's going on at Cedar at the macro level without leadership in place and then really balancing the quality of the data that we put in front of them and what's the flexibility in terms of an abbreviated path to an approval that is beyond the line purely on GRAND CANYON? Is there some flexibility to come in with additional data somewhat in a rolling submission type strategy to really work with the FDA? So that's what we've been exploring with.

And obviously, with the date of a readout in the fourth quarter of '26, so there is kind of -- the timing there is by the time we get the data, if we can get some of the data signed off on, you essentially cut off 6 months off the back end after the GRAND CANYON data. So either way, you win as long as you can move faster.

Yes. With GRAND CANYON, maybe can you remind us, is there anything -- given that it's about roughly 1.5 years out, should we expect any potential interim updates from that program between either in the back half of '25?

No, like many studies, I think we were kind of a little bit back-end loaded for the recruitment of the patients because Europe opened up a little bit later than the U.S. And so getting an interim analysis is not going to be highly productive, I'd say, a 1-year cut. And we're pretty satisfied with the statistical analysis plan right now. Of course, that is not settled and confirmed at the moment, but we'll develop that in collaboration with the agency and different areas of supportive data. We certainly want to shore up natural history studies and do that in a prospective way to augment the placebo groups. So that is another piece of secondary information that's important to the agency in these rare diseases.

Also, I just wanted to point out, we've had a pretty remarkable retention rate at MESA from all the patients who have enrolled. We've had a 99% enrollment. So clearly, something -- these patients are experiencing something and they're continuing to want to be involved with the studies.

Great. We're coming up on time here so maybe just 1 quick 1 which is just on your HFpEF program. Just remind us how you're thinking about this particular patient population here. Who will be included in the treatment experience, just newly diagnosed? And just sort of what time lines for the initial data might be?

Yes. So we initiate the Phase I studies in, say, third quarter of this year. We'll be positioned to run a HFpEF study probably based on the tox coverage, about a 3-month study. The readout is likely to be something along the lines of NT-proBNP. Our bogey for that, as you can see, a 20% decrease in NT-proBNP in a broad HFpEF population that has generally led to, in heart failure, a clinical benefit for those patients. So that's pretty straightforward. I think it would be likely we would run a placebo-controlled study. HFpEF requires larger studies. They have a number of background meds that you need to account for. And then the big question for us, are there 1 or 2 targeted populations where we run a fairly generic basket study and try to use the Phase II study to identify the right population for this drug?

Great. Unfortunately, we've run over so we'll have to end it on that note. My thanks to Kevin and Behrad and Edgewise for joining us.

Thank you, Paul.

Thank you, Paul.