Hello everyone. I'm Joe Schwartz from the Biopharma Equity Research Team at Leerink Partners. It's my pleasure to host this fireside chat with Edgewise Therapeutics. We're pleased to have Kevin Koch and Behrad Derakhshan, CEO and CBO, respectively. Thank you for joining us and giving us an update.
Yeah, thanks for the opportunity.
2024 is shaping up to be a very exciting year for you guys. Maybe we can start by having you lay out the different readouts that you expect to provide over the next several months.
Yeah. No, so very exciting. Really have moved the programs forward quite successfully. And so as you remember, we're a muscle platform company treating rare diseases with an origin in muscle. We have two broad programs, one in skeletal muscle, one in cardiovascular muscle. We'll have readouts in both of those areas this year. So as you might remember, we've advanced our skeletal muscle drug 5506, now called sevasemten. It has a name. And we've advanced that into advanced studies in muscular dystrophy. So there's two primary populations we're studying. One is Becker muscular dystrophy. One is Duchenne muscular dystrophy. So you all know about Duchenne. You're missing a key structural protein called dystrophin. And that has a very rapid progression. Patients lose ambulation in their teens and succumb to cardiovascular disease in their 20s and early 30s. Becker muscular dystrophy is an adult form of the disease.
Patients have a shortened and lesser amount of dystrophin. They can be diagnosed even when they're in their late pediatric stage and they lose ambulation in their 20s or they're typically diagnosed in adolescence. They lose ambulation in their 40s. Both of these are dread diseases. It's not a good outcome to be a Becker patient. We have studied our drug affects the damage to the muscle, which is the driver of the loss of function in these patients. The first studies we ran were in the Becker population. Now this is a 5506. Sevasemten is a small molecule. It affects the damage to the muscle, protects the muscle. By protecting the muscle, you protect the function and slow the progression of the disease. It's disease modifying. We initiated a study in Becker muscular dystrophy. There's 12 patients, open label.
We reported last year of one year data in these patients. What we saw were that there were out of these 12 patients, they had an increase in the measure of function called North Star. They had a dramatic decrease in a set of biomarkers that are indicative of muscle damage. So we decreased the muscle damage and increased the function in a Becker population at 1 year who you would have thought would have lost function by about 1.2 points on the North Star. So that 24-month data is coming out at the Q2 of this year. The second data set we'll have in the Q2, so sometime in probably May-June timeframe, is we initiated a study in Duchenne muscular dystrophy. Now we have a very wide range of patients, 4-9 years of age. We're looking at three months.
We're looking at biomarkers associated with that disease. So that data, we've recruited over 60 patients now in 5 cohorts in a dose escalation. We'll report functional effects. We'll report biomarker effects. And I mean, I think it should be quite an interesting data set. First time you'll see data within the Duchenne population. Third point, let's stick with milestone for the end of the year. So based on the data from ARCH, which was our open-label Becker study, we initiated a controlled phase II study in Becker. That will read out at the end of this year. What we believe this is a study of 3-to-1 randomization, dosed at 10 milligrams, 1-year placebo-controlled. It's powered for biomarkers. So CK is the primary endpoint.
We'll also look at other biomarkers of muscle damage called TNNI2, which is a direct measure of target engagement with fast muscle fibers. And we'll be able to measure North Star, NSAA. We'll have MRI, patient-reported outcomes, and activity measures. So that'll give us a really good understanding of what that particular population brings, what the size of the trial should be. We think it is something that is fileable if we have statistical significance on the biomarker and have a trend in functional benefit. There is no therapy for Becker patients. We think it is a difficult disease. We have fast track designation in Becker. So I think there is an opportunity to discuss with the agency the preponderance of data supporting accelerated approval. Also, we've put ourselves into a position to have a good conversation with the agency because we've initiated a follow-on study.
It should be completely recruited by the time we actually go to the agency with our data package. So it fits within the guidelines of what the agency could find interesting and achievable. So in muscular dystrophy, we have three really very nice readouts in this year. Now let's turn to the cardiovascular asset. What we've shown, so about a year, man, I guess I'd say 15 months ago, we had the initial disclosure of the preclinical data of 7500, which is our cardiovascular asset. Now this is a cardiovascular asset that affects the sarcomere. It affects the contraction of the heart but does it in a very different way than two drugs that you're probably familiar with now, mavacamten and aficamten. So we hit a different molecular target within the sarcomere involved in contraction. That differential molecular target actually is something that's quite different than the CMI.
So the CMI you can think of as an on-off switch. The ATPase is associated with the actin filament or it's not. What we're doing is changing the rate in which the contraction occurs. So not the extent but the rate. What that does is separate the pathologic event, which is the gradient reduction, from the normal physiological event, which is ejection fraction. So what we've observed preclinically is a dramatic change in gradient in the CAT model of obstructive HCM with virtually zero change in ejection fraction in the CAT model. So we think we can disassociate those two effects, which will provide a broader therapeutic window and potentially fixed-dose interactions with the patient. So it's kind of exciting. I think what we'll show in a non-obstructive model at ACC is four-month pig data in a genetically engineered pig model of non-obstructive HCM. So that'll be exciting.
Oral dosing, four months, effect on the heart, durability of the response in non-obstructive HCM. We have completed the SAD study. We had dose ranging from 5 to 300 milligrams. So it's a very broad dose range. We've said that we've completed that entire dose ranging. We've now gone into multiple ascending dose studies in normal healthy volunteers with EDG-7500. And coming the Q2, we plan on initiating our first single-dose obstructive HCM patients. So what do we get out of a single dose in an obstructive HCM patient? So when you look at the prior literature from mavacamten, you'll find that in their early phase I studies, they dosed an obstructive HCM patient with a single dose. And what they saw was about a 50% change in gradient in obstructive HCM but unfortunately a 20% change in ejection fraction. So remember what we saw in the CAT.
We saw a 50%+ change in gradient but essentially no change in ejection fraction. So what we're going to report Q3 of this year is safety data in the single ascending dose studies in healthy volunteers, multiple ascending dose studies in healthy volunteers, and the obstructive HCM data, single dose. That should allow us a couple of things. One is you can look at the slope of the change of concentration relative to ejection fraction in the normal healthy volunteers. For MAVA and AFI, they're steep. For us, we'd like to say we're going to be more shallow. Okay? And then we'll show clear differentiation from mavacamten in obstructive HCM with a single dose. That'll all come out in the Q3, likely September timeframe. Now that study in the obstructive HCM is nine patients, three different dose groups. Okay?
So what we want to show is the breadth of the concentration in which you can have efficacy and a broad safety profile. So if you think about it, when you dose a normal human with a pill, there's about 3x variability in absorption of those pills. So you want to have a broad concentration range of efficacy and an even broader range of tolerability. So that will teach us that we have a reasonable probability of being able to pick a fixed dose for the drug to treat obstructive HCM patients. This is in contrast to mavacamten and likely aficamten that they have a REMS where they have to titrate the dose of the drug to get to the appropriate concentration so that they don't have limiting ejection fraction changes.
That will broaden the market and even broaden it, hopefully, into patients who are not even symptomatic or asymptomatic because that would really open up the market in HCM.
Excellent. Long-winded. Fantastic introduction. Thank you, Kevin. So then let's work backwards and stay with 7,500 for a moment. So right now, you alluded to the need to titrate MAVA. We'll see what happens with AFI. Love to get your thoughts on that and how you think the market could evolve if you achieve your target product profile and do, in fact, have a wider therapeutic index. What do you think the treatment will be like? And how much more could it fit into physicians' practices? Because so far we've seen the uptake has been a bit tepid with MAVA because of the cumbersome nature of using it.
It seems like some physicians are okay with that, find it extremely helpful, and others, it's a nonstarter.
I think AFI, one of the key aspects of AFI is that they don't have drug-drug interactions that are from drugs that are essentially over-the-counter, which I think challenges them to retain whatever titration they've achieved. Right? But I think from a standpoint of is it superior flat-out on safety, I think it is the same mechanism. I think it would be hard for the agency, based on the data sets I've seen, to change the REMS dramatically. It could be somewhat lighter. But I think it would be challenging. So that's my opinion on AFI ultimately. We haven't seen the efficacy data yet. So that would be, I think, might also be a determining factor.
From a standpoint for how we'd like to use 7500, I always view 7500 as I want to insert the drug into standard cardiovascular practice. So what would be standard cardiovascular practice? I would provide the drug to a patient. I might take their measures, echo that same day. Perhaps I bring them back 14 days later or a month later at steady state, look at those same parameters, say, I updose them, I downdose them, or I'm done. That I view as standard practice because patients with HCM are measured every 6 months for echo. So like almost every drug in cardiovascular disease, even things like Lipitor, you get a dose. The doctor asks, "How do you feel? What's going on with your cholesterol?" And you get a dose adjustment. That's where we think we want to go with this drug.
I think your point around the uptake of MAVA is quite interesting. Right? Because I think what we've seen is right now it's being limited to centers of excellence because of the cumbersome nature of the REMS. And I think there's about 60,000-70,000 symptomatic HCM patients. So clearly an opportunity there. You look at PAH, for example. There's four or five different modalities, 40,000-50,000 patients.
Yeah. Interesting comp.
Just inherently being a different mechanism opens up a new opportunity for 7,500. So that's one thing. The other thing is, well, how many of those patients, how many of those 70,000 or 60,000-70,000 HCM patients are sitting at these centers of excellence? Not the lion's share. In fact, a lot of them are sitting in what we call community plus. So it's not your average community cardiologist. It's a more specialized cardiology unit.
Let's say they have a pre-symptomatic or an early symptomatic patient. The kind of drug you want in their hands is one where they can give it to the patient and not have to worry that they're going to have an EF drop that could be problematic. So that's where 7,500 kind of fits a gap. I think the uptake for MAVA has been slow but not too different to Entresto, for example. If you look at that trajectory, they look kind of similar. So I think what's really nice, we know this from looking at me-too drugs, being the first to market doesn't mean you're the best at market. And I think we're going to learn a lot from the market that's going to get created. And we'll know exactly how to position 7,500.
Yeah. We'll anticipate that they're going to expand the market.
Realistically, I think one key aspect of this in both the preclinical models, which you'll see at ACC, or long-term, is that are you affecting the remodeling of the heart? If you can affect that, you would want to provide these drugs as early as possible. As soon as you demonstrated an exercise-induced gradient or a deficit in diastolic function, you would want to dose these drugs because that long-term effect, I think, would be the magic bullet. You want to prevent the progression into, say, a pathologic state in cardiovascular disease. Actually, at this point in time, they're not approved for asymptomatic patients even though these patients with exercise are symptomatic. Remember, we're just talking about two-thirds of the market. There's a third of the market which is the non-obstructive. We've seen the CMI struggle there. Right?
27% of patients see a drop in ejection fraction in that setting. Maybe the drug does work, but you can't test the efficacy to its full extent because you're now rubbing up against having an AE of EF drops. Oh, I see. So that's why something which works before and has a broader therapeutic index could be dosed to a higher level which could drive the efficacy in these other populations. And realistically, from a standpoint of strategic interest in this drug, the HFpEF market is a disease defined by diminished diastolic function. So essentially, the relaxation of the left ventricle. And if we could improve that aspect of that disease, that would take us from an initial foray in obstructive HCM and parallel effects in non-obstructive HCM but would allow us to move into the HFpEF market.
I think that is where, with 4 million patients and still in unmet medical need, I think it would be an exciting place to take the drug. Okay.
Then the data that we'll see later this year, it'll consist of safety, PK, and some
gradient reduction relative to EF changes in an obstructive patient single dose.
Okay. Then as you look forward to proof of concept phase II or pivotal phase III study, what are the most important endpoints to get approval? Is it pVO2? And is that really that meaningful in the grand scheme of things? Or is it more of a regulatory construct? And you mentioned maybe this could be disease-modifying. That seems like it could be really powerful for payers and so forth. Yeah.
So all that data, I mean, realistically, MAVA and for that matter, AFI has laid out a very nice regulatory path where I think we would probably choose the same path of P VO2 and changes in New York Heart Association as the path in obstructive HCM. I think the level of gradient reduction and the breadth of gradient reduction because we probably can beat the 60% of the patients going below the threshold of 30 millimeters. But we'll see. I mean, so I think you can win on efficacy and you can win on safety and ease of use. In non-obstructive, I think the data's still out whether we'll be able to see these benefits, which I think the endpoints have typically been Kansas City Questionnaire. But there is some reasonable definition of what a good Kansas City Questionnaire looks like, what the result is.
So I think we can measure that. And then in HFpEF, I think it's an open question because it's a population that really has a tremendous unmet medical need. So I think there's a lot of options there. But I would follow. I think we can follow the regulatory path of MAVA and AFI very directly and move very rapidly because it's well-defined.
Okay. The key to HFpEF, as Kevin's talking about, is really understanding how to find the right target population. So it's not the 3 million HFpEF patients. It's, "Let's really understand mechanistically what subpopulation makes most sense." And then you can run a feasible trial rather than a 10,000-patient all-cause mortality trial. So we're trying to avoid that. So I think we're spending a lot of time thinking through that carefully.
Okay. Well, we can come back to 7,500.
But I want to make sure that we touch on sevasemten.
Sevasemten, yes. Rolls off the tongue. Save your skeletal muscle. Oh, yeah. Interesting.
So can we talk a little bit about the order of the updates that we're going to get? And what will we learn from the two year open-label ARCH OLE data? And how does that read through to what we can see in CANYON? And what do we know about the patients that are in these studies? Are they the same pretty much in the ARCH OLE as well as CANYON?
That's a really good point.
And what would you expect from placebo?
Yeah. So this patient group, there's been two natural history studies, one of about 75 patients, the other of north of 50, for 5+ years and now more than 3 years.
Both studies, independent institutions, showed that in patients that have a North Star between 5 and 32, so remember, a North Star goes to 34. So patients who are declining on the North Star have a consistent decrease in North Star of about 1.2 points. That was shown in both studies. And that's over a one-year time frame. So when we talk about the 24-month data, you're now not talking about just a year. That could be in the noise of a placebo. Now you're talking 2 years. You would have expected these patients to decline by 2.4 points. I think that's supportive that this is not a placebo effect. And this is a real effect. And I think that reads through to CANYON of what the actual one-year placebo data will look like. And can we see improvements in function at the one-year time point in a controlled study?
So I think that's how it reads to CANYON.
Okay. And still sticking with the same question, what are these patients like in terms of functionality? Put it into perspective because these are Becker patients which might be considered less severe than Duchenne. But they're a lot older. They're not gaining skills like the patients who were in Sarepta's study.
I think the average North Star for the ARCH patients was 50. If you want to equate that to a Duchenne child, that's a 10, 11-year-old Duchenne boy who's about to lose ambulation. So pound for pound, you're looking at a more severe group of patients when you're looking at our Becker population. So that was kind of interesting for us. And half the patients couldn't get off the ground unaided. So it's not a healthy group. It's a declining group.
For that open label, we didn't select based on North Star. It was an all-comer trial. It happened that we naturally selected those individuals who were on the decline.
Think about it. These folks have been traveling to Atlanta from all over the country for now 24 months. If they didn't feel like they were having benefit, they wouldn't have done that. Right? That's just maybe one thing to say. Right? I think it's important to say that these time to rise, which is a measure that is interesting, has been correlated in Duchenne with if you're north of five in Duchenne, within two years, you're likely to decline, lose ambulation over that time frame. These patients had a time to rise of eight. And six of them couldn't get off the floor at all.
So if you look at trying to equate what you're going to see in Duchenne relative to what you're going to see in Becker, these patients are declining. And in Duchenne, I think these older patients that are declining also could definitely benefit from the drug dramatically. I do think that there's been a lot of controversy on the North Star. What does it measure? And is it sensitive enough? Well, interestingly, in the studies that have looked at older kids that are declining and actually follow directions better. The North Star has proven to be successful in showing statistical significance in an older Duchenne population that's declining. So with givinostat, they had a decline in the placebo of 4 points. And the treated patients were 2 points. So they showed a 2-point delta in patients from 7 to 15 years of age.
So North Star in an older population, in a declining population, is a perfectly reasonable method to measure Duchenne. Now, I think the controversy becomes when you're measuring 4- to 7-year-olds. And we all know how can you get a 4-year-old to do anything quantitatively once, yeah, never mind twice. Right? So I think North Star for a 4- to 7-year-old population is a challenging getting a kid to come off the floor, you can get them to do that. But getting them to do 17 different operations is not so easy.
Yeah. That's a great perspective. Yeah. Yeah.
So I think North Star for adults and perhaps even older Duchenne is a perfectly reasonable measure. I think it's a challenging one for the younger kids.
Yeah. So what do you think the FDA wants to see from CANYON?
We've seen that they can be pretty, a different division can be fairly flexible and appreciate the totality of the data and look at the subcomponents of the North Star, like time to rise,
and I think we're in a really interesting situation. Right? Let's say you get to the end of the year in your Canyon data. You hit your primary. All your secondaries are looking positive. Now you've got Becker, devastating disease, no standard of care, no approved therapies. These individuals are declining right at the peak of their adult life, actually. Right? You're a 30-year-old guy. You're playing with your kids. Next day, you start seeing a drop in your gait. All of a sudden, 5 years, you're in a wheelchair. So you can make that argument and say to the FDA, "Look, I've got my pivotal cohort fully recruited.
The totality of the data supports potentially an early approval or an accelerated approval." I think we're going to go up to the FDA and have that discussion. At the least, what it does is it helps de-risk the GRAND CANYON trial. We can go in, based on the data from CANYON, and amend the statistical analysis plan because we'll know what things are more sensitive to change versus others.
It's a proviso. I mean, it is a well-tolerated drug so far. So there's very little downside to providing this drug to a patient. I think the most recent ALS discussion, in some respects, I do think this is the way accelerated approval should be used, though.
If a patient population needs a drug, the agency ought to be open to aggressively treating those patients with something that is safe and well-tolerated and could provide benefit based in a preliminary evaluation. Then you have a study that is larger, provides support for that initial observation. If it doesn't provide support, the outcome is it was too premature to approve the drug. But if it does provide benefit, then you've given patients an opportunity two years earlier, which is important. So it's actually functioning as I would think it should function. I mean, this is saying in the Duchenne community, muscle is time. And so if you've got a Becker patient and the drug is safe, why not let them be on the drug and the totality of the data supports it? And then let the pivotal run its 18-month course. So think about it.
1 point is, "I can't get up out of a chair without help." 1 point is, "I can't step on a curb as I'm crossing the road. I can't get my leg up. I can't even walk across." Yes. It's doing many different things autonomously versus requiring assistance. And 1 point's a year. Right? So with the progression of the disease, that's a real benefit if you start earlier. Right. And 2 points is not being able to do anything at all. Yes. That's right. That's the difference. So it's worth it, I think, to the patient. Yeah.
Well, super exciting time for Edgewise. Thank you for the update.
Thank you. Thank you, Joe. Great.