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

Okay. Hey, good afternoon, everyone. Welcome to the 5:30 session on day 2 at the 42nd Annual Global Goldman Sachs Global Healthcare Conference. My name is Graig Suvannavejh, and I cover U.S. and European biopharma here at the firm. It's a great pleasure of mine to be able to host Edgewise Therapeutics today in a fireside chat. And with me, we've got 3 members of the management team: Kevin Koch, CEO of the company; Alan Russell, Chief Scientific Officer; and Behrad Derakhshan, who is the Chief Business Officer. So with that, I want to welcome the 3 of you, and thanks for joining.

No, thank you.

Thanks, Graig.

It's great to be involved.

Thanks, Graig.

Yes. Great. Maybe to start, and as Edgewise is a relatively newly public company, there are perhaps some folks here on the webcast that may be new to the Edgewise story. So with that, I'm happy to turn it over either to you, Kevin or Alan, if you want to talk about who you are, kind of where you came from and what are the key focuses for the company at this point in time.

Yes. So I'll get started. Thanks a lot, Graig, for the discussion, and great to be part of this presentation. So my background, I started off as a medicinal chemist, worked in Amgen and Pfizer for a decade. And then I was one of the co-founders of Array BioPharma back in 1998. I was the President and Chief Scientific Officer there for 15 years, put over 25 molecules into clinical development. 6 of those are now on the market. And of course, they were acquired by Pfizer a couple of years ago. In between, I went to Biogen as the Senior VP of Research and Translational Medicine for 2 years. And then over the past 5 years, I have been the OrbiMed Venture Partner. Many of you know OrbiMed. And I saw this concept when Alan brought it to our attention about 4 years, maybe 4.5 years ago now. And I really liked it. It really stems from an observation that Alan made in the clinical data in Duchenne muscular dystrophy patients back in the 1980s. And it's turned out to be a very fruitful place to do drug research. So Alan?

Just going to not be muted. Alan Russell, hi, I'm the science guy. I was at Cytokinetics for 9 years, GSK for 7 years. Pretty much most of that time being doing skeletal muscle research. So really, from the early days of skeletal muscle interventions, I've been kind of on the pharmacology pharmacy side. And it's been a great ride. As Kevin said, we started this effort up in 2017 with a very directed approach towards muscular dystrophy, and it's been pretty successful so far. We're quite excited.

Thank you very much for that little introduction.

Okay. Behrad?

It's Behrad Derakhshan, Chief Business Officer, joined the team back in September after spending several years in multiple biotechs, exclusively focused on rare diseases. So good to be here, Graig. Thanks for the invitation.

Absolutely. So the company is really focused a lot on kind of muscle biology, muscular diseases. We actually, earlier today, had, Alan, your predecessor company or Cytokinetics, and just the concept of working on muscle biology I think has had, maybe its share of fits and starts. But maybe walk us through what you find so exciting about -- with your approach on being able to translate what you might be able to -- have been able to show preclinically and how you think that's going to apply at least -- in Duchenne muscular dystrophy at least initially.

Sure. I'm sure many of you guys know about Duchenne and about the genetic kind of lesions that lead to that. It's a very well-known X-linked myopathy. There's a number of other related myopathies. And of course, as genetic disorders, the inclination is always to fix the gene, right? You can't argue against that. But there's unfortunately a couple of things blowing against you in order for you to efficiently do that. One is just the sheer size of the gene. Dystrophin is an enormous gene. And there's many different lesions that cause the disease as they do for Becker. The other is just the sheer amount of muscle that we have, right? It's over 40% of our lean mass. And for AAVs and even for ASOs, it's quite a challenging compartment to deliver to. So we decided to take a very different approach, which really looks at these myopathies as structural disorders. Many of them lack or have dysfunctional structural genes. When muscle contracts, it has muscle injury and breakdown. We view muscle in many ways as a very organized machine. And what we found with 5506 is a small molecule approach that puts a patch on the machine, right? So we're taking a structural rather than a genetic approach. Now why would that be a good idea? Of course, if you can have a small molecule in these types of areas, that's really great. Easy to take, tend to be lower side effects and, in our case, applicable to a wide population. The other is that it doesn't really matter what genetic lesion you have. It doesn't matter if you're on a therapy, we really have a very orthogonal novel strategy that's potentially combinable with a number of different agents. So it's a real-world solution to a very complicated problem.

With that said -- you're still a relatively early-stage company. But with that said, it would be great if you could share with us some of the data that you have generated in support of 5506, particularly not only just in showing the potential in DMD, but perhaps other muscular dystrophies as well.

Sure. I'll take that one. So because of the nature of the strategy, because it's kind of a novel way of looking at it, our emphasis has been on gathering quite a lot of preclinical data to get an idea of what the compound does, right? We've largely worked in models of Duchenne. But as you mentioned, Graig, there's no reason that this strategy shouldn't work in limb-girdle muscular dystrophy or Becker muscular dystrophy. And as Kevin will tell you later, Becker is certainly our opener on the clinical front. But in terms of what data we've gathered preclinically is we've really covered the boards. We've tried to use every model available, every strategy to measure the relationship between muscle contraction and muscle injury. So we've done experiments in mdx mice, the more severe mdx 2J mouse and also the Duchenne dogs as well. And really, there's a consistent theme amongst these data, and that's when muscle contracts in the presence of 5506, it's injured less. There's less membrane injury. There's less membrane rupture. There's less membrane breakdown. And then with longer-term treatment, there's less remodeling associated with fibrosis and inflammation. And in the dogs, you even see decreases in biomarkers coupled with increases in activity. So a very kind of holistic data package that gives us an idea of how much drug we need, how much modulation and muscle that we need and what drug levels equate to that.

I think there is an outstanding or ongoing investor debate about perhaps the translatability of the preclinical animal models into what you might be able to see in the human studies. I'd love to get the Edgewise perspective on how best to use the preclinical animal models and what you can really take from that to inform of your view in clinical studies.

Yes. This isn't the first area of where the models don't always predict the result, right? The classic model for muscular dystrophy research is the mdx mouse. It's a mouse with no dystrophin. When you exercise the muscle, it injures itself. But the big difference between the mdx mouse and humans is that the muscle repair in that model is very, very good. So even though the mouse is getting injured all the time, it repairs that muscle particularly well. So the mouse has a normal life span. Doesn't even have that much fibrosis. Whereas, of course, in the humans, that's what really causes the functional limitations, is the stem cell exhaustion and then the fibrosis associated with that. That's in many ways why the DBA/2J-mdx mouse was developed. So this is a different background strain for the mdx. And what happens in the DBA/2J mouse is that because of problems with TGF-beta pathways, their stem cells really don't work very well. So they're very much more like an advanced human disease in that when the muscle becomes injured, it can't repair itself. And in the DBA/2J-mdx mouse, they get both cardiac fibrosis and skeletal muscle fibrosis and it really limits their function. So in many ways, that's a better model that we view to recapitulate the human disease. And then finally, the dogs, we think, are the best of the bunch, and that's largely because their muscle composition is very similar to us. The progression of the disease is unrelenting, much the same as in humans. And they get really quite dysfunctional muscle fibers, which is coupled to physical weakness. And of course, they're a large mammal as well which helps. So you have these 3 models. Each of them has its own use. Of course, mdx mice are really easy to work with. You can just get them straight over the counter from JAX, for instance. If you think about the way our approach works, what we do is we fix the structural problem. That is the same whether it be an mdx, a DBA/2J-mdx or a golden retriever with muscular dystrophy. So in many ways, all of the models are appropriate for our therapy. But the outcomes in terms of the benefit I think are more relevant in the DBA/2J and the dogs.

Thank you for that.

And really one of the most exciting pieces of data we generated was we dosed the DBA mice, which is a much more severe form, more like humans, for over 15 months and essentially cured the mice, which is I think the first example of that [ than any others ] have actually demonstrated. So it was quite a profound result, and we were quite excited to see that.

Kevin, let me -- since you're a well-respected chemist and have proven track record, I would love to just ask you a few questions just on the molecule itself and kind of how you came to 5506 in terms of that being kind of the optimized molecule. Can you kind of, for the nonchemists here in the room, just describe what are the things that the company did to ensure that this indeed was the candidate? Or is there even a backup that you're considering?

So we took the approach of only working on novel targets and muscle. And our platform was to build phenotypic screens, screen libraries and molecules to identify molecules that modulate functional effects in muscle directly in muscle fibers. We then would use the platform to dissect out the precise molecular target of the particular molecules and use our biophysical platform and assays to understand how we want to position a particular mechanism into a particular disease. That led to the identification of a lead molecule that was related to 5506. We used very classical methodology for lead optimization, which would be classical SAR. But we have very precise ways of going from biochemical to thin fiber isolated filaments to the whole muscle preps to in situ in vivo models and then with chronic genetically engineered models, which allowed us to optimize the drug. So the drug was optimized for a set of properties. One would be high permeability, which is in high solubility, which often guarantees high oral bioavailability. We, of course, look for potency, but we wanted selective potency against the fast muscle fiber relative to the slow muscle fiber, and the slow muscle fiber is the same as cardiac. So we eliminated all cardiovascular effects but wanted only fast skeletal muscle activity. We then take that molecule and run it through a myriad of hundreds of different externalized assays and found that the molecule was very well behaved, was very well behaved in, in vitro P450 assays, induction assays. We did extensive work in multiple preclinical species and ultimately came up with a molecule that is perhaps one of the best molecules I've ever been involved with. This molecule has got high oral bioavailability, solubility, permeability, hits no other targets that we know of. It is a molecule with no P450 interactions that are deleterious. So it is really a pristine molecule. And given the new -- the single ascending dose data in people, it has all been recapitulated in what we saw in the preclinical species. So I think it really is a best-in-class molecule right out of the gates.

Okay. That's a great perspective of the molecule that you're able to build. You are in a multiple ascending dose portion of your Phase I study now. I think you initiated that a couple of months back. Just help us understand what are the key attributes of that study and what you're really hoping to show from there as to be able to inform of the next Phase II trial.

Yes. So as Alan described, when we ran the preclinical models, we looked at target concentrations that we wanted to attain in the preclinical models where we saw the efficacy benefit. We also took tissue samples from the different efficacy species. And so we wanted to understand what were the tissue concentrations, so not only the blood concentrations but how much drug is in the tissue. We then now go back into the multiple ascending dose study and really look at a couple of different things. One would be the concentration. So can we achieve the concentration in the blood that we achieved in the preclinical models? Can we do that in humans? Second, can we achieve the concentrations in the tissue that we achieved in the preclinical models? And then finally, we introduced a pharmacodynamic measure of force reduction in the humans. And so what this is, is an electronic stimulation of the quadricep where you see essentially a twitch response or a reflex response. And what we showed in our previous discussions on the road show was that we could, in a very highly dose-responsive way, decrease involuntary force without affecting any other endpoints of voluntary muscle function. So we measure things like grip strength, strength in the shoulders, strength in the hips. No effect on any of those voluntary functional effects, yet we've had a dose-responsive decrease in the twitch response. So we know that we're engaging the target. And we understand that, that engagement, we have a range in which we want to achieve, that allows us to set the dose for more advanced studies. So the next phase after we work in the normal healthy volunteers will end up being in a Becker population, which we will evaluate later on in the fall. So that will be the first time we get the drug into actually disease population. So we'll select the dose based on our preclinical and our multiple ascending dose data in normal healthy volunteers. Then we'll look for a couple of things. One will be, is a patient with less muscle the same as a normal patient? That's a key aspect of looking at the safety of the drug. Second key thing will be, do we have any effect on the disease markers that we were measuring back in the preclinical models? So do we block muscle damage based on the biomarkers in the blood? Do we have an effect on the overall proteomics of the plasmal compartment in patients in the same way we saw in the dog? And what we believe is that we will see a movement in the right direction. It's only a 14-day study that we're going to get. But we think ultimately, the goal here is to demonstrate that everything we saw preclinically can be ultimately be observed in humans, and that will allow us to set dose for future Phase II studies in both Becker muscular dystrophy and Duchenne muscular dystrophy.

And can you set for us the approximate timing, you think, that we would be able to kind of get a sense of how the compound is performing? Is it by the end of the year? Is it early next year in terms of some of the Phase I data?

Yes. I think we're tracking very well right now for an end of the year discussion, and we'll probably have a -- probably a conference call with the disclosure of the data. There's no natural meeting to talk about that, but we would want to provide that information as an update to everyone about what's going on.

Now there may be a bunch of investors who have been following the DMD space closely and certainly some later-stage programs, some programs being developed by some larger companies. Alan, I think you touched upon earlier, just briefly, kind of how you think this program might fit within the potential treatment paradigm. But I'd love to get the Edgewise perspective of kind of the pluses and minuses you see amongst the different programs, where they are now. And certainly, given the stage development that you're at right now, there could be a discernible gap or maybe the gap isn't nearly as large as people might think. So with that, I'd love to get the perspective from you on how you see the current DMD competitive landscape.

Behrad, why don't you take this and talk about your perspective and how we think about positioning of our mechanism relative to everything else that's out there?

Sure, of course. So Graig, I think maybe the point to emphasize just right off the bat is despite the availability of approved therapies for DMD today, and we're talking about steroid supplementation or the various mutation-specific antisense, [ all arguments that are out there ], the unmet need for efficacious and disease-modifying therapies remains extremely high, right? And this is equally true for BMD because, as you know, part of our strategy is focused on BMD, where you don't have any approved therapies today. So as you point out, Graig, there's also a pipeline of products that are potentially disease modifying at various stages of development. And these include whether it's a gene therapy that's delivering a micro or mini dystrophin, which essentially yields a smaller dystrophin protein or next-generation exon skippers, which are designed to improve like the skipping efficiency vis-a-vis the current LCA compounds or whether it's a CRISPR-based approach that we've seen a little bit of hype about in recent years. I think as we -- Alan and Kevin have both pointed out, our approach for 5506 is really aimed at addressing the root cause of dystrophic muscle through stabilizing the most disease-susceptible muscle fibers, right, the type 2 fast fibers, and then essentially protecting the muscle from damage, independent of the dystrophin status, right? So this is mutation agnostic. This is really focused on the site of the insult which is the muscle. So as such, when we think about positioning 5506, given our slogan of nondystrophin-targeted approach, we see potential for use as either a monotherapy or in combination with approved treatments that are either approved or currently in development. And this could be either as an add-on or in the first-line setting. And obviously, that will evolve and it will depend on kind of the environment when we launch. I think the question around the efficacy and safety of 5506 relative to what's in development, right, and why we believe there's still a tremendous amount of unmet need in the DMD space, I think, Graig, as you know, and more for some of the people tuning in, the current standard of care for BMD is really steroids. And with steroids, you see a very acute improvement in functional decline, but the chronic and the sustained use of these products is associated with extreme side effects. And so it's very much limited to a very small subset of patients where you see an acute benefit, but the long-term detrimental effects are actually quite problematic. And as such, you see a tapering off of kind of steroid use in the later years. If we look at the exon skippers, which there's been a little bit of control obviously in recent years around some of those molecules, there are 4 that are conditionally approved today by the FDA. And today, we've actually seen limited evidence of functional benefit with these approaches. And actually, when you look at the core of it, given that these are very mutation specific, they're only targeting, at best, close to 29% of the total DMD population. So that leaves the majority of the DMD patients still unavailable to exon skipping or not having access to an exon skipper. So I think when you think about gene therapies and what the promise is there, and if they're successful, DMD patients treated with gene therapies are likely to recapitulate more of a BMD-like phenotype. And thus, these patients will continue to have a tremendous unmet need and will require an additional intervention. And I think it's against this backdrop that I've laid out about the standard of care in the pipeline where we've generated probably one of the most complete preclinical data packages, as Alan highlighted, in some of the most rigorous animal models showing that not only are we protective by inhibiting type 2 muscle fibers and allowing a much healthier muscle, but also from the Phase I data that Kevin discussed, we're also much safer. It seems like the safety profile from what we've seen in Phase I appears to be generally well -- safe and well tolerated. So we've got a lot of confidence kind of looking at the totality of what we have today and think about positioning 5506 as really kind of the backbone of future DMD therapies.

I'm sure there are going to be questions around on the efficacy side with your compound, how it might stack up relative to some of these other approaches. Do you think that is necessarily one that's a critical piece to look like -- to look at? Or is it in the totality of the safety and efficacy and the long-term effect you might end up having?

I think realistically, it always is a combination of the risk benefit for any particular therapy. But I think that from what we've seen preclinically, the efficacy has been differentiated and outstanding. What we're seeing clinically with the safety is that it's a relatively benign safety profile to date. And this is really important for certainly pediatric populations and uptake of the drug into the community. So this combination will ultimately win the day for any mechanisms like this. Now I think how we think about this is muscle damage is the hallmark of the disease, meaning, that you're losing muscle to when it functions, it damages itself and then does not repair itself effectively and over time, fibrotic tissue replaces the muscle tissue. It feels to me that the drugs that will actually prevent that muscle damage will be the ones that can be used most broadly. And that ultimately will be the key to our success, and we should know that over the next 18 months.

Yes. It's a very provocative picture that you're painting in terms of what 5506 could look like. And certainly, we'll be interested in seeing the data whenever it becomes available later this year. We've got about 10 minutes of time left, and there's more to Edgewise than just 5506. And you've got a 002 program that looks at some cardiovascular diseases. And again, we did have Cytokinetics on earlier today during our health care conference. But would love to get a sense from you as to where your efforts are focused. And since you are going after hypertrophic cardiomyopathy as well, how do you see -- or how are you trying to design your program to be differentiated from either that compound or the BMS compound mavacamten.

So as we described, we try to focus on entirely novel targets and understand how we might utilize [ these in disease ]. So when we ran our initial phenotypic screen, we had first wanted skeletal muscle selective molecules that were fast myosin inhibitors. At the same time, because of the nature of the screen, we identified cardiac-active molecules because, of course, you have a counter screen to eliminate the cardios, but then you find the cardios. So that leads to now I have novel cardiac active molecules. So we use our platform now to dissect how do they interact with the cardiac muscle. And we were able to find molecules that look very similar to MyoKardia in Cytokinetics in that way we're integrated with cardiac myosin inhibitors. We've found another set of molecules that were not cardiac myosin inhibitors, yet still decreased a force within a cardiac myocyte so they could be utilized in exactly the same way as the myosin inhibitors. Now as we study these more in depth, what we found that they have certain advantages over the myosin inhibitors that Alan probably can describe a little bit better than me about what are the differences and how we might position this particular set of molecules.

Sure. And I guess, first off, it's worth saying that mavacamten and CK-274 are really exciting compounds. I mean it's amazing that you can intervene in this broad HCM population with a single agent, right? But the way we see it is that this is the start of our learnings in many ways of how you can modulate the sarcomere for therapeutic benefit. We had a few false starts, but there's definitely some traction building amongst the few different companies. And certainly, our alternative strategy has a couple of differentiation features just straight off the bat, which are kind of promising. One is that the dose response is different. CK-274 is clearly making some inroads into real-life use by shortening the half-life and allowing for kind of an easier dose titration, right? We're going along that angle, too. We believe that we should have an easier drug to dose in terms of titration. But importantly, you can rescue the effect that we have. With the myosin inhibitors, one of the problems is once you've got an inhibition, it's quite difficult to rescue that because what you're doing is a primary block of the motor, right? We have this alternative strategy and what we've seen in rescue experiments when you have high levels of inhibition is you can rescue that effect, which allows you to dig yourself out of problems if you get to those, right? So we believe that the approach should work similarly in large -- in broad HCM populations, but a few real-world differences that make it easier to use.

That's very intriguing. Can you remind us kind of where you stand in terms of the progress of this program and what we should expect to hear next?

Yes, we're heavily into what we would call a lead optimization phase. So we're into, of course, things -- animal models like dogs where you can measure echo and other cardiovascular parameters very precisely is where we're in right now, teasing apart the exact profile of the drug we want. And now we're hopeful to have a candidate molecule by the end of the year going into GLP tox the first quarter and then sometime in end of '22, beginning of '23, an IND filing for a cardiovascular drug that would have some competitive advantages over the current myosin modulators.

And maybe just a last question on this particular program. As we think about HCM, there are different subpopulations of HCM, and I'm wondering where do you see the utility of this program. Would it be across all or would it be in certain subsets?

Certainly, our data suggests we're in a similar bag to mavacamten and 274, right? Of course, that -- I think it benefits us to do quite -- much the same as with our lead program, quite a lot of preclinical data to fully interrogate that. There's no reason to believe that we shouldn't have that broad effect that those 2 assets have. I think there are some unexpected findings to be had with further digging, and we can probably capitalize on those as well. This is a new area, and there's lots of unexpected things that kind of bubble out. So we're quite excited about that program.

I think what's key is that there's been a renewed interest in targeting genetic subsets in cardiovascular disease. And you've heard from a number of companies who have been quite intrigued by our results that they feel that this is a growth area and that you now can carve up what is a seemingly very large population of people with heart failure into subgroups where relatively small trials with the appropriate mechanism can gain approval be quite valuable. So like everything else, maybe the oncology field led the way of targeted populations that's now becoming the norm in other very large populations in a number of different diseases.

We've got just a few minutes left in this fireside chat. I did want to highlight that a little over a month ago, you did bring on a new Chief Medical Officer. I'm just hoping if you can provide a perspective of what she brings to the company.

No, I think it's great. The interaction with the patient advocacy groups and having run clinical trials of Duchenne muscular dystrophy is a rare skill set. Joanne Donovan not only brings that skill set as the former CMO of Catabasis, but also has a broad background in rare disease development from Genzyme. So some 30-odd years of drug development in rare disease, in general. She is extremely well connected with the patient advocacy groups, the parents of the patients and, of course, had, had discussions with the agency in regards to the development of the NF-kappaB program at Catabasis. So that combined background, plus some of the people that now also joined Edgewise, who were part of her team in clinical development, clinical operations, program management, all of those things allow us to accelerate our development program and not learn on the fly. We have someone who's already been there and done it. So I'm very confident about our team in the clinical side right now.

Sounds like an excellent addition to the management team. Maybe just some last housekeeping questions. And I'd love to ask just -- I know you just completed your IPO, but could you remind us where you stand on cash right now, where does that cash runway take you? And I know we've spoken to kind of the next catalysts or milestones for the company, but could you just summarize those again for us in terms of the next 12 months?

Yes. So we have about $310 million of cash. That will take us through well into '24. And this would mean the most aggressive stance that we could apply for, meaning pivotal studies in Becker, pivotal studies in DMD as well as moving towards proof-of-concept studies with our cardiovascular asset. So -- and also being able to continue doing research on novel targets and fill our pipeline. So I think we're in a really good position right now and very pleased with the fund raise. And we've been very well supported by the investors over the -- since the IPO. I think it comes from both the quality of the team and I think the enthusiasm about the assets that we're creating.

And then milestones over the next 12 months or key events?

What we're looking for, by the end of the year, we will put out a healthy volunteer data, multiple ascending dose and Becker data. We will initiate a Phase II study in Becker, which will be a targeted population that's genetically defined where we think we can see a window of improvement in the Becker population, where there is no standard of care. That will lead to a discussion with the agency in regards to how we would use this aggregate of data in adults to position ourselves for initiating studies in Duchenne muscular dystrophy, which is obviously a pediatric population. And the hope would be that the data we collect in adults will allow us to select defined doses in Duchenne and jump directly into a Phase II/III study or some type of study where we would have an adaptive design with a run-in phase and then ultimately be able to treat DMD patients quite quickly. So that would be essentially a parallel track of both a Becker approval and a Duchenne approval in the '23, '24 time frame.

Okay. Great. And if I could leave you, Kevin, with perhaps one last question that I've been asking all of my companies here at the conference, which is providing a 5-year vision of if we were sitting in Southern California together doing this in person in 2026, what would you hope Edgewise to look like by then?

Well, I would anticipate that 5506 is on the market and approved for Duchenne and Becker muscular dystrophy. And I anticipate we would be in a proof-of-concept study for our cardiovascular asset at that time and perhaps even further with a pipeline behind it of additional novel mechanisms.

Okay. Great. Well, that's a very inspiring vision. I look forward to the progress that you make on that vision. But with that said, I want to thank you. I want to thank Alan and Behrad for joining us today and the audience as well. This wraps up day 2 at the Goldman Sachs Global Healthcare Conference. Thank you for joining, and hope to see you tomorrow. Thanks, everyone.

Thank you all. Appreciate it.

Thanks, Graig.