Okay, great. Good morning, everybody. My name is Kristen Kluska. I'm one of the biotech analysts at Cantor. Very happy to be joined by Fulcrum Therapeutics. They have a big Phase III readout in FSHD expected later this year, so really appreciate you joining us. And from the team, we have Alex Sapir, the CEO, Dr. Iain Fraser, the SVP of clinical development. And new to the team, we have Dr. Pat Horn, who's the CMO. Welcome, everybody.
Thanks for having us, Kristen.
Yeah, so I was just saying a big part of me doing this symposium was really to talk more about other forms of muscular dystrophy, and FSHD is one of them, and there's nothing out on the market for that. So maybe to start our conversation, I'm hoping we can learn a little bit more about this indication, comparing it to other muscular dystrophies that people on the phone may be more familiar with, and maybe tell us about the prevalence and other unique features.
Sure, absolutely, Kristen. And as I said at the outset, thanks so much for having us. Yeah, so FSHD prevalence in the U.S. specifically is about 30,000 patients. It is the second most common form of muscular dystrophy, second only to DM1. And the way that the disease is characterized is this slow yet very relentless loss of muscle function year after year. There's a couple of natural history studies, and maybe we'll have a chance to talk about that in a little bit, that essentially sort of show this single digit, mid-single digit loss of muscle function for these patients year after year. Most of the loss of muscle function for these patients typically occurs in the shoulder girdle, and where it's really impacted is the ability of a patient to sort of reach out above their head.
So while there is some heterogeneity, and I know that's one of the things that you probably want to speak about is just the heterogeneous nature of this disease, I think the two things that I just described in terms of what patients experience, this slow mid-single-digit loss of muscle function, primarily impacting sort of the shoulder girdle, loss of an ability to sort of reach above them, those I think are the two sort of common things that you see in most patients with FSHD.
Okay, thank you for that. What does muscle dysfunction look like in FSHD? We know it involves aberrant expression of the DUX4 gene. What effects does it have in terms of turning on other genes and leading to muscle cell death, fat infiltration, etc.?
Yeah, great question. I'm going to turn that over to my medical colleagues. Maybe Iain, if you want to.
Yeah, thanks, Alex, and thanks, Kristen. So as you mentioned, FSHD is caused by aberrant expression of this transcription factor called DUX4 that is normally expressed only very transiently in the pre-implantation embryo, and then it's silenced for the rest of life. In FSHD, due to a variety of different genetic mutations, that is expressed in skeletal muscle later on in life. And the presence of DUX4 as a transcription factor activates a downstream transcriptional program. So it turns on and turns off genes within the muscle that ends up disrupting the myogenic differentiation program, so the muscles don't develop, and they also don't fuse. As part of muscle development, the muscle cells fuse together. These pathways hit on a number of cellular mechanisms that cause damage, including oxidative stress, changes in the sarcomeric organization of the cells, as well as cell death.
At the end of the day, you get an impairment in the contractile function of the muscle, so the muscle loses its ability to move. The muscle cells themselves die, and then they get replaced by fat. And so that's the pathogenic pathway, all triggered by the activation of this toxic transcriptional program.
Okay, thank you for that. Yeah, I did want to revisit the pattern of heterogeneity here. How does it come into play for the onset and the disease progression? When do you typically see patients starting to see the effects of the condition?
Yeah, so I'll continue on that. So most characteristically, the disease onsets in the second decade of life. It can be earlier, so there are more severe cases that present early on in childhood, and there are also less severe cases that present much later in adulthood. But characteristically, it's in that second decade of life that folks are diagnosed. There's one additional layer of complexity here, and that is about two-thirds of the cases are inherited in a familial autosomal dominant fashion, and about a third of them are sporadic. And so you can imagine in the setting of a family with an autosomal dominant condition, where folks know about the disease, they're more alert to picking up the signs and the manifestations of the disease.
And they'll tell you that they noticed in their infants or even earlier in life that they didn't close their eyes quite as tightly as they might otherwise would have. They noticed subtle weakness in the facial muscles that somebody else might not pick up on. So in the familial cases, there's a tendency towards recognition of the symptoms earlier on. They don't necessarily get formally diagnosed earlier on. And one of the challenges around that is that because there are no therapeutic options for these patients, there's little incentive to make that specific diagnosis of FSHD. And so you'll hear stories in families where they'll tell you, "Yes, I picked up the signs and symptoms in my kids pretty early on, but I've never had them diagnosed because there's been no need to do so up until this time." So I think that's an important component of that.
There is, as you mentioned, when we talk about heterogeneity, there can be asymmetry between left and right sides, where somebody might present with one or the other shoulder, for instance, being more impacted than the other. As the disease progresses, it can do so in an asymmetric fashion as well, with progression on one side more than the other at one particular time point, and then later on it's different. The most characteristic presentation, which Alex alluded to earlier, and which is captured in the name of the disease of FSHD, is in the face and then the muscles around the shoulder joint, so the scapula and the humerus. It's really that with a descending pattern of progression that's most characteristic in FSHD, but there are variations of that.
I think folks might be familiar with the fact that this shoulder weakness is sometimes associated with winging of the scapula. The muscles around the shoulder joint don't hold the shoulder blade in place as well as they normally do, so it wings out. The arm weakness usually involves the biceps and the triceps, so the big muscles around the humerus, and less so the forearms. That's another characteristic aspect of FSHD. You can pick it up in the abdominal muscles as well as the disease progresses downwards. Then in the lower extremities, it's usually manifesting as a distal weakness before it manifests as a proximal weakness. The most common there is a foot drop caused by weakness in the tibialis anterior muscle. You'll sometimes see that patients will have a lot of upper limb shoulder abnormalities, but they'll also be complaining of a foot drop.
There is this systemic aspect to the disease. It's not limited to the shoulder, but that's clearly where the characteristic manifestations occur.
Okay, thank you. And there's also some heterogeneity on the severity. About 20% eventually require a wheelchair. So in terms of available natural history data, what are the expectations around the rate of decline in muscle function?
Yeah, great question. Maybe I'll take that one, Iain. Yeah, Kristen, you're absolutely right. So about 20%-25% of FSHD patients ultimately become wheelchair-bound, and a much larger percentage require full-time care. So they're simply unable to take care of themselves. I would say that there are probably two important natural history studies that look at the progression of the disease using an instrument that we call Reachable Workspace. Reachable Workspace, in its simplest form, is it measures a patient's ability to reach in front of them in four different quadrants: two above the head, two below the head, and then one behind, essentially for practicing good hygiene. There's two natural history studies that have looked at this RWS instrument in terms of looking at patients' decline over years.
The first one was a natural history study that was published in 2019 in Neuromuscular Disorders by Maya Hatch, a small number of patients, 18 patients, and they looked at those patients over a five-year period. And then the other natural history study is our own placebo arm of our Phase II study that looked at 40 FSHD patients over a period of one year. And essentially, what both of those natural history studies showed was essentially the same thing, is that across all of those five quadrants that I mentioned, patients are experiencing probably about a mid-single digit decline, but that mid-single digit decline is happening year after year after year after year.
If you look at the two quadrants that are most impacted, and Iain talked about this, right, really the shoulder joint, and that's really these two quadrants above the head, quadrant one and quadrant three, and you just look at the RWS decline across the patient's ability to reach above their head, that number actually goes up. So it's somewhere in the sort of high single digits, in the 8% or 9%. But I think what's really important to remember is that, you know, while that may seem like a relatively small number within a given period of time, that number, that loss of muscle function that I just mentioned across these two natural history studies, that loss of muscle function is relentless, happening year after year after year.
Okay, thanks. How much of the heterogeneity could be due to diagnosis? When do patients typically enter the healthcare system? What changes do you see in patient journey and the diagnosis side of the equation?
Yeah, yeah, it's an excellent question. So I would say that I'm not sure if we just lost you or not, Kristen, but I, whoop, there we go. You're back on. Great. Yeah, I would say that heterogeneity does impact overall time to diagnosis as well as the overall patient journey. As Iain mentioned earlier, most patients are presenting in their sort of mid to late teens, although there are a number of pediatric patients that present earlier that tend to be much, much more severe. But most patients are presenting in the sort of mid to late teens. The average time from symptom onset to diagnosis is around five years. However, if there is a family history, and Iain had mentioned this, there is, you know, this is an autosomal dominant disease. So if one of your parents has it, it's a 50% chance that you will get it.
The time to definitive diagnosis, as you would expect, if there is a family history, that number goes down quite significantly. You know, the way that you definitively diagnose this is through genetic testing. What's really interesting is that genetic testing is not as common as you would expect. I think that's really for two primary reasons. Number one, it's very sort of clunky and complex from a payer standpoint. Payers are somewhat sort of reluctant to pay for it because it is a rare disease. Probably more important is that once you get that definitive diagnosis with a confirmed genetic testing, there's really not much that can be done with that diagnosis because there are simply no therapeutic options. The question is, why would you get that confirmed genetic testing if there's really nothing that can be done with it?
We would expect that once you begin to see more products come to market for the treatment of FSHD, I think two things are going to happen. Number one, you're going to see a shrinkage of the time from symptom onset to definitive diagnosis. I think you'll also see a much greater utilization of confirmed genetic testing.
Do you think that research has been impeded by this at all? Iain, you want to maybe take that one?
Yeah, I can touch. It's a really interesting question. I don't think it's so much the heterogeneity of the disease, but I can touch on a few of those. I think it's important to remember that the underlying genetic cause of FSHD, which is really quite complex, and the pathogenesis of the disease has only really been known for the last 10 or so years. So it's not known as though this has been something that folks have really known a lot about. So in terms of developing mechanisms to address it, that's not really been possible until relatively recently.
The second aspect of it, and maybe we'll touch on this later, is that even though DUX4 expression is the cause of FSHD, it's very challenging to measure, particularly in vivo, but even in vitro, it has challenges as well, just because of the way it's expressed and it's expressed in a very stochastic and random fashion in cells. And so being able to get a handle on that in order to evaluate your therapies has been challenging. So I think those two factors have probably been the reason why we're only now in Phase III with the most advanced therapeutic option. In terms of the clinical aspects of it, you know, there's some contribution to that. The spectrum of disease severity potentially contributes.
We know that patients who are mildly impacted at one moment in time are less likely to show progression in a relatively short period of time, like a clinical trial about a year or so. We also know that patients who are very severely impacted, who are in wheelchairs, for instance, it's very difficult to measure their progression in the context of that time. There are ways you can focus in on the sort of middle range of patients, which are the ones that are more likely to progress. I think that's something that we've taken on board, and probably others will be doing as well, in defining that patient population. That's across the severity aspect.
And then if you're thinking about the way (and we've spoken about this already) the asymmetry of the disease and the progression and the involvement of different muscles over different times, I think there are some challenges there. So it doesn't lend itself well to testing one particular muscle in every single patient because it's slightly different in every patient. And I think that's one of the things where the Reachable Workspace has been particularly helpful, in that it sort of looks in a relatively integrated fashion around the shoulder joint. So it's agnostic to which specific muscle might be involved. It captures both left and right, so it deals somewhat with the asymmetry, and it's looking at the integrated movement around the joint.
I think there are ways that some of the endpoints now are able to address that and interrogate it in a better fashion than the more traditional single muscle type dynamometry testing or manual muscle testing. That's been an advance, but I don't think that that's necessarily held back the development of drugs, per se.
Okay, great. Thanks for all of that background. I did want to turn to losmapimod. So there's a few advantages here, one being it's an oral drug. So how does inhibiting p38 alpha beta ultimately inhibit DUX4 expression, and thus inhibit muscle cell death and FSHD? And what data do you have to demonstrate that losmapimod inhibits DUX4 expression without adversely impacting myogenesis?
Yeah, no, those are all critical questions. I think just to say at the outset that the exact molecular mechanism by which DUX4 is regulated by p38 is not fully elucidated, but there's a lot that we do know. And I can step through those various pieces and the evidence that we have that losmapimod does indeed decrease the expression of DUX4 and do so in a way that does not affect normal muscle development. So the MAP kinases, as we know, are really important signaling pathways that move signals from outside of the cell to the inside of the cell, and they modulate, among other things, gene expression. And they do that by impacting both transcription factors and also by affecting chromatin modulators, so methyltransferases, for instance, that might regulate gene expression.
So we clearly know that there's a link there between the p38 MAP kinases and gene expression. We also know that in skeletal muscle, there are a number of extracellular signals that activate p38, and these include things like exercise, like differentiation of muscle cells, like exposure to insulin, exposure to reactive oxygen species, and to cell death. So a number of these external things, some of which are very common, like exercise, activate the p38 MAP kinase system. And in FSHD, it's that activation of that p38 MAP kinase transduction system that acts, we believe, through specific myogenic enhancers to lead to the increased transcription of DUX4. So we don't think it's a direct action of the p38 MAP kinase on DUX4 itself, but it's acting to modulate transcription within the muscle cells that then drives the expression of DUX4.
So that's the link between p38 and expression of DUX4. We also know that from work done in vitro with FSHD patient-derived myotubes, so these are, you can get myoblasts from patients with FSHD, and you can differentiate them in culture. The myoblasts don't express much DUX4, but once you differentiate them into the myotubes, they do. In that system, the expression of DUX4 is very, very sensitive to inhibition of p38 MAP kinases. As you titrate in concentrations of MAP kinases into that system, you can see a concentration-dependent reduction in the expression of DUX4. That's really the primary therapeutic effect here. There are some effects on the activity of DUX4 that might already be expressed, but most of the therapeutic benefit is derived from a reduction in the DUX4 expression by itself.
I think one of the important things, and folks often say, but wait a minute, isn't MAP kinases involved in normal muscle cell differentiation, and isn't that a problem? And I think it's a good question, and it's been looked at by a number of folks, including our own scientists, but there are publications from other groups as well, that indicate, because of the sensitivity of the DUX4 transcription to MAP kinase inhibition, the concentrations of MAP kinases inhibitors that ultimately lead to disordered muscle differentiation are log orders higher than what you need in order to decrease the expression of DUX4, 2-3 log orders higher. And that's been demonstrated both in these in vitro myotube cultures, but also in vivo in mouse models. Now, they're not great models of FSHD because it's a relatively human-specific type of disease with unique genetics.
But there's a xenograft mouse model, for instance, where you can show effects of losmapimod reductions in DUX4, preservation of muscle, and no impact on myogenesis in that system. So I think there's both in vitro and in vivo data that speak to that. And then lastly, what I'll just touch on is really related to your question of how do we know that we're actually achieving that? And I think that alludes to the data that emerged from our Phase II study, where we tried to address that directly in vivo by using muscle biopsies in FSHD patients treated with losmapimod and interrogating the downstream transcriptional profile of six genes known to be regulated by DUX4.
The end result of that particular exploration was that the variability in that DUX4 transcriptional profile, both between the patients as well as within the patients from time point to time point, was so variable that it was not possible to detect the signal. I think the reasons for that are twofold, primarily. One is we know that DUX4 is expressed even in FSHD at a very low frequency. One in several thousand myonuclei are expressing DUX4. You're trying to find a needle in a haystack to some extent.
The second piece of it is that when you're sticking a needle into a muscle, particularly a muscle that's affected by FSHD, where there's a lot going on, and we specifically targeted those areas of the muscle that on MRI were showing evidence of disease activity, you end up getting back variable amounts of muscle, fat, fibrous tissue, inflammatory components, and so on. And so there's that variability in the components. You really want to see just the muscle because that's where the DUX4 is expressing, and you have all these other pieces coming along in a variable fashion. And so those were the challenges around that. It's much easier to do that in vitro. You don't have all these additional tissue types. It's easier to measure DUX4.
That's where we've been able to show across a range of myotubes derived from different patients, FSHD type one and FSHD type two, that you get this very clear concentration-dependent reduction in DUX4 in those cells. You prevent the cell death that occurs in those cultures, and you don't interfere with the myogenic pattern in those cells.
Okay, thanks. Let's talk about some of your clinical trial experiences. So you did the 48-week placebo-controlled Phase II ReDUX4 trial. Going into this, what were your expectations around what you would see on Reachable Workspace changes based on the patient population?
Yeah, no, I think that's great. I think we've touched on a couple of these. We'll just say that in the ReDUX4, the Phase II clinical trial, there were a number of functional endpoints that were evaluated either as secondary or exploratory. And one of them was the Reachable Workspace. As Alex mentioned earlier, there are some, albeit limited, natural history data. In the early days, it was based on manual muscle testing, dynamometry, more recently on Reachable Workspace. But the general consensus is that you see, as Alex mentioned, these single-digit, mid-single-digit % declines per year. And so I think that was the expectation going in, is that we'd like to see a recapitulation of that using these functional readouts in the Phase II study.
And I think that's really what we were able to show with the Reachable Workspace, is that it confirmed that mid-single-digit % decline using this novel instrument that has not been used previously for the registration of any drugs. So I think that that was certainly consistent with that. And since then, some of the smaller scale FSHD natural history studies that have used the Reachable Workspace have also confirmed that. So I think that the results on the Reachable Workspace aligned with the expectations of what the disease progression over that period of time was expected.
Okay, and then looking at 18 pairs of muscles, or 36, using whole body MRI as part of this trial, what were you expecting to see on muscle fat infiltration in normal muscles versus those already affected by FSHD? And then in terms of proof of concept, how does this give you confidence into a larger study?
Yeah, yeah, and maybe just step back a moment. We were discussing this issue of heterogeneity earlier. I think that's one of the things where the whole body MRI has a somewhat unique advantage because we're scanning the whole body. We're agnostic to which muscles might be involved, and we can identify muscles based on their MRI signal rather than having to focus in on one or two particular muscles along the way. So, as you mentioned, there are these 18 pairs of muscles that are pre-identified, and that's what we focus in on. Then just to give a little bit of background, we divide the muscles based on their MRI fat signal into three classes. There's a type A, which are the normal muscles. So those are muscles that have less than 10% fat infiltration by MRI. At the other extreme are the type C muscles.
They have more than 50% fat infiltration, and they're sort of considered as end-stage muscles because once you have more than 50% of your muscle replaced by fat, that's not really a functional muscle anymore. And then between the type A and the type C, you have this intermediate group that is showing more fat than you would expect from a normal muscle, but it's not quite reached that 50% level. And so we focus in on those intermediate muscles because I think that's where you're most likely to see the benefits of a therapeutic intervention. By the time you're end-stage, you're probably not going to be able to achieve much, and in the early stages, things might not be progressing as rapidly. And I think the data that we showed confirmed that.
The only real data we had prior to conducting that study in terms of fat infiltration was a large U.K. Biobank study that's been conducted to look at the changes in muscle fat that occur with aging. So they had a population of folks between the ages of 44 and 80 years of age and doing repeated scans. And in that population, the fat infiltration in those normal individuals, as they got older, was about 0.08% per year. So that was really the only benchmark we had on the absolute number of percentage increase in fat.
As it turns out, in the FSHD population, if we look at the placebo group from ReDUX4, so that's kind of our natural history group, if you like, in the type A muscles, the normal-appearing muscles, the fat infiltration in that group was about twice that observed in the UK Biobank natural history study, so about a 0.15% increase in fat. And then again, in accordance with what we thought going into this, the intermediate muscles, those that were more impacted, showed a higher rate of progression, about 0.5% per year. So that's about six times higher than what you'd expect based on the natural history data in normal individuals. And so that confirmed both that the progression in normal-appearing muscles was higher than in healthy individuals, and the progression in muscles that were already showing signs of disease was very much higher in those.
And then, of course, to loop it back to this study with losmapimod, in the losmapimod group, we're able to show that essentially the fat infiltration was flat over the period of the study, the 48-week treatment period. So that's, I think, an important piece, as you started out, the question of confirmation that you're actually impacting the pathology of the disease, which, as we discussed right at the beginning, is muscle dying and turning into fat and showing that, A, we could see an increase in the fat infiltration over time in the placebo recipients, and more importantly, that we could stabilize that in the folks who got the losmapimod.
Okay, thanks. So looking at your ongoing Phase III trial, there's a lot of similarities with ReDUX4. Can you talk about the key learnings that led to this trial design, and then what gives you confidence to get similar rates of change? And then, you know, essentially, can you walk us through powering assumptions and anything else that's important to know?
Yeah, so maybe just touch on the issue of similarity. I think obviously the Phase II trial, as I said earlier, had a number of endpoints, and we were able to evaluate those and pick those that we thought were the most likely to show change, the most meaningful. And certainly, functional endpoints are ones that we know regulators are particularly focused on for good reason. And so we were able to take the learnings from the particular endpoints and transfer them into the Phase III REACH study. In terms of the population, the inclusion/exclusion criteria are really quite similar across the studies. The Phase III study does include some type 2 FSHD patients, which the Phase II study did not. Type 2 is a small minority of FSHD. It's 5% or so. And we've stratified the randomization to ensure that they'll be equally distributed into the Phase III study.
So that's one small difference. I don't expect that to have any difference, make any difference on the expected rate of progression. The clinical manifestations are the same, and our in vitro work shows that those cells from the type two patients behave in the same way as the type ones. So that's one small difference. We have retained in both studies inclusion criteria that exclude the most severely impacted patients, as well as the most mildly impacted patients. And in both studies, we use the clinical severity score called the Ricci score in order to do that, and that's applied the same across both studies. So we're applying that exactly in the same way. There's an additional refinement for the REACH study that was not in ReDUX4, and that's based on a screening Reachable Workspace examination. So in ReDUX4, the first RWS was done at baseline.
We're now doing this at screening, and we have cutoffs established based on Reachable Workspace. Again, exactly the same as we're thinking about REACH, is to exclude the most severely impacted as well as the most mildly impacted, just because that's that middle zone that's most likely to progress and most likely to show a signal in a clinical study. So that's one small refinement that we've made in the Phase III, which just constrains the study population slightly and is intended to focus on a patient population that is most likely to progress. So that's the inclusion/exclusion around that. In terms of the powering, we used the magnitude and the variability, so the standard deviation from ReDUX4 that we saw on the Reachable Workspace in order to power the Phase III study.
When we originally designed the Phase III study, our expectation would be that there would be 210 type 1 FSHD patients, and we powered the study on the type 1s only because they were in Phase II. With 210 type 1s and 20 type 2s, a total of 230, we had about a 93% power with the assumptions from ReDUX4. As it turns out, the study has actually completed enrollment. We enrolled 260 patients rather than the 230. There was a lot of built-up enthusiasm and a lot of patients in queue at the sites when we cut off screening. So we've ended up with 242 type 1s, and there are 18 type 2s. Again, the type 2s are stratified, so they're distributed equally, and that pushes the powering up again based on the same assumptions from about 93% to about 96%.
So that's what we have in the REACH study relative to ReDUX4.
Okay, thanks. Yeah, very big readout for you in the fourth quarter. Given you have the potential to be first to market, what are you hearing from payers, the community, etc., about what potential uptake could look like?
Yeah, great question, Chris. And maybe I'll take that one. Yeah, so I think our assumption right now is, given where we are in development, as Iain mentioned, having completed our Phase III study and reading out that Phase III study in the fourth quarter of this year, we believe that we probably have about a 2-3-year head start over the next closest competitor, so we'll have this market to ourselves. I think the competitor that is most closely behind us is an anti-myostatin, which has not been shown to be successful in earlier studies. So if that anti-myostatin is not successful, it could extend that 2-3-year head start that we have even farther. In terms of what we're hearing both from payers as well as patients, maybe I'll start with patients first. You know, patients are very clear.
About two-thirds of patients are saying that they essentially are looking for a drug that has the attribute to essentially keep them stable or to slow the progression of the disease. I don't think that patients are looking for something that can improve them from their current baseline. I think patients are saying, about two-thirds of patients are saying they just want something that can keep them at that baseline, again, because of something that I said at the beginning, which is it's this slow, relentless loss of muscle function year after year after year. I think in terms of the conversations that we've had with payers, and I'll preface this by saying, I would say that the conversations with payers to date have been somewhat limited.
Based on the results of the clinical results in the ReDUX4 study, we took those results, we put together a target product profile, and then we took that out to a handful of U.S. payers. I think the feedback that we heard from payers is we would expect the pricing of this drug to be in the sort of rare disease pricing scheme, and call it hundreds of thousands of dollars. I think if you're looking for probably the closest comp, I would say that that would probably be the most recently approved product for the treatment of another neuromuscular disease, Friedreich's ataxia, about 7,000 or 8,000 prevalent patients as opposed to the 30,000 that we have. That product is a product called Skyclarys, marketed by Biogen. That product is, I think it's priced at around $30,000 a month or about $350,000 a year.
They've actually been getting really, really good payer uptake at that price point.
Okay, great. Apologies again about the voice, but thank you so much for all of your insights.