Sarepta Therapeutics, Inc. (SRPT) Earnings Call Transcript & Summary

Hey, everyone. Thanks for joining us at the Sarepta fireside chat at the Cowen 2021 Health Care Conference. Hopefully, our first and only virtual health care conference for Cowen. With us from Sarepta, we have CEO, Doug Ingram; CFO, Ian Estepan; and CMO, Gilmore O'Neill, who I'm sure most of you know.

So Doug, let's get started. I guess, Doug and Gilmore. So the last major milestone you have is the top line for the Phase II 102 study from your DMD gene therapy. You missed on the top line NSAA analysis because there's some pretty significant baseline imbalances between age cohorts, which were not stratified for baseline block. One question I've gotten from investors a lot is why didn't you stratify? And if you could, from there, just tell us how you're doing, what's the latest from the crossover patients and how they've been treated.

Yes. So a couple of thoughts on that, and you're going to hear a hint of defensiveness action from this -- in this question. And the answer to the question, look, Sarepta has an enormous amount of experience with the knowledge about Duchenne muscular dystrophy. And actually, it was us that had signaled that in this -- what would have seemed like a fairly tight age range, 4 to 7 years old, there would basically be 2 populations, a 4- to 5-year-old population and a 6- to 7-year-old population. And I think you know that, Ritu, because we've talked about this many times in the past about that. And that's why we built not only to make sure that we've stratified by age, 4 to 5s and 6 to 7s, but that we would have separate analyses of the 4 to 5s and 6 to 7s. We had not anticipated that going even within those 4 to 5s and 6 to 7s that there could be this sort of imbalance. It was, I will point out, extraordinarily improbable that this would have occurred. The p-value -- as I've mentioned before, the p-value on the baseline characteristics in the 6 to 7s, between those 2 groups was 0.004. So this shouldn't have occurred. It should have been only 4x out of 1,000 should this have even occurred. And then in the worst possible way for the study, which is all of the severe kids who are on therapy and the mild kids who are on placebo. So what we ended up having in this trial, ironically, we knew we would have 2 patient populations. We ended up with 3 patient populations, and that made hitting the top line impossible because the 6 to 7s are significant. Now I'll point out -- now I'll get onto the crossover for a second. I will point out that when you get the baselines right in the 4- to 5-year-old age range that with 16 kids, we strongly hit the physical significance, 0.017, a very clinically meaningful benefit over placebo in only 48 weeks. In a heterogeneous population in a degenerative disease that's always had challenges for the first time ever in the history of Duchenne muscular dystrophy, have we seen these sort of results on NSAA in this kind of DMD population. So we're excited about that. But we did miss the top line because of the imbalance. All that defensiveness to one side, I want to be very clear. We are not going to have this happen again. We're not going to rely on just good luck in the next trial. So we will make sure that we are thoughtful about stratification. And like to ensure that the baseline characteristics for the next population are in good order. Now on the crossover kids, the good news, as you know, is by the time we had JPMorgan, all of those kids had then dosed. The -- and then, of course, the other thing people should remember is this is an interesting study, Study 102. That result that we provided in January was only Part 1 of the study. It's still blinded. And in fact, it was quite a debate internally, I will say, and I will give Dr. O'Neill an enormous amount of credit for his pushing in the direction of ensuring that we kept that blinded and did a 2-year look as well. That's going -- ongoing and it's going well. It's all blinded and by the end of this year, we will have last patient last visit, and then we'll have some really interesting data out of Study 102. We'll have some kids that will have been on therapy for 2 years, and we'll have some kids that have had an essentially 48 weeks, but have had essentially 1-year run in to watch their trajectory and then we get to intervene with therapy and see what the therapy does versus the trajectory. And we can look at all of that, it also gets to the actual [indiscernible]. So it remains blinded. It's going well, it's executing, and we should have that data essentially by the very beginning of next year.

And safety and tolerability in the post dosing time period has remained positive, Doug?

Yes. Let me be clear. The study is still blinded. So if you ask very specific things, I won't have them. I will say, broadly speaking, as we saw in the 102 Part 1, we are seeing a very stable safety profile. We know what we are seeing, and we have seen nothing different thus far in the crossover patients. So we feel very -- we feel increasingly confident that we understand the safety profile of this therapy at these doses. As you know, we see some elevated liver enzyme. But we don't see the things that are particularly troubling. We don't see clinical complement manifestations ranging along those lines.

So nothing outside of the already characterized profile?

To the best of my knowledge.

Perfect. Okay. So again, some of the investor pushback has been on the fact that -- or it's been on the topic of if this therapy provided truly clinically meaningful benefit, the baseline imbalance wouldn't have mattered that those 5- or 6-year olds -- the 5- to 6-year-olds would have caught up to the naive -- the treatment-naive patients, and it would have sort of closed the gap, even if they were sort of outperforming almost normal. This therapy does not normalize patients if it's too late.

Yes. That is -- so I'm going to be clear. That is mistaken. I assume I'd say that is the kindest term. That is a mistaken view. It would certainly be the case that, that is true over the long term. This is a long-term degenerative disease. So I will readily agree that that's the case, if you looked at it over a long enough period of time. But looking at it over 1 year, there's no chance that, that is the case. This is -- understand. See, like in the [indiscernible] data, we have much more granular data than this. When you look at like the external data set, and you benchmark against this age range and this NSAA, for the treated group, it was so severe that you could have envisioned those kids going down 3 or 4 or more points over the course of 48 weeks. You compare them against an entirely different, we did. We compare them against an entirely different population of kids who are mild and actually were going modestly up over the course of this year. This is not about the profoundness of the therapy. If indeed those kids in the 6- to 7-year-old age groups should have been going down 4 or 5 points and were stable over the course of this year, this is -- that -- those kids benefited as well as the 4- to 5-year-olds undoubtedly. And that would make sense from a mechanism of action perspective. And this is also not about the size of the study. We would have had 3,000 kids. If you treat -- you compare essentially fundamentally different populations, there's just no way you can tease out over the -- in this short period of time a statistically significant difference regardless of the profoundness of the therapy. So that -- and there's no way around that. I mean, this would be as if you took -- essentially in one regard, age is a very important marker, but it's pseudo-specific. It's not enough. And in it -- so every -- all of us would know, you couldn't compare a 6-year-old to a 9-year-old. No one would expect it. Because you'd say, it's not fully fair because why is this 9-year-old not doing as well as a 6-year old, we would all think that's absurd. And that is essentially, at least qualitatively, the argument here. These kids are so -- the delta between these kids at baseline are so different and trajectory is so different that you just -- you are -- it matched the value of the therapy. That's -- now all of that's great, but of course, it's on us to the fact that the baseline characteristics were different, and we'll address that for the next study.

And how -- so all that being considered and the strength of the 4 to 5 patients such that being considered, what are the chances the FDA is still going to let you file for accelerated approval on the micro-dystrophin surrogate?

Yes. So I don't think anyone that knows us -- I mean, I would have imagined that we're not thoughtful and creative, and we won't have broad-ranging dialogue with the division. But I do think from a planning perspective, I think we should all assume that the next study is going to be the basis for our approval, both in the United States and around the world. I mean that's the most reasonable highest likelihood approach. It doesn't mean we're not going to have discussions. It doesn't mean we're not going to go to the agency and show them our data and the benefits that we've shown -- we've seen in the study for the 4- to 5-year-olds. But I think for all of our planning purposes, if I was, as an example, an investor, and I was making some decisions, I would presume that it's the next study that's going to be the basis for it. Yes.

Got it. Is there still data that you could generate that would -- is it possible that the crossover extension Part 2 of 102 could generate data, especially since you mentioned last night on the earnings call -- this wasn't last night, days are running together, that you're going to be expanding 103 more 4 to 5 patients?

Yes. Well, a couple of thoughts. So sort of without directly addressing the regulatory issue, I will say that, first, understand the following. Between 101 and 102, we already have far more insight and a far more advanced view of our therapy and the way it works and it -- and confidence in the therapy than anyone else could far -- further along than anyone else in that regard. At the end of this year, we're going to have an enormously valuable piece of information, I mean, a set of information; 2-year data, trajectory analysis, analysis gives natural history. I mean quite apart from what that might do from a regulatory perspective, it is going to be a profoundly important moment in the understanding of micro-dystrophin and what it does for children with Duchenne muscular dystrophy across these age ranges. I would say as it relates to the 103, the 103 study, to remind everyone is this study that's looking at the commercial representative material, the processes we induced to commercialize the therapy. And we're going to look at both -- it's really not function-related issue, we're looking at expression and safety. We'll have that data in the second quarter from our first cohort. But as you've rightly noted, we have -- we've expanded the size of that study a little bit to ensure that we have a nice balance of 4 to 5s and 6 to 7s, but it doesn't slow us down. It doesn't slow us down either in releasing the data from the cohort, the first 11 cohort next quarter nor does it have any impact, I think, on the convention of our next study.

Got it. How do you right now look at the relative strength of your gene therapy compared to competitor gene therapies that have kicked off clinical data? And what do you think that might mean for competitive enrollment across the Phase III programs?

So I'm going to be careful everyone knows it and don't want to be unnecessarily competitive, but I will just be actually accurate. We have always been extraordinarily confident in our therapy. Dr. Louise Rodino-Klapac in concert with Dr. Jerry Mendell have been working on this and building this construct and testing this construct long before we ever got to patients for some 14 years. We've always believed our therapy was differentiated in really every regard, safety, expression and then ultimately function, even from the preclinical models, having looked at all of these constructs. And then we're in a place right now where nobody has the -- no one's done as much as we have. No one's been able to confirm the proof-of-concept that we have both in Study 101 and in Study 102. So I will readily admit, Ritu, that I am honest. I am the CEO of Sarepta, but I do think that we have a highly differentiated therapy with a strong proof-of-concept. If you look at Study 102, there's 2 ways to look at that Part 1 data. The first one is, of course, very disappointed that we didn't hit the primary endpoint. Well, although we know why -- then why we're disappointed about that? Because we had had this hope that we could maybe use that as a bridging strategy to accelerate this therapy quickly. But if you put that to one side, and recognize that this means that we'll be -- it will affect a 12-month or so delay, putting that issue to one side, as a proof-of-concept as a Phase II, it's brilliant. We have done something that no one else has done. We've shown -- if you look at the 4- to 5-year-olds, in 16 kids, strong statistical significance, strong clinically meaningful NSAA, we have more insight out of that study than anybody could have right now because of how advanced our approach is. And frankly, it's going to inform study 301 in ways that I think are going to be unmatched. One of the things that others have done, and I think it was wise of them, frankly, they -- others that were looking at using forms of micro-dystrophin, looked at Sarepta as the leader and said we're going to essentially replicate what they did. And so they essentially -- Pfizer's example, really basically replicated their study of 102. We have the opportunity to really look at what's coming out of 102, adapt our thinking to 301 and then get 301 started this year, and hopefully, we roll this out as fast as possible thereafter. So I think we're -- I remain confident that our construct is highly differentiated and our program is highly differentiated.

Going back to the 103 safety for a second, have you actually treated like patient 12 and 13? Or are you still -- or is that going to happen?

It is going to happen. I don't think we'd yet dosed the next cohort.

Does that mean we can still get all of that data all together in Q2, including those [ 13 ] patients?

No, no. We're going to get the first 11 in Q2. The other ones will be a little bit delayed, yes.

Got it. Okay. Got it. When are we -- when are you going to talk to the FDA about the Phase III design? And when are we going to know? Are you going to tell us, investors, at the time of trial initiation midyear? Or are we going to get it beforehand?

You'll get it at around the time of trial initiation. So they've got a lot of work to do. The thing you have to do to have a meeting with the agency is you have to have the data in front of you. So the first thing we have to do is complete our analysis of 102 Study, 102 Part 1. And you might think that, that -- and we're -- we've done that to a significant extent. But there's a lot of insight out of Study 102. So we need to really make sure that Study 301's protocol is fully informed by then. We need to get the Study 103 first 11 patient data, right, and have that conversation with the agency, we have to have the data of both expression and safety out of Study 103, and then we'll have the meeting, do the briefing broken out the conversation with the agency. And then assuming either way, then we will shortly thereafter, communicate to the community where we are with the study. And of course, as you know, our goal is to start that study somewhere around the middle of this year-ish. And then our goal is to try to get all those kids dose as fast as possible with the aspiration of getting oral dose study end of this year. That's at least our current goal.

So we have a question in from a client on the webcast. I'm trying to better interpret it. Do you think that your -- basically, your label and indication will be limited by baseline entry criteria? So like, if you do decide that you're going to go with a certain 6-minute walk baseline or NSAA baseline, are you going to limit your market upon any eventual approval?

Yes. I think that would be -- I know -- let me start with that know, that is not our goal, nor do I think there's no any logic to that and it shouldn't be the case. And the reason for that is straightforward. The reason that we create -- we try to create very rigid guardrails around entrance criteria is because we're dealing with a heterogeneous degenerative disease. And so that the benefits of that therapy can be seen in a short period of time. To the very good question you asked earlier in the conversation, over a long time, it will all come out. If we wanted to wait 7 or 8 years, we would see the benefits of this and we could be far more -- far less restricted in the approach. But I think the agency would understand that the goal of creating a very strict guardrails around entrance criteria is to ensure that you have a very well-matched patient population so that when there's a benefit, you can see it. That's -- and we've seen the hard way, what happens when you don't have good baseline match to 6 to 7s. I am completely convinced that the 6 and 7s benefited from the therapy just as much as the 4 to 5s. But when you compare them against the wrong group, it gets hidden. And so that's the goal. So it shouldn't affect the label. Now -- and our goal with this -- the goal with all of this is to get to the broadest label possible consistent with the data and consistent with the mechanism of action of micro-dystrophin.

Got it. And a follow-up to that one before the next question would work. Client have not finished the second half of your question is coming. Do you have information on detail right now regarding the improvements seen in 102, like skill gains, refinements of like certain movements, et cetera? Or is that something we're going to get at NDA?

If you won't get at NDA, there will be no -- essentially no new data at NDA. There is a medical presentation on 9001 Part 1 of 102, but you've seen it. There's nothing significant and new.

Okay. Nothing new out of the 102 data that you're going to be presenting at NDA. But will we get more granular sort of skill gained after that?

Maybe early next year, but not for the time being, no. I mean what we will -- what we are doing, and we can -- we have a lot of data. We have patient level data. We have all the composite. I wouldn't say the composite. We can bring the composite down. We have time test. We're not going to provide a lot of additional data there, what we're going to do is use it to inform 301. That's what we're doing right now.

Got it. Another question in -- based on the data set that you have right now, what's your confidence that it will work in older patients? I'm assuming, the client means over 7, maybe 10-ish.

Yes. Well, look, I mean, we are confident. I am confident that the mechanism of action of SRP-9001 would indicate that it ought to work across all parts of the patient population. We know what DMD does, and I'll explain what we have to do to get there. We know what causes the Duchenne muscular dystrophy. These kids are missing this structural shock-absorber, dystrophin. Micro-dystrophin replaces that with a functional -- a much truncated functional version of that dystrophin that distributes sheer force. As long as the kid has muscles, cardiac muscle, skeletal muscle, diaphragm muscle, they should benefit from the intervention in the micro-dystrophin regardless of the age. Now there's a separate question for the older nonambulatory kids as well, which is we have to make sure that -- we're going to have to dose them, and we have to make sure that we have the right safety profile for them. And so that's a separate exercise. But from a pure mechanism of action perspective, we are confident that micro-dystrophin ought to benefit all kids across the spectrum. And then no kids should be essentially off the table. I mean, it is for that reason, sorry to kind of go on. But it is for that reason that I find it -- let me say, it is basically silly to envision that in the 102 results, we see brilliant result in 4 to 5s. And for some reason, it just starts -- on your -- on your seventh birthday, it just stops. There's no scientifically valid hypothesis that could explain that. What does explain it is we frankly didn't get the right comparator population. So that's our view. And that's why -- and if you believe that, as we do, then, of course, that's why we got to fight to get the broadest label possible so that we can intervene with these kids across the entire spectrum.

How are you thinking about your limb-girdle studies in light of what happened in Study 102? I mean, I know it's a different therapy. I know it's a different indication. But you picked the subset of limb-girdle to act like -- because it acts as severe as DMD.

Yes. So there's a couple of thoughts on that. One, of course, is all of the building data that we have, both in 2E and in 9001 gives us confidence across this approach to gene therapy. So we're as excited as ever at our ability to create robust expression to do so in a reasonably safe ways that we don't have things like complement. We're confident about the promoter, all of that shared across 9001 and 9003, both capsid and promoter. But if 102 tells us anything, it reconfirms the view that we have with respect to this very rare disease, limb-girdle type 2E, which is this idea that you can do a traditional placebo-controlled trial on a 2E population, I think, is mistaken. And I think if we didn't understand that already, which we did, you would understand even more on that. So I think that's going to be the really important discussion that we have with the agency. It's a much -- it's a far rare disease. It is a -- the native protein. We're making a robust amount of it, both in our low-dose and our high-dose cohort, and we're doing so in a reasonably safe way. And so we should have a very lean approach to that development pathway. It shouldn't include from our perspective, for instance, the approach that we're taking with DMD, for instance. It should be much leaner. But we have to have those discussions with the agency before we're confident that we're alive with the division on that.

Got it. I just want to ask you about -- yes, let's just spend a few minutes on your pipeline. And what do you -- Doug, what's your favorite program from the rest of your pipeline outside of DMD? What's your favorite gene therapy or otherwise? And then talk to us a little more about how you were thinking about gene editing in DMD.

Yes. So a couple of things. I mean one of the things you're going to find from us over the course of this year is this razor-sharp focus. So you're -- it's going to surprise you that we have over 40 programs that I keep hammering on about our top priority, which is AMONDYS, EXONDYS, VYONDYS, 9001, 9003 and [indiscernible]. So in that regard, I almost don't want to discuss the rest of the pipeline because I want to keep everyone focused. I think there's a lot of exciting things in our deeper pipeline. We've got cardiomyopathy and some other really interesting approaches. Probably the most exciting, next-generation element in gene therapy beyond just the disease areas that we're working on is the ability to actually make -- to make what we do even better over time. And so the things that I think are going to be really exciting over time is empowering redosing, right, knocking down preexisting neutralizing antibodies. We have programs in both of those, both internally and with partners, finding down the road, and this is more of a -- much more of a research project for now, but I think it will be important for long run. Down the road either getting even more advanced and improve viral capsids, either immune abating, liver abating, better tropism or moving to even new modalities. You know we have both exosomes and lipid nanoparticles as approaches. That's much more research oriented. That to me is a really exciting area of research, and we want to play a significant role, if not the leadership role in that. Overall, on gene editing, we're really excited about the conscious of gene editing and -- for neuromuscular, neuro and beyond. We have a Gene Editing Innovation Center in Durham, North Carolina. We started building that out some time ago. We have, I think, one of the world leaders in the application of CRISPR-Cas9 in gene editing to the neuromuscular area. And that's, of course, Dr. Charlie Gersbach from Duke, and he's building a team around him right now even as we speak. I think gene editing is going to be -- it's a little bit more down the road than gene therapy. Gene therapy is right in front of us right now, just like RNA. But we're very excited about gene editing over time. I think there's a lot of questions that have to be answered along the way. But I think one of the really interesting questions that has to be answered with respect to gene editing is do we use viral capsids? Because right now, it's all viral capsid-mediated, just like gene therapy. Or if you're going to deal with something like gene editing, do you need to over time evolve into a lipid nanoparticle or an exosome or something along those lines. And it's -- but that's still in the research phase, but we've got some really exciting work going on with some really great scientists working on.

Have you seen anything in the data thus far that sort of -- that's sort of pushing your interest in redosing and technologies that enable redosing? Or is it other people's data?

It's the data of our partners, right? So we've got 2 different approaches, right? We have an approach that you could co-administer with the therapy that would essentially hide your gene therapy from the [ machine ] that would make neutralizing antibody and that could empower dosing. We've got another technology with a partner. I'm not going to mention either one because I might get them confused with these and then I'm going to embarrass myself today. We have another approach where we would knock down essentially ablate the neutralizing antibodies and advance to either dosing people that have preexisting neutralizing antibodies or hopefully, potentially knocking them down to empower redosing. And then we have some other -- we also have apart from that, it's more just all internal. We have some other internal approaches as well. We've done some interesting [indiscernible] work that we haven't yet disclosed that gives us confidence. And I will say this, and I think that the confidence that we have is probably shared by everyone in gene therapy. It's important that we get to a place where we can knock down preexisting neutralizing antibodies. And it -- because that's at least for us, about 15% or more in some other viral capsids and vectors. It's going to be even greater percentage of kids that will be off the table until we do that. And even more important, perhaps even than that is the redosing. And we remain confident that we're -- we all, as a biotech industry, we will solve this issue, and we will get to a place where we can start redosing over time.

Great. All right. We're just a minute over time. Doug, thank you so much. Gilmore, thanks for joining us, Ian, as well. Really appreciate your answers. And looking forward to both the 103 data as well as an update on the Phase III design and that start.

Thank you very much, and don't forget AMONDYS. We just got AMONDYS approved. Do you want to...

Yes. Yes. I completely ran over your actually approved [indiscernible] product. But, we'll get to that the next time. Thanks.

Thank you so much.