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

Awesome. Hi, everyone. This is Maneka Mirchandaney from the Evercore ISI biotech team. Very pleased to be here with Sarepta. I have some important updates coming up early next year and beyond. So from the team we've got Doug Ingram, who's President and CEO; and Ian Estepan as well, who is the CFO. Thanks for joining us, Doug. I want to turn it over to you just to give us a quick snapshot of where Sarepta is right now.

Sure. And I'll be brief given the short amount of time we have. Look, we are exiting 2021 into 2022 with the wind at our backs. Things have gone from an execution perspective this year, very, very well. You'll remember back in February, we had our third approval, which is for AMONDYS. Revenue has done brilliantly. We had our 20th straight quarter of strong growth. We've had to raise our guidance over the course of this year twice. We'll now do between $605 million and $615 million this year, and that growth will continue significantly into next year. That means by the end of this year and into next year we've been growing our revenue at over a 40% compounded annual growth rate since 2017. We were also, as we sit here today, in pivotal trials in both of our lead candidates in 2 of our 3 big transformative platforms. In gene therapy, we've started EMBARK. It's the only Duchenne gene therapy, that's a truly global trial because it includes the United States, and it's a pivotal trial. We're already dosing patients in EMBARK. And we started in SRP-5051, which is our next-generation PPMO. We've started the MOMENTUM Part B trial, which is our pivotal trial for that as well. And then on top of that, as everyone hopefully knows, we've raised some money recently. We now sit on $2.1-or-so billion of cash and cash equivalents that we can use in addition to our robust revenue to drive our pipeline. So we're driving out of 2021 in great shape. And then, of course, early next year, we're going to have a really interesting readout, which will be Part 2 of what's called Study 102, which is our Phase II for SRP-9001 to treat Duchenne muscular dystrophy with gene therapy. So things are going very well right now.

Awesome.

I want to start with that on the crossover data for 9001 coming up early next year. Just give us a sense for what we should be looking for in that data set. Like, realize there's the natural history comparator. These patients also have a well-documented trajectory from the prior year that they were on placebo. So just give us a sense for how we should be thinking about this data set.

So I think the value of this data set, and it's going to be rich is additional conviction around the benefits of this therapy and the transformative potential nature of this therapy that builds on what we've already done. I would remind us all where we are today. We have Study 101, our proof-of-concept study. We've taken that out to 3 years. Those kids are just doing brilliant compared to natural history right now, about a 9-point difference from natural history and a p-value of less than 0.0001. Then we had Study 102 Part 1 in the 4- to 5-year-old kids against placebo properly age and baseline match. Those kids were statistically significantly better than placebo. And that was with 16 kids. 8 on each side. They were still able to hit statistical significance, which says a ton about this therapy in only 48 weeks. And the 6- to 7-year-olds in that Part 1, we had a baseline imbalance, which made it impossible to compare against placebo. But when we look at the natural industry set, those kids were significantly better than what they should have been doing given the severity of their disease. Then we have the 103 data. The 103 datas are commercially representative material, and those kids benefited by 3 points in just 6 months against their own baseline with natural history would have told us they would actually already be declining given the severity of their disease. And so now we're going to look at Part 2 of 102. And the biggest value is that the totality of the evidence sends additional strong signals that this therapy is transformative which should do externally what we already have internally, which is enormous conviction for this therapy and for the success of our pivotal trial EMBARK because if we have great data with 102, EMBARK is going to give us brilliant data because EMBARK is a much larger study, 120-patient study and a very well-controlled study with a lot of tight entrance criteria with that as well. So what are we going to see? We're going to see a ton of data, 41 patients in that study. Some of them have been on the therapy for 2 years. Some of them have been tracked for a year off of therapy and then we get the introduction of the therapy, and we get to see them for an additional 48 weeks on therapy. We have a well-matched natural history control we're going to use against all of those various sets. And our hope, of course, is that what we see out of Part 2 of 102 is what we've seen all the way along, which is a very consistent strong evidence that this therapy is transformational in nature. That's one of the things that excites me most as we think about EMBARK is that the data has been very consistent. It's not like we see one really strong signal, but we see mixed signals elsewhere. We have numerous functional endpoints. We've got a composite NSAA score. We've got time tests. These kids are doing better than natural history on all of these various measures, including the subparts of NSAA and our goal and hope is that we'll see the same thing in Part 2 of 102. And of course, we're confident that we will.

How did you select the natural history cohort? Like what variables were you trying to match for? And maybe like what would you kind of expect to see in this group of natural history patients?

So a couple of thoughts. One thing I want to make sure we all know is that all of our analytics are prespecified. The Part 2 of 102 is blinded. That's important to know because I don't want people to envision that this is supposed to our analysis. It's not. These are all prespecified natural history analysis. The selection of natural history is very rigorous. It includes a bunch of different features to ensure these patients are similar, including baseline characteristics and age as well. And then we'll see in them, we won't know because we don't get to do that analytic until we've locked it all down, and then we can do the analytics against the unblinded data set that is still blinded right now. And so our goal, our hope is that against a very well-controlled, well-matched natural history set, we'll see strong signals of the benefit of this trial, both from the crossover kids, kids with 2 years and the various subsets of the kid 6 and 7 and 4 to 5 as well.

And on the patients who are crossing over, they've got the year on placebo before being dosed. How should we think about the trajectory changes? Are you hoping to see something very clear that you get the gene therapy, and there's a market change in trajectory? Or are there more nuances than that just because we're also talking about an older patient population and patients are aging? Just curious like how we should be contextualizing the trajectory on endpoints as well.

I think the most valuable data set is going to be the natural history match data set. The issue that with -- looking at trajectory analysis, which still will be interesting is exactly the conflating variable that you've just mentioned. But one of the things we often for some reason miss is that Duchenne is a degenerative disease. And it's not a perfectly linear degenerative disease as well. It's not like you can see a kid that's going down a certain number of points every year, and then you could see if that was moderated by the intervention of a therapy. These kids often go through a period of steep decline. And so if you don't -- you will underestimate the value of a therapy if you just simply use them as their own controls is the potential risk here, right? You look at a kid who could be -- who could be relatively stable if they're a milder kid in the sort of 4- to 5-year-old or 5 even to 6-year-old. But when you start getting past that, there's going to be a point at which you essentially start falling off a cliff at a real high risk of actually becoming nonambulatory in that 6 to 7 to 8 range. So to really see the powerful effect of the therapy, you really have to use a natural history set. We'll have all of that data, but I think the natural history set will be one of the most interesting elements to sort of look and frame what is this therapy doing for these kids over time. Because one of the things that we also miss sometimes, but hopefully now with the 3-year data out of Study 101, we're starting to really understand it as a group is that whatever we see -- whatever benefit we see, let's say, 6 months or 48 weeks, that will only strengthen over time, assuming the durability of this therapy because these kids are going to be falling away with disease. I mean anyone who has seen an 8- or 9- or 10-year-old Duchenne kid, will understand how dramatically this disease is and how ferociously it is and its degeneration. And if we can slow that generation down or fully arrest it, then that will be full transformative.

You talked about some of these other endpoints that you'll be measuring as well beyond NSAA. Are we going to get that some of those with the update early next year? And any sense for which ones we should be focused on?

Yes, we'll have all of that. So we have a number of different interesting endpoints, some time tests, rise from floor, all of them are really viable. The thing that we look at most significantly from a regulatory perspective, certainly, is NSAA. But all of these other elements are really important. Rise time is a very important one as well. There's a direct correlate to rise time and loss of ambulation. As an example, children who are -- the rise time falls to being slower than 5 seconds, have a very significant probability of becoming a nonambulatory in the next year or so. And that's one of the problems we have with the 6- and 7-year-olds in 102 is that we had a number of children, a significant. Maybe even the majority of the 6- and 7-year-olds had rise times over 5 seconds. So they were very, very severe as children.

Got it. That makes sense. And I guess if the crossover data looks good, we've seen that the 103 data. What are your latest thoughts on whether or not you might try to file in the younger patients? And I guess what's the downside for trying as long as you're setting expectations that should be a base case scenario?

Yes. So let me start with that. Let me say that I want to make good caution that I really think the most valuable thing that's going to come out of 102 in Part 2 is that the external world will have the conviction that we have about the transformative nature of SRP-9001. And think about how important that is. Because if it looks good in Study 101 and 102 and 103, just imagine what it's going to look like in EMBARK, our pivotal trial, which is a much larger trial, 120 patients, blinded trial with very strict criteria that have all been informed by what we've seen already. So that conviction around EMBARK is important because we're going to -- by around midyear of next year we're going to have EMBARK fully enrolled. So it will not be that long before we have the data from EMBARK and the ability to file a BLA. That's not only base case. That's a really positive base case. The next question is, well, would you talk to the division about a faster or earlier approval and some subset of kids using accelerated approval pathway? And the short answer is we certainly would. And if the totality of the evidence supports it, the totality of evidence today supports it. So the totality of evidence, including the 102 Part 2 supports it, we'll have those conversations. But I am -- and I don't see an enormous downside of doing that unless we all get overly optimistic about the idea of accelerated approval. And we all imagine it has a high probability of success. And then find ourselves disappointed when we should actually be excited about EMBARK. That's my biggest concern here. Because there are 2 things. The data has to come out the right way. Now I'm very confident about that, but the data has to come out the right way. And my confidence doesn't mean a lot because it's a blinded data, let's see what it looks like. And the second thing is that we have to have an agency that it's amenable to that approach. And of course, that's a separate issue and OTAT hasn't to date used the accelerated approval pathway for a seller gene therapy to the best of my knowledge. So that's a big unknown that makes the probability of success difficult to calculate.

Got it. Makes total sense. I want to switch gears to limb-girdle since that's another large opportunity. You guys have shown some strong data in the 2E population, but now kind of focused on CMC. Maybe just give us a sense for what's kind of left on the CMC side? What you think our timelines for getting that ironed out? And also just the latest thoughts on what might be needed for registration? Sorry that's the -- [ products and customs ] in one.

Yes. So let's -- so first of all, you're very right, the CMC is the rate-limiting step to the commencement of a pivotal trial. We have more work to do on the process and analytical development side. And what we want to do before we have a discussion with the agency, we want to do the same thing with 9003 that we did with 9001, which is have our CMC completely [ bind ] down and then go in with one conversation. Here's our CMC, get them impressed with, hopefully, with our CMC and then here's our development pathway and we want to use an accelerated development pathway. It's difficult for me to give you timelines on that because the CMC -- we know exactly what we need to do. It's an engineering project, so we'll get it done. But the exact timing of that is difficult to say, it will be next year. And then we will, if all goes well, have that meeting with the agency next year on the CMC and the development side. The goal on the development side will be an accelerated approval trial, an accelerated approval pathway. And there -- this is a -- there is, in my very biased view. There is no therapy that ought to have an accelerated approval pathway more than SRP-9001 and 2E. This is an ultra-rare disease, doing a placebo-controlled trial would be essentially impossible in my view and meaningful. And it would have enormous bioethics issues associated with it. It is a completely unmet medical need, given that there is not even a trial. I'm not even sure there's research being done right now to help these children if not for our therapy. It is a single gene mutation. It's the lack of a protein as sort of a perfect combination for gene transfer therapy. And we make the native protein. We don't make it -- we make an unaltered, unedited native protein, and we make robust amounts of it, all properly localized to the muscle sarcolemma acting like the structural protein that it makes when we make 50% or more of that. So this is a perfect pathway. The good news right now is that the agency has in writing told us that they see the potential. It's not a certainty, but at least that we're in good shape there, they see the potential for an accelerated approval pathway for this therapy, so long as we can quantify beta-sarcoglycan well, we can. And so long as we have good evidence that this amount of beta-sarcoglycan expression is reasonably likely to predict clinical benefit, we think we have that. So that will be the conversation that we'll have next year when we have the CMC in hand. And then we want to do the same thing with our other sarcoglycan programs as well. Remember, we have -- we've led with beta-sarcoglycan, we have alpha-sarcoglycan and we have gamma-sarcoglycan as well. We'd love to get those things accelerated so we can move them all at the same pace. One of the reasons that we raised some additional money to make sure that we could get our -- all of our limb-girdles moving as fast as possible as well as all of our PPMOs moving as fast as possible.

Perfect. I want to spend -- and this is the -- last couple of minutes on the PPMO side. You've got the registrational cohort up and running for 5051. I guess as we think about this landscape, there are a number of next-gen antisense programs being developed. How do you kind of think about these programs and positioning of your PPMOs versus some of the other technologies that are coming up on the horizon?

Yes, a lot of people that have done things in cell lines and mice. It's very different to do things in people. And we've learned that -- we've all learned that the hard way. We've seen other programs. We saw the Prosensa's data. It looked brilliant preclinically and didn't work in people. We saw Wave, which they had claimed it looked brilliant and didn't work in people, either from an efficacy perspective or a safety perspective or both. So -- and we've had it ourselves. We had a thing called the B-peptide that looks great preclinically until we got to the nonhuman primate, and it wasn't safe. It was toxic by the time you got to nonhuman primates before you got to people. So I would just say that you can't -- there's a lot of really interesting technologies that are early stage that people are working on. You cannot compare that to where we are with the PPMO. The PPMO, we've gone through enormous preclinical work, enormous amounts on the efficacy safety expression side. Then we went into healthy human volunteers to prove that we can do that. Then we did a single ascending dose study to prove we can do that. Then we went to a multi-ascending dose study, which is the first part of our MOMENTUM study. And now we're sitting here tracking into Part B with a therapy that, at least in our early read, has at 20% the dose exposure of eteplirsen, and in half the time of eteplirsen makes 18-fold more exon skipping and produces about an order of magnitude more dystrophin at 12 weeks, which, frankly, we've never even thought to look for dystrophin that early because it seems too early to even looking for dystrophin. So I think there are a lot of really interesting technologies that we'll keep our eye on over time. But comparing that preclinical work to where we actually are with our next-generation PPMO, in my opinion, is a false comparison. That's where we are.

Last question, just we obviously know the timelines for the 51 amenable population on -- how are you thinking about the other opportunities in DMD that run we might hear more about those on [ TALENs ].

We'll come back next year and give you some additional timelines. One of the reasons we raised money, remember -- we raised money a month -- a couple of months ago, at a time when we didn't have to, kind of surprised people. We had $1.6 billion, and we raised money to get us to over $2 billion which gave us an enormous -- not only gave us an enormous runway as an organization because you have to add to that our revenue, and we're outperforming on revenue by significant amounts right now. But it also allowed us to invest more aggressively in some of the areas where we're seeing such positive signals. And the 2 big ones that we really want to start moving as fast as possible is the PPMOs, getting the other exons going in the PPMOs and then eventually finding another disease state as well. But first and foremost, getting additional exons of Duchenne in frame for the PPMO and the limb-girdles, moving these other limb-girdles, moving the sarcoglycans, moving dysferlin, moving type A, getting those ones moving as well. And we'll come back next year and kind of give you some timelines around that. But we're already making -- that we've already put that raise to good use by accelerating our strategy on those other programs as well.

Awesome. I think we're already a couple of minutes over. But thank you so much, Doug, for taking the time and looking forward to follow these updates next year and beyond.

Thank you so much. Thanks for having us, Maneka.