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

Good afternoon, everyone. My name is Gena Wang. I'm smid-cap biotech analyst at the Barclays. Welcome to our fifth gene editing and gene therapy summit. It is my great pleasure to introduce our next presenting company, Sarepta. With us, we have a Doug Ingram, President and Chief Executive Officer. We have Ian Estepan, EVP and Chief Financial Officer. We have a Louise Rodino-Klapac, EVP and the Chief Scientific Officer. With that, I hand over to Doug to give a brief overview before we dive into the Q&A.

Thank you very much, Gena, and thank you for having us today. So I'll try to be brief, but 2021 has been really important and pivotal year for us. In fact, as a result of the progress we've made this year and think about it, the approvals, our third approval, revenue, data, manufacturing, CMC success, regulatory success, we stand right now, I think, in the single most exciting moment in our entire history. Our currently approved therapies, and we have 3 of them, continue to do a wonderful job of serving the patient community. We've had our 20th straight quarter of revenue growth, all of which is organic growth, serving the patients because we've never taken a price increase, I think, as everyone knows. Our long-term compounded annual growth rate, as a result of that, is about 40%. Pretty surprising. That's from 2017. And our guidance for the year, we've had to raise twice, and we'll do about $605 million to $650 million this year. We had great readouts this year, data readouts. We had, of course, the limb-girdle 9003 readouts that were very successful, and 9001 has been, from a readout perspective, this year, extremely successful, 9001, the 101 results. We had 3-year data there where those kids were doing substantially better than natural history would have presupposed that they would be doing with a p-value of about better than 0.0001. We have the 102 data, the 4- to 5-year-olds look brilliant. And if you properly baseline match, the 6 and 7s, they also looked great with a very strong p-value. And then, of course, we have the 103 data, our commercial process material, where from an expression perspective, it looks as good at or better than the clinical material and in a functional perspective in just 6 months, that first cohort was doing 3 points better than their own baseline when natural history would have actually showed them declining. And then finally, of course, we are in 2 pivotal trials on our lead candidates in our 2 development stage platforms. On our next-generation RNA platform, we've got SRP-5051, and we've started that pivotal trial now, and we're moving as fast as possible there. And of course, with respect to 9001, we started EMBARK. The first and only so far, global pivotal trial in gene therapy, including the United States, for gene therapy for Duchenne muscular dystrophy, and we're already dosing patients there. And then finally, we've raised some money. We have over $2 billion available in addition to our revenue to drive a very big pipeline. So we're exiting 2021 with the wind in our sails and a lot to do in 2022, including some really important data readouts in the first quarter of next year.

Great. Thank you, Doug. So maybe I will start with the 9001, the gene therapy program. And early next year, you will have also, before the big Phase III data readout, that will be one of the most important data for lots of investors. And of course, some investors seeing this is almost a binary event would show us how likely the Phase III could be positive. So with that as a bedlam, wanted to ask you some additional -- if you can provide some additional color on what type of data you will share with us regarding Study 102, 2-year crossover data? Will you show also the separation data sets for age 4 to 5 and the 6- to 7-year old in the Part 2 data presentation?

So I don't know if we -- we haven't made all the decisions on exactly the presentation of all the data, but in the broadest of strokes, let me say that you're going to see a wealth of data across, broadly speaking, 2 important cohorts. We've got, of course, all of the kids have now been dosed. So some of the kids have been dosed out to 2 years. And then you have this really interesting group of kids who had been followed for a year without therapy, they don't know it because they're all blinded. And then there was the intervention of therapy, and then we'll unblind it and look at those kids at the end of that year, and that will, of course, happen early next year. And we'll look at those kids against their own baselines. But understand importantly, we'll look at them as well against natural history. And that's important to understand because this is a degenerative disease. So really to get a full view of the benefit of this therapy in these children, you have to consider where would these kids have been, but for this therapy. This is not a stable disease. This is a horrifically degenerative, aggressively degenerative disease. So figuring out where these kids would be versus a natural history cohort is going to be a really interesting element to this.

Doug, so like regarding the natural history, selection as a control, I know you mentioned it will be prespecified, so you will minimize any, say, criticism among the selection bias or anything. But regardless, you still see quite a lot of, say, investors or the -- some doctors question about because of the variability of a natural history. So how would you select -- the criteria will be selecting the natural history so that you were convinced, like showing very convincing data versus the true in 2 arm that we should look for.

I'm going to turn this over to Dr. Rodino-Klapac to provide a little bit of information. But just as a preface, remember, that this is not a post-doc analysis. And I think when people think about criticizing this in advance, I think it's because they have the misimpression that this is going to be some post-hoc analysis. We do a very rigorous job. I think one saw it recently in our R&D Day about how we look at natural history sets. We will nail that natural history set down in a rigorous way before we ever unblind this study so that it is, in fact, prespecified. But with that, perhaps, Louise, you want to provide a little additional color on how we go about it.

I think all I would add to that is that we'll certainly be matching for age and baseline characteristics in our natural history data set. So it will be, in a sense, this is a critical analysis to do to the natural history in comparison because these patients are all now a year later, and so matching them appropriately for their with their age and baseline characteristics is critically important, in addition to the comparison to baseline. But it's all prespecified. None of it will be done post doc in terms of the selection.

And I will say as well, Gena, we did an interesting thing with respect to our R&D Day. We did some analytics against what our natural history predictors would say versus, for instance, our placebo groups would say, and it was a very nice concordance. I mean there are a lot of different ways in which we triangulate to show that the natural -- the approach we take to the natural history is sufficiently rigorous that it does provide, with a standard error that you'll see there, of course, but that will go into all the p-value calculation. Very good concordance in what natural history would say, these kids would do when you baseline, match them in age, match them as well.

Okay. So other than baseline characteristics, I assume North Star and a few other measurements, age. What other things, like historically, say, steroid treatments or other -- anything -- any other you can think of that could minimize any variability that could truly identify the right control for the study?

Well, remember, steroids is going to be standard of care. So these kids are all going to be on steroids. And I think if you started adding additional metrics, you're probably going to create inadvertent bias. So I think the things that we're looking at right now are really the thoughtful approach is baseline characteristics as well as age-matched characteristics. You could imagine you had some kind of longitudinal approach and you could look at it longitudinally and trajectory, that would be interesting, but that doesn't exist. You just don't get that with natural history. So I think, again, every -- we've done a lot of triangulating to get confident about the approach that we're taking, which is important for us. It's not just important for external folks, it's important for us because remember, we're going to do all of this before we unblind the study. So we have to make sure that we've done it very rigorously. So it's not only -- it's not only accurate, but it's also usable.

So Doug, you mentioned that longitudinally. So I'm just wondering like what, in comparison to the baseline of those already, you have the 2 arms. They have a baseline data. They have a year 1 data. And if you compare -- you already have the internal natural history of these patients and wouldn't that there will be also another important data point at the 2-year how they compare to their baseline and year 1 data?

I think it will be valuable, and I think it will be insightful. I think it will likely understate -- or it may understate the full benefit of the therapy. Because remember, one of the things that we forget is this is a moving picture. So if you look at a kid over 2 years, they are a very different child at the end of 24 months than they were at the entry point and even at year 1, right? I think we know -- if you see -- we can all look at the graphs and the long-term charge with Duchenne muscular dystrophies. Kids at some point will just start falling off a cliff. And so you've got to also look at that. So certainly, looking at these kids against themselves is important and interesting and insightful, but we have to also look at natural history because you could easily miss something if you weren't looking at natural history. Look at it when you get -- these kids are getting older. When you get to the 9-year-old range, I mean, you are getting in a place where it's -- you're getting, unfortunately, into that range where you can come off your feet and be permanently in a power wheelchair. So it's important to look at where these kids would have been. And if one wonders what that looks like, look at the 101 boys. The 101 boys, I mean, I think we all remember back in mid-2018, we saw those kids. We saw their initial functional results. I think people were extremely excited. I was beyond excited, to be honest. I was actually a little emotional on that day seeing it. And then you look at it 3 years, and then look at where those kids would have been, these kids are getting in the 9-year range where they would have been versus where they are today. They have actually modestly improved even from then. But what's even more fascinating is where they would have gone, but for this therapy. I mean, out in 3 years, those kids are a 9-point different than natural history. On an NSAA scale, that's 34 points where we showed videos together on what 1 point means in a particular measure, it's heartbreaking, the difference in just 1 point. Those kids are dramatically different than natural history. So it is important to think of this as a moving picture, and unfortunately, never get far away from the fact that it's a very aggressively degenerative disease. That's why I think looking at all of this is going to be important, the totality of all this information is going to be valuable.

Okay. That's very helpful. And Doug, if I think one is a natural history, it's a very important data point. But if we think about comparing to the patient's own and now we are talking about the age is 6 to 7 and then 7 to 8, right? So if we put this in a count in the natural history, how would you think if patients compared to their own baseline, 1 is the drug arm right now, continue on drug for 2 years, the other would be the placebo now on drug for 1 year. How would you think that their North Star improvement should be compared to, say, the year 1 that you think show very clear clinical benefit to you? Of course, we have also the natural history data to backing up. But if we just look at the natural history itself, where do you see that should be?

Well, remember, those kids -- so I'm going to let Louise, in case she wants to provide some additional thoughts, but I would remind us that on the crossover kids that tracked in the Mayo's kids were mild in the first year and stable. So they are going to start progressing. So again, I would go back to say that the most valuable thing to look at is them versus natural history. I don't have much of a thought. I haven't seen the data on what they might look like versus themselves. But Dr. Rodino-Klapac, if you have some additional thoughts.

I think you're exactly right. That's why I pointed the fact that how critical the natural history of data comparison is because they are a year later and now you would be expecting that they'd be stable or declining versus improving. And so that's why it's so critical to look at that. Although certainly, we're going to be looking at baseline comparisons in [indiscernible].

I mean, just to say -- just a -- again, I'd just give more examples of that. You'll look at the 6- to 7-year-old treated kids that were in the first data set that we looked at in last January and those kids were stable at a year. Now if one didn't think about the fact that this is aggressively degenerative disease and these kids are really far more aggressively affected by the disease earlier than other kids to their age, you might think, well, okay, they are stable for a year or so, the therapy wasn't that meaningful. But when you actually baseline match them against kids that are equally severe, and we did this a few weeks ago, as you may recall, they're very statistically significantly better than those kids because they had what -- I wouldn't say because I'm a science driven, but what the parent might say was miraculous. In an aggressively degenerative disease that's destroying their muscles every single day, all of a sudden, these very severe kids didn't go down at all. They were stable for an entire year. So looking at natural history becomes really monumentally important when you want to really suss out what this therapy means because what this therapy means is not this year, it's not next year. If we're right, if we're successful, it's what it means in 3 years, in 5 years, in 10 years, in 15 years, hopefully. And so what we're really trying to do is get a good view of how beneficial this therapy is now, how beneficial it's going to be as these kids progress in life.

Okay. So what factor would determine if you can present the data at the JPMorgan conference or later? And if later, what kind of venue would that be? A press release and a conference call?

Yes. So the only issue with JPMorgan, to be very direct, is just timing. These kids -- one -- the one of the mistake I've made is that you go, well, you had data in early January, and this is 1 year later, so it should just line up. But in reality, there is -- these kids have to get revisited. They have to get baseline. Remember, they're all blinded. So they all have to get biopsies again, they have to get dosed again, et cetera, regardless of whether they had active therapy before because they don't know whether they've had the active or the placebo. And so it's just a timing issue about when we can get the data, then it certainly will be in the first quarter of next year. It's just logistics. I don't think there's a -- I don't think there's an obvious medical meeting venue that we could wait for. So it would likely be a press release and a call, and we certainly have a call and talk through all the data with folks, and we'll do that as soon as we can reasonably do. There are 2 things: get all the data in and unblind them; and number two, of course, do all of the quality controls that we -- one needs to do to make sure that, that is accurate before you release it publicly.

Okay. Great. Thank you. And...

So -- and just -- I do want to make sure I'm clear. I mean that means -- I think that, for instance, by JPMorgan is a bit aspirational right now. But early in Q1 is not. We should have it early in Q1.

Okay. Okay. So now going to the Study 301 clinical trial design. So maybe first, any additional thoughts regarding specification population you excluded? We understand we also saw similar, like the safety issue that was encountered by Pfizer's clinical trial but a little bit, maybe first, like the total, if you can give us a rough estimate percentage of patients that got excluded when you exclude those mutations? And the second question is do you exclude the exon 41? So any thoughts why excluding that particular exon?

Yes. So exon 44 amenable kids, which is the other exon that we excluded, has nothing to do with any issues other than the -- they're one of the -- they're slightly different than the other kids, at least in natural history. Natural history would suggest that they might be a bit -- I don't want to use the word milder for Duchenne muscular dystrophy because there's nothing mild about this disease, but their course may be somewhat different and a bit attenuated versus other kids. So just to ensure that you don't have some bias in the study, they're excluded for that reason only. They won't be excluded from the label. There's no issues associated with them. The 1 through 17 exclusions relate to a specific issue that Dr. Louise Rodino-Klapac has been looking at for some time. And so perhaps I'll leave it to Louise to discuss that for purposes of EMBARK.

Yes. So we're excluding mutations between 1 and 17, which are well -- it's an exclusion in terms of percent of patients for this trial that will be less than 20% of patients. But ultimately, the number of patients, which may be excluded due to this type of reaction will be infinitesimally small. This -- we had 1 patient. As you recall, we dosed 9 patients with mutations between 1 and 17, only 1 with a very large deletion had a reaction. And a good abundance of caution until we got them more data, we're excluding all in that region. But ultimately, we believe it will be a very small number of patients that will ultimately may be excluded. But we'll certainly be doing additional studies to be able to figure out ways to be able to study those patients and ultimately treat all patients with these mutations as well.

Okay. Great. Can you guys still hear me okay? I think my headset just died.

Yes. You sound great. You sound great.

Okay. Good. So another question I wanted to ask, again, all the patients now require stable daily oral steroids for minimum of 3 months at a baseline. And I -- we saw that, that was not the case for Phase -- sorry, for the Study 102. So yes, so wondering once those patients -- once they started weaning off the prophy, will these patients still require to be on those oral steroids consistently throughout the whole study?

Yes. So a couple of thoughts. So historic -- obviously, one of the goals of EMBARK is to absolutely make sure we reduce any potential -- well, just reduce the standard deviation of the trial by creating as much consistency as possible because I don't believe in the end, there's going to be a significant issue associated with recruitment, so you can do that. And these basically, standard of care, all children are on regular steroids, certainly by this age. There is a daily protocol. Some people have this weekend protocol that they use. We, in the earlier studies, we're letting them kind of track in with the protocol they were already on. Just for purposes of absolute consistency, we're just going to say every kid is going to be on daily steroids coming into the trial. And as you said, it will be at a minimum of 3 months. But the truth is these kids have all been on long-term steroids for a much longer than that period of time. Then for purposes of the actual infusion the day before, they'll get a very modest increase in that steroid about -- I think 1 mg per kg and then they'll have -- they'll be on that for 60 days. They'll be weaned off of that back to daily steroids for the remainder of the trial. So that's the -- and it's a very modest change from the past. The only difference from that in the past was, in past studies, we would allow -- people had some other protocols that were using still continuous use of steroids, but with different protocols. We're just having 1 consistent protocol for EMBARK.

Okay. That's very good because some of the feedback from the doctor basically mentioned that steroid could introduce some variability there.

I will say we didn't see it. Just so we're clear. We haven't seen that in our data, but there's no reason to not be consistent. With EMBARK's our pivotal trial there's no reason to invite any risk at all in that regard. So we haven't seen any difference there, but we'll -- just for purposes of consistency, we've moved to that protocol.

Okay. Okay. And I think Pfizer has exactly same, I think, a protocol as well. Yes. Great. So now switch gear. We still have a few more minutes. I just wanted to touch upon quickly on PPMO program. Now you start a pivotal study. So Doug, do you expect protein level to further improve from what we've seen so far? And then would that be sufficient for approval if you were able to show that?

I think based on history and based on the guidances from the agency, I think the dystrophin that we've already seen would be more than sufficient to justify an accelerated approval on the basis of dystrophin production. With that said, we believe that dystrophin levels will almost certainly continue to grow over time. We've seen with respect to EXONDYS, that dystrophin levels have grown, not only over the course of through a year, but all the way through 4 years. We saw about double the dystrophin at the end of 4 years. So the dystrophin clearly increases over time. What was most surprising in that regard about the early data that we've seen in our momentum Part A is that we saw significant dystrophin in the earliest time point we'd ever even considered looking before. This is at 12 weeks, and we saw a 6% dystrophin. So we said 6% dystrophin that's half the time we've ever looked for EXONDYS. And what that -- what that means is we saw 8x more dystrophin in half the time at 20% of the dose and 18x better exon skipping. So we're seeing 6% would, I think, at least based on history, be far more than necessary for the base of accelerated approval, assuming that safety was good. But I do also believe that we're going to see significantly more than that over time. Our modeling would suggest that we should very easily get above 10% over -- over -- if you looked at it over a year. And any of those numbers we just talked about would be, at least according to literature, not merely predictive of significant clinical benefit, but truly transformational for these kids. So we're excited about this. And one of the reasons that we went out and raised money recently, even though we already had $1.6 billion on the balance sheet, was to ensure that we had the ability to start bringing these other PPMO candidates behind 5051 forward, and we're going to do just that.

Okay. Thank you. Lastly, quickly on limb-girdle program. Any -- what is the latest thoughts regarding the typical trial design?

So the first thing to know about -- there's 2 things. The first thing to know is that our big rate limiter isn't the clinical trial itself or even its design, although we do have more work to do there. It's the CMC. So the big issue for us is getting the CMC work done and getting to a place where we have released material for a trial. That's as, I think we all know, within gene therapy, there's no mean feat. We have an enormous amount of experience given the work we've done with 9001, all of which we can apply to 9003. But we do have more work to do. Our goal is to have that all done next year to start the trial. On the clinical trial itself, we're still thinking through a number of approaches. We are strongly of the belief that we should have a development plan that actually is fit for purpose for an ultra-rare disease, where we're making the native protein, which means a very lean approach that would be an accelerated approval approach. Beyond that, we're also considering, although we haven't nailed this issue down yet, we have more work to do there on whether there's a way to accelerate some of the other limb girdles at the same time. Again, one of the reasons we raised money recently was to make sure that we could fully fund the limb girdles because there's an enormous amount of opportunity there, and maybe bring all of the sarcoglycans in any event, together in one basket, so that we can actually get them approved around the same time. Now if we do that, that has implications from a time line perspective on 2E, our lead program and the like. So we've got more work to do on that. But we also have the time to do it because, again, I'd say the biggest rate limiter on this, I mean we're seeing -- I think everyone knows we're seeing great expression, a very stable and consistent and predictable safety profile now that we've looked at both 9001 and 9003 and great functional signals so far. So the big issue for us is CMC, getting the CMC right, not just for 9003, but for the other sarcoglycans as well. So we'll have an update on that next year.

Great. We are running out of time. Thank you very much, Doug, Louise and Ian, and we look forward to the data early next year. Thank you.

Thank you so much. Thank you for having us.