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

Thanks, everyone, for joining us at the Sarepta fireside chat at the 46th Annual TD Cowen Healthcare Conference. I'm covering analyst, Ritu Baral. And with us from Sarepta today is Ian Estepan, President and COO. Thanks, Ian, for joining us.

Let's start with ELEVIDYS and its current commercial ramp. On the Q4 call, Sarepta guided to 2026 net product revenue of $1.2 billion to $1.4 billion, noting it was comfortable with the current 2026 PMO revenue consensus estimate of $900 million. Now this implies that '26 ELEVIDYS revenue will be between about $300 million and $500 million with $500 million floor run rate if nothing else changes, whether that be sirolimus, whether it be label reexpansion, et cetera. What has changed from January to -- for you guys to amend your estimates for 2026 at least?

Yes. Well, first off, thanks for having me. It's a real pleasure. And of course, we'll be making a number of forward-looking statements. So before I get to your direct question, I think it's important to actually take a step back and think about the dynamics because one source of confusion that we've gotten feedback on is really just understanding the dynamic of a onetime therapy versus a chronically dosed therapy. So for example, if you just looked at our current quarter that we reported, we had $110 million, times that by 4, you should be at $440 million, right? And so you really have to understand the dynamic with a onetime therapy where you're starting it from 0 every quarter, right? There is not a built-in installed base.

It's all NRxs.

Exactly.

Queue of NRxs.

Exactly. And so -- and then the other dynamic that's important to realize is the long turnaround time, which is around 6 months. So right now, we're kind of working through start forms that were written during the summer.

Got it...

And beyond...

And so that 6 months, is that longer than it was previously? And if so, for what reason?

It's traditionally been between 4 and 6 months. So it's still within that range. And so that's the dynamic that's really important because when you start from no patients each quarter, we're looking at the run rate from a start form perspective right now and using that as a base case to set our guidance. So we don't from -- so to your good point, where we obviously were comfortable around $500 million at JPMorgan. So we had about 4 to 6 weeks' worth of data from a run rate perspective from start forms, where we currently are and working through the start forms that are already in queue that's working for...

For the next 6 months...

In the next 6 months.

And then you extrapolate like if you sort of map out the next 6 months and then you flatline that, that's your $300 million?

Correct.

Okay. So if you accumulate like you take the area under the curve for the next 6 months, but then you get an inflection, that's the $500 million inflection...

Exactly. And so remember, from a calendar perspective, because there is such a long lead time, even if all the initiatives were hitting the ground running right now, right, you wouldn't see that translate to sales until 6 months later. So you're running out of just time from a calendar perspective. But from an opportunity perspective, everything is intact from the number of eligible patients who are available and the opportunity perspective.

So to that point, that $500 million assumes positive impact from the initiatives that you talked about, the ongoing sales force expansion. Can you walk us through the individual steps, like basically the delta on the sales force expansion, how long it will take to train and deploy them? Any color on where they're coming from? And is there, I guess, a unique element to expanding and detailing a $1 billion potential gene therapy that is different than $1 billion small molecules that we should keep in mind?

Yes. I mean the dynamic as it relates to the education that has to happen right now is as a onetime therapy, you are somewhat precluded from using another AAV approach going forward. Now that being said, there's a wealth of safety data and now with the 3-year EMBARK data, a wealth of efficacy data, which we see that as a drug-modifying agent, we're seeing exactly what one would expect in terms of increasing separation over time. So we're very pleased with the way that the data has panned out. Now from a sales -- so that's an important message these are the types of things that...

So the data that your reps are going out with is only going -- getting stronger over time.

Exactly. So that certainly balances the efficacy conversation. And then from a safety perspective, a lot of education to put everything in the appropriate context, right? The denominator is very important as you think about what is your overall risk from the therapy. So that's critical. And then obviously, just from -- I know that you were asking about the sales force, they're just now kind of hitting the ground running...

So they're trained...

The first wave has been hired and has been trained, but we're also doing a contract sales force that we're looking to expand to also. So about doubling from the number of reps that we currently have, and they'll be more focused on more of the peripheral sites and finding patients.

Got it. Okay. So the 50% more -- the first wave are Sarepta employees and they're -- they have hired a more technical detail. Is that fair to say? And then the doubling is actually this contract force. And you said their focus is.

It's more on the peripheral sites and patient finding.

So the lower volume sites. What sort of patient finding activity?

It's just a matter of, obviously, we've seen claims for a lot of patients, but that does not necessarily mean that they're having the conversations with physicians around -- around gene therapy or exon skipping. And so it's getting those patients in front of the physicians and getting them to a referral site and making sure they have balance in...

Got it. On your Q4 call, you guided to a slightly down Q1. What specific factors are driving that?

Yes. So from that perspective, now especially relatively late into the quarter, I think you know for each prescription has to be written specifically for each child, right? -- filled for each child because it's a weight-based therapy. And so we have good line of sight into the quarter and the kits have to be done about 2 weeks in advance. So we have good kind of line of sight from that perspective. However, what you can't account for is just the sicknesses, cancellations and things like that. And so the down 15% captures that aspect of which we just are just unknown until the patients actually get dosed.

So is that a conservative assumption of cancellations and sickness based on recent precedent like...

Yes. So we just wanted to factor in everything. We don't want to be in a position where we don't disappoint the Street. So we factored in everything in giving that guidance.

So you're assuming cancellations that maybe haven't quite happened yet.

Correct.

Okay. Got it. Got it. That's very helpful. In January, you emphasized that over the year, Sarepta's commercial goal is to detail ELEVIDYS efficacy data sets now, including the 3-year EMBARK data and less time on reestablishing comfort with the safety profile. Has that been the feedback of the sales force right now in the sense that when they go into the field that the pull on questions, the questions posed are more around the efficacy data sets? Or are they still getting questions on safety?

I think there's questions around safety, but it has to be put in context with efficacy, right? Risk benefit is always paramount whenever you're making any decision like this and even heightened as it relates to onetime therapy. So I think the efficacy data is important to balance out, but there could be any potential concerns around safety.

Got it. And what has reception been to that 3-year EMBARK data?

I think it's actually one of the best receptions one could hope for. And so what I mean by that is physicians were thrilled to see the data, but also encouraged that it was very consistent with their own experience, right? So they have been not surprised per se that they're seeing continued separation from what you would expect from natural history, but also to see that it was very consistent with what they've experienced. So you're not seeing necessarily a ton of variability in terms of response where maybe I would have just an outlier of patients doing one thing, but it's not consistent with a larger data set. So people have been very pleased to see that their experience is very consistent with the larger data set that they didn't necessarily have access to.

Got it. Are you still seeing like new prescribers? Or really has everybody who's going to write ELEVIDYS at some point written at least one script?

We have seen new referrals, so not new prescribers. So sites that have not necessarily had a patient that they are now referring them to one of the bigger sites.

I see. Okay. And is this -- this is post 3-year EMBARK data?

This is post 3-year EMBARK data. Yes, Patrick just mentioned it on the call that we're seeing 2 dynamics that are somewhat encouraging not to overinterpret very small numbers, but that a couple of sites who haven't written in a long time have written and then referrals from a couple of physicians.

New referral sites to existing administration sites. Got it. So you previously guided to top line cohort 8 sirolimus pretreatment data in non-ambulant patients for ELEVIDYS treatment. You previously guided to second half. And then on the last call, you noted despite high patient and investor -- I'm sorry, high patient and physician interest, also investor interest, patient and physician interest, no nonambulant patient had yet been dosed, but you received the green light to start enrollment in late November. So is there -- are there gating items to that first treatment? Is it site activation?

Yes, exactly. This is logistics, right? So there's a nuance to it. So the agency agreed to the study design, but you still have to go through the contracting process and IRB approvals. So it's just logistics to get through even though the agency agreed in concept to actually get the sites to be to be able to dose.

How many sites?

Right now, there are 5.

There are 5 sites. Okay. Are they close to dosing?

Yes.

Okay. And the 5 are open...

Yes, they're -- 5 sites that are in the stages of opening. I know 1 site is definitively open right now.

As you think about that data, what do you want top line to include? Is it what liver biomarkers between all the enzymes, ALT, AST, GGT, bilirubin, like what's the most meaningful?

Yes, the real primary endpoint is the rate of ALI.

As defined by the -- the ALI definition is what on the other biomarkers?

It could be 2x, 3x normal at GGT.

And will you also be looking at like ALT, AST?

Yes. We'll be looking at all.

But it's GGT defined ALI. Okay. That's not usually how we think about drug-induced like DILI, right? It's usually like bilirubin and ALT. Is this a unique feature of an AAV associated liver injury that's sort of GGT focused?

Yes. It's GGT focused, especially because you often have variability as it relates to ALT and AST. So that's why don't typically...

In DMD patients.

Yes, in DMD patients.

Understood. That makes sense. What about bilirubin? Or is it like by the time bilirubin moves...

Yes, we look at it much earlier.

Okay. Before the bilirubin starts.

It's actually having an impact.

Got it. And then as part of this, you will be doing a biopsy to see the impact of sirolimus on potential expression. How are you taking the biopsy? How are you assessing the expression levels? And what's the current standard for assessment of muscle content-based adjustments to this analysis?

Yes. So obviously, from our perspective, we want to -- different companies do it different ways, which I think is actually important when you're comparing expression levels across programs. I think it's very challenging to do that. The normal control can be very different than one uses. And so that has an impact on the overall quantification. That being said, we will do it in the exact way that we've done it always in terms of the EMBARK readout and the like, it's going to be completely consistent with the way that we've always done it.

When could we get the first indications of impact on efficacy? Will it be from that first top line data?

I don't know if we'll have the biopsies just because, as you know, it takes time to actually process and get it from both the Western blot and IF perspective. So I would expect that the real key to this, to your good point, we're very interested to see if there's an impact and there's a strong mechanistic reason as to why you might see increased expression when using an immunosuppressive regimen. That being said, we want to get the top line data out from a safety perspective to see if we've had an impact on ALI in the non-ambulant patient population.

What is the probability that this top line data is delayed into 2027?

We feel good from an enrollment perspective. Obviously, it's slated to the really back end of the year. So could you always have some level of slippage? Yes, but I wouldn't expect any major delay.

Was the non-ambulant indication in the label formally removed in the label update? And once you get that Cohort 8 data, could it restore that language? Or what sort of -- restore it quickly without sort of a review process, a formal review process.

We haven't defined that with the agency. Now it was, as you know, you've seen the label. So it was removed from the label. And the pathway to restoring it is not something that's been defined. I think it may have some level of impact. The data may have some level of impact on if the data is sufficient and what the pathway is. So we haven't had those formalized conversations. Now obviously, it could be an sBLA, but we're just going to have to see. So step 1 is just getting the data. Hopefully, it has a significant impact on ALI and then we'll engage with the agency to discuss kind of the pathway forward.

Have you ever had a -- as you were setting up Cohort 8, right, do you set an expectation for ALI or other, whether it's expression for thresholds that would restore the indication.

Not formally with the agency.

Has it been a discussion topic?

I think we want to look to reduce the risk by at least 30%. But again, we'll see where we net out from a...

So moving on to the Arrowhead assets. You noted that your initial FSHDDM1 data is still on track for 1Q. Could we please start by recapping which dose cohorts and what endpoints will be included in the first interim data set?

Yes. So before I do that, maybe let me frame the opportunity a little bit because I actually think you can get more insight from this readout than one would normally expect just because there's been so much work in the siRNA space and in the -- with ASOs that I think there's a lot more information that we can clean.

Specifically within DM1 and FSHD...

Within DM1 and FSHD specifically. Right. So when you take a step back and kind of look at the siRNA approach, which is currently in development, you did not see a dose response curve at all, right, with the mAb. And so there could be 2 reasons why that could be occurring, right? It could be the transferrin receptor or it could be the mAb itself and the interaction with the transferrin receptor. Now when you look at the ASO approach, you actually see a very good dose response curve from a muscle concentration perspective, right? So -- and since that's using the transferrin receptor also.

You mean the transferrin ASO approach?

Correct.

The Dyne approach, right?

I'm not calling people out by name.

I will.

So when -- you see that -- you do see a very good muscle concentration perspective. However, because of the ASO approach that you're dependent on RNAs to be available for knockdown to occur, you actually don't see that translate to a PD perspective, right? But with the siRNA approach with the mAb, you didn't see any dose response curve from a muscle concentration perspective. So of course, from a splicing perspective, you don't see...

Because enough doesn't get anyway...

And so it doesn't...

There aren't enough doors open or whatever mechanism...

You're just not getting enough into the cells, so you can't have more splicing. So you've seen flat from that perspective also. So what we really want to see is really focusing on the muscle concentration with an siRNA approach because if you're able to get more into the muscle, what you've seen from a preclinical perspective, you think that you can get a higher dose response curve, which will drive more knockdown. And I think from a -- I think that's very important because, as you know, especially as it relates to DM1, the level of repeats is correlated to disease progression, right? And so if you have the congenital form or greater than 1,000 repeats, obviously, incredibly severe. If you have the classical form, you're between 200 and 500, obviously severe, but not as severe as the congenital form. So everyone knows that the level of repeat you have is correlated to disease severity. So therefore, getting the highest knockdown will translate into the best overall efficacy. Now whether you can differentiate that at a year or something, who knows? However, but fundamentally, if you're getting the best knockdown, it will lead to the best clinical outcome. So with all that being said, going back to your question, what we're looking for -- and so it's very low doses, but what we're looking for is a dose response curve from a muscle concentration perspective, from a PK perspective to see because this is the big question is with using the TRiM platform and the siRNA approach, can we get more into the muscle, which would actually lead to higher knockdown.

And then mechanistically, the RNA stuff is figured out.

Exactly.

By precedent. Okay. And then the doses that will...

So it's 1.5 and 3 for DM1. And then for FSHD, it's 1, 3 and 6.

Got it. And how long are you treating them for?

So this is just a single dose when we're taking biopsies at 30 and 90 days.

Got it. And you have confidence that the expression will have happened and be sustained over 90 days or.

We'll see. Again, this is proof of concept, and we'll see what the correct dosing regimen will be as long as you're driving the knockdown, that's what's most critical.

Got it. What -- have you seen this sort of increasing muscle concentration with preclinical assays and sort of increasing knockdown with dose escalation in an intracellular basis with DM1 or FSHD, they don't really have good preclinical model. I mean there are some preclinical models...

Yes, [ Ritu ], I think we have good preclinical models, and that is certainly what we've seen from a preclinical perspective, the more you can get into the muscle, the more knockdown you end up getting.

Do you have -- are you going to disclose any updated preclinical DM1 or FSHD data before this?

No, the big focus has been on the clinical data. We haven't been doing a lot of preclinical work.

When you are -- when the program is mature enough for functional data, what do you see as the bar? Where has that been set for DM1 splicing and [indiscernible] in DM1?

I mean I think there's a danger, especially as you look at clinical data. [indiscernible] does respond very quickly, but it's to the point that I was trying to make earlier where because it responds quickly at 48 weeks, I don't necessarily know if you can see differentiation. But if you're driving the most knockdown, you're going to see from a long-term perspective because this is so tightly correlated to the number of repeats you have. And so I think knockdown and proof of concept from a functional measure, whoever has the highest knockdown will ultimately get the most share.

So Sarepta noted ongoing CASI-22 assay development delayed splicing data into second half. So what work remains on that assay? And what potential is there for further delays?

Yes, I don't -- I wouldn't expect any further delays. This is just around the validation. This validation work that's going on. The team has made good progress on that. So we'll have that data with some of the higher dose cohorts.

So Avidity before it was acquired, indicated that it had developed a proprietary DUX4-related gene knockdown biomarker. What bar do you see -- I'm sorry, we'll get to the biomarker next, but just what's the bar for the DUX4 down regulation, and we'll get to the biomarker.

I mean you've seen where they've been in kind of that 20% range. So we'll just have to see how this translates. Again, I think you can ultimately -- and there, you have to really, to your point, kind of compare how the assays that are being currently reviewed and how similar and different companies are doing. But ultimately, if you measure downstream knockdown, I think you'll -- if you get more into the cell, you're going to see more knockdown.

And what about biomarker approach for...

They have...

Yes, they have some proprietary.

Yes. So they have the [indiscernible]. Our team is working on that right now. Obviously, that's early. So...

Your team is working on using [indiscernible]...

I'm trying to validate it. Obviously, we don't have all the information. So the team is working on that now and looking to potentially use that as an assay. So that obviously just came out recently, and the team is just starting to work on it.

What does it sound like time lines will be to full top line data sets?

By full top line data sets, you mean in the.

All the cohorts.

By the end of the year.

Okay. So besides top line Phase I/II data, what else might gate a pivotal trial start in FSHD and DM1?

It's going to be a commercial manufacturing process.

How long do you think that will take?

Into '27.

Into '27. So really a next trial, maybe second half of 2027, assuming success.

In '27, we don't have exact visibility into. There's obviously more work to do from a commercial manufacturing scale-up perspective, but things are on track, and so we think in '27.

I do want to touch on your PMO franchise. Recently, a competitor announced what we see as kind of compelling pivotal data for a direct competitor, EXONDYS. If that drug is approved, what's strategy to counter their launch and maintain EXONDYS share? And what you plan on price?

Just to answer your last question first. I think this isn't a GLP-1 market, right? I don't think price is -- we have to make these viable and with such small populations, I don't think there's as much flexibility on price as real large indications.

Have margins improved with your PMOs?

They've been relatively stable. But as it relates to the competitive dynamic, again, and it goes back to the point I was making earlier just in terms of quantification. If you're using a different control, right, there was a competitor of ours whose drug was very close to ours, and they said that they had a 5% expression. But when you look at the full change...

6% or so, right?

No. So -- but when you look at the full change, it's actually identical, right? So quantification has a big impact. Now I do think on some of the new therapies, the dosing frequency could be important, right? So once monthly versus a weekly lessening the burden on the patient. I think that's where it could be more convenient from a competitive perspective. But I think to your point around what is the team doing in advance of that, it's really educating around the long-term efficacy, the safety profile of the PMOs has been exceptional and driving good benefit from a long-term perspective on all key major milestones of disease progression. So loss of ambulation, time to event survival, mortality. We've seen good data coming out over 5 years from an overall survival perspective. So very compelling data. So really making sure that everyone is aware of that data educated in advance of any competitive entries.

Have you met with FDA to discuss potential full approval of AMONDYS, VYONDYS and EXONDYS?

We haven't had a meeting yet, it will be this quarter.

What's your base case in upside/downside cases for this meeting? -- most likely.

Most likely, I want to be very careful as it relates to -- it's always challenging to navigate the current regulatory landscape. Look, I mean, I think the data -- I'll say it a different way. I think the data wildly supports this drug remaining on the market.

This is in CEDAR, like all of this...

This is in CEDAR. And obviously, there is a very close comp in terms of NS Pharma and the last that they disclosed is that they're still discussing with the agency their protocol. That study didn't read out positively, and that was about 20 months ago. So it actually is current in terms of the current regulatory landscape. And so I think it's as good of a proxy as one could possibly have. So we'll see. But I think the data from a scientific perspective and the way that both physicians and patients have responded, you have not seen any change in prescribing patterns or enthusiasm or utilization of the therapy.

This data hasn't told anybody anything. They didn't know already.

Exactly. We have a long history of experience over 10 years, and the data is very consistent with that.

Got it. Next presentation is mine, so I feel the ability to go over. Tell us a little bit about your third-generation PPMOs.

So the team is actually working on a combination approach using the TRiM platform in combination with the PMO. So they're very, very excited about that potential, but obviously, very early. So -- so we'll see.

Updates over 2026 potentially?

I think...

Was it 2027?

It may be more 2027. Internally, I think we'll start seeing whether it's viable from a preclinical perspective. And then depending on the data, we'll see if it makes sense to share it or not.

Great. With that, we are over time. Ian, thank you for all the insights.

Thank you for having me.

Yes. Look forward to the progress.