Conference Transcript - STOK 2026-06-04

Andrew Tsai, Analyst — Jefferies

Okay, we're going to get started with our next session. I'm Andrew Tsai, Senior Biotech Analyst at Jefferies. Thanks for joining us today. And it's my pleasure to have the Stoke team with me to my direct left, Ian Smith, CEO, and to his left, Jason Hoyt, Chief Patient Officer. Welcome, both of you. Thank you, Andrew. So to get started, would you mind providing a brief introduction about Stoke, Zerva Nearsson, your lead product, where you are, the next milestones, zones, what you're trying to achieve, all that would be very helpful.

Ian F. Smith, CEO

I'm looking forward to the Q&A session more. But let me give you a quick update. First of all, thank you for joining us this morning here in person and also online. I'll start with the safe harbor statement, those things that I'm about to say. Please check our filings in terms of safe harbor language, forward-looking statements. Stoke Therapeutics is a company that's focused on genetic medicines. We have a lead medicine to treat genetic epilepsies, specifically Dravet syndrome. And we're targeting the root cause of that disease, Dravet syndrome, and we're doing it in a way with an ASO. We administer the medicine. It basically boosts the expression of NAV 1.1 via boosting SCN1A, which is the depletion in Dravet syndrome. We do increase the expression of NAV 1.1, and that is the root cause of Dravet. What it results in is a reduction in seizures. But dravite, unfortunately, is a disease where these children grow to maybe 18 months old, two years of age, and then have no cognition or behavioral benefits or growth as they go forward. So they stagnate from two years on. What our medicine does is it provides, because you're expressing NAV 1.1, it does actually provide benefit in terms of more neurotypical development as a child and an adolescent. And so it's a remarkable medicine. You know, if I was just to quickly profile, it's one of the few genetic medicines that actually gives back function. And that's really important. Genetic medicines typically stabilize a disease or slow the decline from a disease. Our medicine actually picks these children up effectively and provides cognition and behavioral benefits as long as they continue to take the medicine. We do have four years worth of data of measuring children on our medicine. That four years of data has shown durable reductions of seizures approximately 75% from their baseline. When I say durable that is 75% and And that's remained durable over a four-year period now. It's actually five years if you include the phase one, two study. They rolled over from a phase one, two study into the OLE. And then we've also measured their cognition and behavioral benefits over that four-year period. And we see a continuous addition of cognition and behavioral benefit in those, each of those four years. In fact, we've done a statistical analysis of those where we show that they have a p-value of less than p01 for each of those years. And so the medicine is actually quite remarkable in terms of how it benefits these children. We're in phase three. We're running a phase three study. We're about to complete enrollment in that study. It's a 52-week study, and that data would read out in the middle of next year. We anticipate doing a rolling filing, which would start in Q1 of 2027, and that would initiate with the submission of CMC details, and Jason next to me is responsible for CMC and manufacturing, and I'm sure he'll talk about that later. But the last submission within that NDA would come with clinical data in the middle of next year and, therefore, give us the potential for an approval in Q4 of 2027 or Q1 2028. We do have other medicines in the pipeline with a similar approach. We approach haploinsufficient diseases. We have a medicine for a genetic form of progressive loss of eyesight called ADOA, and we also have preclinical medicines with the same approach to focus on haploinsufficient diseases, SYNGAP1, which is a neuroscience or a development disorder, and then we have some earlier stage programs as well. Our balance sheet is very strong. We had over $400 million of cash in as of Q1, and our runway is all the way into 2028. And so we're financed all the way through into launch, and so we're in a very strong position.

Andrew Tsai, Analyst — Jefferies

and uh with that andrew back to you very good um point bottom line uh server nuisance has the potential to give back function uh you have four years worth of data and that in of itself is highly differentiating and so starting as we think about the peak sales opportunity of this drug maybe starting with price in phase three if you are able to show cognitive to confirm cognitive benefits, the neurodevelopmental benefits, what does that mean for pricing power? What kind of bookends or precedents should we think about?

Ian F. Smith, CEO

So I'm going to get just to introduce Jason Hoyt. He has the unique title of Chief Patient Officer. That is because Jason is responsible for our commercial organization, medical affairs, and manufacturing. So I'm going to have Jason answer that question. But I would just start with, you know, we've done a lot of payer work already. The reason why we can do this work so early while we're in Phase 3 is because we do have that four years' worth of data where we've been able to measure the patient's benefit in terms of seizure reduction and also cognition and behavioral benefits. We've also had a publication in the New England Journal of Medicine that validates those results. And so Jason's been taking all of this data and over the last three to six months, I would say, has been spending a lot of time with his team with the payers and maybe provide the feedback, Jason.

Jason Hoitt, Other

Yeah, absolutely. Thanks for the question. I think it's timely because I think, you know, historically there's been an assumption that because there are anti-seizure medicines available for Dravet syndrome, that those are an appropriate analog to compare a disease-modifying drug for Dravet syndrome to. and that's just that's not the case I think you know we see other genetically targeted disease modifying treatments as more appropriate analogs for what we're doing with Zareva Nursin and so to Ian's point you know earlier this year we went out and tested with payers and doctors different label scenarios and different elements of the phase three being included or not being included being statistically significant versus directionally consistent and then that relative to the the three years at the time when we first tested this of open label extension data from the phase 1-2 and across both of those audiences what we learned is that the phase 1-2 and OLE data longitudinally are likely to be the most compelling pieces of evidence that we have at the time of a potential approval for both doctors payers and ultimately families and so when we talk to payers to directly answer your question Andrew payers told us we look at the totality of what's in the public domain when we think about reimbursement policy and and and how we're gonna cover a new medicine and so when we asked them this it was before the New England Journal publication it was in late February when we when we had this this piece of market research done so it predated the New England Journal citation and what payers told us then is that, you know, from their perspective, the seizures are driving, the seizure suppression and the durability of the seizure suppression are driving a tremendous amount of value in Zareva Nursen, knowing that it's being dosed on top of the best standard of care medicines available today. And then the Vineland data, the open label extension data on cognition and behavior, to them are the demonstration of disease modification and the demonstration of the fact that Zareva Nursen is doing something different for their members than anything else that they can offer them today. And so our takeaway from that is that irrespective of what the label says on the secondaries for Emperor, we will be in a very good place from a reimbursement perspective, given the New England Journal publication, given the totality of the data. And, you know, to that point, we're deploying our team of national account directors that will be directly engaging with payers, predominantly medical directors, as this will be a medical benefit, to educate them around Dravet, educate them on the profile of Zareva nurse and we have to date so that ultimately, you know, in just over a year when we have top line results from phase three, there's already a foundational level of understanding of the product and a deeper understanding of Dravet syndrome itself. So, you know, as a result of that, we feel really confident in our ability to secure genetically targeted rare disease modifying pricing potential along the lines of a Spinraza, for example, or the Exxon Skippers from Sarepta. I think those are appropriate analogs to think about when you think about where the pricing could land for the reverse.

Andrew Tsai, Analyst — Jefferies

Okay, Exxon Skippers are a million. So, okay, very good. And so then there's price and then there's the volume side. And you've, you know, you've said the prevalence of Dravet in the U.S. alone could be 16,000 patients. You know, when investors try to do their own work, they simply do some kind of incidence math from that to derive prevalence, and they don't get as large of a number. So can you walk us through how you derive 16,000?

Ian F. Smith, CEO

So Jason, he'll give you those same metrics as well, Andrew. I'm just going to start with the simplicity of we have identified 6,000 patients based on coding for Dravet. Those are 6,000 patients that are 24 years younger, and they're on Dravet medicines, so they've been coded already and so there is an immediately addressable patient population with Dravet and then there is a further 10,000 approximately maybe more that we need to identify with screening and that's the simplicity of the market that's just in the US alone but Jason will help you understand how we go beyond that 6,000 and how you get to the 16 to 20,000 in total I will just to also say that given my experience in rare disease I've worked in rare diseases now for for 30 years I was at vertex pharmaceuticals and cystic fibrosis and when you create a disease modifying medicine screening prevails and you identify identify more and more patients and I anticipate that's what will happen here those children that unfortunately is having seizures and maybe developmental disorders, you know, they may not be diagnosed, and they may be just being treated with seizure medicines. I think we're going to have the opportunity to identify them and put them on Zareva Nurse and benefit them in a broader way. Over to you, Jason.

Jason Hoitt, Other

Yeah, to get into how we derive the 16,000, which is a U.S. specific number, right? Before we started the Emperor trial, we wanted to understand, to the best of our ability, the total prevalence of patients both currently and at the time of a potential approval in the seven major geographies where we were running the emperor study so that's the eu4 uk japan and the u.s and in order to do that you know similar to what you were mentioning andrew we we scaled incidence to prevalence but we looked back at live birth rates in each of those geographies for the past 85 years applied the incidence rate scaled it to prevalence obviously included to your point a um you know the the Drabay-specific mortality in there, largely driven by SUDEP in the pediatric cohort, and then looked at WHO life tables because once you get beyond that pediatric period, patients do tend to live relatively normal lifespans. And so that's where we ultimately got to the number of 16,000 in the U.S. And then when you break that number down from an epi perspective, 6,000 of those 16,000 are 25 and younger. And why that's important is because we have a dynamic in the U.S. where just given how patients are typically diagnosed by a pediatric epileptologist and then cared for throughout their childhood, there's a really strong bond that's formed between the family and the clinician, and there's a reluctance to transition to adult care. And so oftentimes, pediatric providers will care for these patients into their mid-20s. And so I think it's important to understand that there are about 6,000 patients that are in the care of a pediatric epileptologist, the subspecialty of doctors that are most familiar with Dravet syndrome, that are most commonly diagnosing and treating Dravet syndrome, and therefore are going to be the most current with the most recent advances in the field. Of those 6,000, 4,000 are pure pediatric patients that are 18 and younger. And we've, you know, to Ian's point, we've cross-checked a lot of this effort with what's available in claims databases in the U.S. today. And so, to Ian's point, the largest claims database out there has about 6,000 identified Dravet patients coded with the Dravet ICD-10 code. If you then do some machine learning and extrapolation based on the known patients and you can predict patients, we feel really confident that based on claims data and the linked NPI numbers to their caring clinician, that we know we're about 80 percent, 70 to 80 percent of the 25 and younger population are being cared for in the U.S. today. And then the greatest opportunity for enhancement of screening is in that the older cohort.

Ian F. Smith, CEO

Andrew, could I just add, I mean, we're focused on the U.S. market. is today while we're discussing it is we have a relationship with Biogen where we have licensed territories or commercial territories outside of North America. When you look at the total population in the major seven geographies in the world, the total population is estimated to be close to 40,000 patients. And obviously we take a royalty return and an economic return from outside of the U.S. with Biogen, and we have a very good partner in Biogen.

Andrew Tsai, Analyst — Jefferies

Got it, got it. So then let's talk about the FDA, you know, at one point starting around second half of last year, you're thinking about going to the FDA to get an expedited approval, all while rightfully so, signaling to the street is it's more of an upside case because at the end of the day you'll have phase three data anyway. Long story short, earlier this year sounded as if expedited approval may not be there based on the phase one two data so instead what are your FDA discussions

Ian F. Smith, CEO

like nowadays what are you guys talking about I use a few words like consistent constant ongoing and we continue to submit data to the FDA I'd say it's all under the banner of education and the medicine does have breakthrough designated status and the reason we went to the FDA frankly in December 2025 to ask whether we could file on the basis of the phase 1, 2 data and the OLE data, because that data was so compelling, and this disease is devastating, and there is not another medicine to treat it in the way we believe our medicine is treating these kids. So we did do that in December. But from then on, it's been a form of education, education around what the disease is, the plausible mechanistic pathway of our drug to treat the disease, which is very important with the genetic medicine, given FDA guidance these days. And we continue to provide them data specifically around how we measure cognition and behavioral benefit, which is with the Vineland. And so recently, about a month ago, we provided the investment community the four-year OLE data of both seizure reduction and cognition and behavior benefits each year for four years, and we're providing that data to the FDA to help them understand how our medicine is working. This is all to help with the pathway of when we ultimately submit the phase three data that it helps them understand how our medicine is working and the benefits it's providing to these children.

Andrew Tsai, Analyst — Jefferies

And speaking of violin, which is the key endpoint here, maybe give us a sense of what you saw actually in phase one to the degree of benefit that you saw and how does that compare to what's

Ian F. Smith, CEO

clinically meaningful out there? Yeah, you specifically asked about Vineland, Andrew, so maybe I could just help you understand what Vineland is. Vineland is an assessment that's done by a neuropsych with these patients, and they actually ask the questions of the patients, sorry, of the patients' families and caregivers. And so if I was to wave my hand at somebody in front row here and say hey can you see my hand and for those of you that are online somebody who's sitting in the front row could see my hand and has nodded that's actually similar to a vineland question where i'm asking a question they're receiving that that that question and then they're acknowledging the what i'm asking them so there's a receptive and an expressive communication those are two of the domains as an example in vineland and that's actually a an example of how you conduct Vineland. There are, we're using five key domains to measure cognition and behavior of these children and they are expressive communication and receptive communication, motor skills, social abilities and things like that. And so that's how it's administered. We obviously for the phase three study we take a baseline score of a Vineland and And then we measure the Vineland score at week 52 in the Phase III study. And given that our medicine is working in terms of the root cause of the disease, we expect that it will provide a benefit and provide these children a gain of function as measured by Vineland. We've also, in the past medical conferences, provided videos of what our medicine has been able to do, baseline video of children with the physician, and then one year on and further on in terms of these children with the physician. And those are available. If you come to the company, we're happy to share them. It's not anecdotal. They've been presented at medical conferences, and we've had children that unfortunately are not functional, and then when they've been on the medicine for a period of time, they've been able to respond to the physician, respond to their name, touch their nose, and just follow actions, ask questions. We have children that unfortunately have a struggle walking but have been able to now kick a soccer ball and things like that. And so these are, they may sound incremental, but these are huge to these children's families. And as I said, our medicine works in a chronic manner, and the four-year data has suggested that these children get better each year, as measured by Vineland. So it's quite remarkable.

Andrew Tsai, Analyst — Jefferies

And so the Vineland benefit that you saw across the subdomains, how much was it in phase one, two? And then how does that compare to natural history, to be clear?

Ian F. Smith, CEO

Yeah, so I'm sorry, Andrew. Yeah, I forgot that part. Thanks for reminding me. So clinically beneficial, defined by caregivers and physicians, is two Vineland points. We've, in the four-year data, we have Vineland scores that are out to 15, 16 in most of the domains. And if you just measure Vineland over a one-year period, when we look at a dosing schema that's similar to what we're running in Phase III, we have Vineland scores that are anywhere between 8 and 12. Our Phase 3 study is actually powered to show a difference of two Vineland points, and it's powered with a 90% probability confidence level for a p-value of 0.01. So we feel as though we're very well powered to these endpoints, and obviously we used all the data that we gathered in the Phase 1, 2, and the OLE to design our Phase 3 study.

Andrew Tsai, Analyst — Jefferies

Great. And then as for the sham side of things, you do have a natural history study. So talk about what you've seen, what untreated Dravet patients can achieve on violin after a year, which is your...

Ian F. Smith, CEO

Yeah, thank you, Andrew. So we did a two-year natural history study. It's called Butterfly. We measured the patients, obviously, baseline. We followed these patients or these children through a 24-month period. And what we've effectively found is that their baseline was the same as their end point at the end of 24 months. I believe the average age of that study was between 7 and 10 years old, somewhere close to there. And so we do have a natural history study that we have published and presented and given to the FDA. And it was shown in the New England Journal of Medicine. But, yeah, it goes to what I said earlier on this call, which was these children grow normally to 18 months, two years of age, and then, unfortunately, they stagnate. And what we're doing is we're lifting them up off that stagnation line.

Andrew Tsai, Analyst — Jefferies

Yep. And so in the U.S. side of things, you know, as we dig deeper a little about success here on Violent, in the U.S. side, my understanding is there is a hierarchy of the five that you mentioned. So for this to be static and fileable, what does that mean on this hierarchy? Do the first two need to be static, for instance, remaining three trending? I don't know how it works.

Ian F. Smith, CEO

Yeah, no, it's a good question, Andrew. So just to be clear, the end point, you're referring to the secondary end point, so first of all, the drug is approvable if we hit our primary end point, which is seizure reductions. The secondary end points provide additional information in terms of statistical significance of how the medicine is working and understand the medicine. In the U.S., the secondary end points does have a kind of a list in terms of priority of secondary end points to hit. The number one, actually, is still seizures, so we have a secondary endpoint at week 52 to show the durability of seizure reductions. Number two and three are communication, expressive and receptive communication, and we need to show a clinical benefit there that's greater than two Vineland points, and as I said, we are powered to a p-value of .01 and a 90 percent confidence level in those. The others do follow the order, so there is a hierarchical order that in the U.S. you need to, as you get through one, you go to the next. And just while I'm talking about it, in Europe, it's more of a combination of those secondary endpoints that supports approval with the secondary endpoints.

Andrew Tsai, Analyst — Jefferies

Got it, got it. And actually, going back to the vinylin benefit that you saw in Phase 1-2, can you speak to intrapatient variability, actually? Were there any hyper-responders driving the overall vinylin effect, for instance, or was everyone very consistent on their vinylin benefit?

Ian F. Smith, CEO

Yeah, it was fairly consistent, the variability. And we've shown graphs with the error bars, so you can actually see that it's actually pretty tight. You know, I don't know whether that's a real word in terms of statistical significance, but no, there's no super responders that sway the Vineland score. And I think the best way that I could provide you the basis of that is we've run statistical analysis on this data and the p-values that we've achieved for year one, year two, year three, year four. So four years, individual years, have p-values of less than 0.01. If you had variability, you wouldn't get a p-value of less than 0.01. So, you know, the drug goes to the root cause of the disease. It helps express NAV 1.1. Our animal models say we're expressing NAV 1.1, maybe to close to wild type. So we do anticipate that this provides this kind of benefit.

Andrew Tsai, Analyst — Jefferies

And so bigger picture, should you succeed on seizures, which I think you will hide probably based on the phase one, and you show this kind of violent score benefit, I mean, the therapeutic index is very high. And so I still have to ask, though, in the phase three, would you expect SAEs, for instance, with this drug? And if so, what kind of SAEs?

Ian F. Smith, CEO

I mean, you're asking me to use a crystal ball and say SAEs. I mean, all I can point to is the data that we have, to be honest, which is we have five years' worth of safety, and we have, and just to give a background, we had approximately 80 patients, maybe a little more, 80 patients that rolled on to the OLE studies from the Phase I studies. We have been able to follow those 80 patients, and 90% of those 80 patients are still on medicine five years later, or four years later in the OLE, but in total nearly five years. We've obviously collected a safety database with that, and the drug's well tolerated. We've administered over, I believe it's now 900 doses of this drug in these 80 or so patients, but we've also now, you know, we're about to close enrollment of 150, 160 patient trial, and so we have a large number of patients that are also being on medicine in our phase three, and we haven't had any dropouts in the phase three. as we reported about a month ago on our Q1 call. And the safety profile from the four-year OLE data shows the medicine is well tolerated.

Andrew Tsai, Analyst — Jefferies

Great to hear. And so actually before the phase three reads out in mid-2027, you're already starting with the rolling submission in Q1. So it sounded like CMC manufacturing to start. Remind us how many modules there are and what's next after that. And then it sounds like the gating factor is that phase three data set, ultimately.

Ian F. Smith, CEO

And so rather than go through each individual module, I'm going to say that the bookends are we start with a CMC manufacturing module, which we anticipate providing that information to the FDA starting in Q1, 2027. The last module is clinical data, which we'd anticipate completing the submission in the middle of 2027. And as I said earlier, that because we're doing a rolling submission, we anticipate based on satisfactory data that we have approval in Q4 2027 or Q1 2028.

Andrew Tsai, Analyst — Jefferies

Great. And so let's just say this is approved ultimately in the label. What will the front page label say? Will the violin 3 data be included in the body of a label, whether it be phase 1, 2 open label or the phase 3 data?

Ian F. Smith, CEO

It's a really important question, Andrew. We actually had a board meeting yesterday. I was talking to our board about this, which is you've got to define what a label is. Because the label allows you to promote the medicine to what you believe the medicine is doing for these children. We want to promote this medicine in a way that we believe what it's doing for these children is it's significantly reducing the seizures on top of standard of care medicines and it's providing gain of function, cognition and behavior and that's durable and it's year after year. To get that kind of promotion of what the medicine is doing, we need a successful phase three, and then we also need the OLE data on the label. The OLE data, obviously, by the time we file, will be five years in duration, and that's why we continue to submit it to the FDA so they understand it, and it was validated with the New England Journal of Medicine. There is a section within a label, just for people's understanding, and this is a detail, but it's important. There is section 14 in a label, and it requires you to disclose clinical studies, results, that help you understand how the medicine is effective and safe. It's a requirement. And so when you have a medicine that's a genetic medicine that treats chronically, and you have data that shows reduction, durable reductions in seizures, and improvements in cognition and behavior over that period of time, we believe that data will be in the label.

Andrew Tsai, Analyst — Jefferies

Great. it's great to see everything coming together the pieces coming together fingers crossed on the face of your success and congratulations on the

Ian F. Smith, CEO

progress yeah thank you Andrew thanks for taking the time today all thank you