All right, good afternoon, everyone. Welcome to Guggenheim Emerging Outlook Biotech Summit 2026. Our next presenting company is Stoke Therapeutics. Myself, Yatin Suneja, I have my colleague here, Delma Caiati. We will be moderating this session. From the company, we have the Chief Executive Officer, Ian Smith. Ian, thank you so much for making time for us. Thank you for attending our conference. Why don't you make some opening comments? Just tell us about the Stoke story, some of the upcoming milestone, and then Delma and I will sort of run the Q&A with you.
Yeah. Thank you, Yatin. Before I start, just may reference you to our SEC filings and our safe harbor statements. With that, I'm Ian Smith. I'm the Chief Executive of Stoke Therapeutics. I've been sitting in this role for nearly a year now. I think the best way to start is to tell you where I see the investor interest and the questions about the company. The last 12 months have been a transitional year, I feel for the company, in terms of the understanding of the disease that it's currently prioritizing, which is Dravet syndrome, and understanding of that disease and the devastation it has on these young children's lives and the families that surround them, and what our medicine may be doing for these children.
That evolution of understanding of Dravet syndrome and what our medicine may be doing for these children has been a function of providing data to the investment community, as well as data to, obviously, in the medical community to understand the impact our medicine is having on these children. The data that is most interesting to the investment community has been the initial phase 1/2 data, where we had a dose escalation, but we moved those patients from dose escalating into an open label study. Those patients have now been followed for 3 years, and as of today, it's close to 4 years post phase 1/2 that those patients have actually continued on medicine.
The findings in that effectively four-year data that we provided last year was that we reduced seizures on top of standard of care seizure medicines by up to 80%, roughly. And that reduction of seizures in these patients has continued now through that three-year OLE period, so it is durable. And despite being remarkable reduction in seizures, it's also durable. In addition to reducing seizures, we are also been measuring cognition and behavioral benefits in these patients. And we followed them over the three-year period in the OLE, and the data that we've provided to the investment community, as well as being supported in medical communities, is that we're seeing cognition and behavioral benefits in these children. It's as measured by a subjective assessment or an assessment called Vineland-3.
I can give you some examples as we go through the discussion today, what certain scores mean in this. But we are causing cognition and behavior to be placed back into these patients, that after the age of 18 months or 2 years, unfortunately, they stagnate in terms of their cognition and behavioral improvement. Unlike us, well, many of us in this room, you know, we develop more neurotypically. These children plateau from the age of 18 months to 2 years old, but what we're seeing is, we're actually, once they start to take our medicine over this 3-year OLE period, we are seeing improvements in cognition and behavior. That's the priority of the investors' attention at this point in time.
The company is transitioning towards focusing on the market, and I would like to say it as, there is now an understanding of the design of our phase 3 program that we're well into, and we expect to complete enrollment in second quarter of this year, second quarter, so next quarter, which means data in mid-2027. We're starting to focus investors' attention on what that may translate to a label. Label translates to good discussions with payers and allows you to promote the medicine the way that you believe it is affecting these children. So that is, that is our objective with Dravet, is to how does our medicine affect these children?
And, just today, we entered the clinic in our second disease area, which is a genetic loss of eyesight, ADOA, Autosomal Dominant Optic Atrophy, and first dose was administered just last week. And that's a dose-escalating study, and we'll keep you updated on that progression. And then, there are other medicines for other diseases that are coming towards IND and into the clinic, most notably SYNGAP1, which is very similar to a Dravet condition, where it's neurotypical development that they're not achieving in these children. And so, happy to talk about all of it today. Yeah.
Perfect. Thank you, Ian. Thank you for that. So as you know, we recently sort of picked up coverage of Stoke Therapeutics, and I think we are collecting feedback. So we're trying to really orient investor toward looking at this molecule completely different. This is not a seizure-reducing medication, right? Trying to get them to or bring them more towards the disease-modifying effect. So two questions for you. One is, like, how do you think investors should think about the magnitude and durability of effect versus standard of care? And then, what supports this notion, which we are also trying to educate investor around disease-modifying potential?
Yeah. So I'll start with seizures in terms of durability and effect. I wanna be very clear that there are good seizure medicines that are treating these kids now with Dravet. Fintepla, Epidiolex are very good medicines. The patients that we've treated are actually on these medicines. 50% of the patients that were in our studies were actually already on Fintepla. And so this reduction of seizures for the patients that are coming into our studies, that is approximately median of 80%, is on top of standard of care medicine. And you may say, you know, "How do you achieve that magnitude of reduction in seizures?" Well, it's because we go direct to the cause of the disease.
The pathophysiology of this disease is a lack of expression of NaV1.1. We upregulate the SCN1A gene, which expresses NaV1.1, and therefore we go to the root cause of the disease, and that's why you get this magnitude of reduction in the seizures on top of standard care medicines. But it's also the reason why you have durability, because once you're at steady state of expressing NaV1.1, you continue to suppress these seizures to this level. And we're now seeing this durability of seizure reduction out to what is effectively four years, which is nine months or so in the phase 1, 2, and they roll over into the OLE, which we've now provided data, three years of the OLE.
So that's the reason we're seeing this durability of response and the magnitude of response or reduction on top of standard care medicines. Once you understand that, that helps you understand, because we go to the root cause of the disease and we express NaV1.1 because of the medicine that you also start to create neurocognition and behavioral benefits. That's less about durability, that's actually more about improvement.
Yeah.
So a lot of these genetic medicines are created to slow the progression of the disease. We're actually turning it around by expressing the gene and causing neurocognition and behavioral benefits. And some of the stories. When I say stories, I want to be careful. This, these are patients that have been presented at medical conferences by leading physicians that are running our trials. And we provided data evidence as opposed to just saying, "Here's the Vineland score and improvement," but we've provided data evidence with videos and graphics to help people understand. And a couple of those patient case studies that are real, and as I say, presented at—by physicians at medical conferences.
We have a patient that is shown doing the Vineland score, where we're asking the young lady's name at baseline, and she doesn't respond. We ask her to look at her hand, or the physician or the neuropsych is asking her to look at her hand, she doesn't respond. She doesn't respond to her name, she doesn't respond to hand gesture. And a year later, and this young lady is 11 years old, a year later, same test, same questions, she's holding her hand up, she's touching her nose, she's responding to her name. And that has never been seen before in Dravet.
Wow!
It is a disease that it doesn't decline, but you just unfortunately flatline from age of 18 months or 2 years old. And so that's the kind of the practical understanding of what we're doing with neurocognition and behavior. And, you know, the technical phrases would be receptive communication-
Communication.
Expressive communication, motor skills, social skills. But underneath those, the questions are, and if I did it with you all here, I say, "Could you all look at my hand?" And you, I hope you're all looking at my hand. When you look at my hand and I do it again, "Can you look at my hand, please?" And you're all looking at my hand, you'd score 2. 2 is the highest score you can get, because you did it every time I asked. If you did it zero, you'd get 0. If you did it often or frequently, you'd get a 1 score. What we're seeing in our cognition and behavioral Vineland scores are scores up to 10 and 11 over a period of 12 or 18 months.
And so that's the importance of the OLE data, the longitudinal data in a genetic disease, that three years' worth of data are helping you understand what this medicine is doing for these kids.
Got it. Very, very remarkable. Two more questions, and I'll hand it over to Delma. How consistent is the effect across the type of patient, the ages of patients? And then what about the safety side of the equation? Like, how is the safety?
Yeah. So we are, we are studying ages 2 through 18. And we have not seen a differential response, whether you are a 4-year-old or a 15-year-old at this point in time. The N of that group that we've studied so far and provided data on is roughly 90 patients. It's now at around 80 patients, I would say. And what I mean by those numbers are 90% of the patients rolled over from the phase 1 studies into the OLE. And what we're doing is, we're looking at those patients, we are looking at the age categories, but we're not seeing any differential result.
The way to think about this is to think about whatever stage you're at, and whether you're a 3-year-old, 4-year-old, 5-year-old, the moment that you come on our drug and stay on our drug for a period of time, you start to spring off that 2-year-old line. So if you're a 5-year-old, you're still 2. If you're a 10-year-old, you're still 2. But the moment you start taking the drug, you have the opportunity to potentially start to develop. So you might turn a, a 5-year-old into a, I don't know, a 6-year-old in 3 years.
Mm.
It might sound as though it's marginal, but if you can imagine having a child that's, you know, 10 years old, and they're like a 2-year-old, but you're able to give them 2 or 3 years' worth of extra cognition and behavioral improvement. That's, that's significant.
Yep.
And so-
Very good.
Yeah.
Delma?
And maybe, zooming in on the EMPEROR study, so the pivotal study that you are running, some technical details there. So the primary endpoint is at six months, and that's on seizure reduction, so a bit different from what we have seen historically. And the secondary endpoints that are also important, as you were discussing before, Vineland, are at one year. So what's the reasoning behind the, the choice of timing, and how that informed from, the previous studies?
Yeah, so you described the study correctly. It is a 52-week study, where the primary endpoint is measured at actually week 28, and that's seizure reductions. And then Vineland is measured at week 52, and that's the secondary endpoint. The reason for the difference in timing is the mechanism of action. It quickly suppresses seizures, and so we can measure the primary endpoint at month 6. However, in terms of cognition and behavioral improvements, the drug needs to get into the system. You need to start expressing NaV1.1, and so we picked week 52 to measure the secondary endpoints to give the opportunity for the drug to get to steady state, and therefore start to see these cognition and behavioral improvements.
So that's, that was the purpose of the design of the study. It's 150-160 patients that are lumbar puncture sham control that is being conducted in the U.S., U.K., and Japan. And then there is another cohort of patients that are needle prick sham control that is being conducted in Europe, the four countries. And the reason for that is Europe preferred the needle prick sham control as opposed to a lumbar puncture sham control for those other... compared to those other countries.
Got it. And how is the study powered? What are the statistical assumptions there?
Yeah, the study is interestingly powered to one of the secondary endpoints. We powered it to a receptive communication endpoint, the secondary endpoint at week 52, for a p-value of 0.01, with a confidence level of 90, and to achieve a Vineland score of at least 2. And so it is conservatively powered when you think about we've powered it towards the secondary endpoints. And obviously, the magnitude of response with seizure reductions is heavily powered compared to the sham. I would just point out in that it's really important to understand the success of this study is driven by the primary endpoint.
Mm.
Let's say the oversize success of the study is actually based on not just Vineland secondary endpoints, but it's also the observed data, and I hope we have a chance to talk about observed data today. When I look at the company going through a transition of 2025 and people understanding the medicine and Dravet more, 2026, I think, is more about understanding how data that you've observed and collected over a 4- or 5-year period actually goes to a label. Which therefore, when you have a label that has primary, secondary, and observed data, how that you position that label with a payer and how it allows you to talk to the medical community and the advocacy community of what this medicine is doing.
The power of that 4-year data, that might be 5-year data by the time the medicine is approved, is what actually drives the uptake of the medicine and also the pricing of the medicine.
Yeah, and turning, focusing on that point that you were highlighting, so you recently had a very important conversation, discussion with the FDA about the opportunity to file for accelerated approval. They denied you that possibility, and they asked you about to add additional data. How are you thinking now about the accelerated approval path? Is that... Is it still open? Is what's your reasoning yeah, in that?
So, so simply stated, yes, it is still open. We met with the FDA in December, as we were required to by the FDA. It was a multidisciplinary meeting, which is your first meeting after you're granted breakthrough designation. And so that meeting, you're required to meet with the FDA to explain the pathophysiology of the disease, mechanism of action of your drug, and provide safety and efficacy data to work towards accelerated pathways to get to the patient faster. So that's what a multidisciplinary meeting was, and you're required to have that on the breakthrough designation. We had that meeting.
Why that meeting was a little different was because we had 3 years of data, 4 years of data, and the magnitude of response and the durability of response made us feel as though we should ask the FDA that question: Does this data support the filing of an NDA? And we did ask them that. It was the first question in our briefing book. We had a good discussion with them. Ultimately, what their pushback was, is that the data we provided them was a post hoc analysis and a cross-study comparison. And in their regulatory framework, they found that challenging to say, "Yes, proceed forward." However, in that discussion, they did acknowledge the data. It's just that it was post hoc, and they asked us to provide more data.
We're going to go back down and have a further discussion with them. We're gonna show them the data for the patients compared to their own baseline, so there is no cross-study comparison. The prior analysis we showed them was the data we provided at AES in December, which was a comparison to natural history, a propensity-matched comparison to natural history. So we're gonna go down and take away the cross-study comparison challenge away and show them against patient baseline. And where that comes out, I don't know. But what I do know is, as we take a week, a month, two months with this discussion with the FDA, the window of benefit-
Yeah
with an expedited pathway for the NDA is reducing. Because at the same time, in the last few months, we've understood the acceleration of the Phase III, and we communicated that earlier this year, that we should complete enrollment into our Phase III in second quarter of-
Yeah
... this year. So that's just around the corner. To think about trying to save six months of timeline to approval, you know, versus having the optimal label because you run a Phase III study, I find that somewhat challenging, and not necessarily just from a pricing perspective. But we want the label to represent what we think this medicine is doing. Not just seizure reduction, but is enhancing the opportunities for these kids to have some neurodevelopment benefit. And that's gonna be very important, not just for pricing, but for how we position this medicine for the patient.
Optimal label is exactly the key for your story because it's where you are differentiating.
Mm-hmm.
How does an optimal label would look like? So shall we think about the indication there, the Section 14? How would that be different from-
Yeah
... a fine line Epidiolex label?
Well, I appreciate you asking the question that way, because, you know, I, I would characterize the optimal label, the most optimal label, obviously, is primary endpoint with a low p-value, one or two of the five secondaries, with a p-value, and then the observed data. And the observed data would be the four-year data to show the curves going from lower left to higher right in terms of the progression in the neuro- more neurotypical development for these, these children.
I say it in that way, it being the optimal label, but even in the event that you get the primary endpoint and you get observed data, which is the four-year data of neurocognition and behavioral benefit, on the label, it allows you to promote the medicine and discuss with the payer that this is a medicine that reduces seizures and improves neurocognition and behavior as evidenced by the data. And so what will be really important is, yes, it would be good to hit those secondary endpoints, absolutely would. And then it will be easier to put the observed data on the label. But we need to make sure the FDA understands the observed data, that it was done with Good Clinical Practice, GCP, it has integrity.
We must publish it, we must, you know, have physicians and the medical community behind it, because that longitudinal data over four years or five years is actually a driver of true value of what this medicine is.
Got it. Good. Very good.
Sorry, and that's the Section 14 observed data-
Observed
... which you pointed out.
Okay, yes.
Thank you.
Yeah. Maybe I'll pick the commercial, and then we'll go to ADOA. Like, how would you define the commercial opportunity here? I mean, obviously, there's about 16,000 patients, but who are immediately addressable, who are the kids? How many are the adults?
Did we switch to ADOA, then? I'm sorry, I-
After that, I'll ask. First commercial on Dravet.
Oh, commercial on Dravet-
Yeah
... thank you. Yeah, so, it's roughly 36,000-40,000 patients, globally. The geographic splits of those patients is roughly 16,000 in U.S., North America, 16,000 rest of the world – well, top six geographies, plus Japan, of 7,000. And so that's how you get to 40,000. Most importantly for us, at Stoke, we have, obviously, the rights for this drug for the North American markets. We retained those in the collaboration we did with Biogen. The patient market and the patient count, the epidemiology suggests there's 16,000, but most importantly, there's approximately 6,000 addressable patients today. And we know the 6,000 because we've tracked the coding of medicines that treat Dravet children. So it's 6,000 that are readily addressable today.
Then there's another 2,000 that are readily addressable today that are aged between 18-24, which we'd anticipate treating as well. The addressable population is approximately 8,000 that's known today. Going from the 8,000-16,000, this market is more about screening and awareness, and we're working on that as well. It's a significant opportunity with a medicine that may bring this amount of value to a patient.
Got it. And then what about pricing? Obviously, disease-modifying can demand different pricing, so what are some comps benchmarks for us to look at?
Well, I think it's your typical disease-modifying medicines. And, disease-modifying, as in genetic medicines, you know, and good analogues would be... I worked at Vertex for 20 years, and, their medicines are good analogues. They change the course of a child's life that has cystic fibrosis. And then you go all the way, keep moving up from there and all the way through to something like Spinraza or even, Vertex's recent medicine for sickle cell.
Yeah.
And, and so it is in that kind of range-
Mm-hmm
... pending the data. As I'm talking this morning, and or this afternoon, sorry, we really believe this is a medicine that not just reduces seizures, but the longitudinal data supports the cognition and behavioral improvements. So it trends more towards a Spinraza-like medicine. And that's how we're thinking about it, as because Spinraza is a wonderful medicine in SMA, and helping those children have, frankly, longer lives.
Got it. Maybe two minutes on ADOA. So you dosed the first patient last week. Why is that an attractive opportunity for you? When are we gonna see the data? And then how should we think about, or how are you thinking about proof of concept there?
So yeah, we press released this morning, first dose in ADOA population. It's a genetic loss of eyesight. What we do there is we express OPA1, which is the gene that has the variant, but we upregulate OPA1. What it does, it increases mitochondrial function in the eye, which increases cell activity or cell energy, which improves vision. All our preclinical studies in non-human primates show that we increase vision based on flavoprotein fluorescence, which is a measure of mitochondrial function. So we went in the clinic with anticipation that we can improve vision-
Mm.
... not stop the declining vision in ADOA, but improve vision. And so we entered a phase I study, phase I, phase II study, this morning. We actually dosed last week. We wanted to see the patient was healthy through the first dose, and we're in a dose-escalating study. The timing and the readout from that study, we anticipate there to be 4 doses, dose escalations. We're starting at 0.1 milligrams injection into the eye, and then the fourth dose is 0.7 milligrams. We anticipate that the efficacious dosing may be around the third dose, and certainly around the fourth dose, and the timing of that would be towards the end of the year or early first half 2027.
How big is the market?
It's around 16,000 in the top seven geographies. So yeah-
It's a...
It's a large market.
One final question. What's the cash position and the burn rate?
We had approximately $400 million at the end of 2025. As we go forward, obviously, we've got a strong cash position, and Biogen funds 30% of the Dravet R&D. Our cash runs all the way through into 2028. So it, the cash runway runs way beyond when we read out data or when we potentially could be approved.
Got it. Very good, Ian. Thank you so much.
Thanks, Yatin.
Thank you.