Stoke Therapeutics, Inc. (STOK) Earnings Call Transcript & Summary

Good afternoon, and welcome to Stoke Therapeutics Second Quarter 2025 Conference Call. I'm Tommy Leggett, Chief Financial Officer at Stoke. Joining me for prepared remarks are Ian Smith, our Interim Chief Executive Officer; Dr. Barry Ticho, Chief Medical Officer; and Dr. Kimberly Parkerson, Head of Neurology Clinical Development. In addition, Jason Hoitt, our Chief Patient Officer, will participate in Q&A. As a reminder, today's webcast slides are available in the Investors section of our website. This webcast is being recorded and will be available for replay later today. Before we begin, please note that today's discussion includes forward-looking statements. These are subject to risks and uncertainties, and actual results may differ materially. Please refer to the filings with the SEC for additional information. With that, I'll turn the call over to Ian.

Welcome, everyone, and thank you for joining us today. I've been working with Stoke Therapeutics for two years as a Director and adviser, more recently as the CEO. There's been a great opportunity to get to know the team, the medicines and the disease areas and to support the company through this growth period. The key priority is obviously Dravet syndrome and zorevunersen as we work to deliver a disease-modifying medicine to patients. What we're seeing with zorevunersen may be new to the field of epilepsy, but I'm struck with the familiarity of the feeling of being part of something very special, again, which is creating a new first-in-class disease-modifying medicine for patients who desperately need it, potentially changing the lives of these children suffering from Dravet syndrome. Let me start by saying that Stoke is in a strong growth position, defined by a late-stage registrational medicine, well-capitalized balance sheet, expanding pipeline of potential medicines and a very strong partner in Biogen with expansive capabilities to support zorevunersen outside of North America. Our Phase III EMPEROR study in patients with Dravet syndrome is off to a strong start with sites initiated in the U.S., U.K. and Japan and Europe expected to initiate in early 2026. The first patient is now dosed, and we anticipate rapid enrollment based on the high level of awareness of the study, competitive participation among the study sites and importantly, the urgent patient need. We continue to generate data that support our understanding of zorevunersen from our Phase I/II and OLE studies. We have prioritized sharing this information broadly as the field begins to transition from a symptomatic treatment to a potentially disease-modifying medicine for Dravet syndrome. We continue to educate around the seizure reductions with our medicine, and those reductions are on top of standard of care antiseizure medicines. More recently, we shared data that was used to inform the assessments of behavior and cognition and the powering of our Phase III study, specifically the substantial improvements in cognition and behavior indicated with a dosing level that is similar to and consistent with the one we are using in our Phase III study. And today, we are sharing with you top line results from the third year of our open-label extension studies. These data support the long-term potential of zorevunersen to modify the course of Dravet syndrome as indicated by durable seizure reductions on top of what can be achieved with antiseizure regimens as well as continuing improvements in cognition and behavior. Importantly, these long-term follow-up data continued to demonstrate a favorable safety profile. Beyond Dravet, we see significant potential to develop disease-modifying medicines for additional genetic diseases. We have advanced STK-002 into Phase I clinical development for Autosomal Dominant Optic Atrophy. Like Dravet, ADOA is a haploinsufficient disease, and we are uniquely positioned to treat by restoring naturally occurring OPA1 protein expression using our antisense oligonucleotide approach toward the goal of preventing further loss of eyesights and possibly improving vision. Now to our collaboration with Biogen. Signed in February. This brings global expertise commercializing high-value disease-modifying medicines for rare genetic diseases and strengthens our balance sheet. The terms of the collaboration retains significant value to Stoke while enhancing our ability to deliver zorevunersen to patients globally. We have a strong balance sheet and are well funded through Phase III readout, which is anticipated in the second half of 2027. And our balance sheet and projected investment support a cash runway through to mid-2028. We continue to build Stoke's internal capabilities by enhancing our leadership team and strengthening key functions, including regulatory, medical affairs and commercial, all fundamentally important functions at our stage of growth. In short, we are establishing a clear trajectory for value creation for patients, for employees and for our investors. With that, I'll turn the call over to Barry, who will discuss our Phase III study design and progress.

Thank you, Ian. This is a very exciting time here at Stoke as we take zorevunersen into this next phase of clinical development and potential registration and approval. Dravet syndrome is a severe lifelong developmental and epileptic encephalopathy that becomes symptomatic around one year of age. For the vast majority of patients, the cause is insufficient levels of the MAb1.1 protein in the brain. There are many medicines available to treat the seizures associated with Dravet syndrome. These medicines have undoubtedly made a difference for patients, but they just aren't enough. Most patients continue to suffer from seizures and few achieve seizure freedom. Furthermore, side effects of the antiseizure regimen also present their own challenges for patients and their caregivers. While Dravet syndrome may be best known for its seizure burden, the effects go far beyond seizures. Nearly all patients suffer from one or more behavioral and cognitive effects. The antiseizure medicines were not intended to address these effects. We intend to change that. Zorevunersen is an antisense oligonucleotide designed to restore naturally occurring Nav1.1 protein levels. As such, it has the potential to be the first disease-modifying therapy for Dravet syndrome. As we designed our Phase III study, we were fortunate to have a large data set available to inform key decisions related to dose level, dose frequency, study endpoint assessments and powering. Here on Slide 8, you see the dramatic reductions in seizures demonstrated in our Phase I/II studies among patients treated with initial doses of 70 milligrams of zorevunersen on top of standard antiseizure medicines. The most substantial reductions were observed among patients who received either two or three doses of 70 milligrams. Based on these data and additional modeling analysis, we selected a two-dose loading regimen Phase III, with seizures as our primary endpoint for the trial. After receiving their last Phase I/II dose, patients were followed for at least six months before restarting treatment in the open-label extension study. When designing our Phase III program, we used the initial eight months of data from the OLE that showed substantial and durable reductions in seizures and a favorable safety profile to inform our maintenance dose. We now have an additional year of data on these patients, which are shown here and support the long-term durable reductions in seizures. Kim will discuss these data later in the call. These new long-term data give us confidence in our loading and maintenance dosing as well as the durability of effects. On Slide 10, you see one of the key analyses that informed our thinking on the Phase III design, which Ian referenced earlier and was presented at EPNS in Munich last month. The analysis shown on the left was performed to assess potential effects on cognition behavior using data from patients treated with dose levels that were similar to and consistent with our Phase III dose regimen. The effects are striking, particularly in the context of the results from a matched cohort of patients followed in our natural history study. Little to no change was detected in the natural history that patients treated with zorevunersen show cognition and behavioral benefits in the five key subdomains that comprise our key secondary endpoints in EMPEROR. On this slide, you can see the phase III design. Patients who enroll in EMPEROR will be randomized 1:1 to zorevunersen or to sham. In both study arms, patients will continue to receive standard of care antiseizure medicines. Consistent with the data I just reviewed, patients in the zorevunersen arm will receive two loading doses of 70 milligrams, followed by two maintenance doses of 45 milligrams. An open-label extension treatment period will allow all patients the opportunity to receive treatment with zorevunersen following 52 weeks of treatment in the study. I will now review the EMPEROR study design and conduct in more detail on Slide 12. Primary endpoint is change from baseline in major motor seizure frequency at week 28. Durability of effect on seizures will be measured as a secondary endpoint at week 52. Zorevunersen is a potential first-in-class disease-modifying medicine. So EMPEROR will also measure effects on behavior and cognition. Powering for these secondary endpoints is robust and designed with the intent to show statistical significance on both individual subdomain and composite assessments. We were pleased with the level of interaction and input from global regulatory agencies and our advisers as we work together to design the EMPEROR trial. Through these discussions, we aligned around a double-blind sham-controlled study with lumbar puncture for all patients. As individual European countries got involved later in the process, they required a modification to the sham arm to proceed. In order to ensure our commitment to patients in Europe, we now plan to add a cohort of at least 20 patients in Europe where sham will be administered via needle press. This cohort of patients will be in addition to and complement the originally planned group of approximately 150 patients who will be randomized to receive zorevunersen or sham with lumbar puncture in the U.S., U.K. and Japan. We expect to activate at least half of the 70 study sites by year-end and European sites to initiate by early 2026. We see significant and growing global interest in EMPEROR, which supports our expectation to complete enrollment in the second half of 2026 with a Phase III data readout in the second half of 2027. EMPEROR is off to a great start. We have had a high degree of confidence in the study design based on a large data set and the ongoing patient need. Our natural history study has proven invaluable in understanding the effects that Dravet syndrome has on patients over time as well as the limitations of the current standard of care medicine. Our experience with zorevunersen in Phase I/II and the ongoing OLEs has helped us understand the initial and ongoing effects of zorevunersen and has demonstrated an encouraging long-term safety profile. Our assessments of behavior and cognition have been validated and our endpoints and powering are informed by our years of experience and data. As announced yesterday, the first patient was dosed in EMPEROR. Study investigators have identified approximately 130 potential participants, and that number continues to increase. More than 10 clinical sites have initiated across the U.S., U.K. and Japan with more coming online weekly. This is exactly the kind of start we were hoping for, and it positions us well for continued enrollment success. With that, I would like to turn the call over to Kim for a review of the new three-year open-label extension date.

Thanks, Barry. Before I share the new data, I want to express my gratitude to the patients, caregivers, advocates, clinicians and stoke employees who have brought zorevunersen to where it is today. Having treated patients with drug-resistant epilepsies for many years in my clinical practice, I can tell you that these data are profound and meaningful. The community has been waiting a long time for something that would address the syndrome, not just the seizures. We remain confident that we may very well see a significant advance in treatment for patients, which would have impacts for them and also for their families. With that, it's my honor to share the new 36-month data. More than 90% or 75 eligible patients who completed treatment in the Phase I/II studies continued treatment with zorevunersen in the open-label extension studies. The high 77% retention to date in the OLEs has provided us with a robust data set to assess the long-term effects of zorevunersen in these patients. Here on Slide 15, you see the three-year effect on seizures observed across patients treated with zorevunersen in our OLE studies following treatment in the Phase I/II studies. I'd like to call your attention to the top blue line, which shows patients treated with less than 70-milligram loading doses in the Phase I/II studies before continuing treatment in the OLE. As patients transition to more stable 45-milligram dosing, you begin to see on the right side of the graph, a trend toward further reductions in seizures. Turning our attention to the orange line, you see the substantial and durable effect for the patients treated with loading doses of 70 milligrams, followed by continued dosing of 30 milligrams or 45 milligrams every four months in the OLEs. For these patients, we observed a median reduction of 59% to 91% in major motor seizure frequency across each time point through month 20. The durability of effect is consistent with what we would expect for a disease-modifying medicine and supports our EMPEROR Phase III registrational study. But we know Dravet syndrome is more than just a seizure disorder. Dravet syndrome is a complex disease that presents daily and life-altering challenges for patients, their families and their caregivers. On the left, you see the impact on patient health and well-being. For most, seizures remain the primary comorbidity. However, Dravet impacts all aspects of life for patients and their families, some of which are shown on the right. On Slide 17, you see a graph that illustrates expectations for the development of a neurotypical child shown with the top line and a child with Dravet syndrome shown with the bottom line. Consistent with findings from natural history data, including results from our two-year BUTTERFLY study, patients with Dravet syndrome experienced minimal improvements in cognition and behavior. Overall, patient development begins to plateau within the first several years of their life. And over time, they fall further and further behind their neurotypical peers and their ability to achieve developmental milestones. The Vineland-3 assessment is helping us measure changes in cognition and behavior in patients with Dravet syndrome. I'll now review what the assessment is and how it works. As summarized on Slide 18, Vineland-3 is a clinically validated and widely used tool for assessing neurodevelopment over time. It is typically administered through a semi-structured interview with the patient's caregiver that is conducted by neuropsychologists or other trained raters. There are four domains evaluated with Vineland-3: communication, motor skills, socialization and daily living. These domains are broken down into a series of subdomains, which are evaluated on a point system scale. We use the Vineland-3 across multiple of our clinical studies, and it is now being used to evaluate key secondary endpoints in our Phase III EMPEROR study. Our natural history study also helped in the selection of Vineland-3 as an assessment for our clinical studies. So now today, we are showing data on Slide 19, which are quite remarkable and striking. What you see here is the progression of Vineland-3 results for patients treated in the OLE studies with zorevunersen over one, two, and three years following treatment in the Phase I/II studies. For this analysis, patients were measured against their own baseline scores at entry into the OLE and demonstrated improvements through year 1 of the OLEs and continuing improvements in year two and year three. The improvements you see here are in addition to any improvements the patients may have experienced during the nine-month Phase I/II treatment period. Notably, some patients received doses as low as 10 milligrams upon entering the OLEs. As a reminder, small changes on the Vineland-3 are considered meaningful to clinicians and caregivers of patients with Dravet syndrome. To give you context, our published survey indicated raw score improvements ranging from 1 to 3 points over a year across individual subdomains would be considered clinically meaningful to at least half of caregivers. The 36-month data show profound changes, addressing the underlying protein deficiency appears to restore function and help patients achieve more of their developmental milestones. In effect, zorevunersen appears to be narrowing the gap between the normal course of disease and neurotypical development that I showed you earlier. While these graphs get clinicians like me really excited, hearing caregivers and clinicians talk about what these data mean in real life only increases the sense of urgency we feel here at Stoke to advance zorevunersen to patients. In addition to compelling efficacy data, we are highly encouraged by the safety profile observed across our studies to date. This safety summary represents over four years of clinical data, including the first year of treatment in the Phase I/II studies, followed by over three years of treatment in the OLE. Over this time period, zorevunersen has been generally well tolerated. 81 patients received at least one dose of zorevunersen. In the Phase I/IIa studies, 30% of patients experienced a treatment-emergent adverse event related to the study drug. The most common were CSF protein elevations, which we continue to observe in the OLEs. Approximately 86% of patients have developed elevated CSF protein levels, of which 45% have been classified as a treatment-emergent adverse event due to the laboratory values. Importantly, no clinical manifestations have been associated with CSF protein elevations. Only one patient has discontinued treatment due to elevated CSF protein levels. Treatment-emergent serious adverse events have been reported in 29% of patients in the OLEs, and none have been attributed to study drug. Across all studies, only one patient experienced SUSARs. To date, more than 700 doses of zorevunersen have been administered across the studies. Patients have received up to 15 doses of zorevunersen with 41 patients having received 10 or more doses. This is highly encouraging as we think about the Phase III design and ongoing treatment. I will now turn the call back over to Barry.

Thank you, Kim. Stoke has long believed in the potential of its platform. And today, I'm pleased to share that we have initiated a Phase I study of STK-002 for Autosomal Dominant Optic Atrophy, the most common inherited optic nerve disorder. From a genetic and mechanistic perspective, ADOA shares a lot of similarity with Dravet. The majority of cases of ADOA are caused by mutation in the OPA1 gene, which leads to diminished protein function or haluinsufficiency related to OPA1 protein deficiency. In a healthy eye, the OPA1 protein plays a key role in maintaining mitochondrial structure and function. Patients with ADOA reduced levels or function of the OPA1 protein, pairs mitochondrial function and results in decreased energy production. Without sufficient OPA1 protein, the retinal ganglion cells cannot meet the metabolic demands and gradually degenerate. The result is optic nerve atrophy and progressive vision loss. Approximately 80% of patients are symptomatic by age 10 and about half of all patients are legally blind. About 1 out of every 30,000 people around the world are estimated to have ADOA with a higher incidence of approximately 1 out of 10,000 in Denmark due to a founder effect. We estimate that approximately 13,000 patients are currently living with ADOA across seven key geographies, including the U.S., EU5 and Denmark. As described on Slide 23, we have generated compelling preclinical findings that support advancement of STK-002 into the clinic. We have observed increased OPA1 protein levels and improved mitochondrial function in ADOA patient fibroblasts treated with STK-002. Increases in OPA1 protein have been demonstrated with STK-002 in vitro and in vivo, including dose-related increases in OPA1 protein expression in nonhuman primate retinal ganglion cells following intravitreal injection. STK-002 has been found to be well tolerated across multiple species. In addition, today, we are sharing new efficacy and safety data from a study of STK-002 conducted in a nonhuman primate model of ADOA that has a mutation in the OPA1 gene, resulting in insufficient protein in the mitochondria. These NHPs have disease characteristics similar to humans with ADOA and therefore, represent a unique opportunity to evaluate STK-002. The new data we are sharing today on Slide 24 show improvement in mitochondrial and retinal function three and five months after treatment with STK-002 in an NHP model of ADOA. Intravitreal injections of single doses of STK-002 were well tolerated in diseased NHPIs. These data suggest the potential for STK-002 stabilize or perhaps even improve vision by restoring functional protein levels. Based on these data, we have initiated a Phase I study of STK-002 in ADOA patients. OSPREY is an open-label single ascending dose study designed primarily to evaluate safe, but which will include assessments of clinical activities. Study initiated in the U.K. last week, and we expect sites in Europe to initiate later this year. There are currently no treatments approved for ADOA. We believe our approach represents a unique opportunity to address the underlying cause of ADOA by restoring naturally occurring protein function. I will now hand the call over to Tommy to discuss our financial summary.

Thank you, Barry. For those following along, I'll be speaking to Stoke's financial results as provided in the 10-Q and earnings release. Stoke is in a strong financial position as we enter Phase III and expand our platform into new disease areas, starting with ADOA. We ended the second quarter with $355 million in cash, cash equivalents and marketable securities, which we continue to expect will fund operations beyond the Phase III readout and into launch readiness to mid-2028. Total revenue for the quarter was $13.8 million, which is driven by revenue from our Arcadia and Biogen collaborations of $10.6 million and $3.2 million, respectively. We expect revenue from Biogen to increase going forward as we continue to execute on EMPEROR and our other zorevunersen related activities. Our net loss for the quarter was $23.5 million or $0.40 per share, slightly improved from the prior year period despite a $6.9 million year-over-year increase in operating expenses. Operating expenses during the second quarter were comprised of R&D expense of $25.9 million, which reflects continued investment in our Dravet program in key areas of our business, namely medical affairs, while also expanding our pipeline. G&A expenses of $15.3 million were largely driven by the continued expansion of our commercial team and capabilities. In summary, our strong balance sheet enabled us to invest in zorevunersen, our pipeline, including ADOA and capabilities while maintaining our cash runway to mid-2028. I'll now turn the call back over to Ian for closing remarks.

Thanks, Tommy. Our recent progress and overwhelmingly positive response to the start of EMPEROR underscore the unique potential of zorevunersen. The era of disease modification is upon us. There is a palpable feeling within Stoke as we work together to create something truly special, a medicine that will transform the lives of patients and their families and realize the potential of our platform. We have the unique opportunity to build a company around the technology, which can create even more medicines to help even more patients. I am energized and committed to doing everything I can to support the team as they embark on this journey. Thank you, and I will now open the call for questions.

[Operator Instructions] Your first question comes from the line of Laura Chico from Wedbush.

I wanted to start with a regulatory question for zorevunersen. Ian, first, you've got three-year OLE data now in hand along with natural history data. Second, if I'm understanding this correctly, you've got 130 patients already in prescreening for a Phase III study that's got a target enrollment of 170 patients. And then third, you've got breakthrough therapy designation. So I guess my question is, looking at the guidance documents for accelerated approval for serious conditions, I'm wondering if you could help us understand how could you explore a faster path to market for zorevunersen?

Laura, thank you for the call. I want to, first of all, start by reminding everybody that zorevunersen has a breakthrough designated class. We applied for breakthrough designation last year in 2024, and we were granted it towards the end of 2024. We were granted breakthrough designation for Dravet syndrome. What that means is the FDA reviewed our safety and efficacy data as of last year and saw the safety efficacy data and efficacy, specifically for seizures as well as cognition and behavioral benefits. That's really important to understand. So the FDA has already seen a lot of data. We chose to wait under the normal process of breakthrough designation to collect further data and the key piece of data being this extension of the OLE being into 36 months now with the hope that the seizures continue to be reduced and durable and durably reduced as well as the cognition and behavioral benefits continue to extend. We were very -- we were thrilled that that's how it played out. We will take all of this data to the FDA. We have a responsibility to the FDA to discuss all of our data to discuss the disease itself and how our medicine may address this disease, and that will all be part of a discussion with the FDA in the second half of this year. We have a responsibility to see whether we can get this medicine to patients as fast as possible given the data that we've seen. We have begun the Phase III study, not changing time lines in terms of recruitment and getting the medicine to patients through a validated Phase III study. And if things change, we'll let you know at that time. But yes, we are very excited about this data. And under the breakthrough designation, we'll have all kinds of discussions with the FDA.

And maybe one quick follow-up, if I could. In terms of what is the range of outcomes then, I guess, in terms of the second half following the meeting, would this be a potential earlier-than-expected filing? Or I guess maybe if you could walk through that, that would be helpful. And I'll hop back in queue.

Yes, it's difficult to answer that question, Laura. Somebody actually asked me a very similar question just the other day internally. And there is no answer because the answer is actually so broad because the way it works with the process is we create a briefing book, we go down to the FDA and we share all data as well as the magnitude of benefits and also safety that we're having in these patients. And you ask a number of different questions to the FDA. And sometimes you get support and sometimes you get pushback and sometimes you get amendments and changes and you continue the process forward. So it's very difficult for me to say what is the potential outcome of those discussions. It is all really backwards and forwards with the FDA. And if anything changes from our current time lines, we'll advise you at that time.

Our next question comes from the line of Marc Goodman from Leerink.

Can you help us understand the magnitude of the cognition and behavior improvements in the Vineland-3 data in this OLE data that you're showing? Just talk about the clinically meaningfulness of the changes. I don't know, context of what these changes mean in real life for the patients, please, and the caregivers.

Thanks, Marc, for the question. I'll try to address that. So really, as regards, I think, cognitive and behavioral outcomes, I think first, it's really important to recognize that patients with Dravet, really, there's not one size fits all for all these patients. I think that they all come in a bit heterogeneous at baseline. I think, that being said, I would say that we're all beginning to learn more about what these changes look like. And I think that's through a few things. One is we certainly have anecdotes from investigators of how these patients have changed over time. And then I think that we had one of our investigators present videos of one patient in December last year. This certainly was something I think that was really emotional for all of us. And I think a lot of that on my mind, I've certainly seen patients in clinic with Dravet syndrome. And really, this change, I think, that we saw, at least in that patient was something that is not something we would expect in a normal course of disease with Dravet. So I know that certainly, this won't help you at the moment. But I think that over the course of our Phase III study, I think we're going to get a good handle on what these patients look like and what they do over time. And I really think that's going to be informative for the community as a whole across a lot of different DEEs and neurodevelopmental disorders. I think if we really focus on the numbers, in particular, the numbers from the data across the three years, in particular for those five key subdomains, but certainly across even others that we didn't see early change in, these points are really 8 to 10-ish points of change. And if we think about the data that caregivers have really given us through the survey we did with them, they identified even 1 to 3 points as something clinically meaningful. So, I think in short, I can't paint a perfect picture, I think, of what before and after looks like. But I think the magnitude of change across these Vineland subdomains will be things that families can see in the daily life of these patients across multiple developmental areas. And I think that with time, hopefully, they'll have significant impact and really the quality of life of patients and their families both.

I mean were you surprised just at the changes from 24 months to 36 months? I mean on some of these domains, it was pretty amazing.

Yes. I think we all looked at it. And I have to say, I think several people internally had almost tears come to their eyes to say, hey, we're really continuing to see improvements here. So I think it was really remarkable for all of us. And I certainly am looking forward to replicating this in a larger group of patients as we really get this Phase III done.

Marc, this is Barry. I'll just add in addition to the cognition and behavior effects that we saw over the three years, what we were seeing is that as the patients were on the maintenance dose of 45 milligrams, those patients started to move more and more towards a greater degree of seizure reduction, a trend towards reduction. And so that tells us also that, first of all, the medicine is working according to the mechanism of action of the medicine to reduce seizures as well as the effects of the Nav1.1 deficiency, but also that as we continue patients for long enough and observe them for long enough, they're seeing a benefit as we get to the higher dose.

Your next question comes from the line of Andrew Tsai from Jefferies.

Thanks for sharing the data. So in the past, you've mentioned how you've done various analyses to help you design the Phase III EMPEROR study. So can you explain what data you've used exactly and give us a sense of how those data help inform your powering assumptions on Vineland specifically?

Barry, why don't you take the question? I mean, focus on seizures and cognition and behavior.

Yes. Thanks, Andrew, for the question. So the data that we used to determine the power calculations for the Phase III partially came from an analysis that we presented earlier this year at the European Paediatric Neurology Society, where we looked at a response in a group of patients who had a dosing regimen that was similar to and consistent with the Phase III study. And when we looked at those responses on the Vineland, we saw that when we modeled it out for one year, those patients had all had substantial improvements in the cognition and behavior based on the Vineland score. That was especially in contrast to a match group of patients in the -- from our natural history study, we showed very little change at all in that score. So that was a substantial contrast and shows the effect. And we then powered the study based on that response, and we did give a change in the point score based on what we expected to be a potential placebo response. So we reduced it by 20% and use that to power. That's where we came up with the 150 patient score. I'll note that we also have long-term data from open-label extension studies, which gives us very much confidence on the safety long term. We have now patients who've been dosed for over four years. We've given almost 800 doses of zorevunersen in our studies. And we have patients who've gotten multiple doses over 15 -- some of them up to 15 doses of the drug. So that safety profile is also helping us with the design of the study. And then finally, for the seizure endpoints, we also looked at the response based on the two-dose loading regimen as well as the 45-milligram dosing. And we also powered that very conservatively with a 0.01 p-value and a 90% power and a generous placebo effect as well. So we're very confident in the powering of the study.

Yes. And let's just say, come second half 2027 when the data arrives. Can you describe what constitutes stat sig in Vineland? How many of the five individual subdomains need to hit stat sig? And for each subdomain, what kind of placebo-adjusted delta do you need to see to be stat sig?

Yes. We've powered the study based on what we think is a clinically meaningful change. So that power is based on response that we had from a survey that we mentioned was a 2 to 3 point -- or sorry, 1- to 3-point change in the raw score of the Vineland. And so statistically significant would be a change in that range. But we think that based on our data that we have, we will potentially have greater changes than that. So, again, this was a conservative assessment with a powering that was used to allow for a generous potential placebo effect.

Our next question comes from the line of Pete Stavropoulos from Cantor Fitzgerald.

Tommy and team, congrats on the progress and the data for the OLE. First question is, are there any trends in terms of degree of seizure reduction and benefit measured by the neurodevelopmental scale Vineland-3. And I'm curious to know if younger patients are driving improvements on the subdomains versus older patients.

Maybe I'll take the first part of that question, Barry, Kim, and then ask you guys to kind of talk about kind of the subsets and individual -- not individual but groups of patients' reactions. One of the striking results from the 36-month data from the OLE study was actually month 24 to month 36, where our low-dose patients, it's on the Slide #15, where the lower dose patients continued to see a reduction in seizures. And what I mean by continue to see a reduction in seizures, they continue to go lower. So they came into the study with reduced seizures, but the seizure reduction for those low-dose patients was not as significant. But as they continued through the one-year, two-year and three-year on a consistent dose of 45 milligrams every four months, we saw those seizures actually further reduce. It's on the slide, as you can see, but it was one of the more striking responses that we see that goes to the mechanism of action of this drug and restoring Nav1.1 in the brain and therefore, causing these children to respond better to our drug over a longer duration of time. That is one of the trends that we were very excited about. As for trends with different patient demographics, Barry, Kim, do you?

Yes. So I think you had a couple of other kind of points in that question. One, I think, had to do with kind of the seizure reduction and the Vineland responses. And I think that's a really difficult kind of correlation, I think, to do, particularly with the overall kind of small sample size in total kind of for a Phase I study across different doses. If you really look at the 70-milligram patients, the vast majority of those patients had over 50% seizure reduction. So then when you try to go to some of the lower doses, there's a lot more variability in their seizure response. And so really being able to correlate seizure responses and viand are really tough right now. I think that's something we can try to do more with certainly Phase III. I think with respect to the younger and older patients, for the Phase III, we've elected to put in patients 2 to 18 years of age because we've seen changes across the spectrum of ages in our Phase I/II study. And so I think that certainly looking across the different sub domains, some do have appear, at least by kind of a covariate to have some preference towards younger and older versus older than younger. But I think that we're confident that we can really move the bar, I think, across all of these patients from young to older. I think that from the mechanism of action of the drug, we're upregulating the fundamental problem of this disease. And there's really no reason to think that maybe it's going to take more time. We don't know. There may be changes in the neural networks, those sorts of things, but there's really no fundamental reason why upregulating Nav1.1 even in a later age cannot produce benefits in these patients. And so that's why we're studying 2 to 18 at least right now. And certainly, the Phase III is going to give us more insights, I think, into that.

All right. Yes. I mean I definitely agree that you may see benefit across all ages, but my assumption is and many investors are assuming that the younger you go, the more sort of dramatic outcome you may have on these neurodevelopmental sort of outcomes. So curious, out of the 150 plus 20 patients that you're enrolling in the Phase III, are you sort of capping the number of older patients or enrolling a certain -- a minimum number of younger patients in the study?

So, Pete, maybe I'll take the kind of the front end of your second question, which is what we're seeing here in the data is definitely a rush to treat patients as young as possible. If you treat a two-year-old, for example, you may have the possibility of changing the course of their lives. Treating a 15-year-old that may have more advanced disease, whether you can recover them may be challenging. I mentioned in my prepared remarks that I have a familiarity in terms of feeling for zorevunersen for my prior role in cystic fibrosis. And we always try to get the medicine to the children at the earliest a as possible because if you can correct the fundamental causal biology of Dravet, then you may put these kids on a more normal path of development and prevent their seizures. And so it is a goal of the program. But as far as saying do patients respond differently at different ages, the fundamental mechanism of action, we should have responses. They may have different responses at different ages. But the way that clinical trial has been designed is in the type group of 2 through 18. There is a slight loading of patients 7 through 10, I believe it is, but it should be beneficial for the patients in the age group 2 through 18.

Our next question comes from the line of Rudy Li from Chardan.

Congrats on the Phase III progress. So I have a question regarding the OLE data. So can you talk about the higher incidence of CSF protein elevations in the OLE part? Any specific concern, especially for the 45% of patients classified as treatment-emergent AEs. Just curious what can be attributed to the higher levels in the OLE study?

Rudy, good to hear from you. Thanks for the question. As far as the CSF protein levels go, first of all, as a reminder, that is a laboratory finding, and it occurs as a standard measure every time a child has or one of the patients has a CSF lumbar puncture done, we do that as a routine test. Most importantly, we've looked specifically for any effects, clinical effects of the elevated CSF protein, and we have not seen any in the patients -- in the 81 patients that we've dosed so far. The elevated CSF protein, those are a class effect. So they're known to occur in other intrathecally administered oligonucleotides, approved ones as well. And so the reason that we see it now later with additional dosing, it may be more of a procedure-related effect. And again, the point is that we now have long-term data with patients dosed for some of them for over four years. And the safety profile continues to support moving into Phase III. And again, we have, from the CSF protein perspective, not seen any reason that, that should prevent us from moving into the study.

Our next question comes from the line of [ Jeanie Kim ] for Tom from BTIG.

I wanted to ask a little bit more about the decision to explore STK-002 in Autosomal Dominant Optic Atrophy. What prompted this decision to pursue this now in this condition? And any comments on the most key data outcomes from your nonhuman primate study that contributed to the rationale would be super helpful.

Jeanie, I'll take the front end of that question, and then maybe Barry can tell you about the data we collected. But over the last six months, when asked by investors and analysts, I've talked about a process that the company went through. And that was a process where you effectively do an evaluation of the opportunity in ADOA. What that means is understanding patients that you may treat and provide benefit to. It's how you may run the clinical studies and what your preclinical safety efficacy package is. And it's a complete assessment of the disease area. And following that assessment and frankly, the data that we received relating to the nonhuman primates, that gave us the confidence and frankly, excitement to go into study ADOA with STK-002. I will just point out that in these genetic diseases where you have a slowly progressing disease and with ADOA, you have a slowly progressing loss of eyesight, you really want to have significant efficacy to be able to run the clinical trial because you're trying to separate from natural history. And so when we saw the nonhuman primate data where we potentially improved vision, it gave us the confidence to move into clinical development, knowing that we can study these patients with STK-002. And maybe, Barry, if you want to specifically talk to that nonhuman primate data, which really was the trigger.

Yes, Jeanie, thanks again for the question. And we are very excited about these data. We're very excited about the opportunity to treat this disease. I'll tell you both my father is an ophthalmologist, my brother is an ophthalmologist. And when I told them that we were treating patients or trying to find a treatment for patients with ADOA, he said to me, wow, that's great because I have several families I follow, and we cannot do anything for them. And since I mentioned it to him, he keeps every few weeks, he calls me and says, I found another patient. So this is a large group of patients that are undiagnosed and have no treatment options at all. It's a very exciting time for us here at the company. And what made us even more exciting with the data from this monkey model, this monkey model was a spontaneously occurring naturally occurring monkey that had a family of monkey that had the same mutation in the OPA1 gene that we find in some patients with ADOA. And they had a very similar profile in terms of some of the testing that we did in our natural history study of patients with ADOA, we found those very similar ones in monkeys with ADOA. So when we looked at a few specific tests, one of them being something called the flavoprotein fluorescence, which measures mitochondrial function, we found that just as we found in people with ADOA, we found in the monkeys that they had high levels of this FPF because the mitochondria were not functioning well. And when we injected 002 into the monkeys, we saw that the FPF levels either stabilized or actually improved within a short time within less than half a year. That gave us a lot of encouragement that we might be able to see this as a biomarker in a clinical trial. And then we also looked at using something called the electroretinogram at the functioning of the nerve because this is an optic nerve disease. And when we looked at the patients we see, we know that those patients have abnormal ERGs. And now when we looked in the monkeys after they were dosed with 002, we saw that the ERG pattern improved. The nerve function was improving. So these are the best measures that we could have in an animal model since we can't actually measure vision, but it gives us a high degree of confidence that when we start treatment in people that we may be able to see actual visual improvement after treatment with 002.

Our next question comes from the line of Jessica Fye from JPMorgan.

I was curious, with 130 patients identified in prescreening and the 170 target enrollment for your Phase III trial in Dravet, can you talk about how you're going to be communicating enrollment updates? Should we expect sort of quarterly progress updates? And how do you think about the possibility of delivering Phase III data prior to the back half of '27, just given how many patients are in prescreening?

Thanks, Jess. I'll take your question. If you don't mind, I'll broaden it in terms of really, you're asking about time lines and disclosures. So we are very excited. The demand from the patients' families to push their children into the trial and the speed that they want to go through screening. And -- but the rate-limiting feature is actually opening of the clinical trial sites. And so we're going through that process, and we're making good progress. As of now, it's still early in the trial. Everything is pointing positively. But at the point that we do see a change in what we've communicated as being recruitment will complete in the second half of 2026. When we see a change from that, we will actually communicate that at the time in an appropriate forum. That same goes for the data. Obviously, it's a one-year trial. So if we've got a one-year recruitment period, that means it's one year from ending of recruitment. So we're still maintaining the delivery of the top line data in the second half of 2027. The other thing I want to refer to in terms of disclosure is, as you're seeing, the company is taking the opportunity, whether it's a medical conference, it's a quarterly conference call here, and we're providing data to you to understand the medicine. We think it's our responsibility to help our investors and the analyst community understand the medicine that the company has created. And so we're taking the opportunity to provide data as we move along, including the 36-month OLE data today, which is striking. And we have medical conferences coming up. We've got a medical conference coming in Labor Day weekend, and we have one later in the year that we will be present at and we'll be providing further data. And so please expect us to continue to communicate the benefits and safety around this drug as we progress. And as far as time lines are concerned, if they do move, we will communicate at that time.

Your next question comes from the line of Yaron Werber from TD Securities.

I got two. Maybe the first one, Barry, for the Phase III for EMPEROR, the 20 patients in Europe specifically, I imagine you can lump them right into the overall study and treat them sort of indistinctly. And do you need to enroll 1:1 in the EU just given the requirements of the way the sham injection is going to be? Is it going to be 1:1 drug versus placebo? And then for the ADOA, so it's a single injection out to 48 weeks. So clearly, you're expecting from your nonhuman primates, very stable sort of protein expression. Do you think it's going to last longer than 48 weeks overall long term? And how many patients would you enroll in that Phase I?

So, Yaron, I'll take the first question in terms of the regulatory authorities and ask Barry to respond to you on ADOA, and it's good to hear from you again. Maybe if I give you the kind of the broader picture, this was a study that was agreed and aligned with the major regulatory authorities globally. And what I mean by that is the U.S., Europe, Japan and U.K. And it's a long arduous process that includes getting clinical trial designs approved in individual countries as well. Later on in the process, we actually had feedback from certain European countries that required a needle prick sham control to be part of the study as opposed to a lumbar puncture sham control. And so what we simply did is we added 20 patients in a separate cohort in those four European countries. and we just increased the number for lumbar puncture and maintained our 150 patients. What we're doing overall is kind of maintaining the integrity of the study in the event that there is a separate analysis that is required, but we want to maintain that powering of 150 with lumbar puncture and patients being blinded through a lumbar puncture sham control. And so simply put, we added 20 patients, at least 20 patients in Europe. It doesn't change our time line, and it doesn't change our regulatory filings nor the powering of the study.

And I'll just add. So, Yaron, your question about 1:1. So yes, it is 1:1 balance, so 1:1 for sham and for active. And that applies to the number of patients in Europe as well. Sorry, there was one other question -- you had one other question about the 002. And as you know, Yaron, these oligonucleotides have -- do have long half-lives, and we have already some data from our animal model showing that the oligonucleotide can have a half-life of nine months or longer. So a single injection could have an effect over that full 12 months of the clinical trial. And so that's why we're going to be following these patients and observing them that entire period of time.

Our next question comes from the line of Joseph Stringer from Needham.

This is Eddie on for Joey. Congrats on starting enrollment in EMPEROR. Just a couple from us. When do you anticipate beginning some of the North America commercial build-out? And what do you expect the sales force composition and cost to be at peak? And then elaborating a little bit more on the ADOA I -- Phase I program. Is there any requirement for OPA1 genetic screening? And what might be the cadence of some of this data after you have a 48-week duration for the primary endpoint, but when might see some interim data or final data and then progression into Phase II?

Yes. Thanks for the question, Eddie. This is Jason. On the commercial front, right now, we're a relatively small team. We have expertise in marketing, market access and new product planning. We intend to slowly build the team over time as we start engaging in efforts like a disease awareness campaign later this year, starting to educate the market and better understand the market through insight generation. At peak, we anticipate somewhere in the neighborhood of about 20 salespeople with additional kind of cross-functional support functions in the field to support reimbursement, patient education and site activation for the intrathecal administration on the commercial front. So a relatively modest overall field-based infrastructure for this rare disease.

I'll just add to Jason's comment in terms of being involved with medicines, disease-modifying genetic medicines like zorevunersen and other CF medicines. You don't sell a medicine. You actually help with access and reimbursement for patients and families. The medicine by the time it's approved is usually very well understood and known because as you conduct your trial, you're actually utilizing most of the treatment centers globally for your major geographies. And so the commercial build is focused on access and reimbursement and understanding the market. And it's why I made a comment in my own remarks that medical affairs is fundamentally important for the stage of the company right now where we educate both families and advocacy groups and also the physicians as terms of the medicine. And so it's more of a science education than it is a commercial cell. So, I just want to reiterate that. And that's why Jason refers to the kind of the low build. As far as the ADOA study is concerned, it's a Phase I study, which is dose escalating and it goes through multiple doses. And so I always view those studies as no news is good news because primary endpoint is actually safety as you escalate the dose. But at the point that we may see data that causes us to act beyond just the study we're running, then we'll inform you. That could be over the next year. It could be over a longer period. It all depends on what dose we get to that we start having safety -- maintaining safety and start having efficacious outcomes. And so we'll let you know when we see that.

And I'll just add on the -- your question about the genetic testing. So OPA1 gene is included in many panels for inherited retinal disease and other vision issues. So it does not require a separate test. The issue is that many of these patients either never get tested or even if they do get tested, there's nothing available, so they get lost to follow-up. So we're hoping to do an extensive education campaign, and Jason is going to be part of that and also encourage genetic testing for children who have vision problems when they're young.

I will now turn the call back over to Ian Smith for closing remarks.

Yes. Thank you. And thank you for all that participated in the call today. Thank you for the questions, and thank you for those people that were on the line and listening. We're very happy to provide this update to you all, and we will be available in our offices post call to answer any further follow-up questions. Thank you for attending the call today.

Ladies and gentlemen, that concludes today's call. Thank you all for joining. You may now disconnect.