Xenon Pharmaceuticals Inc. (XENE) Q3 2025 Earnings Report, Transcript and Summary

Good day, everyone, and thank you for standing by. My name is RJ, and I will be your conference operator today. At this time, I would like to welcome everyone to the Q3 2025 Xenon Pharmaceuticals, Inc. Earnings Conference Call. [Operator Instructions] I would now like to turn the call over to Colleen, Senior -- or Xenon Senior Vice President of Corporate Affairs. Please go ahead.

Good afternoon. Thank you for joining us on our call and webcast to discuss Xenon's third quarter 2025 financial and operating results. Joining me today are Ian Mortimer, President and Chief Executive Officer; Dr. Chris Kenney, Chief Medical Officer; Darren Cline, Chief Commercial Officer; and Tucker Kelly, our Chief Financial Officer. After completing our prepared remarks today, we will open the call up for questions. Please be advised that during this call, we will make a number of statements that are forward-looking, including statements regarding the timing of and potential results from clinical trials, the potential efficacy, safety profile, future development plans and current and anticipated indications, addressable market, regulatory success and commercial potential of our and our partners' product candidates, the efficacy of our clinical trial designs, our ability to successfully develop and achieve milestones in our clinical development programs including the anticipated filing of INDs and NDAs, the timing and results of those filings and our interactions with regulators, our ability to successfully obtain regulatory approvals, anticipated timing of the top line data readout for our clinical trials of azetukalner, and our expectation that we will have sufficient cash to fund operations into 2027. Today's press release summarizing Xenon's third quarter financial results and the accompanying quarterly report on Form 10-Q will be made available under the Investors section of our website at xenon-pharma.com and filed with the SEC and on SEDAR. I will now turn the call over to Ian.

Great. Thank you, Colleen, and good afternoon, everyone. Thanks for joining us on our call today. We're excited to share the considerable progress we have made over the past quarter as we remain focused on our 3 critical priorities: first and foremost, completing our Phase III X-TOLE2 study of azetukalner for the treatment of focal onset seizures, the top line data readout in early 2026, followed by the filing of our first NDA for the approval of the azetukalner in the U.S.; second, broadening the therapeutic opportunities for azetukalner beyond epilepsy with potential neuropsychiatric indications where we have identified strong preclinical, clinical and genetic evidence supporting the development in major depressive disorder and bipolar depression; and third, expanding our pipeline through the advancement of our promising earlier-stage Nav1.7, Kv7 and Nav1.1 ion channel programs with recent progress of our novel Nav1.7 and Kv7 modulators moving into Phase I studies. I will focus most of my comments on our azetukalner, or AZK, Phase III epilepsy program, and Chris will provide additional details across our clinical stage portfolio. As a reminder, AZK remains the only Kv7 channel opener and the only ASM in development that is backed by long-term efficacy and safety data from clinical studies of patients living with epilepsy, having demonstrated a highly compelling placebo-adjusted efficacy in focal onset seizure patients in our Phase IIb X-TOLE trial and durable and sustained efficacy over time through our open-label extension study, with greater than 800 patient years of exposure and safety data. As we disclosed in today's press release, the final patients in our X-TOLE2 study have completed the baseline period. and all patients have now been randomized. The final number of patients randomized is 380, which is a significant milestone, and we remain on track for top line data readout in early 2026. As a reminder, X-TOLE2 was designed and powered to randomize approximately 360 patients. So we are very pleased to have randomized more than the target in the study design. This will result in good power across the critical endpoints in this study. From the outset, we have prioritized working with high-quality, experienced clinical sites to maximize study success while diligently monitoring key metrics throughout the study. These metrics are tracking as we expect and as we disclosed previously, patient baseline demographics and the open-label extension rollover rate are consistent with our successful Phase IIb X-TOLE2 study. Therefore, we remain confident in X-TOLE2 and share the epilepsy community's excitement as we progress towards top line data readout. Two topics that we often get questions on with respect to X-TOLE2 are the final steps between now and top line data as well as our expectations going into this important readout, so I'm happy to address both of these topics. As I mentioned, the final patients in X-TOLE2 have recently been randomized. That means all patients have completed their 8-week baseline period and the randomization visit. These final patients are now in the 12-week double-blind portion of the study. For those patients who complete the double-blind portion, they have an opportunity to enter the open-label extension. The OLE rollover rate has been high in X-TOLE2, consistent with X-TOLE, where we saw greater than 95% of patients roll over to open label. For those patients who don't enroll in the OLE, there is an 8-week safety follow-up visit. Therefore, the final timing of the top line data will be determined based on the last few patients and whether they enter OLE. After the final patients have completed the double-blind period, we will finalize data cleaning and lock the database, complete the statistical analysis and medical review and be ready for top line data release. We will be in a position to narrow guidance about the specific timing for top line data in the coming months. We are optimistic for a positive outcome, and we believe that X-TOLE2, together with the strong results from X-TOLE, will serve as the basis for a new drug application for AZK in focal onset seizures. As we prepare for the X-TOLE2 readout, we have completed a detailed review of prior FOS studies, and we find that there is high reproducibility of results from Phase II to Phase III. ASMs that have strong efficacy results in earlier studies demonstrated the similar positive results in subsequent Phase III studies, although there is some reduction in effect size, which is not unusual when moving from Phase II to Phase III. Over the last 20 years, anti-seizure medicines that have been approved in adult FOS in the U.S. have shown a placebo-adjusted seizure reduction percentage ranging from the teens into the low 30s. Interestingly, some of the more successful ASMs, including Vimpat, are on the lower end of this range and often, the drugs on the higher end of the range had other challenges, either around tolerability or an onerous titration or DDI profile. This reinforces what we consistently hear from physicians. Although efficacy is an important component, the overall profile of the ASM drives prescribing decisions to address a broad range of unmet needs for their patients. And it is this overall profile where we believe azetukalner is differentiated and has a compelling set of attributes. At launch, we believe AZK will be an only-in-class Kv7 mechanism of action with strong short- and long-term efficacy, QD dosing with no required titration, no adjustments for DDIs, the potential for mood benefit and an overall favorable safety and tolerability profile. It is this profile that we believe will drive adoption and commercial success. So again, we have high confidence, and we expect that a positive X-TOLE2 readout, combined with the impressive efficacy from our X-TOLE study, will form compelling profile supportive of our NDA submission. We remain excited as we look forward to the potential of bringing an important new medicine to the epilepsy community. So I'll now turn the call over to Chris, who will share more details on our clinical development programs across epilepsy, depression and pain. Chris, over to you.

Okay. Thanks a lot, Ian. I'll begin with an update on our epilepsy program. As Ian already said, we're really pleased to have completed randomization in our Phase III X-TOLE2 clinical study of azetukalner with a total of 380 patients, which exceeded our original goal of 360. Our team's focus is now on completing the study to deliver top line data in early 2026, with the shared goal of the positive impact we could have by providing a new treatment option for these patients. We're also placing a great deal of effort into the other 2 studies of azetukalner in epilepsy including our Phase III X-TOLE3 study in focal-onset seizures and our X-ACKT study in primary generalized tonic-clonic seizures. While we advance our various studies in epilepsy, we are also focused on scientific exchange and education around the profile of azetukalner with health care providers. This fall, we had a strong showing at the International Epilepsy Congress, or IEC, in Lisbon, where we had an opportunity to present 4 posters while meeting with various health care providers as we highlighted the 36-month data from the ongoing X-TOLE open-label extension study of azetukalner in patients with focal-onset seizures, which demonstrates sustained monthly reduction in seizure frequency, impressive seizure freedom rates and a consistent adverse event profile suggesting long-term efficacy and tolerability of azetukalner. We also presented data from our X-TOLE study showing the efficacy of the azetukalner in certain focal-onset seizure subtypes as well as presenting a targeted literature review outlining the comorbidity burden in focal-onset seizures. In addition to these clinical presentations, we presented findings from our early-stage Nav1.1 program with data from preclinical models specific to Dravet syndrome. The energy at the meeting was high and excitement continues to build around the long-term data and continued scientific evidence generation. Looking ahead, we continue to generate data from our azetukalner open-label extension study and will present new 4-year long-term data at the upcoming annual meeting of the American Epilepsy Society, or AES, in Atlanta early December. AES is a critical meeting for us to engage with the epilepsy community, and Xenon is currently an emerging leader in the field. We look forward to significant scientific engagement. With 7 abstracts accepted for presentation, we're looking forward to showcasing a number of presentations, including updated long-term data from the ongoing azetukalner open-label extension in focal-onset seizures, study centered around depression and the impact on epilepsy patients as well as preclinical data from our Nav1.1 program. In addition, we look forward to interactions at our various booths, one-on-one meetings with physicians facilitation of ongoing scientific exchange through a dedicated scientific exhibition and symposium. So in summary, considerable momentum is building in our azetukalner epilepsy program with important milestones in the near term with the presentation of the 48-week open-label extension data at the American Epilepsy Society followed by our X-TOLE2 Phase III readout in early 2026. Now turning to Xenon's efforts to expand azetukalner's use into neuropsychiatry, an area where we believe the differentiated profile of azetukalner could really benefit patients. We hear from physicians that they are interested in new therapeutics with novel mechanisms of action, potential benefits on anhedonia, rapidity of onset along with a potentially differentiated tolerability profile. Our clinical development teams have made great progress with X-NOVA2 and X-NOVA3, 2 of our 3 planned Phase III clinical trials evaluating azetukalner in patients with major depressive disorder, which are underway and enrolling patients. In addition, X-CEED, the first of 2 planned Phase III clinical studies evaluating azetukalner in patients with BPD I and BPD II depression is also underway. Effective treatments for depression in bipolar disorder are limited, and many patients are non-adherent due to side effects and other factors. There remains a significant unmet medical need for safe and effective therapies to treat patients with bipolar depression, and the physicians that we have spoken with are keenly interested in azetukalner's differentiated profile. Beyond supportive physician feedback, a number of key factors informed our decision to expand our clinical development of azetukalner into bipolar depression, including an in-depth review of the existing literature outlining genetic links between BPD and Kv7, evidence of Kv7 down regulation in BPD as well as clinical studies that explore the use of Kv7 potentiators in depression, including results from our own proof-of-concept study in MDD. We've also generated preclinical data showing an antidepressive effect of azetukalner. Considering the current treatment landscape, azetukalner's novel selective Kv7 mechanism of action, potential benefits on anhedonia, rapid onset of effect and differentiated safety profile are particularly attractive in BPD. As a reminder, our X-CEED trial, is a multicenter, randomized, double-blind, placebo-controlled clinical trial to evaluate the clinical efficacy, safety and tolerability of 20 milligrams of azetukalner administered orally with food over the 6-week double-blind period, as monotherapy treatment in approximately 400 patients with bipolar I or II depression, with an opportunity to increase the sample size to 470 patients based on an interim analysis. The primary efficacy end point is the change from baseline in the MADRS score at week 6 in patients who received azetukalner compared to placebo. Upon completion of the double-blind period, eligible patients may enter an open-label extension study for up to 12 months. We're incredibly excited by the potential of azetukalner and its Kv7 mechanism in neuropsychiatric indication such as MDD and BPD. And I look forward to providing updates as we leverage azetukalner's pipeline and a mechanism potential across multiple streams of late-stage clinical development. Looking at our early-stage programs. As Ian mentioned, both of the lead molecules in our Nav1.1 and Kv7 programs, XEN1701 and XEN1120, respectively, are now in Phase I first-in-human studies in healthy volunteers. In October, we hosted an investor webinar focused on Nav1.1 and Kv7, which has garnered much interest. We received insightful questions about our approaches, including our focus on leveraging mechanistic insight, especially around ion channel function to target pain at its source and develop precision therapies that can address both the complexity and chronicity of pain. When we engage directly with clinicians, we hear a strong desire for opioid-sparing therapies that can meet the everyday realities of pain management without compounding the problem. Physicians recognize the limited efficacy of current options and remain concerned about the substantial risk of abuse and dependency tied to opioids. Even when opioids are used appropriately, their long-term safety profile is far from ideal. Chronic NSAID usage can also be problematic for different safety and tolerability issues that may arise. So these physicians are looking for alternatives that are both effective and well tolerated over the long haul, and importantly, they're interested in ion channel blockers as a potential transformative class of therapies. We know that analgesics can act along multiple different points of the pain pathway and interrupt the pain signal on its way to the brain. This is why we are excited about the potential for Nav1.1 inhibitors and Kv7 potentiators as these channels play important roles at multiple points in the pain signaling pathway, including through the initial transduction of pain stimuli into pain signals, the transmission of those pain signals along nociceptive neurons and the relay from peripheral sensory neuron to spinal cord neurons within the central nervous system. Starting with Nav1.7, we believe it is the best genetically validated pain target with striking genetic data in patients with loss of function mutations who have no ability to feel pain. Gain of function mutations have also identified -- have been identified that drive pain disorders, further underscoring the critical role now Nav1.7 plays in pain signaling. Our lead Nav1.7 inhibitors are CNS penetrant to enable global inhibition of Nav1.7 to better mimic the human genetics. They also demonstrate good free fraction and tissue distribution to achieve high levels of target engagement. And lastly, we have identified molecules that have excellent potency and selectivity to safely achieve target therapeutic levels of Nav1.7 inhibition. We believe we have solved for some of the critical limitations of prior Nav1.7 compounds and continue to build a strong pipeline of optimized Nav1.7 inhibitors for development in pain. With our long history with Nav1.7 and our deep ion channel drug discovery expertise, we are well positioned to deliver a novel and differentiated Nav1.7 compound profile into the clinic, one that has never been tested before. Kv7 is also a compelling pain target to modulate neuronal hyperexcitability at multiple points along the pain pathway, and we believe Kv7 potentiators have the potential to decrease neural hyperexcitability for the treatment of a range of pain conditions. This is supported by high levels of Kv7 expression throughout the pain pathway, and our data shows that Kv7 is enriched in the C and A delta pain subtypes of sensory neurons. In addition, Kv7 openers can block action potential firing in both DRG and spinal cord neurons, thereby significantly inhibiting pain signals from reaching the brain. Additionally, evidence supports that dysfunction of down regulation of Kv7 activity has been observed in altered pain states. And lastly, a clinical compound previously approved for the treatment of pain, flupirtine, has a mechanism of action that involves potassium channel opening, providing further validation of this approach. So in summary, we're excited to have both XEN1701 and XEN1120 now in Phase I first-in-human studies in healthy volunteers. And our goal is to initiate Phase II proof-of-concept studies next year, and we'll provide more details as we get closer to those important milestones. I'll now turn the call back to Ian, so he can cover our Nav1.1 program. Ian?

Great. Thanks, Chris, and thanks for sharing the significant momentum across our pipeline. We are proud of our extensive knowledge and development expertise in potassium and sodium channel therapeutics as well as the focus and investment in pain, neuropsychiatry and epilepsy. Rounding out updates with our Nav1.1 program, which continues to progress as we generate preclinical data that suggests targeting Nav1.1 could potentially address the underlying cause and symptoms of Dravet syndrome. Data shows that dosing with an orally available small molecule CNS penetrant and highly selective Nav1.1 potentiator suppressed induced seizures and improved motor performance supporting the potential for improvements in Dravet patient motor function. Further, in these animal models, chronic dosing suppressed spontaneous seizures protected against sudden unexpected death in epilepsy, or SUDEP, and increased long-term potentiation, a potential cellular correlate of learning and memory. We anticipate presenting preclinical data from this program at AES and expect that a lead Nav1.1 candidate can enter IND-enabling studies later this year. Finally, also coming out of our lab, a promising selective dual inhibitor of Nav1.2 and Nav1.6 sodium channels is now in a Phase I study as part of our collaboration with Neurocrine Biosciences. Neurocrine has guided that this first-in-human study will evaluate safety, tolerability, pharmacokinetics and pharmacodynamics of the investigational compound, NBI-921355 in healthy adult participants to support its development for the potential treatment of certain types of epilepsy. As our diverse pipeline of early-stage drug candidates continues to mature, I'm incredibly proud of the considerable progress we are making across multiple programs targeting ion channels. Before offering some concluding remarks, I do want to take a moment to introduce Dr. Kelly as our 1new Chief Financial Officer. Tucker recently served as the Executive Vice President and CFO at Deciphera Pharmaceuticals, where he oversaw the growth of the company as it advanced from discovery to direct commercialization in the U.S. and abroad. He built and strengthened the company's investor base and led strategic financial planning related to corporate strategy and pipeline, and this culminated in its $2.4 billion acquisition by ONO in 2024. Before joining Deciphera, Tucker also served as CFO of various public and private life science companies and also spent time as a life science investment banker. His experience will be incredibly valuable to our team here at Xenon, where Tucker will be instrumental in our strategic approach to building out the necessary functions, strategies, systems and infrastructure critical to our future commercial success as we await top line data from X-TOLE2 and prepare for our first anticipated drug approval. I believe he has already made a positive impact and look forward to continuing to collaborate with Tucker as Xenon evolves into a commercial stage company. So with that, I'll turn it over to you, Tucker, to say a few words, and then I can -- and conclude with our financials.

Thanks, Ian. I really appreciate the warm welcome. I'm thrilled to join Xenon at such a pivotal time as the company progresses X-TOLE2 with the goal of delivering positive top line results early next year and planning for the anticipated launch for azetukalner in epilepsy and beyond. I'm excited to apply my experience and expertise driving corporate and financial strategy for U.S. and international life sciences companies to Xenon and working with the team here as we build for commercialization and the impact we could have as a fully integrated biopharma company with the aspiration of delivering life-changing therapeutics to patients. With a healthy balance sheet and solid foundation, the future looks bright for us as we plan for a successful commercialization of azetukalner and our long-term growth. I have already been out on the road to begin connecting with our investors to share our vision for Xenon to become a leading company in neuroscience and in pain. Briefly turning to our financial results. Cash, cash equivalents and marketable securities totaled $555.3 million as of September 30, 2025, compared to $754.4 million as of December 31, 2024. Based on our current operating plans, including the completion of the azetukalner Phase III epilepsy study and supporting late-stage clinical development in MDD and BPD, we anticipate having sufficient cash to fund operations into 2027. Given our strong balance sheet and fiscal management, we are well positioned to support multiple registrational programs for azetukalner and the continued maturation of our early-stage pipeline. I'd refer you to our press release and the 10-Q filed today for further details on our financial results. And with that, I'll turn the call back over to Ian for closing remarks.

Great. Thank you, Tucker. I hope today's call reflected the excitement and relentless drive that permeates the whole team of Xenon. As we continue to progress our Phase III X-TOLE2 study of azetukalner with the anticipated top line data readout in epilepsy planned for early 2026, we are focused on the preparation of our NDA with the intent to file with positive top line X-TOLE2 data and advance azetukalner towards commercialization, bringing us one step closer to delivering a new antiseizure medication for patients still struggling with seizures. As I mentioned earlier on the call, once the last patient has completed the double-blind portion of the study, we will have visibility to the final time lines, and we will be able to narrow guidance at that time. To round out our azetukalner programs, we see the immense promise of applying azetukalner in other neuropsychiatric conditions and serving other patient populations in need and are proud of the progress with the X-NOVA and X-CEED programs. And while earlier stage, the excitement around our discovery pipeline is tangible. As we apply our ion channel expertise across multiple targets and therapeutic areas and grow these programs, we are taking important steps towards becoming a fully integrated neuroscience-focused biopharma company. So with that, I'll pause and operator, we can now open the call up for questions.

[Operator Instructions] Your first question comes from the line of Paul Matteis of Stifel.

Appreciate it. I was wondering if you could just kind of set the stage for the top line data release. How much should we expect to be disclosed on efficacy and safety? And once you have those data in hand, if possible -- if positive, what would be rate limiting to filing?

Thanks, Paul. I can start, and then, Chris, please add your perspective as well. So your first question, just on top line data. Yes, there's always this balance, as you can appreciate, between a Phase III clinical trial, where we generate huge amounts of data and what we can actually just realistically get done in a reasonable period of time to get out of a top line press release and then what would come out of subsequent medical congresses. So if you look back at our X-TOLE data, I think that's a pretty good proxy for what you'll see in X-TOLE2, so obviously, the key efficacy end points as well as our overall comments on safety and tolerability. So I think in our previous top line press releases, we've tried to have a fair bit of information in there and good balance between both efficacy as well as safety and tolerability. And I don't think anything would be different for X-TOLE2. In terms of prepping for the NDA, I can start and then Chris, please add your perspective. So it's really the efficacy results from X-TOLE2 that are on the critical path. As we all know, there's a huge amount of work that goes into filing a new drug application. A lot of that work is ongoing and continues to be -- we even have sections that are written and completed today, and we'll continue to do that over the coming months. And the rest of the package and dossier will come together over the course of 2026, including, obviously, the data from X-TOLE2. But Chris, do you want to provide any more granular comments?

Well, just as you can imagine, I mean, we don't wait to start writing the NDA until the top line X-TOLE2 data comes. So a lot of work is ongoing and I know a lot has already been completed. So the critical path was your question. It's basically defined by what Ian just said, so incorporating X-TOLE2 into the story that's already being told from a clinical perspective from X-TOLE to create the integrated summary of safety and integrated summary of efficacy. So that's it, and we're well on our way already, Paul.

Your next question comes from the line of Tess Romero of JPMorgan.

Welcome to the team again, Tucker. Are you able to disclose where your screen failure rate ultimately landed for X-TOLE2? And generally, when screen-outs occurred, were they for reasons as expected from prior experience? And then my second question is just how far behind do you think the results of X-NOVA2 will be from X-TOLE2? Think it makes it into 2026?

Thanks, Tess. Again, I'm happy to start, and then Chris can add his perspective. So when we talk about -- just want to be clear on some definitions upfront. So we will, at the appropriate time, give the screen and baseline failure rate. We have that as a combined number. So those are patients that may have dropped out during the screening period as well as the baseline period prior to randomization. So again, we're -- with an ongoing study, we don't go into very specific details across a number of different parts of the study, including this. But I would say that it's kind of trended and tracked as we would have expected in the Phase III program. And Chris can go through probably some of the reasons that you lose patients due to either baseline seizure burden or BMI or compliance with diarrhea or a variety of things that people drop out during the screen and baseline period before randomization. So Chris, do you want to do that? And then I'm happy to address the second question, which is just the timing of X-NOVA2.

I'm happy to do that, but I think you kind of covered it, Ian. I mean, the screen failure rates, largely, it's a reflection of insufficient seizures. And then we have a number of other inclusion-exclusion criteria. And so there can be kind of a smattering of other reasons that follow behind that. But it's been consistent with what we would expect from Phase II, Tess.

Thanks, Chris. And then your question on the MDD program and specifically as it relates to X-NOVA2, so this will be the first Phase III readout from the psychiatry program. We haven't yet given guidance on it. That study, that Phase III study started right at the end of last year, kind of really got up and running in the first quarter of this year as we got most of the sites up and running. We haven't given guidance, and I think we've generally said that, in our experience, if we look at our Phase II X-NOVA study, and we extrapolate forward, these studies often take kind of 2, 2.5 years. So as we progress over the next few quarters, we'll be in a position to provide guidance to top line data.

Your next question comes from the line of Brian Abrahams of RBC Capital Markets.

This is Joe on for Brian. On the commercial side, you talked about overall clinical profile being important. Just wondering how much of that -- of the efficacy docs are willing to trade off for other positive benefits like tolerability, ease of use and some of the other benefits there? And what are some of the learnings from how cenobamate launched and has been performing commercially as of late?

Yes. I'm happy. Thanks, Joe, for the questions. I'm happy to start. And Darren is here as well and can provide his perspective. Darren's now been here a number of months and had the opportunity to attend one of the big medical congresses in Europe as well as interact with a number of key physicians in the space. So yes, as we talked about in the prepared remarks on a placebo-adjusted basis, we've seen efficacy kind of range from the teens into the low 30s. And so there is quite a range. And it doesn't seem to be predictive of where you are in that range to commercial success. And I think that's, to your point, Joe, that there are a number of these other attributes. You specifically referenced cenobamate. Cenobamate is on the higher end of that range from an efficacy point of view, but we do know that cenobamate in terms of the titration, over 12 to 16 weeks. And as you push that dose higher, there are a number of adjustments that need to be made because of DDIs and tolerability. And so it can be a bit of a more challenging medicine for prescribers and their patients. So again, I think that really reemphasizes the point that we see in the data that efficacy is part of the picture but not the complete picture. And I think Darren's perspective on this would be really helpful.

Yes. Thanks, Ian. It's -- yes, I think it's the -- each focal-onset seizure patient is a different one and will respond to different types of therapies. I think with AZK and the attributes we provide that we've outlined on the call today, provide another option for patients. And if you think about physicians and particularly the general neurologists who treat the majority of these epilepsy patients, these attributes translate into a simpler, safer and really more reliable care decision. And on the patient side, AZK has a potential to really meaningfully reduce seizures burden without the trade-off of titration, as we've mentioned, or the cognitive or mood side effects that often limit some of these current therapies. So in my 5 months here at Xenon and having the ability, as Ian said, to interact with physicians, AZK, which will be the first -- the next branded drug in almost 8, 9 years since the launch of XCOPRI, there's a lot of excitement around the attributes that AZK is going to bring to patients, their families and caregivers.

Your next question comes from the line of Brian Skorney of Baird.

This is Charlie on for Brian.

Are you there, Charlie? We can't hear you come through.

Brian. I think you're on mute. Are you still there?

I think he's back. So Charlie, we -- yes, we -- you cut out, so maybe you can start your question from the beginning.

Okay. Apologies for that. Yes. So it was on the X-CEED trial. Can you hear me?

Yes.

Okay. So on the X-CEED trial, just thinking about the differences between the 2 types of the -- of bipolar disorder, given the higher predominance of depression in type II as well as why you decided to go with the MADRS scale versus HAM-D like you did in the MDD trials.

Chris, do you want to address both of those, just the BP I, II and then also using MADRS versus HAM-D17?

Sorry, the first one was -- what was the question about BP I? I apologize.

It was just a question around -- I think, Charlie, it was just around the differences between bipolar I and bipolar II and including both of those patients, I think, patient populations in the X-CEED trial.

Yes. I mean, so the bipolar...

And difference in depressive dominance in bipolar II.

The difference in depressive symptoms between BPD I and II?

Yes.

Yes. I mean, the largest difference that we're going to see is just the propensity towards a true manic state versus hypomanic state. And because of the potential for differential treatment response, we don't know that for sure. We've decided to stratify BPD I and BPD II. To the extent that there could be a different response in depressive symptoms, I guess, we'll have to kind of see what the study shows. The decision on MADRS was largely driven just by the -- Ian, was that about depression? That was the MDD or the BPD?

Just changing the -- we have the HAM-D17 end point in MDD, Chris, and moving to the MADRS end point in bipolar depression.

Thank you. Yes. I mean this is largely driven by the fact -- so let me just kind of explain what happened in MDD just to set the stage. So what happened in MDD was there was an ezogabine proof-of-concept study that showed improvements in depressive symptoms they could use MADRS. And that was the precursor to our X-NOVA study. And so we did the same. We used MADRS as the primary end point. Fortunately, we also looked at the data, the depressive symptoms with the HAM-D score. And ultimately, when we looked at that study, even though there was essentially a 3 point improvement in both scales, there was much less variability within the HAM-D, and so it was significant. And so FDA guidance allows you to use either, and so we decided to switch from MADRS to HAM-D with FDA support. So bipolar -- the reason why I say all that is because bipolar is a different situation where basically there is largely a precedent of focusing on MADRS for the primary end point, and this is the first study that we've done. And so we don't have data that would suggest one or the other, and so we leaned heavily upon the precedents for how things have been done in most bipolar studies up till now. So that's the main logic.

Your next question comes from the line of Jason Gerberry, Bank of America.

Congrats on the quarter. This is -- sorry, this is Dina on for Jason. Congrats on the quarter, and thank you so much for taking our questions. First one is just more of a clarification question on X-TOLE2 enrollment. Just wondering what the reason was for enrolling 380 patients instead of at initially planned 360. And do the additional 20 patients randomized like impact your powering assumptions at all? And then just wanted to touch upon the earlier stage pipeline. Can you provide just any color on data disclosures from the Phase I XEN1120 and XEN1701 SAD, MAD studies, maybe what like an initial update might look like and when we can expect it? And if you could also maybe frame what you kind of define as success from those programs.

Great. Chris, do you want to take the first one on X-TOLE2 enrollment and powering? And then I'm happy to do data disclosure around 1701 and 1120.

Yes, sounds good, Ian. Thank you. Yes. So thanks for the question. You have to keep in mind that when you're shutting down a study, you have several factors that are occurring. The -- sometimes there's an increase in recruitment. Sometimes it stays the same. Sometimes, it can even like unusually go down a bit. And then you have a screen failure rate, which you have been seeing for a while, which may remain the same or may go up or may go down. And so there are a certain amount of variables. And so when you decide when to kind of shut down screening, it's an imperfect science. And so when we chose the date on the back end to stop screening, there was a significant bolus of patients on the back end that brought us from 360 to 380, largely driven by the interest in azetukalner and its differentiated profile, which we've already gone through. So you could have ended up being 360. It ended up being 380 because of that increase at the end. And then as far as the power goes and just as a reminder for everybody, the powering for 25 milligrams versus placebo in the Phase III study in X-TOLE2 is quite high, like 99%. And so the study is also powered over 90% for the 15-milligram group versus placebo. And so if you go up in the number of patients, you get an even higher bump in power, I wouldn't really say that I think there's an appreciable impact on power going from 360 to 380, but whatever it is, it's certainly a little bit higher than it would have been at 360. So we're feeling confident as confident as we can, particularly because of the translatability of Phase II data in [ epilepsy ] Phase III at least historically.

Thanks, Chris. And then, Dina, your second question just on data disclosure, yes, just as a reminder, we have these 2 programs in Phase I now, XEN1701. This is the selective Nav1.7 inhibitor; XEN1120, which is our Kv7 modulator that we're also developing, both of them for pain. So they're both in traditional Phase I studies. So these are healthy volunteer, what you would expect dose escalating through single ascending dose and MAD cohorts. So those are ongoing. We believe those will wrap up probably in the first part of next year at some point, and then we would be in a position, depending on the data, to support moving into a Phase II proof-of-concept studies for both molecules. So in terms of the Phase I data, what we're looking for and what success would look like is that we'll get through the dose escalation, and we can -- based on our preclinical modeling for 1120 and also based on the genetics for 1701 or Nav1.7 is that we want to make sure that we have high enough exposure, that we believe we're going to see an analgesic effect in a human proof-of-concept study. So that's based on our preclinical modeling or we're trying to, with Nav1.7, really mimic the human genetics, so we can look at things like a modeling of receptor occupancy. Obviously, we want to look at overall safety and tolerability. And so it will be that profile in totality in Phase I that would give us confidence to move into Phase II. We haven't yet decided how that information will be disclosed publicly. But I think needless to say, I think once we have that information in hand, and we're ready to move to Phase II, we'd be happy to give information supporting our decision for future development.

Your next question comes from the line of Cory Kasimov of Evercore ISI.

This is [ Adi ] on for Cory. In the recent few months, there have been early Phase II readouts from competitors. If these readouts hold in larger studies, how would that change how you see azetukalner being used? And just another question on how should we think about the operating cost into 2026 given you have to plan for a launch and other Phase IIIs are also planned for next year.

Thanks, [ Adi ]. I'm happy to take the first one and then pass it to Tucker for the second one. So yes, there's been -- I think, overall, seeing more innovation in epilepsy is great. That's good for the epilepsy community. It's good for patients. I would actually -- Darren mentioned earlier on one of the questions that there hasn't been a branded launch in quite some time, I would argue there hasn't been a lot of innovation in quite some time. So to see more innovation into focal-onset seizures, I think, we're a bit of a drive for that, and that's good for the epilepsy community. Specifically, always challenging to compare across trials with different programs. I'll also say that the other programs that have released data this year, we haven't seen any placebo-controlled data definitely kind of as we see in the X-TOLE program, either in the X-TOLE or what we're doing in X-TOLE2. So one, I think we set an incredibly high bar with the attributes of azetukalner; and two, we have -- these other programs are significantly behind with no -- none of the other programs having run a double-blind, placebo-controlled study as of yet. And we're in this position that we're going to share at the American Epilepsy Society meeting next month, where we now have patients have more than 5 years of dosing. We'll show our 48-week data on efficacy and open-label extension. So we have a huge amount of information on azetukalner and the attributes and feel really comfortable with our position. And I think we set an incredibly high bar as others are coming behind us. OpEx, Tucker?

Yes. So on the commercial side, so I think we've made some targeted investments already, which we think have been really important for Darren and his team to get prepped for either the readout and ultimately commercialization. So we've made those investments so far. And obviously, on the back of data early next year, we'll continue to prep for launch and the OpEx will reflect that. But when we look at really a 2027 launch time frame based on the estimated readout and obviously time to NDA submission, the bulk of the cost in terms of bringing on the sales force and the like will likely fall outside of '26. But yes, we will certainly have an increase on the SG&A side in '26 in the back of positive data to get prepped for '27.

Your next question comes from the line of Marc Goodman from Leerink Partners.

This is [ Fatima ] on for Marc. For the first question, could you please remind us again whether you're planning to assess HAM-A or MADRS in X-TOLE2 in patients who have comorbid depression? And do we have any idea what's roughly going to be the proportion of patients? Have you looked at the blinded data of how many patients have comorbid depression? Second question we have, could you provide more color on the selectivity of Nav1.7 compared to other channels, the selectivity to 1.7 subunit versus other channels. You only disclosed information about receptor occupancy versus off-target effect. Are you going to have any more information about the selectivity? That's it for us.

Great. Thanks for the questions. Chris, do you want to -- I think it would be helpful maybe just to walk through the exploratory end points in X-TOLE2 as it relates to the psychiatric comorbidity, maybe the end points and obviously that it's an exploratory end point. Why don't you start there? And then I can add any other comments, and then I'll -- I can address the Nav1.7 selectivity question as well.

Okay. Sounds good, Ian. Thanks. And thanks for the question. So in -- not only in X-TOLE2 but in all the Phase III epilepsy studies, we are following a patient-reported outcome both for depression and anxiety for all patients and all visits in the study. And so yes, so it's being done in X-TOLE2, but actually, there's a really large body of data that we're gathering on this topic throughout the program. That's the first comment. The second is that the scale that we're using are patient-reported outcomes. Specifically for depression, it's the Beck Depression Index. And then for anxiety, it's the GAD-7. You had asked a question about the percent. I mean so we haven't shared baseline characteristics, and so these are sorts of things that you keep an eye on, but we haven't been sharing them. Just suffice it to say that not the entire population is expected to have depression and/or anxiety because we're not enriching for that. We're enriching for a certain degree of seizures. That said, there are such common comorbidities. We do think that there will be sufficient numbers that we'll be able to look at data and see if there's a readout. Ian?

Yes. Maybe I'll just add to those comments. These are exploratory end points, so obviously, not powered. And as Chris said, not sure exactly how impaired the population's going to be. And we're also not stratifying. So this is an epilepsy study. And so we may get some imbalances across the treatment arms in the psychiatric comorbidities as well. So I just want to provide the appropriate caveats there. On Nav1.7, I think you're right. We haven't provided all of our preclinical profile there. I think we can provide more over time. Some of that, we do want to keep for competitive reasons. But needless to say, I think you probably heard this in the pain webinar, that we feel that these molecules are very selective from Nav1.7 over the other sodium channel isoforms. So we feel very comfortable with the profile that we're -- that we have moved for 1701 into clinical development, and we have a number of molecules that are coming behind it as well. So these are highly selective for 1.7. As we said, we think also from a free fraction point of view and a distribution point of view that a profile of a molecule like 1701 has never been tested clinically before. So we're really excited that that's now in a Phase I study and hopefully next year moving into a proof-of-concept study.

Your next question comes from the line of Joseph Thome of TD Cowen.

Maybe on the Phase III epilepsy study, can you talk a little bit about your expectation for the change in cenobamate use in the Phase III versus the Phase II given that that's been on the market, obviously, a little bit longer now? And maybe how should we think about that when we see response rates in the placebo and the active arms or the discontinuation rates due to AEs? Is that going to be a consideration? And then maybe just one on MDD. Can you talk a little bit about why you don't have an interim analysis in the MDD studies? And is this a consideration with the third Phase III that would start given that you did incorporate 1 in the bipolar study.

Thanks, Joe. Chris, why don't I start on the cenobamate question and I can share maybe some of our thoughts on preclinical data there as well that may be relevant? And then you can add your perspective as well as answer the question on the interim analysis on MDD. So Joe, yes, we expect that cenobamate usage -- so we saw some cenobamate usage in X-TOLE in the Phase II study, but remember around that time, cenobamate was just going through kind of getting approvals and then being available commercially. So we do expect cenobamate usage to be higher in the Phase III study than we saw in Phase II, and we're just going to have to kind of see what those data tell us when we unwind. Obviously, in our Phase II program, because patients are on 1, 2 or 3 background antiseizure medicines and there's lots of these drugs available, there's actually a huge number of permutations of different kind of combinations of background medications that it becomes quite difficult to tease all that apart. But I think that as we get deeper into the Phase III analysis, I think you've raised an interesting question that we've been thinking about as well. I mean I can share with you some preclinical data that maybe gets to your question a little bit of a different way, which is we've looked at azetukalner in our preclinical epilepsy models in combination with all of the commonly used mechanisms and medications, including cenobamate. And whether we combine azetukalner with cenobamate or levetiracetam or lacosamide or lamotrigine, sodium valproate, like we've looked at a whole bunch. We've looked at a panel. And we don't see, when you add the 2 drugs together, you get a benefit of efficacy, and we don't see changes necessarily from the tolerability perspective. So I think we feel we've -- at least based on our Phase II data, is azetukalner, based on the novel mechanism and the profile, we think plays really well with others. And I'm not sure that's going to change in Phase III, but we'll know better when we unblind the data. Chris, I don't know if you have anything to add on the cenobamate side and then maybe you can address the MDD question.

I would say you covered cenobamate really well, so we'll just have to see what the data shows. We're not expecting a difference, but we'll have to take a look and find out. On depression -- on the MDD question, so why not have an interim analysis, I sort of already, I think, laid the groundwork for the answer from a previous question. The bipolar program really is different than MDD in a few different ways. But in particular, we don't have a precursor study to base the data on. So with depression data, we had the ezogabine proof of concept. We had our own proof-of-concept X-NOVA study. And so we really went into the Phase III study with a pretty good idea of how we thought the drug would behave in a larger study. We don't have that data in BPD. And so we've made some assumptions based upon what happened in MDD and what's happened with other drugs that have been tried in both indications. But ultimately, there's a little bit more ambiguity in the BPD program than there is in the MDD. And we decided to compensate for that by conducting an interim analysis to allow for the study to be increased in size should we need it. And so yes, we don't see any need to do an interim analysis on the MDD program.

Next question comes from the line of Andrew Tsai of Jefferies.

It's Brian on for Andrew. Maybe just a follow-up on the interim for the Phase III BPD. What could be in the various scenarios for that interim? And then if you could just share the kind of placebo-adjusted deltas that you'd like to see associated with those scenarios. And then maybe just one more on X-TOLE3 timing. In a very unlikely worst-case scenario that X-TOLE2 doesn't succeed, how much further behind is X-TOLE3 at this point in time? Do you think you'll still be able to file for an NDA in '26?

Thanks, Brian. I think we got them all. Chris, I can do the X-TOLE3 question, if you want to do the bipolar depression interim analysis and options there. So yes, we -- as we've spoken about previously, we have prioritized X-TOLE2, both in terms of that. It was the first Phase III study that we initiated. We did bias more of the X-TOLE clinical sites into X-TOLE2 and biased more of our U.S. clinical sites into X-TOLE2. So there is some delay between X-TOLE2 and X-TOLE3. I mean I think I share what you said in your question, which is I think it's unlikely, given the confidence we have going into X-TOLE2, that we're going to need it. But yes, if for whatever reason, then we would do everything to accelerate the time line as best we could to get to X-TOLE3 data. Chris, do you want to address the bipolar depression interim scenarios?

Sure. So I mean, first of all, the -- how did we come up with a study with 400 patients? That's largely based upon the data that we do have in MDD. So more specifically, we're powering at greater than 80% to detect a 2 point difference in the MADRS for 20 milligrams versus placebo using information that we got on data variability, specifically the standard deviation from X-NOVA. The scenario -- the way the interim analysis is going to be done is very binary. You do -- you firewall off the data. You take a look at the powering of that study and if you need more power to have a favorable outcome, then the number of patients has increased from 400 up to 470. So there are so many different possible ways. There are a lot of different ways it could go, but ultimately, it's broken down to a binary question, which is do you need more power for a successful study, and if so, then there's an increase from 400 to a number north of that between 400 and 470. I hope that's helpful.

So that ends our Q&A session, and we appreciate your participation. I will now turn the call back over to Ian for the closing remarks. Ian, please go ahead.

Great. Thank you, operator, and thanks, everyone, for joining us today. If we did not manage to get to your question during the allotted time, we apologize. We did run out of time, and we will reach out directly to you to connect. We look forward to continuing to provide updates as we continue to advance our late and early stage programs as we deliver on critical milestones over the coming months and quarters. So thanks, everyone. Thank you for joining the call. Operator, we can now end the call.

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