Thanks very much. It's my pleasure to be moderating this chat with Ian Mortimer, President and CEO of Xenon Pharmaceuticals. I'm sure most folks know Xenon well, but maybe Ian can give us a brief overview to set the stage, and then we can do Q&A. Thanks, man. Take it away.
Yeah, thanks, Paul. Thanks for hosting us. Thanks to Steve Fogg. Yeah, I'll give a quick high-level overview, but I think the more informative part is the Q&A. As you mentioned, I think many people know the company well. There's kind of three key themes that we're talking to investors about. One is the lead program, azetukalner in epilepsy. Obviously, critical is we've got near-term phase III data. We're fully enrolled on our phase III X-TOLE2 program, and we've guided to data in early 2026. We can talk about azetukalner in epilepsy, expectations into data, as we think about that also as a commercial opportunity. The second thing is expanding the opportunity for azetukalner into psychiatry. We've got a large phase III program ongoing. Two phase III clinical trials are up and running in major depressive disorder.
We have just recently initiated a phase III clinical trial in bipolar depression. The third thing that I think we are getting a little bit more attention paid to recently is our emerging early-stage portfolio in pain. We have two phase one assets now on two different targets. One is NaV1.7, which we just think is such a phenomenal genetic story. The second one is on Kv7, which I think has some really nice clinical pharmacology validation. Those have entered two phase one studies over the last couple of months, and we hope to be in a position to put those into phase II proof of concept studies in 2026.
We have phase III epilepsy data in the near term, and then behind that, a number of catalysts with additional phase III readouts in epilepsy, psychiatry, and then the proof of concept data in pain.
Okay, great. As it relates to the bar for X-TOLE2, you were asked on the earnings call the bar, and of course, you're talking about what people would expect, right? Like good safety, statistical significance. I think sometimes when people hear you say statistical significance, they worry. Is that talking down the bar? Is there an expectation that the phase II data is not going to look the same as the phase III? How would you respond to that line of questioning and your confidence that this X-TOLE2 readout is going to be robust?
Yeah, let's talk a little bit phase II-B, the X-TOLE study, and then we can talk about phase III, the consistency and how we've approached phase III. I think there's also providing a little bit of education, what we see with other epilepsy assets, and then I'll come specifically back to our expectations going into data. We have an incredibly strong profile as we stand today. If we look at our phase II-B data on a placebo-adjusted basis, it was the best efficacy ever seen in a focal onset seizure study, and it was in quite a severe patient population. We look at severity or kind of the refractoriness of these patients based on how many drugs they failed, how many background medications they're on, as well as the baseline seizure burden.
It was quite a severe population, and I think the outcome was really good. What we've also been educating is just the rest of the attributes and profile of azetukalner. This is a novel mechanism. We have not seen a novel mechanism launch in FOS in a long time. We see early onset to efficacy, so we see a reduction in phase II. We saw statistical significance at week one. We do not have to titrate the drug, so both on the titration up as well as it has a good safety and tolerability profile, as you mentioned, and we are not making adjustments for DDIs. When you look at that, you can kind of put a lot of those attributes in this ease of use. How easy is it for a prescriber to prescribe, and how easy is it for a patient to comply?
I think if we look at that, we're incredibly encouraged by the profile. A lot of those things I've talked about remain, right? Regardless of the phase III data, the mechanism, the profile of the drug remains the same. We have high confidence going into the phase III readout. When we've looked at other epilepsy drugs over the last 10 to 15 years, and you look at those that were robust in phase II and moved to phase III, you get a little bit of a coming in of the effect size. The effect size comes in a little bit, but I wouldn't say, I think if we were talking about some psychiatry indications, we would think about it a little bit differently. I think we have high confidence that you don't see a massive change when you go from phase II to phase III.
We've also communicated over the last 10 to 15 years, if you look at that placebo-adjusted efficacy at the high dose on the primary endpoint, and the primary endpoint for these epilepsy studies is called the MPC. We're looking at median percent changes on monthly seizure reduction is what the endpoint is. We see a range from the teens into the low 30s. The high end of that range is phase II-B data. Just behind us is cenobamate and their clinical data. At the low end of the range is Vimpat. There's kind of everything in between. What I think is interesting, and we've pointed this out, is that when you look at where the drugs are in that efficacy range, it's not a good predictor of commercial success.
That really comes back to the point, I think, on some of the other attributes of the drug. Again, our confidence hasn't changed. The way that we've taken an approach to phase III is with the success in phase II, we haven't changed anything. Inclusion-exclusion criteria is the same. We've gone back to a lot of the same clinical sites that we worked with in phase II. We're working with investigators that have experience with the drug. We also have many more of our overall clinical sites in the phase III program in X-TOLE2 or in the U.S. All of those things, I think, give us confidence going into the phase III readout.
Yeah, it makes sense. Do you want to talk about just the population you've enrolled and how similar it is to the phase II population?
Yeah, so what we haven't commented in specifics in terms of what those baseline demographics are. We will provide those at the time that we unblind the phase III data. As the study has progressed over the last couple of quarters, as we were getting closer to the end of the study, we've communicated that the phase III population looks similar to the phase II population. Again, I'm just saying kind of in the ballpark of when we look at the number of drugs that these patients have failed, the number of background medications, how long they've had the disease, the baseline seizure burden, very consistent, very similar with the phase II population.
Yeah, okay, okay. Data are positive, let's hope, right? What exactly are the next steps between then and filing?
We are going to file, and I think this is maybe I'll take this opportunity to comment a little bit on your last question, then bring it forward to this question, is we will file, and we've communicated with FDA and had regulatory interaction on filing on X-TOLE and X-TOLE 2. I think the reason I say that is because our belief is that X-TOLE is part of the filing package, which means it would be one of the studies on label as well, which I think is important. Critical path to us filing is X-TOLE2. If you think about everything else as it relates to the NDA, from non-clinical to CMC to clinical pharmacology, we've made really good progress.
Next steps are phase III data, getting obviously the clinical study report from phase III, putting all that information together and filing the NDA next year.
Okay, okay. What else would you be doing alongside that and after that on the launch prep side?
We've started. I think part of your question comes down to how can we make this a successful commercial launch? There's been some very successful drugs in focal onset seizures, and I think our profile supports that this could be a really important medicine for patients and an important commercial opportunity. We have a Chief Commercial Officer in place. We have a lot of that senior leadership on both the medical side and the commercial side already in place, including we have MSLs out in the field. We are talking to the epileptologists and the general neurologist right now. The nice thing from phase II-B study, the X-TOLE study, 97% of those patients went into open label extension.
We continue to follow those patients, and so we have an opportunity to continue to engage in the epilepsy community as those data continue to mature. In a couple of weeks, we'll be at the American Epilepsy Society meeting in Atlanta in early December. We'll have our four-year data. That is maturing nicely. We are doing a lot of the things that we need to do today to engage with the epilepsy community. That part of the commercial organization, raising the profile both of Xenon and of azetukalner, will just continue quarter by quarter as we get closer to launch.
Okay, makes sense. We've talked about this before, but I think on the one hand, right, drugs like Keppra and Vimpat would be certainly nice comps, right? I mean, they would imply success. On the other hand, those drugs in their first couple of years on the market, I don't want to say they got off to slow starts, but they didn't get off to fast starts at the very least. Is that just like a testament, or not a testament, but is that just like a benchmark for the reality of how an early launch goes in this category? To what degree is that in your control?
Yeah, I think we have to recognize in focal onset seizures is there's a number of drugs available, right? We still believe that of the 3 million Americans that have epilepsy, the 60% that have focal onset seizures, up to 50% of them not getting good seizure control. We add in some adolescent and pediatric patients that also have FOS. There's probably about a million patients that are not getting good seizure control. It is a significant number, but there are a number of generic drugs available. The question is, when is a patient switching to a new medication, either because of breakthrough seizure or because of tolerability concerns? I agree with you. Overall, we generally see these as more gradual launches.
I think the opportunity here, based on the profile of azetukalner, is to be very communicative and be out there talking to the epilepsy community. As we are ready to launch this drug, there's a huge amount of enthusiasm around the molecule.
Maybe it's kind of like not the easiest question to answer, but when we look at the best case comps in FOS, right? You and I have talked about this. It's less about efficacy. It's more about ease of use and safety and the comfort of kind of the non-epilepsy hardcore specialists, the comfort of the general neuros to prescribe the drug. When does that start to happen for these kinds of drugs? When do you as a company start trying to push the drug more broadly?
As we think about, as you're describing it, there's the epileptologists, right? And there's probably, our estimate is between 2,000 and 3,000 epileptologists in the U.S. A lot of these are working at the larger academic centers, what we would call the level three and level four epilepsy centers. I think they're the ones that are attending AES. They're the ones that are going to know the data better. They're the ones that often we talk to. They're the KOLs and the leaders in the field. I think those are going to be early adopters for sure. We are not discounting the general neurologist at all. More like 15,000-20,000 general neuros. Not all of them have large epilepsy practices.
All of the commercial mapping and work we're doing now is those that have the larger epilepsy practices, those that have experience with branded epilepsy drugs and raising the profile there. I think you kind of mentioned it in your question. I think part of this is educating on the ease of use attributes of azetukalner. As we talk about the safety and tolerability profile, the no titration, the no adjustments for DDI, the novel mechanism, regardless of whether you're a specialist or a general neurologist, all of those things are going to be important attributes for your patients. We do know those drugs that are being a bit more challenging to use, whether because of the titration scheme or because of the DDI challenges. Those are the molecules that have been harder to get into general neuro.
I think that's where we can differentiate.
Okay, makes sense. Anything else to cover on FOS?
I don't think so.
Okay, great. You want to talk about your efforts in MDD and just sort of set the stage there?
Sure. So we ran a proof of concept study called the X-NOVA study. We had data at the end of 2023 where we saw clear separation between active and placebo. We saw a dose response. In that study, it was a smaller study, so truly a proof of concept study. We looked at the 10 milligram dose, 20 mg dose versus placebo. At the high dose, we had about a three-point separation between active and placebo on both MADRS and on HAM-D 17. So we looked at both clinical scales of depression. We took all of that information from the phase II program and designed a large phase III program. Two of those studies have started now. The first that we call X-NOVA two started at the end of last year, but I would say the vast majority of sites were up and running in Q1 of this year.
About six months later, about mid-year this year is when we got X-NOVA3 up and running, which was the second MDD study. We are currently in two phase III MDD studies. We have made some changes from phase II to phase III. it is a one-to-one randomization. We have chosen to move ahead with the 20 mg dose. We have changed the sample size materially in a much larger study. We are looking at HAM-D 17 as the primary endpoint. We have increased the entry criteria to bias it to a bit more of a severe population. There are some other adjustments that we have made as well, but those are some of the larger ones. We are partway through the phase III program. In terms of execution, those phase III trials have been going fine.
Yep, okay. What's the timing there of those readings?
We have not, yeah, we have not yet provided guidance to top-line data. If we look at our phase II data, or sorry, our phase II study, how long it took, the size of that study, the number of sites, and we extrapolate into the phase III program. In rough terms, these studies probably take about two, two and a half years to run.
Okay, okay. The tolerability profile in MDD is surprisingly good relative to the healthy volunteer study. What did you make of that?
Yeah, and I thought you were going to actually compare it maybe to the epilepsy tolerability.
I mean, I guess it looked better than epilepsy too, right? It's so hard with the con meds. Yeah, I mean, I think many people thought it might look worse.
Yeah, no, yeah, it was interesting. When we look at our epilepsy data and then with all the caveats of a cross-trial comparison, and I think you've hit on maybe the most important point, but we just don't know, is it seems that the drug in depression, so in a disease state and as a monotherapy, seems to be better tolerated than the drug in epilepsy. Granted, more than 50% of the patients in our epilepsy study were on three background anti-seizure medications. And so difficult to kind of tease that apart. Yeah, I mean, I think when we looked at the totality of the safety data in MDD, one, we saw the same safety or adverse events that we saw in epilepsy. I think we're seeing a consistency of the mechanism in the adverse events.
I think it was on the softer side when we compare it to epilepsy. But as I mentioned, it's a cross-trial comparison and across two different disease indications. The other thing that we hear from psychiatrists is bringing a novel mechanism and an opportunity for a novel mechanism is important. The opportunity for the drug to work more quickly. We've seen that in epilepsy and we've seen that in depression, where we do see separation between active and placebo early on. That's important in psychiatry as a lot of the standard of care takes longer to work. And then also it looks like the drug is having an impact on anhedonia. But to the adverse events, not only did the profile look a little bit softer, but when we compare it to other mechanisms in depression, we're not seeing the weight gain and we're not seeing the sexual dysfunction.
To come with a different adverse event profile with a novel mechanism, with rapidity of onset and the opportunity in anhedonia, I don't think we've seen a profile like this being developed yet in psychiatry.
Yep, yep, makes sense. Maybe one more before we move on to the pipeline. I guess it's a little bit going back to FOS, but it's become more of a competitive category. How do you think about your relative positioning and is there room for four new drugs coming over the next couple of years, potentially?
Yeah, I mean, my opinion is that it's longer than the next couple of years. But we have seen that the category can have multiple branded drugs and multiple successful branded drugs. You had mentioned some of them earlier in the questions, but if we look at probably two of the largest drugs that have been really successful, were Keppra and Lamictal, right, that were branded at the same time. So I think we feel comfortable that when we launch the two branded medicines, it'll be us and Xcopri, adding another branded medicine, I don't think is a concern to the overall category. That said, I think we've set an incredibly high bar, right? Everything we've talked about from the profile of our drug, a novel mechanism, there are no potassium channel modulators on the market. And we look at some of the competitive landscape.
So far, we have not yet seen randomized controlled data from any of the other drugs in development. Everything has either been open label data or no data at all. Let's see how that evolves. I think we've set an incredibly high bar on the efficacy profile, safety and tolerability profile, and the overall ease of use. I think that that timeframe between kind of where we are and the rest of the field is quite significant.
Makes sense. Let's talk about some of the pipeline stuff in the last 10 minutes here. So NaV1.7, what makes this target compelling and why has it been so hard?
We need to go back to talk about the genetics of NaV1.7. We have been in this field really longer than any company. For those that know a little bit of the background of Xenon, actually, before we were called Xenon Pharmaceuticals, we were called Xenon Genetics. We were a genetics company cloning genes and identifying these patient populations that we believed had a single gene mutation that had a large impact on biology. One of the early genes that we cloned more than 20 years ago was this gene, which was SCN9A. SCN9A encodes for this protein called NaV1.7. What was remarkable about this gene is we found patients that felt no pain, regardless of noxious stimuli. These were firewalkers. They were circus performers.
When we did the genetics and we cloned, we found that some of these patients were completely null or homozygous loss of function for NaV1.7. They had none of the protein in their body and they did not feel pain. Over the last kind of 10 to 20 years, it was a target that many people in pharma were trying to prosecute against, including us. We had multiple molecules that we had developed against the target. No surprise, I think some of the early efforts on any new biology is hard and takes some time. I think what we have identified is there were kind of three limitations. Depending on the company and the molecule, it was a little bit different. In the early days, it was tough to get a selectivity profile.
When you want high receptor occupancy of any target, you need a good selectivity profile. In the early days, we did not get good isoform subselectivity across the different sodium channel isoforms. I think we have changed that now in some of the chemistries. Some of the early drugs we do not believe had the right pharmaceutic properties, enough free fraction of the drug to interact with the target. We also believe because 1.7 is expressed globally, both in the peripheral nervous system as well as in the central nervous system, that we need a molecule that is going to get across the barrier and is going to interact with the channel centrally to better mimic the human genetics. We were at this for many, many years. Our most recent kind of iteration of the chemistries has been over the last couple of years.
We have a molecule in a phase I healthy volunteer study with a chemistry and a profile that's never been tested before as we think about selectivity, free fraction, and the biodistribution of the drug. We are really excited to obviously get through phase I, but importantly, get into a proof of concept study so we can really start understanding, can we mimic the human genetics in a patient population and start to see this type of analgesia as a non-opioid approach to treat pain?
Yep. And so from a safety perspective, is it selectivity and off-target tox for other isoforms that we're optimizing for? Is it also on target?
Yeah, so I know where you're taking the question. It's a little bit of both, right? I think on the selectivity side, there are certain targets that we absolutely want to stay away from, right? We want to stay away from NaV1.7, which is expressed in the heart. We want to stay away from NaV1.2 and 1.6, which are expressed centrally. We also believe, although it's a hypothesis, that if you hit NaV1.7 really hard, you may see some effects as well. There were molecules developed both by Merck as well as in one of our collaborations with Genentech where they saw some orthostatic hypotension and some syncope. The hypothesis there, although we don't entirely know, but the hypothesis is if you have a very, very quick T-max and you hit the target hard, that you may see some of this cardiovascular effect.
That actually may be on target. What's interesting is, at least with the Genentech experiment, as they slowed down the absorption of the drug, they didn't see that effect. We also know that the genetic population doesn't have the CV effects. And we also know based on our chemistries, we haven't seen it in non-human primates. I think there's kind of a body of evidence that suggests that you just don't want a T-max and you don't want to hit the target too hard, too fast. There's probably some compensating mechanisms there if there's a slower inhibition.
So you have a SAD MAD safety study that you'll look to see if you're getting to a model target exposure?
Exactly.
Is that right?
Yeah.
Okay.
Yeah, model target exposure where we can comfortably say that we believe, based on our modeling, that we have high enough receptor occupancy. I think one of the things that I didn't mention earlier on the genetics was there's this population that is homozygous loss of function. Actually, some of the more recent genetics and some of the literature that's been published over the last 10 years is this isn't work done by Xenon, but by others, is they found patients that are more like 75%-85% loss of function. They also didn't feel pain. I think that's an important genetic population and an important phenotype because, as we know, trying to get close to or at 100% receptor occupancy is going to be incredibly difficult. Having to get above 75%, 80%, 85% receptor occupancy, I think, is very doable in a drug discovery program.
Yeah, a couple of things in the phase I data. We want to get to exposures that we can extrapolate that we're getting high enough in terms of high enough RO and also looking at, obviously, the safety profile of the drug.
Okay, okay. What do you see as the best population for proof of concept here?
I think we're going to probably go to a traditional proof of concept post-operative acute setting. I mean, I think ultimately the real opportunity here is to have an oral chronic non-opioid pain drug. I don't think that's where we go first. Although we have not yet fully designed the phase II proof of concept study, we're leaning towards studies like bunionectomy or abdominoplasty.
Yep, okay. On pain, you also have NextGen KV7 programs. What is the rationale and thought process there?
We know that KV7, which is obviously a target that we're interested in from an epilepsy and psychiatry perspective, but we also know that KV7, this is a channel that's involved in pain signaling. There was a drug that was developed a number of years ago called Flupirtine that had a potassium channel mechanism that showed some really nice analgesic effect, and the drug was approved. It was eventually removed from the market, or at least in Europe, it was removed. It is still available in some other geographies, but it was removed from the market for hepatotoxicity, not due to efficacy or not due, it's really a chemistry of the molecule, not of the mechanism. What we've seen preclinically in some of these pain studies, that this is an incredibly potent mechanism. It behaves actually quite like an opioid preclinically.
We've got clinical pharmacologic validation. I think we have a really interesting preclinical pain package as well. Yeah, you're right. We've got two of these molecules. We've got the NaV1.7, which we call XEN 1701. That's in a phase I healthy volunteer study. Then we've got a Kv drug that we call XEN 1120, also in a phase I study. Both of those should be in phase II proof of concept studies next year.
Okay, great. You want to briefly talk about the NaV1.1 as well in the last few minutes?
Sure. The third target, part of this may be to bring kind of full circle to who we are as a company. I think we've built this expertise that we have an incredibly talented team at drugging ion channels and the CNS. We've talked about the KV target. We've talked about NaV1.7. We've also got work on a target called NaV1.1. The kind of on-target approach for NaV1.1 is a rare epilepsy. It's a developmental and epileptic encephalopathy called Dravet syndrome. The children that have Dravet syndrome have 50% loss of function of NaV1.1. Our approach with an oral small molecule, which is a very, which I think ultimately could be the best approach because we know there are other companies out there that are looking at different ways to interrogate the target.
What we've done with an oral small molecule is we can bind and hold the channel open longer. So we potentiate the channel. That increases current and gets current back to wild type.
How do you avoid going beyond wild type?
You can absolutely do that with titration or dose escalation. We can get back to wild type. When we get back to wild type, we can see a number of things in these genetic animals that have Dravet syndrome. There are animal models where they are 50% loss of function in NaV1.1. They have spontaneous seizures, and they have death associated with SUDEP. What we found is that we can protect these animals from spontaneous seizures. We can protect them from death. We can provide survival in these animal models. We are also seeing some interesting data on what we call long-term potentiation, which is a correlate for memory and cognition. I think we have fascinating molecules. We will show additional data at the American Epilepsy Society meeting next month.
Our plan is to hopefully get one of these molecules into a human clinical trial next year.
Okay, great. And cash runway, Ian?
At the end of the last quarter, we had north of $500 million in cash, which we're guiding gets us into 2027.