Ovid Therapeutics Inc. (OVID) Earnings Call Transcript & Summary

Thank you very much for coming. Those of you here, wonderful to see you. And those -- many of you online, I really appreciate you taking the time to spend with us today. And hopefully, you'll visit with us one-on-one. Today, we're going to talk about biomarkers in epilepsy and our 329 program. I'd like to really encourage you to dig in. This is an important moment, not just for Ovid, actually for the industry. Our pipeline is something very unusual. We have readouts coming later this year. We felt at this moment, it was the right time to share more about what we're thinking and how we think the future of epilepsy drug development will be shaped. And central to that will be biomarkers. As a public company, of course, these are our forward-looking statements. Before we begin, I want to really take a moment to introduce you to the team who's going to be walking you through this. This is the powerhouse that drives Ovid. First, I'd like to introduce you to Meg Alexander. Meg is Ovid's President and Chief Operating Officer and actually is the architect of what you're seeing right in front of you today. She's been instrumental in shaping our strategy, building our team and aligning operations around a very focused, extremely efficient, high-conviction portfolio. Her leadership brings a discipline and clarity that you need in this kind of an environment to translate bold science into decisive, measurable progress. And what we have working with her arm in arm is Dr. Amanda Banks. Amanda is a practicing physician. And not only that, she is our Chief Development Officer. So she has hands-on experience in seeing patients as they come into the hospital. And at the same time, she actually shapes our clinical development and neurology drugs. She's leading our clinical programs with a very clear focus. And we're deeply honored today to have with us Dr. Alex Rotenberg. Alex is somebody very special. Alex is somebody that understands how biomarkers play an instrumental role in teasing out the causes and roots of hyperexcitability, how you can measure it, how you can take it and take a clinical learning from a normal healthy individual and find that useful in drug development, going slow upfront, thoroughly, inexpensively and then translating that into medicine. Alex is an epileptologist, sees kids, and he'll be talking to you and he's a professor of note around the world. But today is not just an opportunity for us to explore that. We also want to give you a broader view. That broader view is simple. This is a paradigm shift in antiseizure medication. What you're going to see today underlies that. We'll also share an update in 329, the clinical development and opportunity that, that presents and in addition, discussing how biomarkers derisk a program like this, perhaps applicable to many other epilepsy drugs. And we have top line readout coming. So don't forget that, that's important. We're not just talking about it, we're doing it. So at Ovid, we really are focused on one fundamental thing, solving one of the greatest opportunities in brain health. That is teasing out uncontrolled neuronal hyperexcitability. And the way we do that, we basically take foundational biological targets, GABA-AT, KCC2. They play a central role in the electrical balance of the brain and are relevant across multiple different neurological disorders. We build selective small molecules, oral, IV, that's what interests us. They get into the brain. And we have completely differentiated mechanisms of action. What we're doing today is special. We focus on hyperexcitability. Our programs in OV329 and the KCC2 portfolio target that area. They are differentiated, and they are unique. And with that, I'd like to turn over to Meg Alexander, who's led the strategy and will now take you through our programs.

Thank you, Jeremy. Good morning, everyone. Thank you for being with us. So as Jeremy just mentioned, our focus is really targeting the underlying neural hyperexcitation that's at the root cause of so many different conditions of the brain. Seizures are a big piece of this, also psychosis, pain, substance abuse and addiction. And I'm going to tell you today about our strategy to quell neural excitation and specifically our strategy for the OV329 program. So as Jeremy said, we are going in areas where others don't play with bold, exciting, differentiated mechanisms of action. Our pipeline is deep and differentiated. We've got multiple programs ongoing. But today's focus is going to be on OV329, our GABA aminotransferase program. You'll hear about why we believe this is a next-generation and best-in-class drug. And why we're telling you about it today, as Jeremy alluded to, is we have an exciting milestone coming up in just a couple of months. At the end of Q3 of this year, 2025, we'll have a Phase I readout. Yes, it will be a traditional Phase I in the sense that we have safety, tolerability, pharmacokinetic data, but it's much more than that. And that's why we have you here today. We have one of the most comprehensive biomarker programs that has been studied in an antiseizure medicine. The way the study is designed, we should understand target engagement, pharmacodynamic effect. We should know if we're getting into the brain and doing what we need to do. So that's what we're here to talk to you about. But before we dive into that deeply, as Dr. Rotenberg and Dr. Banks will do, first, we want to take a 30,000-foot view of the industry that helps shape the strategy that you see in our pipeline in this particular program. When you look at the field, as we often do with a competitive lens, we're commonly looking at the drugs that are in development. And what you see is a sea of sameness and many of the drugs that are out there in development to try to target not just epilepsy to treat, but other diseases of neuronal excitation. And it looks something like this. If you look at this diagram behind me, essentially, what you're looking at is all the drugs that are in development from the bottom to the preclinic all the way through to Phase III and the dots represent drug development programs. What do you see when you look at it? It's pretty busy, right? And then if you start to look down the individual columns, these are different biological targets that programs are trying to drug. And what you also see here is that certain columns are pretty dense, right? In the sodium channels, in the potassium channels and some of the cannabinoids, what we're seeing is a lot of dots. And what that says to us at Ovid as part of our strategy, so when we think about trying to enroll trials, which Amanda is doing, we have to think about do you have a mechanism that's different that a patient is going to want to enroll in? Or are you going after the same mechanisms that are already available to them? When you think about getting to commercial approval and needing to differentiate, having to differentiate on slightly nuanced pharmacology relative to 3 or 4 other programs, not ideal. And finally, if you're a patient, and so many of these patients are uncontrolled on the medicines that exist today, you want something different. You want a step change in your seizures and your epilepsy. So that's why we've really focused on the fundamental targets that Jeremy alluded to, and it's part of our strategy, not just for 329, but the other programs in our pipeline, which, as you can see, are in a less dense space. But let's double-click on this and go to the indication, which will be our first indication for OV329 that we're disclosing today, which is drug-resistant epilepsy. Drug-resistant epilepsy is defined as people who have failed 2 or more antiseizure medicines. And if you look at how many medicines there are for seizures today, there's a lot. In fact, over the last 15 years, there's been 30. But how many of those have been novel mechanisms? 2 in the last 30 years -- or the last 15 years. So that's why when we look out at the environment and the unmet need, we know today that 30% to 40% of people living with drug-resistant epilepsies still continue to experience seizures. So what they need is medicines that are safe, that are well tolerated and that bring a differentiated mechanism of action to bear. You don't need more of the same. So that's what we really are endeavoring to do with OV329. OV329 is a next-generation best-in-class, we believe, GABA aminotransferase inhibitor. And it is focused not just on seizures, but we think there's a broader opportunity here even beyond that for other conditions, again, where neural excess excitation is at play. And the story with OV329 is we actually found this in Northwestern, and we went to get it from the man who invented Lyrica. And it was rationally designed to address many of the challenges that exist with GABA-acting drugs today and a very significant challenge that was in effect for the first-generation GABA-AT inhibitor. So we know with many GABA-enhancing medicines, there's challenges. Many of them are short-acting. A lot of them are heavily sedating. And the first-generation medicine that tried to interdict GABA-AT and inhibit it was a drug called Sabril, that was a brand name. The generic name was vigabatrin. And there was a major challenge with this drug in the sense that it had ocular changes in a proportion of patients that took it that caused irreversible blindness, not okay. So OV329 was rationally designed to be able to optimally tune GABA to hit the sweet spot of GABA to deliver seizure reduction, but durable seizure reduction, and to do that with, of course, an improved safety profile. So what was important to us is that we didn't have titration. Importantly, we didn't have the ocular changes that were expected. And then finally, that this plays well in a polypharmacy regimen because that's what we need as rational polypharmacy for these patients. So how it works is pretty simple. We know it's a validated mechanism, as I said, from the first generation. But what it does is it allows ambient levels of GABA, the major inhibitory neurotransmitter in the brain to increase in the synaptic and the extrasynaptic region. And it does this essentially by inhibiting the enzyme that catabolizes GABA or that more simply eats GABA. So if you look at that gray blab in the image behind me, the cartoon behind me, what you'll see is GABA aminotransferase. So by inhibiting that, we know that we can raise ambient levels of GABA in the synaptic and the extrasynaptic region. And most particularly, Zhong Zhong, our CSO, has done a tremendous amount of work characterizing the nuances between 329 relative to the first generation. And we know that we're able to deliver basically an optimally tuned GABA between that synapse and the extra synapse in such a way that we can hit the sweet spot, which is really the goal of an antiseizure medicine. You want to create an overall inhibitory neural milieu so that you're not having seizures, but you don't want to sedate people, right? We don't want to be dropping people on the floor. And that's what we know OV329 can do, and we know we can do this better than the first-generation medicine can. And we've characterized this very deeply. We know there's 5 core differences between us and the first-generation GABA-AT inhibitor. So you can see at the top, OV329 and vigabatrin, completely different compounds, different constructs. We're much more potent, a thousandfold more potent than the first generation humans. We irreversibly bind to the enzyme, but we very quickly clear the tissue. And that is important because it's allowing us to carve a therapeutic index. And let me tell you why. We have a very fast PK, but with a couple of very precise doses, just a few repeat doses, we know that we can maximally inhibit the enzyme relative to our pharmacology target, and we can keep that enzyme suppressed so that we can use very, very potent and effective doses to keep the enzyme down and to quell that excitation, which we believe will lead to anticonvulsant behavior. So we have this therapeutic index where vigabatrin does not. So if we are effective, we think this is not going to be just important for patients, we think this is going to be a significant opportunity. In the United States today, nearly 2.5 million people have epilepsy. Of them, about half continue to experience seizures after 2 medicines, so drug-resistant epilepsies. The majority of those patients have focal seizures. We know this mechanism of action works well in focal seizures from a range of different areas of evidence. So if you have very conservative assumptions, which we do, we're early in development, and so we put in place very conservative assumptions. If we penetrate the market by 10%, and we assume list prices as they are today in antiseizure categories, this is more than $1 billion opportunity even in a crowded marketplace, which seizure medicines are. And that's what if you look at the right, you can see medicines that are being used to prescribe first, second line for focal seizures. But importantly, look at the color coding. There's not many other medicines that work on GABA in the way that we do that are safe and well tolerated and that will have the target product profile that we expect to. And we've done a lot of work because we think the opportunity with OV329 is bigger than just seizures alone and focal seizures alone. So while we're interested in going into focal seizures and drug-resistant epilepsies as our first area of exploration with patients, we also believe this should work in developmental epileptic encephalopathies that have focal seizures as well as in pain and finally, in substance abuse and alcohol withdrawal and addiction. And we have very strong animal data supporting all of this. So this is our belief for the future of 329, but let's talk about what we came here today to discuss, which is biomarkers and specifically the biomarker strategy that we're applying for OV329. So we benefit from not only having superb advisers like Dr. Rotenberg, but we benefit from the technology improving. Over the course of the last 20 years, imaging and electrophysiology is allowing us to have much better prediction early in drug development about how a drug is acting in the brain. Why is that important? A number of reasons. Mostly -- well, in many ways, it's great because we can, in a more cost-efficient way, get to a proof of concept and demonstrate that the drug is getting to where it should be and that it's working. But we also can start to explore comparability against different mechanisms of actions and other antiseizure medicines. So what this allows us to do is to demonstrate target engagement. We're able to understand pharmacodynamic response. We can get early signs of efficacy, and we can even quantify dose dependency. So that's a lot out of a Phase I program. And because of the power of this, we're seeing other antiseizure medicine drug developers use similar tools. So if you look behind me, this is a chart of how some of the leading pharmaceutical drug developers in epilepsy are applying different technologies to get to biomarkers. And you can see there's a reason why they're doing this across the top, this magnetic resonance spectroscopy, transcranial magnetic stimulation and EEG. So many of them have used 1 or 2 of these to derisk their programs. The second thing I want you to look at is the 2 rows at the top that we have highlighted here. So the first-generation Sabril or vigabatrin was a generic name, used all of these different tools to assess biomarkers for target engagement and PD effect. So we have nice comparability in many places where the technologies haven't changed. And importantly, for OV329, we have what we think is the most comprehensive exploration of an antiseizure medicine that's been done using biomarkers. So this should give you a lot of conviction and a lot of derisking when we have the data in late Q3 this year. And we'll tell you why. Dr. Rotenberg, who is the expert, will explain why. So we designed our Phase I strategy with healthy volunteers. We're measuring them pretreatment at their baseline, and we're measuring them post treatment. So we can compare each of these healthy volunteers, these participants against their own natural baseline in terms of how they perform on these biomarkers. We also have placebo participants that we can compare against. We're using those 3 tools that I just told you about. And in some cases, we're able to compare against the literature of how therapeutic doses of vigabatrin and other antiseizure medicines performed against these same parameters. So if there are GABAergic effects, we can see if we're working in the right way. And this is all designed to detect directional signal. And if we see signals across multiple of these biomarkers, this should be very convincing to us. And importantly, we're also measuring safety, visual safety. Amanda will explain how we're doing that. So in total, when I say comprehensive biomarker strategy, this is what that looks like. Alex is going to walk you through -- Dr. Rotenberg is going to walk you through all of this, but essentially, what it's going to tell us at the end of the day and what you should expect to see from us in a couple of months is the magnetic resonance spectroscopy will tell us target engagement. Are we getting into the brain? Are we raising levels of GABA where we should be? Transcranial magnetic stimulation will explain our pharmacodynamic activity. Specifically, are we operating on GABAA, GABAB receptors. Are we seeing GABAergic activity? And also, are we not moving glutamatergic parameters, right? We want to make sure we're not doing something funky that a control would tell us about. And then finally, with EEG, we're looking at brain waves that have been linked to GABAergic or inhibitory activity. So we'll be able to triangulate all of these to be able to show is the drug doing what it should be doing, where it should be doing it and having an effect, and that gives us a lot of conviction going into a Phase II program. And we can do this in a dose-dependent way. So now I'm going to turn you over to Dr. Alex Rotenberg, who is one of the leading experts in this area of antiseizure medicine biomarker strategies. He is a prolific researcher to the point that he can't update his CV anymore because I think there's more than 150 publications that he's authored. But he's also importantly, an incredible pediatric epileptologist who treats these patients. Alex?

Thank you very much, Meg. Thank you, everybody, for the opportunity to talk about the utility of superimposing a series of biomarkers on a Phase I study. My day job is as an epileptologist, and I also run a laboratory focused on developing antiseizure therapies and then on identifying biomarkers that can tell us that the brain has been altered in a way that would suppress seizures. In this particular study, we'll focus on 3 categories of biomarkers derived from 3 tools: MR spectroscopy, transcranial magnetic stimulation and electroencephalography. And let me zoom in on these because each of these can tell us a story about whether or not an intervention has increased the magnitude of GABAergic tone or has avoided to increase the magnitude of glutamatergic excitatory tone in the brain. The first one we want to discuss is MRS, magnetic resonance spectroscopy. This is a noninvasive way to measure concentrations of prespecified molecules in a patient who is inside an MRI. This is highly relevant to OV329, but is not particularly relevant to other antiseizure medications. The reason for that is that OV329 is an intervention that will produce a metabolic event. It will ultimately interfere with GABA catabolism and should increase the ambient concentration of gamma-aminobutyric acid of GABA. Other antiseizure medications will bind to a sodium channel or potassium channel or the GABAA receptor. They will not change the amount of neurotransmitter in a circuit. And so if we are to try and detect whether or not an intervention has changed the amount of ambient neurotransmitter, MR spectroscopy stands out as one of the least invasive and most effective ways to measure whether or not your intervention has changed the amount of available GABA for the circuits that ultimately suppress seizures. With each of these, I'd like to show you what the operator sees, what the investigator is actually looking at on their screen when they're doing the study. And the right panel of this image actually shows it all. MR spectroscopy ultimately composes a signal into a series of peaks. Each of these peaks correspond on the Y-axis to the amount of a particular molecule and the x-axis sorts the individual molecules. There's a conspicuous GABA peak that could be identified by MR spectroscopy. And this is, of course, the peak that we expect to be enhanced by an intervention that increases the amount of available GABA. That's MR spectroscopy. The next set of biomarkers are derived by transcranial magnetic stimulation, TMS. And TMS is a technique from which we can obtain any number of metrics, and there's 5 of them listed here. Some of them reflect the magnitude of GABA-mediated cortical inhibition. Some of them reflect the magnitude of glutamate-mediated cortical excitation or the stability of the membrane that's mediated by the voltage-gated sodium channel. Not surprisingly, for OV329, we want to focus on those measures that will detect for us an enhancement of GABA-mediated GABAergic tone. But before we talk to how to detect GABAergic tone by TMS, let me introduce TMS and take us back to high school physics just for 30 seconds because it's an unusual technique. Everybody has seen MRI. Everybody has seen EEG. Not everybody has seen transcranial magnetic stimulation. What transcranial magnetic stimulation is, is a means for focal and noninvasive electrical, the M in TMS is actually a little bit of a misnomer. The magnetic field is used to produce an electrical current. But it's a means for focal, noninvasive, cortical electrical stimulation based on back to high school physics, Faraday's principle. If you recall, this is what sometimes called the right thumb rule, which is where if you produce electrical current that spins in one direction, you point your thumb into perpendicular to the direction in which the current is moving, you'll identify the north pole of a magnet. The reason why it's relevant that moving electrical current produces a magnet is because that magnetic field produces then electricity somewhere around it. The sketch from 1831 from Michael Faraday in the middle of this panel actually shows what he identified, which is if you pass current through an insulated wire, let's say, wire B, wire A will detect electrical current even though they're not touching. This is the same way that an induction stove works. There's a magnetic field. It generates electrical current in a pot, that pot heats and you get spaghetti. But although TMS has been embodied in a medical device, not many people have heard of it. And so if I may, let me explain Faraday's principle in a medical system. So this cartoon actually shows how it works, and it's very straightforward. In TMS, there's a handheld electromagnet that generates a magnetic field. That magnetic field fluctuates rapidly. It's actually a very strong magnetic field. It's in the same territory as an MRI, but it's very confined in space and only turns on and off very rapidly. That generates electrical current everywhere around it, including in the brain, if we place this magnet next to the scalp. Now we can introduce electrical current to anywhere in the brain. But the part of the brain that will talk back to us immediately is the motor cortex because if we introduce electrical current to the motor cortex, something moves. And the magnitude with which a muscle contracts reflects the magnitude to which the motor cortex has depolarized, has fired. So this gives us an electrophysiology something extremely valuable. In a walking, talking human, noninvasively, we can interrogate the motor cortex and ask how excitable is the brain. And the output, again, what does the operator see on their screen is what's called the motor-evoke potential. It's this electrical signal of a muscle contractor that's recorded by surface electrodes usually on somebody's hand, it could be any muscle. And from that is produced something called the motor-evoke potential, where you can quantify how big is it. And if you know how much energy went into the cortex and you know how much output you got from the muscle, you have something very valuable for measuring cortical excitability and input output curve. Now the way it appears to the investigators is something like this. These are just photographs from our laboratory of some of our staff and a healthy volunteer baby. But what the operator sees is in the right panel. They see a reconstruction of somebody's brain. They see a site where the electrical current landed. And the most important part of this figure is what does the operator measure. And that's the squiggle in the bottom right corner of this screen. This is the motor-evoke potential. This is the electrical output, a very conspicuous signal that we get out of a muscle after stimulating the brain. And with this motor-evoke potential, this motor signal that we get out of a muscle after stimulating the brain, we can identify how intact or how altered has been a range of inhibitory or excitatory circuits. So let me zoom in on the inhibitory circuits and how we measure them by this technique. I'll focus on 2. One is what's called the cortical silent period. And I'll focus on this one because it's altered by the precedent molecules, OV329 by vigabatrin. The cortical silent period is what it sounds like a little bit. In the figure here, you can see on the left side what we ask the participant to do. We ask them to squeeze their index and thumb finger. While recording from a muscle, usually what's called the first dorsal interosseous muscle, it's the subtle muscle, on the base of the index finger. Not surprising, if they're squeezing their fingers constantly, that muscle produces a constant electrical signal. In the right panel, what you could see is after a stimulus, an electrical stimulus is applied to the motor cortex, there is a conspicuous large wave. That's an excitatory event that's forced into the muscle by electrical stimulation of the motor cortex while the patient is constantly activating a muscle. Now for every excitatory action in the brain, there is necessarily an inhibitory compensatory reaction. If there wasn't, then with every excitatory event in the brain, we would seize. So what happens in the electromyogram, that signal that we are recording from the muscle is that signal is interrupted for a prespecified length of time, usually in the range of 100 milliseconds or so. And what the panel on the right shows is in a control condition, a length to which the muscle signal was suppressed and then the muscle signal resumes. Now the length to which it's suppressed in time is a function of how much GABA is available to compensate for this excitatory event. The more GABA available, the longer is this duration of suppression. Our jargon for this is the cortical silent period. The longer the cortical silent period, the more robust is the GABAergic capacity to inhibit the motor cortex and with it inhibit the muscle. So not surprisingly, in a publication that measured this with patients, healthy participants actually, who were exposed to vigabatrin, the cortical silent period is predictably lengthened. This is what we would expect to see with OV329. There's another set of GABA measures by transcranial magnetic stimulation. I'll walk you through one of them. This is something called paired pulse inhibition. So if we stimulate the cortex once, again, just like before, with this activation of excess excitation, it's an unnatural amount of excitation that we forced into the cortex by electrical stimulation, there is an activation of the compensatory GABAergic inhibitory circuits, again, to keep the brain from seizing. For every excitation, there's going to be a surround inhibition that dampens the cortex. And if we look at the diagram in the right side of the panel, we can see what happens when we typically stimulate the cortex twice, an initial conditioning pulse and a second test pulse. If these are spaced closely together in time, the first stimulus produces a wave of excitation that triggers a wave of inhibition that dampens the response of the cortex to the second stimulus. So if we produce 2 consecutive responses, the size of the second of the 2 responses is never as big as the first. It's predictably smaller. And the ratio of the amplitude of these 2 signatures that we get from a muscle corresponds to the magnitude of GABAergic cortical inhibition. Again, just like the cortical silent period, it's not a surprise that there is an enhancement of the magnitude of this paired pulse inhibition in patients who are exposed to vigabatrin and this kind of enhancement is what we would expect to see with OV329. What the data here show on the Y-axis is the ratio of these 2 consecutive motor responses. The closer is the ratio to 1, the less inhibition there was of the second of the 2 responses. So the closer the value is to 0, the more inhibited is that second of the 2 responses, the greater is the magnitude of GABAergic inhibition. And in this figure, the open circles correspond to the state of the participant after they've ingested a therapeutic dose of vigabatrin. And not surprisingly, with therapeutic doses of vigabatrin, there is an enhancement of paired pulse GABA-mediated cortical inhibition that we can measure by this technique of noninvasive focal electrical brain stimulation and recording the output from a hand muscle. Now I won't belabor the remaining TMS markers. It's a lecture in and of itself. But suffice it to say, there are some of them that are contingent on the voltage-gated sodium channel, which is not affected by OV329 or GABAergic drugs. We wouldn't hypothesize that to change. There are some of them that reflect glutamatergic tone. Again, we're not affecting the glutamate receptor. We don't expect those to change, et cetera. So we can make rational hypotheses about these sets of biomarkers that we can derive by MR spectroscopy and by TMS as to which ones would move and in what direction they would go. And let me wrap up by describing one more set of biomarkers briefly, and these are ones that we can obtain by the electroencephalogram. The EEG, as you may know, is essentially nonnegotiable clinical test that all of our patients with epilepsy undergo. By EEG, we can identify seizure susceptibility by recording pathological brain wave, things we call spikes and sharp waves, but also even at rest, a normal EEG can be decomposed into fast and slow oscillations. Now these oscillations, which are there even at rest in a participant are produced by, not surprisingly, neuronal activity. Now if we coarsely divide neurons in the brain into excitatory and inhibitory, they actually have very distinct firing patterns. In our brain, the foot is always on the brake. And the GABAergic interneurons are constantly firing and at a very rapid pace. So not surprisingly, the faster frequencies that we can detect by scalp electrodes correspond to the instances where a GABAergic inhibitory interneuron released GABA and caused a subtle fluctuation in voltage in the cortex that we've now recorded with a scalp electrode. You can imagine the more GABA is available to these fast spiking inhibitory interneurons, the more robust will be the faster frequencies on EEG. And this is, in fact, what we find even clinically in somebody who's on a GABAergic drug, we can see an enhancement in the fast frequencies on the electroencephalogram. Again, this is something we would hypothesize will happen with OV329. Now ultimately, the question is for us, why do we need these extensive biomarkers in a Phase I study. And this is actually an opportunity that's we're thinking about for Phase I studies period. In other words, notwithstanding that the Phase I study is really designed to test safety, a practical question is, does the intervention move cortical excitability in the direction that we would expect to suppress seizures. So if we have one metric, let's say, MR spectroscopy move in a favorable direction to detect the metabolic event that we've had more GABA, we could say this is already favorable, but only one metric might be just luck. If we have 2 metrics, let's say, MRS and one of the paired pulse measures move in the appropriate direction, we could also say that's good luck. But if we have MR spectroscopy, GABAergic, TMS measures and EEG move in the direction that tells us that we have favorable target engagement that we have not just safety of a product, but we have a shift in cortical excitability toward an increase in GABAergic inhibitory tone, then chances are we have something meaningful for patients. And if the other metrics, which are negative controls, ones that measure voltage-gated sodium channel dependent properties or glutamatergic properties don't move, then we can say we have some specificity, then we have a drug. And this is ultimately the value of superimposing biomarkers onto a Phase I study. Now why it matters is because in our day job, we're still encountering a major problem in epilepsy, which is a large minority of patients with seizures do not have their seizures adequately controlled by the 40 or so available antiseizure drugs. These patients essentially invariably have to be treated with polypharmacy. There aren't that many classes of drugs that are available to our patients with drug-resistant seizures. And so trying a drug from a class that they've tried before is impractical. And so what we're looking for in my day job and others are new classes of antiseizure medications that we can add to improve seizure control in our patients. So thank you for this drive by through biomarkers. And I'll hand this over to Dr. Banks, who can talk about the clinical trial.

Thank you, Alex. Hi, everybody. Thanks for being here this morning. I'm Amanda Banks. I'm really excited to talk to you about our strategy for OV329 and how we believe we can bring this impactful medicine to patients who need it. I'm going to start by talking about 3 things. I want to talk about where we are in the development program. I want to underscore what we expect to see in the Phase I study and when we expect to see it and then share with you our vision for moving the program forward. So behind me, you see a high-level time line. The Phase I study that we are currently completing, we expect will read out at the end of the third quarter of this year. We'll then move into a Phase IIa study in patients that will read out, we expect in the beginning part of 2027, and that will set us up to move forward into the Phase IIb. So let me talk for a moment about the Phase I trial and the study design. It's a really comprehensive Phase I study. You've heard that already today. We're looking, of course, at safety and tolerability. And in particular, we're looking at visual acuity and ophthalmologic measurements that I'll talk about in more detail in a moment. We're looking at pharmacokinetics across a range of doses, and we're looking to demonstrate target engagement and biological activity, pharmacodynamic activity in the participants. If you look at the top 2 boxes that are shaded differently than the rest, we're completing that one at the very top, which is the final MAD cohort, but you'll notice that they look a little different than the others. That's because we've obtained biomarkers using the tools, all of the tools that Alex described a moment ago in both of those cohorts. So when we think about what we expect to see from the biomarker portion of the Phase I study, I want to talk about a couple of things. What we expect to see, of course, why we expect to see that and why it matters. So we know that the first-generation GABA-AT inhibitor increased each one of the metrics that are listed here on the left in this chart. GABA concentration by MRS, 2 measures within TMS and high frequency waves in EEG. We expect OV329 to do the same. If we look at each one of these and talk about why those matter. By MRS, if we see an increase in GABA concentration, it's a direct measure of the concentration of GABA as a result of giving 329. We will have validated target engagement, and we will be able to show that we've set up the participants for the pharmacodynamic effect. What will that increase in GABA do? And that's what the next 3 measures are going to show us. With TMS in both the LICI-150 and the CSP, we expect those measures to increase. That tells us that the GABA that we've increased in the brains of the participants is doing what we want it to do. it is reflective of an inhibitory milieu that has been introduced by increased GABA concentrations. And finally, and similarly with EEG, we expect to see the same increase because of that pharmacodynamic effect that's been set up by the increasing GABA in the brain. These things together give us tremendous conviction about moving into the patient study. In addition, from the completion of the last cohort in the MAD, we have excellent safety from the Phase I study, and that's shown here. We've seen no treatment adverse events or serious adverse events. The most common adverse event that we do see is mild headache. We have looked extensively at visual measures and ophthalmologic measures, and we don't see any changes. And I want to spend a minute just to talk you through what we're actually looking at in this portion of the study because I think it's important. The study is set up, as Meg told you, to look at baseline and at post treatment -- 30 days post treatment at a series of ophthalmologic measures. The first looks at visual acuity, does the vision change? The second looks through a dilated people directly at the structures of the eye to look at the health of the structures of the eye and if those have changed. The third one looks at visual fields. Has anything changed in the participant's ability to detect visual fields, anything in their visual field. We look at photography. We take pictures of the back of the eye. And the last measure is actually a way for us to look at in a cross-sectional manner, every level, every layer of the retina and look at the health of the retina. So it's extremely comprehensive. And like I said, we do this at baseline and 30 days post treatment. And we've not seen any visual changes. So extremely safe profile. Taken together with the biomarker data that we just walked through, we are really, really excited to be able to bring this medicine into our first patient trial, which is set up as a Phase IIa to look at up to 2 doses of OV329 in adult patients who have drug-resistant focal onset seizures. The point of the trial, of course, is to evaluate the safety, the tolerability of the drug, but also to show antiseizure effect of OV329. And this will give us critical information to shape the rest of the program as we move into Phase IIb and beyond, particularly as it relates to effect size and the most effective dose that we want to take forward. We anticipate being able to do this and share results with you in the first part of '27. And importantly, also, all of this data that we're generating will contribute to the safety database that we'll be building along the way for approval. So I want to hand over to Meg. I want to underscore again what Alex said. We're here because we -- he and I take care of patients. We're all here because our goal is to bring medicines that actually matter, really make a difference to people who have drug-resistant epilepsy, but just to bring really effective medicines to patients and to their families and to the people who take care of them. So that's what drives us, and that's what we think OV329 can be for us.

I think that's why we wanted actually to take -- we're going to take questions in a moment, I promise. But we wanted to take a minute to acknowledge a really important family in our community. These are not people that you probably commonly hear us talk about in an analyst meeting or for at a non-deal roadshow meeting or conference with one of you. But Tracy Dixon-Salazar and her daughter, Savannah Salazar, who are pictured behind me, are not just partners to many companies, us and beyond, but they were leaders in their field. So Savannah has a form of developmental epileptic encephalopathy or she had a form of DEE that was treatment resistant. Her mother, Tracy, and her father when she was diagnosed as a toddler and had terrible all kinds of heterogeneous types of seizures, decided to go out and change drug development. They took it into their own hands and they said, we're going to help lay down the tracks for research so that the world better understands the symptoms that our children are experiencing to try to produce better medicines for their kids. And they did that. They've done it very well and well beyond just the particular DEE that they happen to experience. But this is just an important family to us. I was working on a drug for the indication that Savannah had years ago now, which has since been approved. And I remember sitting across the table from Tracy and these kind of patient advisory boards and she said, "I sleep with my daughter every night because I'm worried she's going to die in the night" because that's pretty common with a lot of people who have treatment-resistant epilepsies and seizures. You're worried about that. It's called sudden unexpected death in epilepsy or SUDEP for short. And many of you in the room are parents, you can understand what that's like to sleep with your child worrying they're going to die in their sleep, what that means to your family, your job, your life, all of it. Savannah, about 3 months ago, did die of SUDEP, so a seizure in her sleep in the night. And she's really representative to us, one, of what a tenacious family can do; but two, why new mechanisms are needed for all those drugs, not enough for kids like her, adults who have treatment-resistant epilepsies. So while we couldn't get there fast enough for her, we will get there, and that is very much our commitment. So what should you expect from us about what comes next? You hear our commitment and our focus. So as we said, expect the top line readout late in the third quarter of this year. If successful, we'll be coming back to you showing proof of target engagement. We'll show that we're delivering GABAergic or inhibitory activity. We'll give you comparable results on some of those parameters that Alex walked through to the first-generation medicine where we can. In a couple of places, the technologies have evolved a lot in 20 years for the better. That's a good thing. And importantly, we'll bring good safety and tolerability. As Amanda said, we're following these healthy volunteers, these participants a month out. That's really important because as we start to move into the Phase IIa program, we know and Dr. Rotenberg will tell you, if you were going to experience ocular safety, you typically see it within the first couple of months. So we'll be starting to build our safety database with the Phase II that we're launching in the first quarter of next year. We'll be taking that again into patients who have treatment-resistant epilepsies. We're starting it in Q1. And at the end of the day, we're on track to be able to deliver what we believe will be a very differentiated, powerful anticonvulsant medicine that's safer, well tolerated and should work rationally within polypharmacy regimens that these patients need, patients like Savannah and so many others. So with that, we'd like to open the floor to questions.

How we'll do this? I see Tom Shrader in the front. There may be one in the back. I think they beat you, Tom, but I'll come to you a second. Stephanie has the microphone in the room. She'll be coming around. And if you are on the webcast and you have questions, Eric will be feeding them. Okay. We had one, sorry, then we'll come to you, Tom.

We do have a question from online. You showed us how competitive the field is. How do you expect to compete?

I think we just told you is how we expect to compete. So not only do we have a differentiated mechanism of action, which is needed in polypharmacy regimens, but we expect to ideally have a gentler profile. And what I mean by that is to be safer, well tolerated. So from a commercial perspective, that's how we hope to differentiate. From a clinical development perspective, one of the things that Amanda has done very well along with her team is they've gone out and they've looked at every site around the world, working with the epilepsy study consortium to understand what are the absolute best high-quality sites to study a trial in and who enrolls well. And that's really important given that this is a busy drug development landscape. So both commercially and from a drug development perspective, we feel like we have a good strategy in hand. Now Tom, from BTIG.

Tom Shrader from BTIG. First, I commend you because the only way to get anyone to care about biomarkers is to tell us beforehand what they're going to be and what you expect. So...

A wise analysts may have told me that once.

Really remedial questions for Dr. Rotenberg. How many of these biomarkers are going to correlate only with vigabatrin versus antiseizure drugs that work? And I'd be curious with correlates with KV7 inhibitors, which is another exciting class. And then finally, do you get any hints on safety? Are you going to precursors of somnolence, things like that? Just a little bit more detail on what we might see?

Sure. So I'll answer all three of those, if I can remember all three. So the question of which of those are unique to OV329, I would say that MR spectroscopy in particular. Other agents that are in pipeline should not affect the amount of available GABA. This is the only agent in pipeline that will modulate the amount of available neurotransmitter. So MR spectroscopy is a very custom fit for this mechanism of action. The transcranial magnetic stimulation markers are a little more versatile. And the ones, the 2 that I described will detect an increase in GABA-mediated inhibitory tone, which ought to come with OV329, they may come with other agents that modulate GABAergic inhibitory tone. The KV7 modulators, XENON1101, et cetera, should not affect those markers. They should affect the TMS markers that I didn't discuss in only glance pass, but those are the TMS metrics that are contingent on the integrity of the cell membrane, which is dependent on the voltage-gated sodium and potassium channels. The third question is somnolence and whether or not we should have a physiologic metric of an adverse event in a Phase I. And the best instrument for that will probably be the EEG. The EEG signal in a healthy participant fluctuates depending on their level of arousal. If we are asleep, the EEG looks very different than when we are awake, but it also looks different when we're drowsy. And so if there is a signal on the EEG of drowsiness, that should correspond to a participant's subjective sense of drowsiness. But if it doesn't, and it's there, it might nevertheless prompt the investigator to ask whether or not that participant is drowsy. So of these sets of biomarkers, I would say, the MR spectroscopy is a very custom fit for OV329. The range of measures that you could obtain by TMS is not a single measure. It's several of them. Some fit well into 329. Others fit well as negative controls. You wouldn't expect them to change with 329. And the EEG is a very acceptable measure to detect a physiologic readout of an adverse event like somnolence.

I think, Tom, just building on everything that Dr. Rotenberg just said, specifically the 2 parameters he talked about in TMS, so the cortical silent period and the long-acting intracortical inhibition at 150 milliseconds has published research on vigabatrin. That's what he showed you in here. So that's one where you can easily compare against the first-generation medicine. No, they're different based on the mechanism and that's essentially in the biological target, which is essentially what Dr. Rotenberg was saying. But there are a couple of very good papers on this including one that he wrote that gave a pretty good landscape assessment of different seizure mechanisms and biological targets and which TMS parameters can appropriately characterize their pharmacodynamic response. So that's cited in this deck, but we can also give it to anyone here. So there's some good kind of landscape views of this in the literature.

Boobalan from ROTH Capital. So two questions. So firstly, great presentation. I wanted to talk about the blood-based biomarkers. And then I wanted to juxtapose with some of the noninvasive ones that you talked about. So there's a lot of literature, they're talking about neurofilament light chain and then neuronal specific enolase and S100B. So I wanted to kind of get your thoughts on why certain approaches prefer over the other based on your experience? If you could comment on that, that would be good.

I think Dr. Rotenberg and then Dr. Banks, do you each want to expand on that? So broader than just antiseizure medicine biomarkers.

So blood-derived biomarkers, of course, are quite useful. They're more useful in disease than they are in Phase I. In other words, where we have participants, we don't expect some of the markers that would be signature for brain injury crossing through the blood-brain barrier and making their way into serum to be as useful. Of course, in disease, if the disease is severe enough, some of these invasive blood biomarkers may have a utility. The blood metrics that, of course, are useful in early phases of drug development are the one that tell us about the pharmacokinetics and pharmacodynamics. In this case, the pharmacokinetics and pharmacodynamics are uncoupled though because this is a permanent impairment of GABA transaminase, the enzyme. So notwithstanding that the serum might clear the drug, which is a favorable thing, so it doesn't accumulate in the retina, the pharmacodynamic, what the drug does to the body is relatively static and prolonged. So the serum markers in this case may have a utility, but the noninvasive physiologic and imaging biomarkers is, from our perspective, what should really be the primary focus in the biomarkers that we derive from the Phase I.

Yes. And then just one more very quickly. So the single most important parameter with respect to GABA-AT is the visual defects, right? Correct. So I wanted to know at what point you'll be able to convincingly tell the market this drug does not cause any visual field defects. Do you think the Phase I time duration is sufficient to do that? Or you wanted to wait until like IIa?

So we looked at this 2 ways. The Phase I is not necessarily designed to be the definitive study to show ocular safety. We do monitor it, and it's been exceptionally clean, and that helps us to start building our safety database. But it's really the Phase IIa that Amanda described, which should be one of the more definitive answers to ocular safety, which we believe we will deliver. So if you talk to physicians who actually prescribe vigabatrin, which Dr. Rotenberg does, what they'll tell you is for the patients who take vigabatrin, their likelihood of actually seeing a change, if you see it, is within the first couple of months of monitoring very closely. So what we're starting to do with the Phase IIa design, we obviously have very strong safety from our Phase I. But for the Phase IIa design, we'll start to build our safety database. We'll keep the patients on open label to be able to accumulate that over time. And I think you can tell by Amanda's description, we are measuring in the deepest way possible, not just visual acuity, but any changes happening in the back of the eye. So if there's something happening with these healthy volunteers in these patients, we will see it. We have the definitive way of being able to ask and answer that. But again, we feel from all the work that we've done, the safety we've seen in humans as well as the toxicology work that we've done that we have a very definite therapeutic index that we feel confident with.

This is Dylan for Laura Chico from Wedbush. This is for Dr. Rotenberg. So with the absence of vigabatrin as a control in this study, could you talk about the best way to compare the OV329 results versus the historical vigabatrin data?

So perhaps talk about the Parentosi paper, Dr. Rotenberg. I think that's...

The best way, of course, is to compare it to published data. There is one manuscript that describes well how exposure to vigabatrin affected in healthy participants at therapeutic doses a range of TMS measures. These include the paired pulse inhibition that I described and paired pulse inhibition metrics that I did not describe, which are ones that would be negative controls. These are ones that should not be affected by GABAergic drug as well as other measures, something called the resting motor threshold and something called intracortical facilitation. Those metrics should not move and they haven't historically in published literature with a GABA transaminase inhibitor. And indeed, they don't at therapeutic doses of vigabatrin. So that will be the best way to compare. Analogous on EEG, a separate manuscript describes what happens to the EEG in healthy participants with exposure to vigabatrin. And sure enough, like with GABAergic medications, we expect an increase in the faster frequencies in the baseline EEG. And that's, in fact, what happens. So the best way to compare is to compare. There are published data out there, and we can see how the data from OV329 compared to the published data.

If you're looking for those references, they are in our slides. So when we post them, there's 2 to 3 really foundational studies that have looked at vigabatrin. We use very similar methodologies wherever we could so that you actually can compare side-by-side using those TMS parameters like the cortical silent period and the long-acting intracortical inhibition at 150 milliseconds like Dr. Rotenberg spoke about. Yes.

Myles Minter from William Blair. Thanks for the presentation. Very helpful. One of my questions is just on the inherent noise and variability in some of these biomarkers. I think investors in the epilepsy space, they like to look at preclinical models and then maybe a photosensitive epilepsy seizure model in early clinical stage. And I just wanted to hear your comments, maybe we can pick the paired pulse inhibition considering we're talking about that. These are healthy volunteers. Is there variability in the cortical silence period there between everyone? And sort of how do you accommodate for that to make sure that this data is robust?

Yes. So I think maybe Dr. Rotenberg and Dr. Banks, what we can speak about is what we've done a variability in a natural -- just human, if you can speak to that Dr. Rotenberg. And then Amanda, maybe you can cover some of the things that we've done just operationally because these are very sensitive tests. So we put a lot of very thoughtful methods in place working with Alex to be able to try to minimize variability.

So you're absolutely correct. There is variability in every electrophysiologic measure. Humans are not identical. And the way to accommodate that is twofold. One is to really optimize technique and to ensure that the techs administering this study are placing the electrodes in the same way with the same impedance in the same location, et cetera. And technique is actually very important in every electrophysiologic measure and putting constraints on technique is what helps. The other, of course, is to deal with the biology and variability among participants. The way to deal with that with modest -- in a Phase I study is to have more than one metric. This is why it's useful in this case to have a half dozen metrics because if any one of them moves, it just might be good fortune. But if all of them move in the predicted direction and the ones you predict do not move, don't move in a predicted direction, the statistics for that are such that it's unlikely to have happened just by chance. And so we can mitigate the natural variability and statistical noise in every measure by having multiple measures across multiple participants. We necessarily have to deal with error bars in statistics, but this is how we accommodate them.

Can I sneak another one in, if possible?

Yes. Just quickly on it. The contribution potentially from operational variability can be very high. So we've stripped it out. And we did that and we actually -- Dr. Rotenberg went to the sites, trained the people who are doing the TMS, looks at the data in real time for quality assessment. We put into place multiple different layers of quality assessments and training so that we can take that piece of it away. We can cover the second question.

Now we can cover the second.

No worries. It's more a strategy, just your Phase IIa in focal onset seizures. Any reason why that sort of slightly broader indication and maybe not going after like a medial temporal lobe epilepsy that has direct GABAergic interneuron dysregulation there and sort of proving out that thesis in a smaller population before going broader?

So we've considered all of it, and Amanda will probably chime here in a moment with me as well because we know the mechanism of action exactly, to your point, does have potentially broad therapeutic activity across a number of seizure types and potential epilepsies and beyond, as you heard our ambitions state. But we like focal seizures. We like treatment-resistant epilepsies with focal seizures because focal seizures afford us a number of things. They allow us to clearly identify the seizures and count. That's important when we think about derisking a program. We also want to enroll quickly because when we start to look out again at the competitive landscape of drug development in some of the DEEs, where we believe actually very strongly that OV329 would have a therapeutic effect, they're actually slower to enroll these days. And even though you might be able to get to registration quickly in a couple of these rare indications, there's many trials running at the same time. So the way when we looked at the regulatory need, importantly, the clinical unmet need still in focal seizures, which you heard us describe, but then the ability to derisk a program by going after seizures that we can count, we can enroll quickly, we felt this was the best first indication. Is there more opportunity? Absolutely. I think we may have time for one more question, and then we're going to have to wrap.

This is Kalpit Patel from B. Riley. I have two questions for Dr. Rotenberg. So first one is what's an acceptable range of GABA on MRS as it possibly relates to sedation. And the second question is that I know we have vigabatrin data for TMS, but what numerical cutoff would you predefine for LICI reduction and CSP prolongation to call this upcoming Phase I study a success. If there is a range for each of these, that would be useful to know.

As far as what to expect on the GABA peak, I think the data will have to tell us. But ultimately, a single-digit percent statistically reliable change will be adequate to say you've had the metabolic event in question. You've increased the amount of ambient GABA. The prolongation of the cortical silent period, those data have been published. It's low double-digit percent from baseline. And the enhancement of long interval intracortical inhibition, again, could be in the low double-digit percent per published data. That's what we would anticipate based on what we know from healthy participants. If we are in that range with the drug and in the absence of change in TMS and other biomarkers that are related to glutamatergic signaling or to sodium channel modulation, et cetera, we should have good confidence that we've had target engagement.

Okay. Very useful. So I have a last question for Meg. So do you believe that you need confirmation from all 3 biomarkers to warrant further clinical development?

No. The short answer is no. So I think actually Dr. Rotenberg did a really nice job of describing this earlier. Across all 3 tools, right, between the MRS, the TMS, the EEG, each of them has multiple different biomarker parameters underneath of them. And if we're seeing directional movement across multiple, we feel very good about that. That is a sign where we feel very comfortable moving forward. If we do so in such a way that's comparable to the first-generation medicine, great. If we do so in a way that's also stat sig on top of that despite relatively small ends in these cohorts, that's gravy. We know we're having a strong effect. So we would feel encouraged by seeing direction on any one and multiple is terrific. I think the last question goes to Athena, and then we have to conclude.

You actually answered my questions.

Okay. Well, good. I'm glad we were efficient for you. She said it answered her question. Well, everyone, we just want to thank you again for your time today. Dr. Rotenberg will be here for a while. If you had a question that you wanted to ask one-to-one, feel free to approach him or Jeremy, Amanda or Zhong, our CSO; or Jeff, our CFO, in the back if you have any questions that we didn't cover. And thank you. Have a good day.