Thank you all for coming to New Haven, Connecticut. It's great to see so many of our investors and analysts, as well as our collaborators, you know, here today, for our annual R&D Day. And it wouldn't be an R&D Day if me and the team weren't up in the middle of the night, pressing refresh on the lab portal from the Mayo Clinic and downloading the latest data for you that we shared this morning. As we get started, I just wanted to thank Yale Ventures and Yale School of Management for hosting us at this venue. It's particularly, you know, really, you know, salient for us because we were initially a Yale spin-out company. So we're grateful that they continue to host us for our R&D Day. And being here at Yale reminds me of the importance of our academic collaborators. You're gonna hear from some of the top academic KOLs in the field today. They're gonna be part of the panels. You're gonna see how important that collaboration is to Biohaven's R&D engine. What we've kind of created here at Biohaven, I think, just to summarize in this slide, is these really 6, you know, main areas of focus. I'm really proud of our team at Biohaven, our small but mighty team who's put together, I think, a pipeline that rivals some of the most innovative and large pharma companies to date. You're gonna also hear from some of the actual inventors of the technology. So Dr. David Spiegel from the Spiegel Lab here at Yale, as well as Dr. Voets, whose work on TRPM3 ion channels will be highlighted during the ion channel platform. You'll see we've tried to incorporate these external KOLs and their science into Biohaven, and trying to really advance paradigm-shifting treatments for patients. So I'm not gonna go over all the areas and all the KOLs who are here, but that is our identity of translating academic science to patients. And the collaborations that we have with our KOLs who are here today are critical to that. Before I turn it over to Irfan, who's gonna walk you through the whole day, I did wanna comment on the importance of our press release this morning with our degrader platform. And what was critical about that is I want you to pause for a minute and think about this science, right? David Spiegel had this idea that he could create a bispecific that would pluck out any autoantibody or immune target or other antigen-specific target in your bloodstream, redirect it to your liver, and get it to degrade it. That's never been done before, right? That technology doesn't exist. It is not the same as FcRns. We believe it's better than FcRns, and it's modulator and has greater implications. And the data we put out yesterday showed that we get reductions in humans in our target within hours. We are seeing the liver remove IgG within hours. And we put out our first 96 hours of data 'cause that's what we have on our first four cohorts. Literally, the 96-hour data just came in last night. And when you look at what we're doing by 5 days, 40% reduction of IgG by 5 days. You don't see if you look at the FcRn data, you know, they're hitting about that at 10 days or plus. So we're hitting it sooner, and we're just at the beginning of our therapeutic doses. And so dose cohort four really represents our first modeled therapeutic dose of right around the 40% lowering. And we are confident in the data we've seen that we're gonna be above the 70%, and we're only gonna do two more dose cohorts in the SAD, and we're gonna accelerate flipping to the MAD. So this type of data we put out as paradigm-shifting, it represents a brand new technology that has potential for diseases like type 1 diabetes, autoimmune dilated cardiomyopathy, IgA nephropathy, and others. So it's really an exciting time. And I think this platform highlights what you're gonna see across all of these platforms. You're gonna see a team that's worked tirelessly. I'm so proud of the Biohaven R&D team that has innovated across all of these areas. And so although I know a lot of investors are focused on the Degrader platform and the data coming out, I think you're gonna be impressed by the science across all 6 of these various platforms. And you know, looking forward to hearing your questions, having your dialogue, and the feedback that we'll receive from our KOLs. And so when you look at this sustainable pipeline here and the number of innovative programs, you know, we have a potential to change a number of treatment paradigms. And I'm gonna turn it over to Irfan now, who's gonna walk you through the various panels today and areas. And we'll have time for questions. We'll pause for that. Then also, so you know, as we get through each, you know, presentation, we will push the slides out live, so you'll be able to see each segment as it occurs. So Irfan, we'll turn it over to you. Thank you, Vlad. I'd love to add my welcome. Thank you. Program, as Vlad described, I wanna start with just a reminder about our agenda today. Each session is going to have a number of talks and presentations that will be relatively brief. Those presentations will be given both by our internal experts as well as our thought leaders who, as Vlad said, have been close collaborators or are very involved in our programs. We're very lucky to have them here to bring their expertise. After each session, we'll have a panel discussion that will be led by one of our esteemed moderators, analysts from the investment banks. All of that is in your agenda, which you have in front of you. I'm not gonna go through it in detail, but we're gonna start this morning with our Kv7 and TRPM3 ion channel program. You're gonna have lunch, and then we're gonna come back later and go through some of the other programs. I did wanna give you just a little bit of housekeeping information before we get started. If you want to go use the restroom, you go outside, step to your right, go down the hall, you'll see the restroom. We have coffee outside. Lunch will be downstairs in the forum on the first floor. We'll give you that information again later in the day. If you need to step outside, we have coffee and refreshments all day. I know it's a long day. We will have a little bit of a break in the afternoon, probably around 3:00 P.M. We're just sort of trying to make sure that we have a good cadence and are able to give you an overview of our exciting pipeline today. So with that, I will go ahead and get started, and I will introduce our panel and presenters for the Kv7 platform discussion this morning. I will start with our experts. So Dr. Mike Rogawski, I'd like to welcome you to come have a seat here. For those of you who don't know, Dr. Rogawski, Dr. Rogawski is a leading epilepsy expert. He is Distinguished Professor of Neurology and Pharmacology at UC Davis, where he previously served as the chair of the department. Prior to being at UC Davis, he led the epilepsy research at the NINDS. And his expertise is really in ion channels and in neurotherapeutic development, focusing on Kv7 and other very exciting mechanisms. So very pleased to have you here, Mike. I'm also very happy to introduce Dr. John Krystal. Please have a seat here, Dr. Krystal. Dr. Krystal at Yale needs no introduction. You know, Dr. Krystal is the chair of psychiatry here at Yale. And, you know, he has done groundbreaking research in neuropsychiatric disorders and in the mechanisms, for example, which he elucidated of the rapid action of ketamine in depression. And so we're very pleased to have Dr. Krystal here as well. We have our internal experts, Dr. Mike Bozik. Mike is a neurologist, and he is the president of our ion channel research program, Mike. He, he'll be kicking us off today. And then last but not least, we have Dr. Steven Dworetzky. Steven is a biologist who 25 years ago discovered and cloned the Kv7 genes, including Kv7.2. So we're very pleased to have Steven and Mike, who've been working on Kv7 modulation for the last 20+ years. I just wanna highlight that the esteemed Dr. Rogawski is also a proud alumnus of Yale University and the Aghajanian Laboratory. Yeah. Absolutely. We have many special connections. You know, Dr. Rogawski is a Yalee. He did his MD/PhD here. And so, you know, very pleased to have him. So with that, I'll hand it over to Mike to get started. Okay. Well, thank you so much, Irfan, for that time. I think there's 2 mics. Oh, Mike Bozik. Sorry about that. No worries. You're happy to turn it over to the real expert. So, want to also, like Vlad and Irfan, extend my welcome to everybody here. This is a really exciting day for Steven and me in particular. You know, Vlad highlighted the remarkable science that's done at Biohaven. But the other aspect that I think is a defining characteristic of Biohaven is the clinical performance of this team. And 25 months ago, our company was acquired by Biohaven, and we had not filed an IND on the molecule you're gonna hear about today. Today, we have initiated and dosed patients in four clinical registrational trials in focal epilepsy, two in focal epilepsy, and major depression and bipolar disorder that you're hearing about. And it's really around a transformational target today. You'll hear more about KB7 and why we think it's a transformational target. And we think that we have and we'll show you and hope that we can convince you that we have a best-in-class activator that is unique in that it selectively activates the Kv7 channel. And that confers some really important clinical advantages that we'll walk through. We also have the opportunity to highlight and leverage sort of nature's experiment, unfortunate experiment in some cases with genetic mutations that conferred disease phenotypes. And when that occurs as a result of mutations with the Kv7 channel, we have an opportunity to actually reverse those phenotypes and demonstrate the activity of the molecule in ways that open gateways to broader channels. So this is the target. And Doc you'll hear a lot more about this from Dr. Rogawski. But Kv7.2/7.3 really is the sort of master switch regulator of neuronal hyperactivity and conductance and action potentials and what the brain does. It's broadly expressed in the CNS. It's you'll hear more about the M current today. And it really acts like a rheostat that, as the voltage changes in the brain around depolarization events, you can finely tune it. When the brain doesn't do that well enough on its own, we have a drug that actually helps augment that function, sort of restores the rheostat. You'll hear, or you've heard previously that when we. It's really a great pleasure to be here and to see the incredible progress at Biohaven. And I have to say, as one of his mentors, incredible to see Vlad's leadership of this company over the years. So I'm John Krystal, and I'm a psychiatrist and neuroscientist. And I'm gonna be talking about depression and bipolar disorder. Depression, yeah, as you know, has become even more common as a result of the pandemic, global pandemic. And even before then, the chances of developing depression was about 1 in 5. As a result, inadequately treated depression is one of the most disabling global medical conditions. And it's also inadequately treated depression is also one of the most lethal conditions. It shortens life expectancy by up to 10 years, and it increases the mortality of nearly every medical disorder. For example, it doubles the mortality of cardiac disease. It's really a critical imperative to develop new and more effective treatments for major depression. Here you can see some more data about it. One of the shocking pieces of data is that when you prescribe a medication for somebody like Prozac, you expect them to get better. But in the real world, in the Large STAR-D study, only about 30% of people remitted on their initial antidepressant medication. About 30% of patients will not respond even after multiple trials of standard antidepressant medications. There are aspects of depression, like anhedonia, that commonly do not respond to the typical antidepressant treatment. There are substantial opportunities to improve antidepressant medications. Because of the side effects and the slow onset, adherence to antidepressant medications is a big problem. It is one of the most common strategies for trying to enhance the effectiveness of standard antidepressants: to add an antipsychotic medication. These medications also have their side effect burden in terms of movement disorders like tardive dyskinesia, metabolic dysfunction, weight gain, increased risk for diabetes, and just the feeling that many people have on these medications that they're dull, and they don't like taking them. One advance that was mentioned that I've been particularly pleased to see reach the market is esketamine, which is an S-isomer of ketamine, i.e., esketamine, which we studied way back in the 1990s. And it's a very effective treatment. But you do have to come into the clinic in order to get it. And esketamine is an intranasal insufflation. So there are clear gaps in the treatment of depression, which, as I said before, ramifies in our society because of the incredible burden caused by public health burden of inadequately treated depression. So how does Kv7 fit into this story? Well, it's, it's a little bit of a theme. But the story kind of begins with one of my, colleagues here, former colleagues here at Yale, Eric Nestler. And Eric studied and his and his laboratory studied a social defeat model of depression, which basically boils down to if you're a mouse and you get beat up, you get depressed. But what was striking is that, that when you when these, animals got depressed, that there were changes, in the ventral tegmental area, an area of the brain that, feeds dopamine to the higher centers of the brain. But about 50% of the mice didn't get depressed. One of the characteristics of those mice, those resilient mice that didn't get depressed, was high levels or increased expression of Kv7. Now, there are other signs that Kv7 can be linked to depression. So, for example, what I'm showing you on the left-hand side of the screen are the levels of KCNQ or Kv7 in mice that have undergone social defeat stress. And the TD Tomato Plus is a marker of glutamate neurons in the ventral hippocampus. A part of the hippocampus is a part of the limbic system. The ventral hippocampus is the particular part of the hippocampus that's involved in mood regulation, in establishing emotional context. And so that you see if the stressed animals downregulate Kv7 in the glutamate neurons in the ventral hippocampus. On the right-hand side, you can see what treatments do to it. So ketamine upregulates in these same neurons, Kv7 levels. And if you add a Kv7 activator, in this case, ezogabine, to ketamine, you get an even enhanced upregulation of Kv7. However, our standard antidepressant, in this case, escitalopram, had no effect on Kv7 levels in these glutamate neurons. You can see this theme emerging in a variety of behavioral data. So here is a five-choice operant model. Ketamine increases operant behavior in that task. Oh, yeah. Look at that. Biohaven 7000 has the same effect of ketamine. And if we look in animal models like this, the swim test, a very common kind of behavioral assay that is somewhat relevant to depression. You see that ketamine has an antidepressant effect, and that antidepressant effect is enhanced by co-administration of ezogabine. What's interesting is that this is not an artifact of ketamine specifically. And so that if you have a different kind of an NMDA receptor antagonist, in this case, a drug called lanicemine, which is also being licensed by Biohaven, that you get an enhanced antidepressant effect with the combination of BHV-7000 and lanicemine as well. Okay. So what about people? So as we think of ezogabine effects on brain function in people with depression, the first study was kind of a pilot study where there were clinically significant changes in depression scores that were associated with a reduction in functional connectivity on fMRI in brain structures associated with mood regulation, limbic system. On the right-hand side of the figure, you see a study from James Murrough. And Costi was the first author, which looked at reward anticipation brain activation in during reward anticipation. Although the data didn't reach significance, you saw a really interesting pattern suggesting increased engagement of areas involved in reward processing, such as the ventral striatum. These data are very consistent with the clinical trial data we've seen with Kv7 agonists. The Costi study, which from James Murrough showed a noticeable reduction in depression scores relative to placebo in about 20 patients per group. A much larger study conducted by Xenon with XEN1101 is less robust. But what is interesting to note, building on what Mike said earlier, is that in the Costi study, 20% of patients had dose-limiting side effects that may have prevented them from getting to an effective dose of ezogabine. And similar issues around tolerability emerged in the Xenon depression study. That means that there is a potential major opportunity. You've seen this slide before. But, but it's really important to highlight that getting the right dose is exquisitely important in getting effective treatments. And if you can't get to the right dose because of dose-limiting side effects, in this case, dizziness, somnolence, and other kinds of side effects, you can't really get an optimized treatment. And BHV-7000 seems like it offers a much more favorable tolerability profile, which has the potential meaning that more patients treated with this medication can get to a dose that might work for them. So why is why is BHV-7000 antidepressant? I would say this is a story that is kind of unfolding. But I just wanted to introduce a term, for further discussion, which is called homeostatic plasticity, which means that these neurons and circuits are exquisitely sensitive to changes in the tone of glutamate signaling. Too much glutamate signaling, the networks downregulate. Signaling becomes less efficient, and you lose dendritic spines and synapses. And if you can prevent that overstimulation, and Ewa Kavalali and Lisa Monteggia have shown that ketamine prevents this overstimulation by blocking NMDA glutamate receptors. KB7 agonism may prevent this overstimulation by downregulating excitatory tone. So you have a convergent effect of ketamine and KB7 agonism on homeostatic plasticity that may help to restore normal synaptic function associated with chronic stress, depression, and related conditions. Okay. Let's switch gears to bipolar disorder. Those of you who know people with bipolar disorder, who have treated patients with bipolar disorder, know that in some ways, in terms of the management of individual clinical patients, it's an even greater challenge. One of the reasons that it's a greater challenge is that we have relatively few treatment options. We haven't had a really fundamental advance in the treatment of bipolar disorder in about 20 years. Patients are often non-adherent with the medications that they get for bipolar disorder, and so even if they would have worked, they don't like the side effects. They don't make them feel right. They stop taking them. And that's a huge problem because bipolar disorder is the most lethal psychiatric disorder that we have, the one that's associated with the highest rate of successfully completed suicide attempts. It's a psychiatric disorder that has the greatest substance abuse comorbidity, that has a tremendous burden on the lives of people. So what is the link between Kv7 and bipolar disorder? Well, some of these links come from human genetics. Bipolar disorder is a highly heritable disorder. About 80% of the risk for bipolar disorder comes from familial sources. A gene called ANK3 or Ankyrin-G turns out to be a risk gene for bipolar disorder, and it turns out to be associated with Kv7. There are other studies that more directly implicate both Kv7.2 and Kv7.3 in the genetic risk for bipolar disorder. In postmortem tissue, there's evidence of transcriptional, epigenomic, and proteomic changes in Kv7, in both postmortem brain tissue and in peripheral tissues as well. In preclinical models, Kv7 activation shows some evidence of treatment benefit. I'm going to show you some of those models now. What you see on the left-hand side of the graph is a dose-related attenuation of amphetamine-related hyperactivity, which is an assay of kind of revving up psychomotor activation, which, of course, is a very prevalent feature of bipolar disorder in the manic phase, but also in the mixed phase of illness. Ezogabine also attenuates brain hypermetabolism in animals as measured with FDG, fluorodeoxyglucose. One of the features of bipolar disorder, interestingly, in both the manic and the depressed phases, is an elevated, increased impulsivity of decision-making. They make bad choices in their life because they leap to certain kinds of decisions, compromising their judgment and reflection. What you can see in this animal model, five-choice serial reaction test, is a reduction in impulsive decision-making in animals. So I mentioned earlier on that the medications commonly used to treat bipolar disorder have a number of side effects. And so that even if they're, in theory or in general, very effective and helpful to many patients, we have to be very careful about these medications because of the risk of side effects. Lithium perhaps is the one that people know best, which can damage your kidneys, can dysregulate your thyroid, can alter your skin, make your hair brittle, have produce a slight modest reduction in IQ. And so these are medications where the side effect profiles are often getting in the way for patients, even when these medications could be helpful. So there's incredible need for more effective, better tolerated medications for bipolar disorder, just as there are for major depression. So with that, I will stop. And Steve, you're up. Welcome, everybody. I'm excited to be here, to continue the discussion of not only BHV-7000, but the opportunities with the Kv7 mechanism. And I think John and Mike made my talk a lot easier. They talked about the mechanism and how it works. They talked a lot about the genetics. And so in this schematic, you could really start to think, well, where which way can we go? What's going to add value? What's going to bring more value to the compound, but also better understanding of the mechanism? And so there's the genetic component. So, Dr. Rogawski mentioned, BFNC, benign familial neonatal convulsions. So that was found where there are those haploinsufficiency mutations that are causative of seizures in neonates. Now, interestingly, those children go on to remit on their own, and most of them lead healthy lives. But there's also a component where there's a dominant negative effect, and you'll see that in the next slide. Dr. Waxman cannot be here today. We have a collaboration with Yale University and Dr. Waxman, where he's done some work with an inherited erythromelalgia. And then I'm revisiting the literature, and given Biohaven's experience in migraine, we can rethink about the literature and my previous experience in other molecules that sort of bring the rationale to the migraine concept. So here's a paper that talks a lot about haploinsufficiency, loss of function, dominant negative effects. When you have a dominant negative mutation in the KCNQ2 gene, these children are born seizing. They're non-ambulatory. They're non-verbal. Their lives are very challenging, as well as to the parents. So what we wanted to do was we wanted to look at similar to the Kalydeco story in Vertex, we not only wanted to look at one particular mutation because trying to recruit for that patient population would be very challenging, but we wanted to try and see how the compound interacted with as many mutations as possible. And to address this, we worked with Dr. Al George at Northwestern, where he uses APC, automated patch clamp technique. And he can transfect in the dominant negative mutations, the different ones seen. Let's see if I can get the, there it goes. You can see on here, we have a list of different mutations of about 50. You can see compared to the control, which is the vertical dark line, that the currents are shifting to the left. So it's doing what you would expect it to do. It is poisoning the current. And when you look on the right-hand side, when you add BHV-7000, you can actually activate or increase the current out of these mutated channels. And so this really is the ability to restore current to almost wild-type levels or beyond wild-type levels, which gives you the potential that 7000 can modify the disease phenotype in patients with DEE. And that's really the gold, you know, the brass ring that we want that the field wants to understand is in these children that you want to change the course of the disease if you can treat the root cause of the disease. This is a recap of Dr. Waxman's talk from last year, but this is very, very fascinating. He's got several families. I'm just going to point out to one of them. And here in the first family, you can see that this is a mutation in the sodium channel 1.7 gene. He can make sensory neurons induce pluripotent stem cells. And when he does that, and he tries to understand, the mother has the mutation, but the son is in a lot more pain, both in awakenings, in the number of attacks, in the time in pain per day. And the mother is not, but they have the exact same mutation. Why is that? When he what he discovered was that there was a gain of function in the Kv7.2 gene that offset the pain problem in the mother. And the father doesn't carry any of the mutations. So what that says is, if we have a molecule that can activate Kv7.2/7.3, we can act like that gain of function. But how do you study that? What's the translational model? So I'm talking about life cycle activities for Kv7. And here is one of them where you can see in his experiment, these are multiple electrode arrays, and these are recordings. And there's a lot of heat maps associated with it. And in the next slide, we took the same type of experiment, and I have a video here, and let me call your attention. You can see these cells are spontaneously firing, and there's a lot of activity. And yet when you add BHV-7000, and these are tens of thousands of events, you can see a significant reduction in the number of action potential firings. And so this really gives you that translational component. When you look at the potency of BHV-7000, you can see a really nice concentration response curve. The IC50 value in this particular patient's cells is 88 nanomolar. So the compound really shows an amazing potency in the context, in the pharmacologic context of the sensory neuron. Then the last slide I have is we went back into the literature, and these are three separate compounds. This is getting into the field of migraine. If you look at a compound, like ezogabine, and if you go and you do preclinical work and you look at the basal level, you can look at the central and peripheral effects, how ezogabine blocks CGRP release. You all know the CGRP release story with Nurtec. This was a paper that I had published decades ago. This was an old Bristol Myers Squibb compound. We did something very similar. Instead of looking at the peripheral effects, we looked at the central effects. We looked at cortical spreading depression, which is a surrogate for the aura associated with migraine. You see a nice dose-dependent effect with this compound. Then if you go to the European literature, there's a compound called Flupirtine, which is an analog of Ezogabine. The brand name is Katadolon, not approved in the US, but they had run a small clinical trial where you have a proof of concept that the Flupirtine compound can reduce the pain associated with migraine. So you can think of a Kv7 mechanism as both a peripheral and a central-based effect. I'm going to turn it back to my colleague, Mike, to talk to you about the trials. Thanks, Steven. Well, we could talk all day about Kv7, but Vlad wouldn't let me, and so, I'll be very quick. I think this sort of overview slide really underscores what I talked about at the beginning of this session, which is, you know, 25 months ago, a molecule that had not even had an open IND, and now 5 phase II, III trials underway in epilepsy and mood disorders. So I'm going to go very quickly through this because we have run over a bit. Both of our focal epilepsy studies are ongoing. We will be activating over 100 clinical trial sites per study. And, you know, Biohaven has this history of catching the runner that's in front of it. And, that's what our goal is, not only to be best in class, but we'll see if we can be first in class as well. Late breaker, we've initiated our pivotal study in idiopathic generalized epilepsy. So three studies across two different types of epilepsies. Very excited about the potential of this drug in epilepsy, in generalized epilepsies, which is a different form of epilepsy, but one that's quite severe and one where there's a high risk of sudden death. Very excited. You heard about the significance, or the morbidity that's associated with an unmet need in major depression. I can, I'm very excited to tell you that we enrolled our first subject in this study yesterday. And we did the same in our bipolar study, yes, just yesterday, first patient. And these are studies that were extremely optimistic are going to make differences in patient lives. And I think, I hope that you've come away from this session recognizing how important this channel is in the brain. And we have a really, really special molecule that's going to change the lives of patients with these disorders. So thank you. Oh, wait, there's more. We are also, as you saw, Biohaven has this history with migraine, changing the paradigm treatment with migraine. And so we think there's a role for KV7 in migraine as well. Thank you. This just says everything that I just said. Well, thank you. KV7, as you heard, we could talk about all day. It's a transformational target, and we are really excited about the potential for BHV-7000. We've heard about the great unmet need in epilepsy, in mood disorders, in areas like pain. And we're aggressively moving forward where we know that we have proof of concept in epilepsy. We're moving forward in bipolar, in depression, in a more measured way, trying to seek proof of concept. But those are also pivotal studies. We have these other exciting areas like pain that we're exploring and migraine. I want to also tell you about another very exciting, potentially transformational target where we have an amazing asset. That's TRPM3. And so I want to introduce our speakers for this part, and I'll invite them to come up as well. I'm going to start with Professor Thomas Voets. Dr. Voets is professor at KU Leuven, and he's group leader in the ion channel research group there. His research is on TRP channels. And he discovered the TRPM3 channel and has characterized and developed BHV-2100. And several years ago, we partnered with Thomas and his team. It's been a phenomenal collaboration to drive this really exciting program forward. We also have Dr. Richard Lipton, who will be joining us virtually. Dr. Lipton, unfortunately, had COVID, so he didn't want to come in person. But Dr. Lipton, no one, I can think of better to talk about the potential unmet need that we can make with BHV 2100 in migraine. Dr. Lipton is a Vice Chair and professor at Albert Einstein in New York. He's the director of the headache center there. He is the, you know, sort of epidemiologist of migraine headache. And we're very pleased to have Richard, who will be joining us as well to give that part of the talk. And last but not least, we have, Dr. Volkan Granit, who, trained at Einstein and, is, one of our medical directors who's working on the TRPM3 program. Before I hand it over to Thomas, I, I just wanted to show a slide and talk a little bit about Biohaven and our legacy of success in migraine. I think you all know this, but I'd like to point out a few really important facts. Number one, CGRP is a peripheral target. We were able to treat very effectively migraine by targeting CGRP in the periphery with Rimegepant and Zavegepant. Interestingly, there is a really resurgent interest in peripheral targets for ion channels in particular to treat other forms of pain like neuropathic pain and acute pain. We've seen recent data from other companies like Vertex, which has been potentially transformational because these drugs can potentially not have CNS side effects, sedation, the types of things that limit other drugs in the past, like pregabalin. And so that's one of the reasons that we are so excited about the potential for TRPM3, which is and BHV-2100, which is a selective and peripherally restricted compound that we think is highly effective and it is non-sedating and it's non-opioid. So with that, I'll hand it over to Dr. Voets to give us an overview of BHV-2100 and TRPM3 biology. Thank you, Irfan, for the introduction. And I'm really pleased here to speak to you about the biology of TRPM3 and how we came to it to consider it as an interesting target to treat many indications regarding pain and migraine. Okay, so my lab is focusing on ion channels of another class than the class that was discussed before by Dr. Rogawski. It's TRP channels or transient receptor potential channels. And you could consider them a little bit the yin and yang of ion channels in neurology because, as nicely was explained, KV channels, they mediate the efflux of potassium, which makes neurons hyperpolarized and inhibits their activity. Whereas TRP channels, they are permeable to sodium and calcium. And when they open, they have the opposite effect. Sodium flows in, calcium flows in. This causes excitation of neurons. And in this sense, it would, in a central neuron, lead to things like epilepsy. And a peripheral neuron initiates things like pain. TRP channels are relatively well known or have been in the news. This is very complex here. In the news by the Nobel Prize, it was awarded in 2021 to David Julius and Ardem Patapoutian, who discovered actually three TRP channels, which are relatively well known to, let's say, the larger public. It's the TRPV1, the capsaicin receptor, which mediates heat responses, the menthol receptor, TRPM8, and the wasabi receptor, TRPA1. So they're all involved in sensing temperature and also in sensing specific chemical components. They are expressed in sensory neurons, which explains why, for instance, menthol gives a cold feeling or wasabi gives a prickling feeling when you eat it, capsaicin is hot. These have also been considered as interesting targets to treat pain. However, at this point, they have not really yielded any significant drugs. And the main reason for that is that, for instance, TRPV1 and TRPM8 also have important roles in setting the body core temperature. So if you mess up with their function, you also can induce things like fever, which are, of course, very unwanted side effects for a pain treatment. So my lab has focused on another, a bit less known, member of the family, which is TRPM3. So what do we know about TRPM3? It's a channel which is activated by heating up, so by warm temperatures, and also by neurosteroids, like, for instance, pregnenolone sulfate. And these two stimuli actually reinforce themselves. At room temperature, neurosteroids need micromolar, very high concentration. But if you have warm temperatures and neurosteroids, you get like an additive activation of the channel. When the channel opens, as explained, it's not potassium that flows through, but it's sodium and calcium, which has two effects. Sodium depolarizes the neurons, leading to excitation. Calcium also depolarizes the neurons, but also can lead to all kinds of calcium-dependent effects, like the release of neuropeptides, such as, for instance, CGRP. So these are some of the original experiments that we did when we discovered the role of TRPM3 in pain signaling. Here on the left, you can see experiments where we actually measured TRPM3 in nociceptive neurons from mouse. So if you isolate them and you give the agonist, pregnanolone sulfate, you see that you get an inward current, so a depolarizing current, which is specifically localized in the smaller sensory neurons, which are the neurons involved in pain, where they're also co-expressed with TRPV1, the capsaicin receptor. If you look at the results here on the right side, it actually just gives a clear indication that activation of TRPM3 leads to pain. So these are animals which are injected in their hind paw with pregnanolone sulfate, the TRPM3 agonist. And this induces a short-lasting but acute pain response. This pain response is no longer seen in animals that do not express TRPM3, TRPM3 knockout animals. These animals, by the way, are healthy and fertile. You don't see any specific problems to them. They have a normal motor behavior and so on. But they lack TRPM3-mediated pain because they do not give a pain response to agonist of TRPM3. They still can sense pain. For instance, if you inject capsaicin, the agonist of TRPV1, you see a normal pain response in these animals. So from this, we can conclude that TRPM3 is expressed in nociceptive neurons. If you activate it, it leads to pain. As said, it's not only permeable to sodium, leading to depolarizing currents and pain. It also leads to calcium efflux. This is why, if you activate TRPM3, as seen here on the left in the mouse skin, so you can induce the release of CGRP with the TRPM3 agonist only in the wild-type animals, much less or not at all in the TRPM3 knockout animals. So we see this with pregnanolone sulfate, which is an endogenous but not super selective agonist. But we also see it with CIM0216, which is a very potent synthetic TRPM3 agonist that we developed in the lab, so which can induce CGRP release from nerve terminals in the skin to similar levels as, for instance, capsaicin. And here on the left, these are results that come from our lab in mice. But here, very recently, it was also shown that in human skin, you can also induce release of CGRP from nerve terminals with TRPM3 agonists in a TRPM3-specific manner. So activation of TRPM3 not only induces an activation of neurons and pain signaling, but also leads to the release of CGRP. As said in my initial slide on TRP channels, so the development of antagonists for, for instance, TRPV1 has been limited by the fact that there are some side effects there, which are directly related to the function of the channel. One effect is that if you inhibit TRPV1, it leads to hyperthermia because TRPV1 is important in sensing environmental temperature. The other thing is that you inhibit TRPV1, this can also lead to heat hyposensitivity, meaning that you have a reduced sensitivity to acute heat, which in some phase I studies showed up as people that would burn themselves on drinking coffee because they wouldn't sense that it was too hot. So TRPM3 is also involved in acute heat sensing. And we did a study. We found actually that TRPM3 is together with TRPV1 and TRPA1 are three heat sensors. But TRPV1 seems to be the first in line. So as long as TRPV1 activity is preserved, inhibition of TRPM3 does not affect acute heat sensitivity and also does not affect body core temperature. So it's much safer to inhibit TRPM3 compared to, for instance, TRPV1. So what happens under pathological conditions? So we have seen in many different animal models, and this is one example, an animal model of inflammation in the hind paw. So what we do in mice, we cause an inflammation by the injection of complete Freund's adjuvant. We do this in one hind paw. The other hind paw is a control. And then we trace back, as shown here in red, the neurons that innervate this inflamed paw or the healthy paw. And what we then see, and the green dots here, if you can see, the green dots are RNA for TRPM3. We see that neurons that innervate specifically the injured paw increase their expression of TRPM3. So there is more expression of TRPM3 specifically in those neurons that innervate injured tissue. And we see this not only in this inflamed hind paw model, but we have also seen very similar results in, for instance, neurons that innervate an inflamed bladder, also following chemotherapy-induced neuropathic pain. So TRPM3 activity is upregulated in various pathological pain modes. In line therewith, we also see that if you have a TRPM3 knockout animal, so an animal that doesn't express TRPM3 at all, that they are protected to develop hypersensitivity in different pain models. So here on the left, this is the inflamed hind paw model, so the CFA model. What in a normal mouse happens is that if you cause inflammation of a hind paw, they become hypersensitive to, for instance, warm stimuli, which is indicated here in this graph by a reduction in the latency to withdraw from a hot plate. If you do the same in a TRPM3 knockout animal, they don't develop this heat hypersensitivity. And very similar results were seen for other types of pain models. So this is an oxaliplatin model, so a model of chemotherapy-induced neuropathic pain where animals and also humans that take this drug develop, for instance, cold and mechanical hypersensitivity. You see the development of the cold hypersensitivity in the wild-type mouse, but much less so in a TRPM3 knockout mouse. And this is data from another team where they did a nerve injury model, the chronic constriction model. Also there, what you see is that a knockout animal does not develop hypersensitivity when you make a constriction of the nerve. So overall, TRPM3-deficient mice seem to be protected in various pathological pain modes. So because of this protection, and the fact that TRPM3-deficient mice are healthy, fertile, and show no obvious abnormalities, we think that TRPM3 is a very interesting target as a peripheral pain sensor, but also very important in the setting of hypersensitivity in all kinds of pain models. So this led us to start developing TRPM3-specific TRPM3 antagonists that work in the periphery for treating pain conditions. And as a result of many years of work, we finally developed BHV-2100, so a novel peripherally restricted TRPM3 antagonist. It has really very great in vitro efficacy, so, low nanomolar IC50s to inhibit TRPM3, very good selectivity across all other TRP channels, more than 1,000-fold selectivity over other TRP channels, and also overall a very clear selectivity profile. We didn't see any specific activity in other targets and many other targets at concentrations up to 10 micromolar. Also, the pharmacokinetic and toxicology findings were very positive. So there was a moderate clearance, no SIP induction, very good oral bioavailability, and all the toxicity studies in animals were without any signals. So how does this compound act then in animal pain models? So we tested them in many different models, and it seems to work very well. So this is a very acute model. So the model I already explained. You inject an agonist in the paw of a mouse. Dose-dependently with BHV-2100, we can inhibit the pain response caused by acute TRPM3 activation. We also see efficacy in other and more relevant pain models, like the PSNL model, so a model of nerve injury, where we see that following dosing, you get a reduction of the hypersensitivity to mechanical stimuli, for instance, which is comparable to the efficacy of pregabalin. You have to note here, at this dose of pregabalin, the animals are partially sedated, whereas the BHV-2100 did not cause any sedative effect or no side effects that we could measure. Similar activity is also seen in a model of chemotherapy-induced neuropathic pain model. Here, the readout was cold hypersensitivity. There we see, again, a dose-dependent reversal of the cold hypersensitivity with an efficacy that was comparable to an opioid. In this case, it was tramadol. We also see efficacy here, finally, in a diabetic neuropathy model, where we see that also dose-dependently, BHV-2100 reduces the hypersensitivity to mechanical stimuli, again, to a similar extent or even better than positive controls like pregabalin. So these are all models where the pain is mediated by sensory neurons that come from the dorsal root ganglia, so close to the spinal cord. But there is also a very good rationale to use TRPM3 antagonists for the treatment of migraine. So there is, it's very clear that migraine is primarily caused by, the trigeminal system, so the trigeminal neurons, which are located in the skull. We know that TRPM3 channels are expressed there, that they can be sensitized, that they are involved in CGRP release. Then there is also human genetic evidence, GWAS studies, that give a link between TRPM3 and migraine and also other forms of pain. And finally, there is some preliminary clinical data that show that if you target sensory neurons, for instance, TRPV1-positive neurons, that this could have positive effects in migraine indications. So that's why we believe that TRPM3 antagonists could be a very new and very interesting application for treatment of migraine. So just a little bit more background there. So the migraine pathophysiology. So the trigeminal vascular pathway is actually the system that induces migraine. So you have the trigeminal sensory neurons that innervate the face, but also other parts of the skull, the meninges. And it's known that when they are hyperactive, they can cause release of CGRP. And this lies at the basis of all the symptoms that we see in people with migraines, so the pain, but also the emotional behavioral changes in migraine. They are all initiated by activation of the trigeminal vascular pathway. So TRPM3 is highly expressed in the trigeminal neurons. So these are some calcium imaging that we do. So if we stimulate mouse trigeminal neurons with the agonist of TRPM3, you see very robust calcium responses there, which are largely gone in the TRPM3 knockout. So at least half of these neurons do express TRPM3 in the mouse. If you activate the trigeminal neurons in the mouse, you also get pain. So these are pain—this is then pain which is located to the head, to the face. So if you inject the TRPM3 agonist, mice exhibit long-lasting ongoing pain, which is caused by the trigeminal ganglia and which is then absent in a TRPM3 knockout mouse, as you can see here. So you get pain behavior induced by pregnanolone sulfate or CIM0216, the agonist of TRPM3, which is abolished in the TRPM3 knockout. But you can still have a pain evoked by capsaicin. So activation of TRPM3 also causes trigeminal-mediated pain in the mouse. Then this is a study that was done by another team where they looked at TRPM3 specifically in the trigeminal afferents that go to the meninges, so the layers in the brain or surrounding the brain that are involved in initiating migraine. And there also it was seen that you get activation of TRPM3, which leads to neuronal firing of these specific neurons. And interestingly, this activity actually has some sex specificity. So it's the higher levels of overactivity that are seen in female mice compared to male mice. But the bottom line is that activation of TRPM3 in the trigeminal ganglion neurons that innervate the meninges leads to strong firing of these neurons. So what about the human trigeminal ganglia? So there are recent studies that look at the expression of all genes and all cell types in the human trigeminal ganglia. And there you can see that in the neuronal part, so in the neurons which are involved in pain sensing in the trigeminal, that there is high expression of TRPM3. But interesting, we also see TRPM3 expression in other cell types in the trigeminal ganglia, like the satellite glial cells, which surround the neurons. And at this point, we don't really know yet exactly how they contribute to pain sensing. But it could be that they have a further reinforcing effect on, for instance, hypersensitivity of trigeminal ganglia under pathological conditions. And if you look then at specific cell types, so this is from the same study, if you analyze the data, you see that there is TRPM3 expression here on the top line in almost all the cell types, but particularly in the cell types which are involved in migraine, for instance, the somatostatin neurons, the neurons that express TRPM8. And also there is co-expression with CALCA and CALCB, which are actually genes encoding for, for instance, CGRP. And ADCYAP1 here, which is a gene involved in expression of PACAP, which is also another protein, neuropeptide, that is important for initiation of migraine. So overall, these data suggest that TRPM3 could be a very new and important target for the treatment of migraine. With this, I think I pass on to Dr. Lipton. Thank you. Okay. Yes. So I wish I was there with you in person. I tested positive for COVID both yesterday morning and this morning and thought it would not be a good idea to share my viral load, even if I very much want to share my ideas with all of you today. I guess I will share my screen. Give me just one second, please. So I assume you see my slides now? Yes. Yes. Yes. Yeah. Yeah. Fantastic. So I've worked with the Biohaven team for a very long time, played a role in helping with the development of Rimegepant and Zavegepant, and want to begin with an error that I made actually shortly after the launch of Sumatriptan in 1992. So I was speaking with the guy who was in charge of Sumatriptan marketing for Glaxo. I had decided I wanted to devote my career to headache medicine, and I'd just moved to the Montefiore Headache Center, which I currently direct, to be a headache clinician. And I remember saying to Steve Skolsky, this Glaxo marketing guy, that I was afraid that Sumatriptan was going to ruin my career because the treatment was going to be so effective that all needs were going to be met. And he actually laughed at my face, which is unusual behavior for a marketing guy, and explained to me something that has proved very true and perhaps obvious. And that is that for conditions that are underdiagnosed and undertreated, this treatment gets better. More and more untreated people are drawn into the treatment pool, and clinical needs and clinical opportunities and pharmacologic needs and drug development opportunities actually expand, not contract, as therapeutics get better. And I think that remains true. So this is an overview of what I'm going to talk about over the next 5-8 minutes. We're going to talk about the high and stable prevalence of migraine in the United States. We're going to talk a little bit about burden and disability despite advances in treatment. We're going to talk about the need to make better use of existing and emergent treatments, and then just a bit on the promise of TRPM3 as a promising target. So as most of you know, migraine affects 1 billion people worldwide. And if we look just at the United States, it affects 47 million people approximately, an extraordinarily highly prevalent disorder by any standard, and also in terms of time lost due to disability, overall the world's second most disabling disorder. And then if we look at the age and sex distribution of migraine in the United States, we're plotting one-year period prevalence here as a function of age. What you see is that prevalence rises with age from the postpubertal period to middle life and then declines thereafter. And you also see that at all postpubertal ages, migraine is more common in women than it is in men. Part of the reason the economic burden of migraine is so great is that it is a condition that affects people during their peak productive years. It's perhaps worth saying that migraine is comorbid with, probabilistically associated with, a large number of psychiatric disorders, including depression and anxiety and bipolar disease, among others. Then if we look at chronic migraine - and these are the people who have 15 or more headache days per month - about 7% of all migraine is chronic migraine. Just from a prevalence and commercial opportunity perspective, chronic migraine is actually a little bit more common than epilepsy, which is, of course, an extraordinarily common neurologic disorder. There is a widely used measure for migraine-related disability called MIDAS, the Migraine Disability Assessment Scale. And here, what I'm plotting are serial results that look at the proportion of people with migraine who have moderate or severe disability, starting all the way back in 2001 and extending to 2018, which is some of the more recent data that we have. And we're looking at total disability, then women in green, and then men in blue. And what you see is that migraine disability appears to have increased and has remained stable over many years. And the most recent data is also compatible with this. So despite advancements in treatment, levels of reported disability remain high. And I suspect part of the reason for that may be that migraine is less stigmatized than it used to be, and people may be more willing to report the life impact of migraine. So the American Headache Society has issued some consensus position statements on migraine. When they define the goals of acute treatment for migraine, they talk about rapid and consistent freedom from pain and associated symptoms, especially whatever symptom the patient designates as most bothersome other than pain. That is most often sensitivity to light or sensitivity to sound. We also want treatment where the headaches don't return. Restoration of function is very important because short-term activity limitations are a hallmark of migraine. We want to have a minimal need for repeating doses or using rescue medication. We want to optimize self-care and reduce utilization in less optimal settings and more costly settings, like the emergency room. Obviously, low adverse event rates is important. One of the major needs reasonably well addressed by the gepants is their very favorable tolerability profile. The AHS reminds us that cost matters and also reminds us that many people with episodic migraine transition to chronic migraine, so they go from less than 15 to 15 or more monthly headache days. Suboptimal acute treatment has been shown to be a risk factor for migraine progression. Part of the argument for delivering cost-effective acute treatments is to relieve pain and disability today. Part of the argument is that we may prevent worsening of headache tomorrow. In headache subspecialty practice, perhaps 60% or 80% of the patients I see have chronic migraine, and then I have to work really hard to help reduce their monthly headache day frequency. There's, of course, no one-size-fits-all treatment for migraine. Clinical practice, especially in refractory patients, may include multiple treatment trials in order to find the optimal regimen. Switching within and between classes and using combination therapies is often needed. Of course, what's really needed is a more personalized approach to treatment where we have a rational basis for selecting the treatment we give to a particular person based on the probability that they'll respond to that treatment. TRPM3-targeted small molecule receptor antagonists are certainly a mechanism-based treatment for migraine. There's lots of reasons to believe this will be effective. Because the mechanism is distinct, that raises the hope that it'll be effective in people who don't respond to the currently marketed acute treatments. It should be relatively free of the cardiovascular contraindications that severely limited the use of triptans and of ergots. A drug with a long half-life may cause less recurrence. A drug with a short Tmax should be associated with a more rapid onset of action. And again, that need for a rapid onset of action is a priority for many people who have migraines. This slide is intended to be a summary of how we're doing, or at least how we were doing in 2013, in terms of delivering care to people with migraine. So in a population study, we defined a group of people who had episodic migraine and headache-related disability. And we first asked, "Are they currently seeing a doctor?" And at the time we did the study, 45% of people were currently consulting. The most recent data says now that perhaps 60% of people are currently consulting. But of people who are currently consulting, many never receive a medical diagnosis of migraine. And of course, in the absence of a diagnosis, people aren't eligible for effective treatment. Of people who are consulting and diagnosed, when you sort of multiply this cascade of probabilities, only about a quarter of people are consulting, diagnosed, and treated in aggregate. This is just intended to reflect the high level of unmet need. Of course, as therapies improve, what we need to do is make sure that the people in need of treatment get the treatments that they need. This is data from a paper published by Seymour Diamond more than 20 years ago, showing the promise that Civamide, a TRPV1 receptor modulator, may be efficacious in the treatment of migraine. On the left side of the slide, we're looking at improvement in headache pain severity. In this small study, we see that two hours following dosing, 56% of people who were treated with either a lower dose or a higher dose had a decrease in pain severity in comparison, and that 22% of people were pain-free, and that the proportion of people responding is higher with active treatment than placebo. In the right side of the slide, what we see is that there is relief of some of the associated symptoms of migraine, nausea, photophobia, and phonophobia over time as well. Again, the key point here is that there is promise that TRPM-targeted therapies may be effective in migraine. That includes some human data and extends beyond the clinical data we saw already. Then in summary, again, we've seen migraine prevalence remains very high, 1 billion people worldwide, 47 million people in the U.S. That disability and burden remain high despite advances in treatment. That there's a real need to make better use of existing treatments and a need for more effective treatments and better treatments. That in that background of huge unmet need, TRPM3 is a promising target. With that, I would like to pass the mic on. Thank you. Okay. Do I have control back? I think so. Are you seeing Dr. Lipton or myself? Let me see. Okay. Yes. Okay. You're able to see my slides. Okay. So we talked about the potential of TRPM3 targeting in pain, biologically, both in migraine and in pain in general, because TRPM3 receptors are upregulated in inflammation. They're expressed in the trigeminal ganglia. So they are a rational target in the treatment of pain. Okay. So that's great. To reiterate, there is a huge unmet need in both migraine, but in pain in general, 1 in 5 people in the world live with chronic pain. About a quarter of them are suffering from neuropathic pain, most commonly diabetic neuropathy. There is a huge unmet need in chronic pain. Probably the best proof of that unmet need is the opioid epidemic that we've been dealing with in this country. The potential is great, but do we have a drug to address this potential? I want to show you the early data we have with BHV-2100, our TRPM3 antagonist. As Dr. Voets showed, in preclinical studies, highly targeted to TRPM3, selective. But how does it do so far in human trials? And what are our plans for the upcoming year? So starting with migraine, one question would be, if we're going to target migraine and we have target engagement in migraine, can we treat acute migraine rapidly? People want to get relief from their pain rapidly. So do we reach plasma concentrations target engagement fast enough to treat migraine acutely rapidly and effectively? So this is our pharmacokinetic data from our first-in-human study where we studied BHV-2100 ranging from 25 mg up to 500 mg. And I will show you the tolerability data in the next slide. But what you notice here is that our EC90 data, 90% inhibition of pain coming from our preclinical data, was exceeded even in our first lowest dose. And by the time we came to the 500 mg, we were many, many folds above that EC90. And how long did that take? Even in our first PK analysis, which is 20 minutes, we have seen in most doses that we were above that by 20 minutes. So this gives us the potential of acutely treating migraine very rapidly. And how did the patients do or healthy volunteers do in our first in-human study? I'm showing the SAD data, currently single ascending dose. The MAD dosing is completed, but we have data pending coming out. But it's been going very well. So the phase 1 study so far has been going very well. So we have had no dose-limiting toxicities, no serious adverse effects, really mostly mild and what's expected in typical phase 1 studies kind of AEs. There's been no trends in vital signs. Very importantly, what plagued, let's say, the TRPV1 antagonist, the temperature-related problems have not occurred so far. No trend in our vital signs, including temperatures. So we had only two AEs in more than one person. And this was dizziness, and one of them in the context of loose stool, and fatigue in two people. Really, we have not seen a pattern and excellent tolerability so far. Just to add, this safety tolerability and preclinical promise was recognized this year by the AAN. And the 2100 presentation got the abstract of distinction. So I'm really excited that I'm getting to present these data to you. So what are we planning to do later this year? So as mentioned, we have a potential in migraine. And we want to go right into migraine with a proof of concept study in migraine where we will study a range of doses. So you'll see from 25-150 milligrams in a 200-person study. So important to note that we have demonstrated very good safety and tolerability so far, up to 500 milligrams. All of these doses that we have chosen to study in our proof of concept study exceeded EC90 by several folds. So this will be a typical acute migraine study. Patients will have 45 days to have one migraine attack. They will be receiving one of these doses of BHV2100 in a double-blind manner. We will measure pain relief and relief from most bothersome symptoms, of course, also safety and tolerability. How about pain? What do we do with pain? So we have designed a really what we feel is a very innovative study in pain. We're going to plan later this year a proof of concept data proof of concept study for pain. I want to spend a minute on this proof of concept because we're very excited about this. So what we intend to do is to do a laser-evoked potential study. So we have seen that in animal models, 2100 works, and it inhibits pain. But in the pain world, you want to see human data. Does the pain reduction translate into humans from mice? And can we show this in an efficient manner and understand dose-response relationship and disease pathophysiology relationship? So lo and behold, we learned that laser-evoked potential. So if you use a CO2 laser to rapidly heat the skin, you generate a sudden rise in temperature. And that stimulates exactly those neurons that are responsible for pain. So the free nerve endings in the skin, without injuring the skin, you create this rise in temperature, and you stimulate the nociceptors in the skin. So this gives us a possibility of giving a standardized dose and duration of pain. Then when we test a drug with it, we can measure the response thereof. And how do we measure it? Is there a way to quantitatively and objectively measure pain response in humans beyond saying, "Hey, how much did this hurt?" So it turns out there is. So if you record an EEG recorder - so that's actually an evoked potential - and give the person the laser-induced pain, you can filter that EEG and measure quantitatively the pain that they feel. And then you give a medication that effectively reduces that pain. And the amplitude of that waveform decreases depending on how much pain reduction you can do. So not only this gives us the ability to measure pain reduction, it gives us the ability to do so in a quantitative way and in a placebo-resistant way because this does not change based on placebo. And it gives us to get a measure or reading of our pain reduction in a very rapid manner. So this is our design of our proof of concept study in laser pain, again, to do later this year. We will, again, study a broad range to understand how a given dose of BHV-2100 works. We will give four different doses. And as an active control, we will use ibuprofen that's been demonstrated to work in this pain model. We will have a small sample size, which has been demonstrated already to be actually much more than enough to get this proof of concept data with the laser-evoked potential. And we will basically create a neuroinflammation state by giving UV burn into the participant's backs where basically like a burned skin after a sunburn, if you take a hot shower, it says, "Oh, it burns." So basically, it simulates a state of inflammation. And then we test it with the laser, as I mentioned to you in the earlier slide, to see if we can reduce inflammatory pain with BHV-2100. So we're very excited about these two studies, the migraine to show us the path to move forward in migraine, and in the evoked potential study to design an efficient way of testing our potential in pain and understanding which pain models and studies we want to go to moving forward. And with that, I will lead to the panels. Yeah. Thank you so much. I want to invite Tyler Van Buren to join us from TD Cowen to moderate today's panel. I know we're running a little bit late. We have a little extra time built in for that. So we're going to try to keep on schedule. So Tyler, welcome. And we have a lot of exciting things to talk about. So I'll hand it over to you. Cool. I I have till like 11:40 approximately. Yeah. All right. I'll do my best to be efficient with the questions so I can get you guys to lunch. But thank you very much to Biohaven for having me and trusting in me to moderate the session. It's a privilege to be up here with the distinguished individuals and world-renowned KOLs. Those scientific presentations were fascinating. There's going to be a quiz at the end. So I hope you guys were paying attention. But we'll start with BHV-7000 and Kv7. So you guys have dialed out GABA. And for some folks, I think it might be difficult to understand how you could separate efficacy from some of these adverse events that we're seeing with some of the other compounds. So it'd be helpful to hear you guys elaborate on how you did that and the structural differences. Thanks for your question. So the concept of dialing out is a med-chem approach. You make about 5,000. It's an iterative process. You make about 5,000 compounds or 2,000 compounds. And each compound that goes through your screening criteria, you see whether or not that you have that activity. And so all compounds are screened against either a binding-based assay, which sometimes does not relate to function. So we decided to use a functional-based assay. So the concept of dialing out, for me, is a med-chem iterative process to ensure that we don't have that activity in the molecules that we bring forward. Your second question really goes down to the heart of the program about the hypothesis here is obviously, and Dr. Ogassi talked about this, there are a lot of different mechanisms for antiseizure activity compounds. So if we did not have a polypharmacology molecule, maybe similar to ezogabine because it was not med-chem derived against that particular target, when we did it, we thought, "Well, will we have the efficacy that we need? And will we start to see the separation preclinically?" And the answer to that is yes. And as Dr. Ogassi also talked about, the clinical utility of ezogabine showed the narrowed index. And we could replicate that preclinically. So it was almost like a reverse engineer. So the idea was something more selective, less polypharmacology would enable better efficacy, wider tolerability. Yeah. I think Steven has done a really good job answering that question. I'll just elaborate by saying that if you remember, in my presentation, I pointed out that Ezogabine had been developed as an antiseizure medicine before we really understood the Kv7 mechanism. And as people tried to understand how it was exerting its antiseizure effect, people studied the Kv7 channel. But other groups, including Steve White's group at the University of Utah, discovered that Ezogabine had GABA-A receptor positive modulatory activity. Now, the GABA-A mechanism is one that is generally useful in the treatment of acute seizures. But it hasn't really been exploited for chronic treatment of epilepsy because of the sedative side effects that accompany positive modulation of the GABA receptor. So at the time, we hypothesized that this could be a problem with Ezogabine, the fact that it has these two mechanisms, that if we could get a cleaner drug that just targeted Kv7, we might be able to dial out, in essence, the side effects that occurred with Ezogabine, which was developed through an unbiased screening mechanism and without attention to the underlying mechanism. Thanks for that. Dr. Rogawski, it'd be helpful since you're involved or you were involved with the Xenon program, it'd be helpful to hear you elaborate on some of the most compelling aspects of the efficacy that has been observed to date and what unmet needs in focal and generalized epilepsy that addresses and why you might be excited about BHV-7000? Well, so first of all, I just want to reemphasize that there is a very significant unmet need in epilepsy and the treatment of various forms of epilepsy, including focal epilepsy, which affects about 60% of patients with epilepsy and is probably the most challenging of the epilepsies, apart from the DEEs, which are rare but are very, very significantly challenging for patients and their families. But even among the patients with focal epilepsies, we're actually only able to treat about 70% of the patients. There is a very large number, 30% of patients, who we can't control seizures with any of the medicines that are available today. So there's a continuing interest in the academic community and in industry for molecules that can add to our pharmacological armamentarium for epilepsy. And as I mentioned, at the moment, there isn't a Kv7 activator that is available on the market. Ezogabine ran into problems because it causes a bluish discoloration in the skin and also in the sclera and in the eye and even in the retina. The FDA was concerned about this. It turned out not to really be a medically significant problem, but it dissuaded its prescribing. The manufacturer removed it from the market. It's no longer available. Now, in terms of the Xenon program, they believe that XEN1101, their molecule, has eliminated this skin discoloration problem, which they believe is due to dimerization of the molecule and precipitation in the skin and in the sclera of the eye. That's an advantage. When they took it into the clinic in their phase two study, they saw a very nice dose-dependent inhibition of seizures at the 10, 20, and 25-milligram dose. Unfortunately, however, there was also a dose-dependent incidence of side effects, particularly dizziness and somnolence. I believe that they really weren't able to take it further than the 25-milligram dose because of the dose-limiting toxicity. As you heard and this is maybe a general comment to make - we're limited by side effects in the ability to use our antiseizure medications effectively. We believe that if we could push them, we would be able to treat those 30% of patients that are currently refractory. We can't do that because of the side effects that occur with many of the medicines. The concept here is that if we can get to greater impact on the target without the dose-limiting side effects, we could cover more patients. Okay. So what's the probability that the relatively pristine safety profile BHV-7000 holds up in phase 3 from what we've seen in phase 1, in your opinion? Gee, I left my crystal ball in California. I hesitate to speculate on this. But as I mentioned in my presentation, the test that we use to assess side effects and there's more or less predictability of those animal model tests, the rotorod and so forth. I should have mentioned this perhaps a bit more aggressively during my talk. But if you recall that slide, there was no toxicity seen in the animal model at any dose of BHV-7000, whereas you got nearly complete protection in the maximal electroshock test. And this is, to me anyway, looking at the data, a bit unprecedented. Normally, you do see the occurrence of toxicity. And with BHV-7000, you didn't. So if that's any reflection of what the experience is going to be in the clinic, I think it bodes well. Tyler, the only thing that I would add, and Dr. Rogawski certainly has the clinical experience here. But often in the phase one studies, what you see in phase one often translates into what you see in the phase two and phase three studies. And the side effects, somnolence, dizziness, they don't take months to emerge. So what we saw in our MAD study over two weeks, I think augurs well that we anticipate that we will see a similar, if not identical, profile. Okay. That's helpful. And you guys have reported encouraging or Biohaven encouraging EEG data. Some of the feedback from investors is that's obviously not seizure reduction, right? How can we get confidence that that's going to translate to seizure reduction? So for those of us not skilled in the art of interpreting EEG, can you just elaborate on why that gives you confidence that we'll see good seizure reductions? Happy to. The purpose of that study was not to demonstrate that or to imply that activation of the EEG will translate. It's not an established biomarker in that sense. That said, what we were intending to do and what we were pleased to see was demonstration of a pharmacodynamic effect of the drug. So again, without seeing any somnolence, a reasonable question is, how much are you getting into the brain, if any? And what we showed was that and I think this is important because pharmacology, if it's real, will be both time and it'll be concentration-dependent. And that is seen both in terms of increasing doses and evolution over time. And we saw both of those effects. I think between that and the preclinical data, we're very excited about this. And I guess the only other comment that I would make is that as Dr. Rogawski really explained well historically, this is, in medicine, if you can amplify or augment what the body does naturally, you have a really good chance of restoring sort of the homeostasis of health. And I think that's what we're excited about with the way that 7000 activates the Kv7 target. Okay. That's helpful. And you mentioned pharmacology being real. So maybe you could just elaborate on that point. Have the preclinical results lined up with what you've seen in phase one? And in the planned studies, you're testing 25, 50, 70 mgs in focal and 75 in generalized. And obviously, 50 and 75 are way above EC50. So do you think you could get enhanced seizure reduction by going up to those levels? Yes. I think that is the goal. If we can one of the things that and all credit to Xenon because this is they demonstrated the activity, the really terrific efficacy, early onset of effectiveness. If we can replicate that with a clean safety profile, it'll be a big win for patients. That said, they're achieving those results, sort of briefly getting to the EC50 in their preclinical models. So we're guardedly optimistic that we will have improved efficacy and maintain the safety profile, which obviously would be a game changer for patients. I can just add to that. So when you talk about the pharmacology, we're talking really about the exquisite potency of the compound in a systems-oriented approach. So one of the reasons why we were doing the IEM or the sensory neurons and why I continue to do some additional studies in a we went to the screening cells of highly expressed cells. That's not going to be reflective of the potency of the molecule. And you could see from the MES data and Dr. Rogawski showed this, showed our data, how exquisitely potent it was. And so when you can actually look at how you can alter action potential firing with the potency of the compound, it increases the confidence with the EEG that we're going to have a really nice effect with the large dose range that we're testing in the clinic. Okay. So Tyler, I just wanted to have the group when you review your slides and look at the EEG data that was presented. To focus on the activation of the delta frequency band in the case of the Xenon compound versus the lack of that in the case of BHV-7000. I think that's really important to pay attention to because the delta frequency band activation is a reflection of drug effects on the EEG. This is a standard sort of EEG biomarker of drug toxicity on the brain, if you will. So barbiturates and antipsychotics, when they're overdosed and so forth, can cause activation of that delta frequency band. And we're not seeing that in the case of the Biohaven compound. Okay. That's helpful. As we think about the phase 3 program, you've mentioned that 110 global sites have been selected. So is that for both the focal and the generalized phase 3 trials? And Michael, you talked about catching the runner in front. So how long will it take to get those sites activated? And how is enrollment going? So as you pointed out, it's actually over 100 sites for each focal study. And we've activated the majority of our U.S. sites. The European sites are going to be coming online very shortly. So we anticipate through the summer that we will be largely fully ready to go. Okay. And enrollment? Have you guys started enrolling patients in those sites? We have. Enrolling, as you know from the design, there's a screening period. So we started enrolling patients in March and are already randomizing patients, so. Okay. Great. And when do you estimate that enrollment could complete and when we might receive top-line data? Yeah. Well, we have not disclosed exactly when we expect to complete that. But I think that our goal is to I think Xenon has said that early next year, they will have completed enrollment. And we don't expect to be that much farther behind them. Okay. Let's move to migraine and bipolar disorder, Dr. Krystal. So I recall when the Xenon data came out in the fall. It didn't blow people away, but there clearly is something there. So it would be helpful to hear you elaborate on that data and why you're excited for this class of drugs and what unmet needs it could address in MDD specifically. Sure. I think it's hard to estimate the true magnitude of the effect of the Xenon drug from that single trial. But I think one of the issues that I think is coming up as a theme of this morning is that dose-limited efficacy could dose-limiting side effects can constrain efficacy. So certainly in the COSTE study, where the group separation between drug and placebo was greater or at least appeared greater than in the Xenon trials, even there, you had 20% of patients that had dose limited by side effects. And that was also an issue that came up in the Xenon trial. So I think one of the issues that emerges is that if you can dose higher and aren't limited in individual patients by the side effect profile, then you can have more consistent efficacy. We really can't tell. I can't tell from the data that I've seen about the variability of the response, whether the people who could achieve the optimal dose had better efficacy than those who did not. But that would be certainly a hypothesis to explore. Okay. With reference to the low and high-dose data from the EEG study and CNS target engagement, could you just elaborate on that and how you expect it to translate into efficacy and why you're using the 50 mg dose in MDD and 75 mg in BPD? Yeah. Well, I think I certainly would defer to Dr. Krystal from the efficacy or what's seen in the precedent in bipolar and MDD. But as you saw from the pharmacodynamic or PK data that we showed, we are well beyond the EC50. Acute mania is an acute illness. We wanted to get maximum dose. There is an opportunity to down-titrate if there are tolerability side effects. So I think in the case of bipolar disease, we were looking to especially in acute bipolar mania, we were looking to maximize the effect. We're very confident that with a 50-milligram dose, that's the robust dose. That's the dose that's being replicated across the two focal studies. We think that's the sweet spot. Okay. Maybe to round out the MDD and bipolar disorder discussion, Dr. Krystal, do you believe the probability of success is higher for one versus the other at the moment? Do you have a favorite between the two indications? Do I have a favorite? Do I have a favorite on this? What I can say is that I showed you two figures from two trials where you had positive data already from MDD. As they say, past performance is predictor of future success. Having two studies in the bag is a good start for MDD. I think that the need is so great for bipolar disorder that building and the data are different. You have that genetic data and the animal data, the transcriptomic data. All the arrows are pointing in the right direction. So I think we can have a high level of excitement about testing the drug in bipolar disorder. But it's nice to have two studies in the bag, really, already to help to motivate further studies in MDD. Okay. We'll end with a discussion of TRPM3 and move to Dr. Lipton. So Biohaven, clearly, Biohaven 1.0 was very successful with the initial migraine franchise. I've got Nurtec in my backpack. I learned from the presentation that I'm 7.4% of males, unfortunately, with migraine. But Dr. Lipton, it was announced that there will be a Kv7 or 7000 migraine study as well as the TRPM3 migraine program. So I'd be curious to hear your thoughts on the remaining unmet need in migraine and if you have any initial thoughts on how these two agents might address that. Sure. Well, the ultimate issue in migraine and in epilepsy, of course, and I assume in depression and bipolar disease as well, is that each of those categories is more than one disorder. The genetics tells us that there are multiple genetic types of migraine. And the ultimate and when we look at how well any given acute treatment works in migraine, rimegepant, which is a wildly successful drug and one that has changed my life as a clinician and the lives of my patients, the two-hour pain-free rate to rimegepant is in the 20% range. Though 60% of people get pain relief in two hours. People with migraine also have tremendously high expectations. They want rapid relief. They want relief of pain and associated symptoms. They want relief across multiple attacks. And ultimately, I think success will arise as we develop personalized medicine approaches for treating migraine. And remember, when you have a condition with 47 million Americans who suffer from that condition, if you have the treatment of choice for 10% of them, that is a gigantic home run. So what I look forward to is a day when we move past trial-and-error medicine. I pick drugs based on comorbidities. I pick drugs based on patient preference for route of administration. I pick drugs based on the patient's willingness to tolerate or not tolerate certain side effects. But I don't pick drugs from among those that work based on what will work for the patient in front of me. And so the exciting thing about the CGRP era and the exciting thing about what I hope is the beginning of two new eras is that we'll learn enough about the biology to move past trial-and-error medicine to personalized medicine. Okay. That's helpful. For the TRPM3 or 2100 phase 2 program that you guys will be kicking off, once you get that enrolled, can you talk about how quickly we could get to initial data with respect to antipain and antimigraine effects? So I think the key differentiator of that study is that it's an efficient study. It's a small study. It's easy to recruit. Biohaven has the expertise in running migraine studies. Our sites are already knocking our door. "Hey, when are we starting our study?" So I think we're starting later this year as we announced. It's a 45-day study. Our patients will have 45 days to have a migraine. We expect a quite rapid recruitment. So can't tell you exactly when we're going to release our results, but I expect quite efficient, rapid enrollment and data. Okay. And just wrapping up with a couple of questions with TRPM3 and neuropathic pain, can you just speak towards the proof of concept achieved with the NaV1.8 inhibitors like Vertex's program and how you expect 2100 to differentiate? Indeed, an interesting comparison. So the NaV1.8 and the TRPM3 programs, they have some similarities, but also some differences. So the NaV1.8, the role of NaV1.8 in pain signaling is very specific. It's involved in the action potential that is necessary to transmit the pain signal from periphery to the central nervous system. And if you inhibit that or if you lower that activity, you expect to reduce pain signaling. TRPM3 is involved both in sensing the pain stimulus. So by reducing that, you may reduce the sensitivity to all kinds of stimuli. But what is also a very important distinction is that it also is involved in calcium signaling. So it lets calcium in, and the influx of calcium in nerve endings is not so much important for initiating the pain signaling, which is important for all the other consequences that you have like neurogenic inflammation, the release of CGRP, and other modulators. And I think that in that respect, TRPM3 would have additional value on top of inhibiting the action potential propagation, which you expect with Nav1.8 antagonists. If I may, I want to add on it. So while we're comparing maybe the Nav1.8 to TRPM3, what the differences are. But one important, I think, thing to appreciate is that the Nav1.8 data suggests the peripheral inhibition of pain has a chance of success now in clinic because there are some naysayers. They'll say, "If you do not inhibit the central pain generator, peripherally blocking pain is not going to work." Well, now we have human data showing that addressing the excitability of the sensory nerves has a role in pain. And the advantages, you have much less of a chance of addiction and sedation and other central nervous system effects like sleepiness and etc. that we would get if we got a high dose of any centrally acting drug. So I see it as a proof of mechanism. Thank you. Okay. To wrap up the pain discussion for the proof of concept, sunburn on the back study that I'm not sure I'm going to be signing up for. So you've got this laser-induced potential endpoint, and then you have various pain endpoints, which I imagine are more patient-reported. How are you going to determine success with that initial proof of concept study? What do we need to see? So there is a large database of almost every known and approved drug has been actually tested in that method. So I think success is demonstrably lowering the action potential that we objectively measure, as I showed in that graph. But what we haven't mentioned is that those measures do correlate with what the participants tell us. So we hope that we'll be able to first demonstrate target engagement, second, a dose-dependent increase over time in the reduction of pain quantitatively. And then there are other aspects on that waveform. We can see what works. So we'll be doing the inflamed skin and normal skin. So that gives us the ability to understand, do we work better in inflamed tissue, or do we reduce the pain threshold or normal tissue? And we predict based on mechanism, we're going to reduce work very well on inflamed tissue. So we'll be able to, I think, gauge success that we're engaging target. We're able to demonstrate dose-response relationship, and we can do this fast. Okay. Excellent. Could I make just one final comment? In listening to this, it occurs to me that there's sort of a general concept here that perhaps is worth reflecting on. And that is that in the treatment of epilepsy, we very frequently end up with polypharmacy. It's very rare that my patients are going to be able to be maintained on only a single medicine. We often have patients on two, three, four medicines for difficult to treat epilepsies. As a neurologist, in addition to epilepsy, it turns out I also treat patients with migraine, and I find that this is a similar situation. Even for acute migraine, Nurtec is a great drug, but it doesn't help everybody, and it doesn't help everybody completely, as Dr. Lipton pointed out. So you add nonsteroidals. You add triptans. You add other agents in the cocktail, if you will, to help patients eliminate their headache symptoms. I think in pain as well, there's no reason why the Vertex drug couldn't be used in combination. The fundamental concept is you want to have complementary mechanisms. You want to have different mechanisms. You don't want to have just another drug with the same mechanism, but spread it out. And that's the way that you get enhanced efficacy. So what you're doing is you're adding to the market. You're not sort of taking market share from somebody else. But what you're doing is you're increasing the pie for all of the products that are in the space. Okay. That's a great point. Thanks to the panel for all your thoughts. With that, we'll go ahead and wrap up. Okay. So thank you, Tyler. And I want to thank all of our panelists. And I think, Mike, you said it great. At the end, really, what we're trying to do is find new treatments for patients by targeting novel mechanisms of action that are complementary to those that exist with potential for superior efficacy, safety, better combinability. I think the ion channel program that we have built here really puts us in a leadership position in ion channels and really proud of that. So thank you. I know you see the panel here. I will introduce them in just a moment. I'm first going to go to give you a very brief update on our glutamate modulator program. I think you guys are all familiar with troriluzole, which is our longstanding glutamate modulator program, which we're studying in obsessive-compulsive disorder and in spinocerebellar ataxia. So briefly, for OCD, I think it's really important to know that there's not been a new treatment for OCD for more than 20 years. The unmet need for OCD is very significant. People are still getting ablative neurosurgery for the treatment of refractory OCD, and up to 60% of patients do not respond to the first-line treatments. So it's a huge unmet need. We have phase 2 data with Troriluzole, which is very exciting. We're running a phase 3 program. I think you guys have heard that the phase 3, we have two studies. They're identical studies. The first study we're announcing now will have top-line results in the first half of 2025. The second study will have an interim analysis in the second half of 2024. The first study already passed the interim, and now the second study will have its interim in the second half of this year. So that's the update on OCD. I think, again, we're really excited about this program. The phase 2 data looks good, and we're moving towards a pivotal study readout that could be transformational for people suffering from OCD. I'm also going to give you a very brief update on our spinocerebellar ataxia program. As you know, SCA is a rare neurodegenerative disease that is fatal with no treatments available. We've been studying SCA at Biohaven for many years. We announced last year, of course, that some of the trial results from our phase 3 program. We filed an MAA application in Europe that's currently under review. We had filed or submitted an NDA application in the US. We are now in conversation with FDA about potentially resubmitting a new NDA regarding SCA with additional data and analyses. That's a brief update on the glutamate program. Of course, if you have questions, let me know. We can answer some of those questions perhaps a little later. Without further ado, I wanted to go ahead and introduce our neuroinflammation platform program. We have a number of presentations regarding the program, again, from internal experts and external experts. I'm going to start with introductions. The panel is all here. I'll start first with Dr. Cynthia Lemere. Dr. Lemere is a professor at Harvard. She is a scientist who studies immunology and Alzheimer's disease. She's really an expert in amyloid-related imaging abnormalities. That's what she's going to be focusing her conversation and presentation today. We have Dr. Steve Salloway. Dr. Salloway is a professor at Brown University. He's a leader in clinical trials in Alzheimer's disease. I think you probably recognize his name. He's the first author on many papers in Alzheimer's disease, including New England Journal and Nature papers. He's been interested in some of the appropriate use for the new anti-amyloid treatments that we have for Alzheimer's disease. And he's not going to be giving a presentation, but he'll be on the panel speaking about his experience with ARIA and what the potential unmet need is for a treatment like BHV-8000, which could potentially treat or prevent ARIA. We have Dr. Mark Albers, who's a neurologist at Harvard and at Mass General. Dr. Albers is a translational scientist as well. He's done a lot of non-clinical work with TYK2, JAK-STAT pathways, including work with BHV-8000 and other inhibitors in that class, both preclinically and clinically. And so he will be on the panel to speak about his thoughts and perspectives on the role of BHV-8000 in neurodegenerative diseases. So those are our external experts. And then our internal experts, I'm going to start with Dr. Nick Kozauer. Nick is Senior Vice President at Biohaven, but you may recognize him. He spent 13 years at the FDA prior to coming to Biohaven. The last several years, he was leading DN2 of neurology as the director. So we're really pleased to have Nick here at Biohaven. He'll be kicking us off. We also have Dr. Lindsay Lair. Lindsay is a neurologist's VP of clinical development, and she'll be speaking about BHV-8000 for MS. And then we have Dr. Pete Ackerman. Pete is VP of clinical development here at Biohaven as well, and he'll be on the panel speaking as well. So with that, I'll hand it over to Nick. Thank you, Irfan. Good afternoon, everyone. So I'll get us started speaking about BHV-8000. So BHV-8000 is a brain-penetrant selective inhibitor of tyrosine kinase 2, or TYK2, and Janus kinase 1, or JAK1. TYK2 and JAK1 are two of the four members of the JAK family of enzymes that are highly involved in inflammatory signaling via the JAK-STAT pathway and therefore provide a target with broad therapeutic potential across multiple neuroinflammatory disorders, including Alzheimer's disease, Parkinson's disease, multiple sclerosis, and others, as well as for the prevention of amyloid-related imaging abnormalities, or ARIA, which can occur in early Alzheimer's patients who are starting anti-amyloid therapy. You'll hear a lot more from Dr. Lemere about that today. Additionally, a selective inhibitor of TYK2 and JAK1 is expected to have a safety profile that is much more favorable relative to a number of the approved JAK inhibitors for peripheral autoimmune conditions that target JAK2 and/or JAK3. I'll provide more detail about that as well. I'll also give an update on our phase 1 program, including some encouraging data from our safety data from our phase 1 program, as well as some encouraging early preliminary pharmacodynamic data and target engagement. So as I get started, I do want to just take a minute and highlight two very important recent developments across two of our programs, both of which relate to FDA interactions. So firstly, with regard to the prevention of ARIA, this is a novel program. This has not been done before. As you'll hear from Dr. Lemere, ARIA is a potentially serious complication of anti-amyloid therapy in patients with early Alzheimer's disease. And in some rare cases, it can even be fatal. So we met with the FDA about our program and had a very productive meeting on several fronts. So the first is FDA absolutely understands the unmet need that ARIA represents currently and the challenges that it presents in the management of patients with Alzheimer's disease. And they also understand that that unmet need is only going to grow exponentially as the use of these agents becomes more widespread in practice. They made it clear that there is a path to an approval for this indication, which was great because no one's done this before. And so again, our program is very novel in that regard. And they also gave us very favorable feedback on the design of our planned phase 2/3 clinical trial to support that indication. So a really productive meeting on all fronts there. Additionally, for Parkinson's, we are proposing a novel primary efficacy endpoint for an early disease-modifying trial in Parkinson's that's based on a time-to-event analysis. We believe that this endpoint addresses a significant gap in the tools that are available to detect clinical benefit in disease-modifying trials of Parkinson's. As I'll talk about a little later in the presentation, this endpoint should save upwards of 300 patients per study in terms of sample size relative to the current approach. We met with FDA about that endpoint, and they also gave us very favorable feedback there as well. I'll talk more about that a little bit later on. Very good news on both of those fronts. We look forward to working closely with the FDA on those programs. Why study a brain-penetrant TYK2, JAK1 inhibitor for neuroinflammatory neurodegenerative diseases? The reason is there's a rapidly expanding body of literature, including non-clinical data, clinical data, genetic evidence, epidemiologic data that all speak to the central role that inflammation plays in driving neurodegeneration and progression in these conditions. And so a brain-penetrant inhibitor of TYK2 and JAK1 is ideally designed to target three of the key cellular drivers of neuroinflammation, including abnormal microglia via type 1 and type 2 interferon signaling, abnormal astrocyte function via type 1 and type 2 interferon signaling, and abnormal lymphocyte function, particularly TH17 T cells, which have been repeatedly demonstrated to be important components of inflammation in these disorders. This is kind of a distilled visualization of what we're hoping to accomplish with this mechanism. So in most of these diseases, you have a number of factors that combine to trigger neuroinflammation. And that neuroinflammation leads to neurodegeneration. That neurodegeneration, then, it kind of creates a loop because that neurodegeneration leads to neuroinflammation. It becomes a self-perpetuating cycle. So particularly with a brain-penetrant inhibitor of TYK2 and JAK1, we believe we can break this cycle and have the greatest potential to slow and even halt disease progression in these conditions. So this is a table. This is a partial list of some of the approved JAK inhibitors for peripheral autoimmune conditions, not an exhaustive list by any means. But really, what I want to use this slide to point out is that although these drugs that are not specifically selective for TYK2 and JAK1 are highly effective for really a long list of peripheral autoimmune conditions, they are not without safety concerns. So most of the drugs in this class have an FDA boxed warning for major adverse cardiac events, malignancy, thrombosis, and serious infections. What I do want to draw your attention to is the last two rows of the approved agents table. Abrasitinib, which is a relatively recently approved drug for atopic dermatitis, is a selective JAK1 inhibitor. FDA, although they tend to be conservative in how they approach labeling in these cases, and I can say that from experience, if you look at the data from the development program, it is fundamentally different than the agents that target JAK2 or JAK3 in terms of its safety profile. Additionally, Deucravacitinib, which is a relatively recently approved drug for moderate to severe plaque psoriasis, which is highly selective for TYK2, does not have a boxed warning. The point here really is that for BHV-8000, which would be selective for TYK2 and JAK1 at the doses that we plan to bring forward to the clinic, we anticipate a safety profile that would be much more aligned with abrocitinib and deucravacitinib relative to drugs that target JAK2 and/or JAK3. Okay? Then I can provide a quick update on our phase one study. We've completed dosing in three single ascending dose cohorts. These are in healthy volunteers, of course, 10 mg, 20 mg, and 30 mg in three multiple ascending dose cohorts for 14 days of 6 mg, 10 mg, and 20 mg. From a safety standpoint, drug's been very safe and well tolerated. There are no SAEs or severe AEs, only a handful of very mild AEs that all resolve spontaneously. No abnormalities in terms of ECG or vital signs, and no adverse laboratory trends related to drug. Then we also wanted to use this study to try to at least begin to understand the pharmacodynamic activity of the drug. Now, keep in mind, these are healthy volunteer patients who are not in a de facto inflammatory state. However, even given that caveat, these two markers, which are well-established markers of inflammation, so high sensitivity C-reactive protein, which is a known inflammatory marker, as well as interferon beta, which is a highly pro-inflammatory cytokine, we saw significant reductions in these markers over just a 2-week period in healthy volunteer subjects. This is further evidence that this drug has a potent immune suppression or anti-inflammatory suppression effect over this period, even in healthy volunteers. We're going to continue to look at additional pharmacodynamic data to inform dosing for our clinical programs. I'm going to stop there, and I'll turn it over to Dr. Lemere to speak about it. All right. Thank you very much. I would like to thank Biohaven for the opportunity to speak here today. It's been very, very exciting and interesting. I will move forward. I will be talking about anti-amyloid immunotherapy-related ARIA. And ARIA is amyloid-related imaging abnormalities. But first, I just wanted to go over what's out there right now. In June 2021, the FDA granted accelerated approval to let me see if I can get this to work. Okay. To Aducanumab or Aduhelm, the trade name. And this was a drug from Biogen and Neuroimmune. And this antibody recognizes fibrillar beta amyloid. This was quite a controversial approval based on different outcomes in two phase three clinical trials. However, in January 2023, the FDA gave accelerated approval to Lecanemab, which is from Eisai and Biogen. And this antibody recognizes protofibrillar beta amyloid. This is a sort of a prefibrillar form of beta amyloid. This antibody received full approval in July of 2023, so almost a year ago. This is now rolling out in the clinics. It's available at our hospital. It's an important antibody that is still on the market right now and really moving forward quickly. I should mention here that Biogen did pull Aduhelm from the market. Eli Lilly has an antibody called Donanumab. This is an antibody that specifically recognizes a pathogenic form of beta amyloid. It's called pyroglutamate 3 beta amyloid. It's an N-terminally truncated and modified form that's highly resistant to degradation and is present at least if you look at unfixed tissue; it's present in all plaques and vascular amyloid. This antibody is actually going before the FDA Advisory Committee on June 10th. What is ARIA? As I mentioned, ARIA stands for amyloid-related imaging abnormality. This is something that's detected by MRI. This typically shows up as ARIA-E, which stands for vasogenic edema and effusion, or ARIA-H, which is typically microhemorrhage and superficial siderosis. These are shown on the right side here. This is ARIA-E, and on the bottom is ARIA-H. These types of events can occur as part of the natural history of Alzheimer's disease, especially if vascular amyloid is present. These are definitely increased with amyloid-modifying therapies, in particular antibodies. ARIA occurs very early in the initiation of anti-amyloid immunotherapy, typically after 1-3 doses of treatment. While most ARIA-E is symptomatic and transient, these effects can be severe and life-threatening. So this is a table showing the ARIA risk rate for each of the three antibodies that I mentioned, Aducanumab, Lecanemab, and Donanumab. Overall, if you look on the second column there, overall, you can see that the incidence or the risk rate of ARIA with Aducanumab was about 35% compared to 2.7% in placebos. With Lecanemab, it was 12.6%, so substantially lower with 1.7% in the placebos. Lastly, Donanumab showed an ARIA rate of about 24%, so somewhere in between. In placebos, there was 18%. One of the things that became very clear in these clinical trials is that ApoE4 allele carriers are at increased risk for ARIA. ApoE4 is a risk factor in general for Alzheimer's disease, and it typically results in more amyloid deposition a bit earlier and, in particular, a lot of vascular amyloid. So the risk of ARIA can be complicated. It can definitely affect the benefit-risk assessment of the anti-amyloid antibodies. At this point, the anti-amyloid antibodies are the only disease-modifying treatment for Alzheimer's disease. As I mentioned, there can be severe ARIA and even fatal ARIA. This is a report of a woman at Northeastern who's sorry, she was treated at Northwestern University. She was part of a clinical trial there, the Lecanemab Clarity phase three clinical trial. She was a 65-year-old female, early AD stage, and she was homozygous for ApoE4. Prior to treatment, she had MRI. She had no microhemorrhages and no edema. She was on the placebo group. However, during the open label extension, she went on to the drug. After three doses and Lecanemab, I should mention, most of the antibodies are dosed once a month. Lecanemab is every other week. These are IV infusions. Four days after her third dose, she had stroke-like symptoms, including aphasia and seizures. And her family took her to the emergency room, and they thought she was having a stroke. And so they treated her with TPA. And unfortunately, she succumbed to this massive hemorrhaging event. So you can see the MRI in the upper left; you can see that there's a lot of microhemorrhages. In C, you can see that even macroscopically, you can see these hemorrhages. But I think what's interesting is that when you look down here, you can see that the amyloid she had a lot of vascular amyloid, and it looks very fragmented in the blood vessel wall. Keep in mind, she was only treated with three doses. But importantly, she had a tremendous amount of this cellular infiltration. These are immune cells coming up to the blood vessel wall. And so there was a big inflammatory response. So ARIA is a mixed inflammatory response to vascular amyloid. And on the lower left there, this is a figure I like to show because it shows a normal artery and a vein. And in a normal artery, which includes the leptomeninges and the large penetrating vessels, which is where most of the vascular amyloid in the Alzheimer's brain occurs, you can see that those arteries have the lumen and then I'm not going to point with this the lumen and then the endothelial cell layer, a smooth muscle cell layer, and then collagen, and then perivascular space before you see the end feet of the astrocytes that form the barrier. And if you look down below, you'll see that the amyloid actually intercalates right within that blood vessel wall, and it can go all the way out to the lumen. So it's not just sitting on the outside of the collagen, for example. It's actually going right in. And you can see degradation of both smooth muscle cells and the endothelial cell layer. However, in the smaller veins, the capillaries, for instance, there's no smooth muscle cell layer. The collagen layer is thicker, and the amyloid is really on the outside of the collagen layer. So there's no direct access from the blood. So we and others have now shown that these anti-amyloid antibodies, at least in mouse models, seem to bind directly to vascular amyloid deposits and that they seem to be initiating this inflammatory event in the perivascular space. And this is a mixed immune response. So it involves both microglia and astrocytes that are brain-resident immune cells, as well as peripheral immune cells. So we're seeing recruitment of perivascular macrophages, monocytes. Others have shown T cells in the space. And so it's a very localized immune response. So in general, when people have been talking about potential mechanisms for ARIA, the initial idea was that these antibodies are getting in, they're binding to the plaque, they're dissolving the plaque, and then as that dissolved A beta starts floating around, it's getting bound to ApoE, and then it gets carried to the vasculature, induces this immune response at the blood vessel wall, and then some of this amyloid's getting stuck in the blood vessel wall. However, we have been working on a different hypothesis because we were really curious. I have a whole program on complement in my lab. So part of the complement system's role is to remove immune complexes. The way that happens is antibodies bind to their antigen, and then C1Q comes in, and there's a C1Q binding site on the FC region of antibodies. So C1Q comes in, that activates the proteases C1R and C1S, and that then starts the whole classical complement pathway. The three main roles of that pathway are phagocytosis of this immune complex in particular, anaphylatoxin effects of C3A and C5A binding their receptors, and that causes a huge recruitment of immune cells to that site of complement activation. Then finally, if the complement cascade goes all the way down to the end, you get the formation of the membrane attack complex, which pokes holes and punches holes in cellular membranes and kills cells. And so at least in mice, we've seen all three levels showing up in the blood vessels where we see ARIA, where we see microhemorrhages, where we see binding of A beta antibodies to the vascular amyloid. But what's really interesting is that C3a and C5a can actually activate the JAK-STAT pathway. And as I mentioned before, there's activation of both microglia and astrocytes in the CNS, as well as this local activation of peripheral immune cells all the way out in the periphery that come in and are right at that blood-brain barrier junction. And we think that this whole event is what's leading to the effusion and edema, and that eventually that may become as the blood vessel wall becomes destabilized, may lead to the microhemorrhages. So corticosteroids and other immunosuppressive drugs have been effective in treating ARIA thus far after it occurs. This is a paper from Eli Lilly late last year showing that in old mice, you have to wait until these mice develop vascular amyloid, which often takes a long time. So in this particular mouse model, they were looking at 23-26 months of age. These mice have a lot of vascular amyloid. And so they injected them with biotinylated 3D6. 3D6 is a murine precursor of bapineuzumab, which was the first antibody to induce ARIA in humans. And it very reliably induces ARIA in mice with vascular amyloid. And so what they showed was that this on the bottom there, you can see right here, you can see that in green, that's the biotinylated antibody binding to the vascular amyloid shown in blue. And then in turquoise, that's the staining of CD169, which labels perivascular macrophages. There are a whole slew of other cell types coming in there. They also show that peripheral monocytes are coming in and migrating to that site where the antibody's binding. This is shown primarily in the leptomeningeal vessels and the large penetrating vessels that come down from the pial surface. So it looks like this TYK2/JAK1 inhibition can block both central and peripheral cellular signaling, and maybe that would be effective in preventing ARIA pathogenesis. Why might that be? Well, because this inhibitor will work on both the central CNS glial cells, the microglia, which express interferon gamma type one and type two molecules, as well as astrocytes. Both of these are very intimately associated with the blood vessels that are containing amyloid, and especially in the case of anti-amyloid immunotherapy. But then also, it would work on the lymphocytes and other leukocytes. As Dr. Kozauer showed, really focusing on TH17 T cells and IL23 and IL17 signaling. So with that, I will throw it back over to you, Dr. Kozauer. Really good. Thank you. All right. Thank you so much, Dr. Lemere. So I'm going to build on what Dr. Lemere just presented and provide an overview of our planned phase 2,3 clinical trial in ARIA. As I mentioned earlier, we met with FDA about this program. They're very supportive, and they have also given us very positive feedback about this design. So the population would be patients with early Alzheimer's disease who are eligible to begin treatment with a monoclonal antibody towards amyloid. And they would be ApoE4 carriers, so either heterozygotes or homozygotes. And as you just heard, those are the patients who are at greatest risk of developing ARIA. Participants would be then randomized to one of three treatment arms, either high-dose BHV-8000, low-dose BHV-8000, or placebo. Participants in each of those arms would have a one-week period with daily dosing of BHV-8000 or placebo, which would be a priming week. They would then continue on the BHV-8000 or placebo for 12 additional weeks, but at which point they would be started on an anti-amyloid monoclonal antibody according to the recommended dosing for that particular agent. And again, that combination would be then dosed for 12 weeks. That duration is selected because that is, as the data show, the period of highest risk for developing ARIA once you initiate treatment. The primary endpoint would be rates of ARIA-E during the double-blind period, but a number of different imaging endpoints, clinical endpoints, as well as inflammatory and Alzheimer's-specific biomarkers would also be assessed. Following the double-blind period, BHV-8000 or placebo dosing would be stopped, but participants would continue on their anti-amyloid regimen for an open label extension period. The idea there would be to ensure that any reduced rates of ARIA that you see during the double-blind period were maintained low after the discontinuation of BHV-8000. We believe that this is a really efficient way to understand the impact of BHV-8000 on preventing ARIA. Again, we've gotten very good feedback from FDA. We look forward to working with them on this program. With that, I'm going to shift gears to Parkinson's disease. I'll start with a brief overview. Parkinson's is very common. About 1 million people in the US living with Parkinson's disease. 10 million worldwide. It is the second most common neurodegenerative disease in the US, second only to Alzheimer's. Although there are a number of symptomatic drugs that are approved, there are no available disease-modifying drugs for Parkinson's. For the purpose of this presentation, I'm not going to go through this graphic in a lot of granular detail, but the point really is that Parkinson's is a relentlessly progressive neurodegenerative condition. It is defined by the abnormal deposition of alpha-synuclein. And that the disease itself, when the alpha-synuclein is accruing, has a very long pre-diagnostic period where there can actually be signs or symptoms, things like hyposmia or REM sleep behavior disorder that result from alpha-synuclein deposition, but is typically not recognized as Parkinson's until you begin to have motor symptoms and signs. So tremor, rigidity, bradykinesia. That's when Parkinson's is diagnosed. Those motor signs and symptoms will progressively worsen over the course of the disease and have greater and greater impacts on daily functioning. Other motor signs and symptoms can occur later in the disease as well, like freezing, falls, dystonia, etc. Most Parkinson's patients will also have non-motor manifestations of their disease, neuropsychiatric symptoms, or other body systems being affected. Those tend to be most severe in the more advanced stages, but can occur really at any stage of the condition. The other wrinkle with Parkinson's is that patients at some point typically start levodopa or other dopaminergic therapy in the early stages. Most patients do pretty well at first, but then the medications themselves start to become less effective over time and have their own complications like wearing-off periods, dyskinesias, etc. For our development program, we are focusing on patients with early clinically diagnosed Parkinson's disease who have not yet started on dopaminergic therapy. And the idea there really is, as I mentioned earlier, with our mechanism, we're trying to stop inflammation and slow the neurodegeneration. We want to intervene at a stage of the disease, at least in clinically diagnosed patients at first, who have the largest amount of neurons to protect. And that would be the earliest stages before dopaminergic therapy is started. So I'm also not going to go through every point on this slide. And there are a lot of familiar themes with what you just heard from Dr. Lamara. So the point really here is that, again, Parkinson's is driven by abnormal alpha-synuclein deposition in the dopaminergic neurons in the brain, as well as in the enteric nervous system and the gut. That alpha-synuclein is recognized as foreign and triggers an abnormal immune response involving both the innate and adaptive immune system, both in the periphery and in the central nervous system. And you can see even at a glance, and again, this mirrors what Dr. Lamara presented, the same cell types that I mentioned in the earlier slide as driving neuroinflammation in these conditions are the same cells that are depicted in this graphic. And so the point really is that a designed TYK2 JAK1 inhibitor is ideally positioned to target both abnormal innate and adaptive immune activity, both peripherally and centrally. We believe that all of those components are necessary to have the maximum potential for benefit in Parkinson's. So this is just kind of a biopsy slide of some of the different sources of clinical data that speak to the role of neuroinflammation in PD. So for example, there are post-mortem data that show higher levels of inflammatory cells in the brains of patients with Parkinson's, like reactive microglia. There are a number of in vivo imaging tools now, like TSPO PET imaging, which is a PET ligand for microglial activation, so a surrogate for inflammation. And this is a relatively recent study in early PD patients relative to controls that shows higher TSPO signal in areas of the brain that are known to be critical to PD, like the putamen and the substantia nigra. And then there have been a number of studies - this is a recent meta-analysis - that have looked at levels of cytokines, pro-inflammatory cytokines in the blood and the CSF of Parkinson's patients relative to controls. And really, almost generally across the board, you see higher levels of these pro-inflammatory cytokines in Parkinson's relative to controls. When we were considering this mechanism, we wanted to think about all the different sources of data we could look at to support this hypothesis. And so one of the things we did is we commissioned a very large healthcare claims database analysis using the Komodo Health Claims Database. There are over 300 million patients in this database over a slightly greater than 8-year period. And what we looked at was patients who had a range of different autoimmune conditions: rheumatoid arthritis, inflammatory bowel disease, ankylosing spondylitis, and others. A very long list. And looked at in those patients over this time period, patients who were exposed to either an anti-TNF agent or an anti-IL-17 agent, which in essence mimic much of what a TYK2 JAK1 inhibitor does, what is their incidence rate of Parkinson's disease relative to the untreated controls. What you see is that in both cohorts of patients that are exposed either to an anti-TNF agent or an anti-IL-17 agent or both, you see highly statistically significantly lower rates of Parkinson's over that time period relative to the untreated patients. Then my last slide on Parkinson's, I want to provide an overview of our planned phase 2/3 clinical trial. As I mentioned earlier, we are planning to target early clinically diagnosed Parkinson's patients who have not yet started on dopaminergic therapy. Right now, we're planning for 2 active doses of BHV-8000 relative to placebo and with a 48-week double-blind period, and then rolling over to an open-label extension. The 2 things I really wanted to highlight is that FDA has this requirement that a clinical endpoint in Parkinson's trials demonstrates a functionally meaningful benefit to patients, so how they're functioning in their daily lives. The issue is that the tools we have available to do that are very limited for the purpose of a clinical trial. And so for an example, one of the scales that FDA often refers to is the Movement Disorders Society Unified Parkinson's Disease Rating Scale, or the MDS-UPDRS Part II, which is a functional scale. The issue is that scale moves very slowly in early disease, and it's very hard to show a treatment effect. And to even attempt to do that, you need very, very large sample sizes that become unwieldy. So we wanted to really understand how we could address this issue. So we spent a lot of time looking at registry data from the Parkinson's Progression Markers Initiative, or PPMI, which is a well-established Parkinson's registry, as well as placebo arm data from a number of Parkinson's trials that we have access to through our membership in the Critical Path Institute. Based on our analyses of those data, we are convinced, and we have determined that if you find an endpoint that's at time to a clinically meaningful confirmed worsening on the MDS-UPDRS Part II, you can show a clinically meaningful functional benefit, which is what FDA wants. You can do it in a clinical trial that requires 300-ish less patients per study than trying to even show a mean change difference on the scale. So we think this really has potential to add a tremendous amount of efficiency to an early disease-modifying trial in Parkinson's. We've met with FDA. This Part II position they have has been very longstanding. We met with FDA, though, and they were very supportive and encouraging about our endpoint. We are very excited about that and really think it'll add a lot to the design of this study. Finally, I just want to end by noting another endpoint that's going to be an important secondary endpoint in our study. It's something called the Parkinson's Disease Composite Score, or the PARCOMS. This is a scale that we are developing. It's based on established methodology that FDA is very familiar with that was used to develop the Alzheimer's Disease Composite Score, the ADCOMS, which is used in the lecanemab development program. We also developed the scale leveraging PPMI data and placebo arm trial data. And really, in a nutshell, the approach is the development of this composite uses statistical methodology to look at commonly used scales in a particular neurodegenerative disease, so in this case, Parkinson's, and select for the items that are most responsive to change over the period of time you're interested in. And then you create a composite based on those items. And so you end up with a composite that's much more highly responsive to change in that population relative to the contributing original scales themselves. And so we believe that this will be an important supportive secondary endpoint in our study as well. So with that, I will turn it over to Dr. Lara. Thank you. Good afternoon. Multiple sclerosis, or MS, is a chronic inflammatory demyelinating neurodegenerative disorder of the central nervous system. Myelin is the fatty substance that covers the nerve fibers. It actually helps impulses be transmitted quickly and efficiently. When that myelin gets damaged, demyelination, it actually slows down the impulses or stops them altogether. That leads to the devastating neurological symptoms that people with MS experience: weakness, bowel and bladder problems, problems with vision, cognitive problems, tingling, numbness, pain, difficulties walking. There are devastating effects at the age of onset, 20 to 40 years old typically. These are people that have young children in the midst of their careers, maybe taking care of older family members. It affects women about 2-3 times more so than men. There are relapsing forms of multiple sclerosis, or RMS, and progressive forms of multiple sclerosis, or PMS. With the relapsing forms, you have the symptoms. There may be a period of improvement, remission, and then relapses again. It kind of cycles through. With the progressive forms, you have continued and progressive disability without remission. There's a greater impairment in function, greater issues with quality of life, and a higher burden with the progressive forms of MS. About 15% of those living with MS have what's called primary progressive MS, or PPMS, where the disability accumulates from the beginning. This is the area of highest unmet need, these progressive forms of MS. In fact, there's only one drug that's currently approved to treat PPMS, and it doesn't cross the blood-brain barrier. There's increasing evidence that shows that with progressive forms of MS, it's really important to target locally in the central compartment, the local cells that cause that inflammation. That's why we're really excited about our TYK2/JAK1 inhibitor, BHV-8000. It's a potential treatment for multiple sclerosis: selective, potent, and as you've heard, brain penetrant, acting with a different mechanism than the currently approved therapies out there. There is genetic, non-clinical, and clinical data to support the rationale for the use of a TYK2/JAK1 inhibitor in MS. There was a recent study that showed that there's a protective genetic variation in the TYK2 gene. So in response to interleukin-12, interleukin-23, there was actually decreased signaling capacity. So this reduced the function of TYK2 and resulted in a reduction in risk for developing MS. In non-clinical models, typically what is used is an EAE model. This is an experimental autoimmune encephalomyelitis model. It resembles MS. And it's important to have the TH1 and TH17 cells in order to develop this EAE. A JAK1/2 inhibitor, baricitinib, was used in the EAE model. The mice were treated with this, and the mice that were treated with this JAK1/2 inhibitor actually had delayed onset of EAE, reduced time of symptoms, and overall lower symptom burden. You can look in the bottom left-hand panel there. The control bar is the black bar, and the blue and sort of brownish color were those treated with the JAK1/2 inhibitor. Here, lower is better, so you can see better clinical scores in those that were treated with the JAK1/2 inhibitor. If you look in the middle panel from a pathologic tissue injury perspective, on the left-hand side, the control, you can see at the tip of the arrowhead there, the darker blue color and the paler color on top, those are myelin stains. There was tissue injury, demyelination, in those that were untreated. In the treated mice with the EAE, the ones treated with the JAK12 inhibitor in the middle and right-hand sides there, specifically the higher dose on the right, you can see that there was much more robust tissue there, so less tissue injury. Then from a clinical perspective, data supports that there's presence of abnormal immune activation in MS patients. If you look at the right-hand side here, these are MRI scans looking through the brain horizontally. Those white spots that you see on the left-hand side there, that's abnormal. Those are areas of demyelination. That's disrupted myelin tissue, and those often not always, but often correlate to the symptoms. Even when patients aren't having symptoms, those areas of abnormality can accumulate and then cause symptoms and dysfunction later on. So in this study, Secukinumab, which is an IL-17A inhibitor, was used to treat these patients for five months. And you can see on the right-hand side there that there was a lot less of that white area, that there was a regression in some of those lesions. So the old way of treating MS was for many years, treat the relapses, reduce the relapses. And the thought was, if you can do that, then you can really be effective. And there are a lot of therapies available that are lower potency that do treat the relapses, but it's not enough. We know that there's a high unmet need, especially in the progressive forms of MS. And the field has moved to, how can we do better? So the new paradigm is no evidence of disease activity, or NEDA. Stopping relapses, it's not enough. We need to stop the relapses, but also stop the disability progression, stop the quality of life deterioration, the cognitive impairment. Over time, the brain actually can atrophy, especially in the progressive forms of MS, or shrink. Can we stop the brain atrophy as well? As I mentioned, there's only one therapy that's approved to treat PPMS. It does not cross the blood-brain barrier. Everyone recognizes this high unmet need. There are a lot of folks out there studying other types of medicine. You might have heard BTK inhibitors, really promising, but then there seem to be some safety and tolerability issues. The field recognizes this unmet need, but we need to be able to work in a different way than the current therapies are and potentially cross the blood-brain barrier to achieve success in this area. Our selective potent brain penetrant, TYK2 JAK1 inhibitor, or BHV-8000, we hope to target the pathophysiology of MS, modulating pathogenic microglia cells, which we think are responsible for the cognitive and physical disabilities that occur with MS, and also through inhibiting the TH1 and TH17 cells, lymphocytes, which drive the neuroinflammation. So with this, we hope that our BHV-8000 could offer a novel therapeutic approach for not only the relapsing forms of MS, but also the progressive forms of MS. Here we show a phase 2 proof of concept study for adults with relapsing MS for our BHV-8000 program. It's an imaging study which is very typical in the phase 2. As we talked about before, you saw those MRI scans. You can see very well the lesions that accumulate in MS in imaging studies. That's the gold standard of looking at the brain tissue in MS patients. What you see in MRI scans is often a very good predictor of long-term what you might expect to see clinically in longer-term studies. So here we're proposing a study with a screening period of 8 weeks and then a 12-week treatment phase with high-dose BHV-8000, low-dose BHV-8000, and then placebo. And looking at imaging endpoints, looking at those MRI scans and the lesions on those MRI scans, and then an open-label extension as well to look longer-term. So while we have sort of this typical imaging study that you might expect to see in phase 2, we're also being very innovative in our approach, talking to our thought leaders. And we're going to be looking at other potential imaging in the central compartment over the longer term for effects that we might expect to see in progressive forms of MS as well. In that way, setting up a strong foundation for hopefully bringing BHV-8000, our selective potent brain penetrant, TYK2 JAK1 inhibitor, as a novel therapeutic approach for not only relapsing, but also progressive forms of MS. Thank you. Great. Thank you, everyone. I'm going to invite Charles Duncan from Cantor to come and moderate this panel. I think we have a lot of folks up here. And Charles, I know you probably have a lot of questions across these different indications. I think we can take the full 30 minutes for your section, and we'll make up a little time later. Thank you. Thank you, Irfan. Thank you, Vlad and team, the Biohaven team, for inviting me to participate. I appreciate this. Last year was a pretty interesting year where I saw a much broader Biohaven than I had anticipated. And this year, I'm equally impressed, or more so. It seems like you've really put some meat behind some of the clinical trial plans that you have. So because of that, it's good that you've actually given me an hour and a half to do this panel because I have a lot of questions. We'll stay here until 6:30. Sure. Super. Yeah. So yeah, this is an area of part of Biohaven that I hadn't done a lot of diligence in and frankly hadn't included in our model. So it gave me an opportunity to do a fair amount of work in preparation for this. Now, we have many good panelists here to talk, but unfortunately, we didn't actually hear from Doctors Albers and Salloway. So I thought maybe what we do is start with kind of an overview of why they're involved in this panel and what they think about the opportunity set with this particular platform. So, Dr. Albers, can you give us a sense of your perspective on this basic mechanism and the importance of neuroinflammation? Certainly. Good afternoon, everyone. I'm particularly excited about TYK2 as a target for neurodegenerative diseases. In the slides, I agreed with everything on the slides, but I would add another cell type, and that's neurons. I think that the role of TYK2 in mediating inflammatory signaling within neurons themselves is very important. And by inhibiting TYK2, then not only can you block inflammation, but also block neuronal death. In my lab, we've been able to identify one of the root causes of inflammation that accumulates within neurons. It's cytoplasmic double-stranded RNA. We see that associated with TDP-43 pathology. And if we then treat neurons, human neurons, we get inflammation and cell death. We were fortunate to be able to use 8000 in our lab and show that that compound is potent at blocking inflammation and blocking neuronal death in cell-based models. We're very excited about the idea of moving into animal models to test the role of these compounds to block neurodegeneration. So I think my view is a little bit more forward-thinking, that there's disease-modifying potential in ALS, where we see TDP-43 in 97% of patients. In Alzheimer's disease, where about half the patients have TDP-43 pathology. Next week, we're presenting data at the Keystone meeting that implicates TYK2 in Alzheimer's disease. We have a machine-learning-based platform that shows that targeting this pathway is specific for it shows signals in the form of Alzheimer's disease where there is evidence of TDP-43 pathology. That's through the expression of cryptic exons. So, I'll stop there and happy to address further. So I have one more question before we go on to Doctor Salloway because I wanted to ask you specifically, when you think about the target product profile of 8000 and the putative mechanism, how would you compare that to some of the other agents that are TYK2, JAK1 inhibitors, or dual inhibitors, or other, call it JAKinibs, JAK inhibitors? How do you feel about this mechanism? So we're engaged right now in so just in general, I think Nick summarized it very well, the safety concerns in JAK inhibitors with JAK inhibitors. And so I think that the safety profile is quite exciting. And I would amplify that. Not only do the SNPs that reduce TYK2 inflammation, picking up on Lindsay's talk, not only do they reduce the risk of MS, but they also were not associated with increased risk of infection, fungal infections, tuberculosis. So that's different than the JAK inhibitors. So there's, I think, both a safety as well as efficacy positivity there. In terms of other TYK2, JAK1 combination inhibitors, I'm not aware of those as I think this is a relatively unique molecule in the way that it's acting. So I think that's why the presentation was like, let's take two different things and look at them in combination. There's just not that many molecules that are able to accomplish what this special molecule can do. Certainly, the brain penetrance is a clear differentiator. And we can talk about some of the others as we get into this. By the way, if anyone has any questions, just yell out my name and I'll try to grab it. I want to make this interactive. Doctor Salloway, you and Doctor Lemere know a little bit about treating Alzheimer's patients. So I just wanted to ask you, the company did a great job overviewing the need for treating ARIA, as did Doctor Lemere. And I'm wondering if you could provide a little bit of perspective on this mechanism and the unmet need for treating ARIA. Sure. Well, I want to thank Biohaven. And I trained here in New Haven. It's always great to be back and to see old friends. Well, I think we're at a turning point in Alzheimer's treatment. There really hasn't been anything new available. Obviously, it's a very large unmet need. I don't need to make that case. There was really nothing new approved for 20 years. Now we have the first disease-modifying treatments that target a key component of the pathology of Alzheimer's disease. Now, they have modest benefit, but there is a benefit, fortunately. It took a while, many years of testing, to demonstrate that there is a clinical benefit. There also is a significant risk. You've heard about ARIA from Cindy and others. Usually, ARIA is transient and asymptomatic, but it can be serious and it can be fatal. It really factors in in the risk-benefit equation for each patient. It would certainly put me at ease if we had a way to mitigate the serious component of it. I don't know if this strategy will work. This is something novel. Certainly, as Cindy laid out, there is a key inflammatory component. Cerebral amyloid, APOE is the biggest risk factor for ARIA, following with cerebral amyloid angiopathy and drug. There is clearly an inflammatory component to the serious cases. If we can mitigate that, that would be fantastic. That's number one. Number two, I think this is a reasonable drug to try. Then the second component, which this company didn't get into, is the whole role of inflammation in Alzheimer's disease. Could there be an independent therapeutic benefit for that? Again, TBD. Certainly, targeting inflammation is a good idea. That last point is one that I want to come back to in terms of a treatment paradigm. With regard to, call it a combination treatment, Dr. Salloway, are you treating patients with lecanemab? What kind of percent change in ARIA risk would you like to see that you think is clinically meaningful, or any positive change, especially in those patients who are either homozygotes or heterozygotes, would be positive? Right. Well, I think it's wise to target APOE4 carriers because the rate of ARIA is so much higher in that group. I mean, aducanumab is no longer available, but for the homozygotes, it had a rate of 66% of ARIA. With aducanumab, it's like 43% with donanumab. It's one-third with lecanemab. So it's really common. Even lecanemab has the lowest rate. But I want to make the point. It's really the serious cases that matter. Because if all of it was mild and transient, clinicians can deal with that. But if it's going to be serious, requiring hospitalization, presenting like a stroke, which it can be a stroke mimic, getting thrombolytics, Cindy showed you that case. That's not the only case. There are other thrombolytic cases with a disastrous outcome. That's not acceptable. So especially with the degree of clinical benefit from the drug. So those are the ones we really need to reduce. Doctor Lemere, do you have any perspective on a percent change that would be clinically meaningful in those different patient cohorts? What would you like to see? I'd like to see no ARIA. That's what I'd like to see. And I think if you could we get there with this mechanism? I think potentially. Yeah, I do. I think the evidence for the amount of immune cell recruitment, both in the CNS and in the periphery, is just so strong now that this is a local inflammatory response. So I think if we can sort of tamp that down a bit, then I think we have a very strong chance. And if that works, then I would say there is a different form of a similar type of event called CAA-RI, or CAA is vascular amyloid. RI is related inflammation. So there are patients that don't necessarily have Alzheimer's disease that have vascular amyloid. And they have these episodes where they have a very localized immune response, inflammatory response. And by the way, there's evidence now that those are actually induced by binding to autoantibodies. So it's much more of an autoimmune type response and disease. When you think about this mechanism, do you feel like the target product profile is well served by this mechanism? And is there anything in particular that you would point to in this area that gives you confidence in that mechanism? Yeah, I would say that the fact that this drug, first of all, is brain penetrant, I think is really key. But also that it's addressing a mixed inflammatory response. So there's CNS cells as well as peripheral immune cells that are involved. And so it would be very helpful on both of those ends. So I do think it would be just asking to pick you up on a question. Yes, please. Dr. Lemere, why did we see the increased incidence of ARIA within the first 12 weeks, right? And after that, the incidence simply goes down. And that question relates to, obviously, we're focused on trying to prevent it in the highest risk period of time. Why doesn't it happen at the same incidence later after treatment? This is all hypothetical. But my feeling is that the antibodies, the reason I was showing—oops, sorry. Thank you. The reason I showed that picture of the arteries and the pathological arteries is to make the point. I've given many talks where people said, "There's no way those antibodies are going to bind to vascular amyloid." And I would say that that's absolutely untrue. And that's why I like that figure because it shows the Aβ breaking up the blood vessel wall and really totally intermeshed within the blood vessel wall. And at least in our mouse models, we can see the antibodies actually binding. Lily Xavier Taylor's paper showed that the antibodies combine directly to the vascular amyloid. We're actually doing a time study right now to try to figure out how many doses do you need. Right now, I can say certainly after three-weekly doses in mice, we can see it binding. So I think that what's happening is you're giving this large bolus of antibody. The antibody, the first thing it's going to see is this vascular amyloid that happens to be in the leptomeningeal vessels and in the large penetrating arterioles from the pial surface. So it's binding to the beta amyloid there. And then I think it's activating complement, but we don't know. But it's definitely setting up this very local immune response. And then once that's moved in the first 3 months or so, you just allow some risk for it later in the year? Right. I think we are probably removing vascular amyloid. But then you're going to have more vascular amyloid perhaps accruing as you get more antibody in. I think one of the interesting questions is, if you have more ARIA, do you actually get faster clearance of plaque amyloid? So that's another area of investigation right now. But there's also a huge upregulation of matrix metalloproteinases, which are meant to remodel the vascular wall. And so I think it's this ongoing, remove the stuff, try to fix the blood vessel wall. Thank you. Sure. That's helpful. Sir, go ahead. I mean, jumping off Dr. Salloway's comment about the difference between sort of asymptomatic ARIA and posterior risk ARIA. Doctor Kozauer, have you discussed with the FDA in terms of looking at pivotal approval? Can you get approval on just an overall ARIA-E, or do you actually have to show reductions for serious ARIA-E? It's a great question. I'm actually going to let Doctor Ackerman, who's leading the ARIA program, respond. Yeah, thanks for the question. So our discussion with the FDA did include preliminary discussions around endpoints. Our focus was around statistically significant reductions in ARIA-E at that 12-week time point. This was acceptable in our discussions with them. Maybe just a follow-up as to Doctor Lemere and Salloway. Do you need to see sort of statistically significant reductions in severe ARIA for utilization, or would you assume severe ARIA would be reduced if you saw statistically significant reductions in overall ARIA-E? Well, I mean, to me, I'm not working for the company. So these are early days. And so this would be really a proof-of-concept study, even though it's billed as a phase 2/3. So can you reduce the rate of ARIA? Can you reduce the rate of serious ARIA? I would give a suggestion to the company that there are imaging programs that use machine learning that can measure ARIA, at least so far, pretty reliably. So you could get a clear quantitative readout of how you're influencing in the treatment versus the placebo group and not just wait for the clinical event or you actually could measure the clinical event. So I think it would probably be a great to answer your question, a graded response. I think the first trial will give you an idea. Can you get a signal that you're reducing ARIA? And does that eventually lead to a lower rate of clinically significant reduction? Great, great question. Nice to hear you just add a comment to that too. So the nice thing about the study as well is that, like Dr. Ackerman said, it is a study that in design, we're absolutely going to learn from. But it also is designed in a way that FDA would find acceptable if we show an effect. So we're also going to learn that it's a proof of concept in a study that can be capable of supporting approval. So this could prove to be one of two pivotal studies? Okay. And then Peter or Nick, just you mentioned this study is you plan to do this study. When do you plan to kick it off? And I guess maybe to Dr. Salloway, do you think that there will be interest by the patient community in adding another drug on top? And I guess one other question. Are you going to do this study only with Lecanemab or perhaps with Donanumab if it's approved? Yeah, a couple of really good questions there. So I'll take the last one first. It was the question around Lecanemab versus Donanumab. Our intention in the initial trial is to concentrate on a single anti-amyloid agent, and it would be an approved anti-amyloid agent. But in the broader program, we intend to include all anti-amyloid therapies that are approved by FDA. We believe that if we can demonstrate benefit against any one anti-amyloid therapy, we will be able to against other anti-amyloid therapies. In terms of the start time, it's a discussion that's ongoing internally, and we haven't really made that public yet. And I'm not ready to do that here. Okay. Matt? This came in on the QR code. Oh, I don't know how to work that. BHV-8000 contains a JAK1 inhibitor. And given abrocitinib as a selective JAK1 as a black box, does BHV-8000 have similar safety liabilities from its JAK1 component? What was the SAD method format? Sure. So I can answer that. I mean, that is sort of along the lines of that table I presented in the presentation where it does have the black box. But the agency, as I mentioned, they tend to be very conservative in how they approach that. But if you look at the data from the program, it is definitely a favorable safety profile relative to the JAK inhibitors of target JAK2 or JAK3. And for the data we have in our SAD method, it was very, very well tolerated. Nothing of concern in that regard. Yeah, that's a question I wanted to ask. And you kind of jumped the gun, but we'll get there in more details. Tom? Really quickly for Doctor Lemere. For your complement model, do the antibodies interact with complement differently? Because the model that ARIA is rate of plaque removal is kind of all over the place. Donanemab's incredibly aggressive, but ARIA is not as high. Aducanumab and Lecanemab doesn't remove a lot of plaque, but you get a lot of ARIA. So do you have an explanation to these subtle but real differences? Yeah. I mean, I'm of the school of thought that it is the antibodies binding to the vascular amyloid first that's causing ARIA, not to plaques first. And I think there's increasing evidence for that, some studies that were presented recently at a meeting. I also think different antibodies bind to vascular amyloid differently, and that may also impact. And I just got a grant from NIH to look at that. But we're trying to sort that out. Why is it that some antibodies have higher rates of ARIA? And we think that it may have to do with their binding potential to binding dynamics to vascular amyloid. Helpful. Thanks, Tom. Any other questions at this point? Good job. Just to answer the earlier question. So regarding whether patients would be open to being in a trial to reduce ARIA, always that decision depends on the risks and the benefits. So patients don't want to have brain edema or brain hemorrhage. And especially if they're at a higher risk group, being an APOE4, APOE testing should be done for everyone. And standard will become that patients are informed about their risk for ARIA, and the likely rate, and then they can decide if they're a good candidate, whether or not to be treated. And many patients are really concerned about that. And if there was a way to reduce that safely, many patients would be open to that. And Nick, I think you mentioned when you were presenting the slide that it would be in carriers only, and you'll be offering the testing service. Yeah, again, I'll let yeah, the initial study is planned in individuals that are carriers for APOE4, both homozygous and heterozygous. But in the broader program, we don't intend to exclude anyone, including non-carriers, in that development program. Okay. Eventually. Okay. And if I could just make one comment, Charles, just in terms of there's a sad irony, I think, around all of this. And it is that we've waited for a long, long time to have disease-modifying therapies available for people living with Alzheimer's disease, which obviously is devastating for families and communities and nations and now globally. The irony is that it's available, and those that are in greatest need are at greatest risk for this side effect. And it's confounding the ability to implement the one disease-modifying therapies that are out there for these individuals. Yeah, certainly based on genetics. Dr. Salloway, do you see that as a modulator to uptake, or is it perhaps greater access issues? I'm sorry. Could you ask that again? The question is, for Lecanemab, has ARIA risk been a modulator to uptake, or is it broader access? Yeah, I think the rollout has been slower than expected. This is a novel therapy. It involves a change in our standard of care. None of the elements are standard. IV treatment is not standard for Alzheimer's. Amyloid testing, APOE testing, safety MRI monitoring, ARIA management, all new. And so there's no revenue model. Each institution has to figure out on the fly how to get this standing up and going. So those are the main limitations. But ARIA definitely factors in there because if it was a benign treatment with a benign side effect profile, it would be much easier to roll out on the risk-benefit side. Now, Nick and team, you talked about Parkinson's in the next 36 minutes that I have to talk. Why don't we talk about that briefly? I'm quite intrigued with the use of or deployment of the mechanism in Parkinson's and then especially for the pre-dopamine therapy patient population. So I guess what is the evidence that you have that suggests that this drug might be particularly useful for reducing alpha-synuclein in that patient population and getting to a clinical endpoint that you can interpret? Yeah. Well, that's a great question. I think, again, a lot of it ties back to some of what I discussed earlier. I don't know that we're suggesting it's going to lower alpha-synuclein levels necessarily. We'll try to understand what we can about that. But I think the idea is that the inflammation that you see as a result of alpha-synuclein deposition is driving neurodegeneration. So we think that by stopping neuroinflammation, we can have a neuroprotective effect and limit the neurodegeneration that you see in diseases like Parkinson's. So MDRS type 2, change of 2 points, have no idea of what to expect in terms of this planned enrollment. So kind of what timing would you expect, or would it have been better to do a milestone analysis, a response rate at 6 months or something? Yeah. No. So we spent a lot of time looking at PPMI data and placebo arm data. And so the nice thing is the actual operationalization of the study in terms of the endpoints doesn't change. You're still getting the MDSUPRS part 2. And one thing maybe just wasn't clear. Patients in the double-blind period will all be in the double-blind period even after they have the event. But we are very confident that if you do a 48-week study, which really does also limit the number of patients who start dopamine or want to start dopamine during the study and allows them, even if they do, to fully contribute to the primary endpoint, we can do that study with around 300 fewer patients per study relative to looking at a mean change in part 2. So more capital-efficient than could be. But what kind of change would you anticipate in that timeline in terms of progression to MDS part 2? Yeah. Again, I mean, the endpoint is time to have confirmed 2-point or greater worsening at 2 consecutive visits. But I'm sorry. I didn't ask the question correctly. What percentage of patients do you think will progress in that timeline? Yeah. I don't know if we're disclosing a specific expectation. And not everyone will progress. But we're confident that in a sample size of roughly under 600 patients, that enough will progress to allow you to show a statistically significant effect on the endpoint in 48 weeks. Okay. And same question with regard to Alzheimer's. You mentioned that this is a plan. Is it a plan to operationalize in a reasonable period of time? And what can we expect in terms of time to data, roughly? Specifically to Alzheimer's disease? No, no. Parkinson's. Parkinson's. Yeah. Oh, wait. I'm sorry. Can you ask that again? I know you're asking that. Well, do you plan to operationalize this study soon, and how long will it take to get to data? Same answer as PK before that. We've gotten great feedback, but I'm commenting specifically on the timelines for now. Okay. Could we expect by year-end? I'm kind of keen to press, but yeah. Okay. We can move on. I got it. All right. Maybe last question, Charles. Yeah. So we are nearly out of time, and I would like to take more time talking about this area. But I guess the last question, maybe for someone who can provide a perspective on how you became interested in this mechanism, this particular molecule, BHV-8000. I don't know if it's for someone on the panel or out in the audience. But then is this a platform? Do you have other candidates that could be deployed in other areas of neuroinflammation? Maybe I can take this one. I think it's a fantastic question. And I think one of the things that you'll see as a theme is that we like to think about the field and where there's great unmet need. And I think at Biohaven, we've been committed to patients who have neurodegenerative diseases. That's something we care deeply about. And we want to try to attack those diseases in a way that is focused. So we've thought about neuroinflammation because it could be applicable across many different diseases and be complementary to targeted treatments like anti-amyloid or anti-alpha-synuclein treatments. We also think that the immune system is sort of the key and something that we can use to unlock some of these different treatments for different disease areas. Given that we have a lot of experience, many of us at BMS in the past, we thought about immune targets that are validated. We know that there are TYK2 inhibitors that are peripherally restricted that work and are safe. We know the same thing about some of the JAK inhibitors, although they have liabilities. To us, we looked at some of the immunology, and we thought this combination was fantastic. We went out and we sought it, and we found it, and we worked with a company that we licensed this asset from. Do you have more than one agent of this? We have a collaboration with them. We haven't talked about the details of that. This is the clinical asset, but potentially. We didn't talk about MS. We didn't talk about ALS. A lot of opportunity. A lot of inhibitors, inflammation involved in neuro or psychiatric disorders. Yeah. So stay tuned. Okay. Charles, I might just add Gil L'Italien, who runs our health and epidemiology group. I have to give him a lot of credit because when we had these hypotheses, he went out and bought several million patients' worth of data to analyze if you're currently on agents that target TNF-alpha or IL-17, do you have a risk reduction for Alzheimer's and Parkinson's? And he's now replicated that twice. The answer is yes. When you look at real-world evidence, it suggests these pathways are tied to the pathophysiology of Alzheimer's and Parkinson's. And so I think that's another important data point that brought down risk for us as well. So you have a lot of data reliant from that perspective. We do. do. Yeah. Genetics, human data, other data. To you. I think we're set. Okay. Awesome. Thanks so much. Really appreciate it. Thank you, everybody on the panel. We're going to move into the next session. I'm going to invite my colleagues to come up and have a seat here on the table. Please have a seat. I'll start with brief introductions. This is a session focused on our nascent oncology program. I'm really excited to introduce our three speakers from this program who are all internal experts. As you'll see, we have a tremendous amount of experience in the oncology space and opportunity within the oncology space at Biohaven. Our first speaker, Brian Lestini, is going to kind of give that vision. Brian is a leader in oncology. He has 15 years of experience, including 10 years as a colleague at BMS. While at BMS, he led the initial development and approvals of a drug called Opdivo in lung cancer, as well as another checkpoint inhibitor, Opdualag, which is a LAG3 inhibitor. Brian has a tremendous amount of experience that he'll be sharing some of the vision for oncology, leveraging the resources that we have, and talking about some of the exciting milestones we've achieved. We have Nushmia Khokhar. Dr. Khokhar is our Chief Medical Officer for oncology at Biohaven. She led the Darzalex program at Janssen. She also has served as CMO and head of development for Umoja Biopharma and Autolus Therapeutics, which designed several registrational programs, including a recent BLA for Obicel. Really excited to have Nushmia here. She's going to be talking about her TROP2 program. Last but not least, we have Dr. Gene Dubowchik. Gene was also a colleague at BMS where he was head of, excuse me, at Biohaven, he's head of molecular technologies there. He was a chemist. He led the development and patented the technology that underlies the Seattle Genetics linker technology. And so a tremendous amount of experience here at Biohaven that we are going to now leverage. And I'll hand it over to Brian to speak about that. Great. Thank you, Irfan. If I could get the first slide, please. Oh, I can just advance to the title slide? Perfect. Thank you. All right. Thank you. It's a pleasure to be here today. I'm excited to share with you the work that we're doing at Biohaven to build a leading oncology organization that's squarely aligned with the mission and the legacy at Biohaven of delivering transformative medicines to patients in areas of unmet medical need. The oncology strategy that I'm going to share with you capitalizes on Biohaven's track record of innovation and execution and leverages our core pipeline platforms as well as the end-to-end capabilities and expertise that we've assembled across discovery, manufacturing, and clinical development. The main strategic focus that we'll discuss today is the antibody-drug conjugate portfolio, where you'll see that we have tremendous flexibility to design and optimize each ADC asset depending on the target and the disease of interest, and to generate a diverse and sustainable portfolio of highly differentiated ADCs. As Dr. Khokhar will tell you in a moment, we're extremely excited this morning to announce that our lead program, BHV-1510, which is a highly differentiated next-generation TROP2 ADC, has now dosed the first patient in a phase 1/2 trial. We're also advancing multiple additional programs that are positioned to enter the clinic in the next few years. Now, the other strategic focus that we won't cover today leverages the innovative targeted protein degradation platform of Biohaven, where we believe that there are significant opportunities as a modality that's becoming increasingly important for several indications in oncology and hematology. Now, the recent approvals and the expansion of activity in the ADC space speaks to the powerful therapeutic potential of this modality. However, the core unmet need with current-generation ADCs that we are looking to address with our portfolio is the high rate of dose-limiting toxicities that lead to narrow therapeutic margins, which ultimately has an adverse impact on efficacy. Now, this is driven largely by the poor stability of the ADC and therefore premature leaching of the free cytotoxic payload that leads to off-target systemic toxicities and suboptimal exposure of drug at the site of the tumor. As Gene will show you, with our approach, we've been able to demonstrate superior preclinical profiles across a broad range of targets and payloads when compared directly to competitor compounds. And as such, we've developed a robust pipeline of first-wave assets that are now poised to enter into the clinic. As I mentioned, our TROP2 program is now in phase 1, and we're preparing for several additional INDs over the next 2-3 years. Now, we're also driving future innovation and sustainability in our pipeline by looking to how we can best optimize each component of the ADC that will lead to the highest level of differentiation and competitiveness of the total product profile. So this includes careful selection of targets, but it also includes optimizing the functionality and the performance of the antibody component itself. It also involves thinking carefully about the choice of the payload, whether a differentiated cytotoxic that can overcome resistance or synergize with other mechanisms like immunotherapy, or even looking to novel non-cytotoxic payloads. Now, one of the things that sets us apart from many other companies who are developing ADCs is the set of flexible and complementary approaches that we have at our disposal, which position us with multiple routes to achieve a common, superior safety and efficacy profile. The modularity that we have in terms of being able to utilize native off-the-shelf antibodies, flexibility as to the choice of the linkage sites and the drug-to-antibody ratio, or the DAR, and the choice of payload demonstrates the breadth that we have to specifically design and optimize each component for a particular ADC. Both the technology we're utilizing in BHV-1510 and also our MATE platform programs result in a highly stable linkage of payload to the antibody through formation of irreversible peptide bonds, site-specific conjugation with a highly homogeneous DAR, and then finally, the novel differentiation of the payloads that in particular can synergize with standard-of-care agents, especially checkpoint inhibitors. Now, all of this allows us to generate a broader therapeutic margin that may improve clinical efficacy by optimizing dose and time on treatment and allow us to pursue not only fast-to-market opportunities as monotherapies, but also to move into earlier line settings, for example, with immunotherapy combinations. So now, to tell you more about the first ADC in our portfolio to enter the clinic with such a profile, BHV-1510, I'm pleased to turn it over to our CMO, Dr. Nushmia Khokhar. Thank you, Brian. So good afternoon, everyone. Thank you again for being here. I'm going to take the next few minutes and talk about the TROP2 landscape, the TROP2 ADCs that are in clinic, also some of the shortcomings that we've seen with these first-generation ADCs, and really how BHV-1510, which is our novel TROP2 ADC, is differentiated and has a potential to be a best-in-class ADC within this target. So TROP2 is a very attractive target in oncology. It is expressed on a majority of epithelial tumors. And not just that, a majority of patients will express it at high levels. There's only one TROP2 ADC that is approved thus far. That is Trodelvy. It is approved in breast cancer and urothelial cancer. Another TROP2 ADC called Datopotamab deruxtecan or DS1062, as you'll see on the subsequent slides, has advanced in non-small cell lung cancer. It has shown benefit in terms of overall response rate and PFS when you compare it to standard-of-care chemotherapy, namely docetaxel. However, it has not shown overall survival benefit. The major limitation of these first-generation ADCs is that they have issues regarding their linker payload stability and chemistry. These are labile products that fall apart in the plasma. A lot of the toxicities that we see with these ADCs and the narrow therapeutic index that we see are related specifically to the toxin or the payload and are not related to the target. So as an example, Trodelvy has black box warnings for neutropenia and diarrhea. These are specifically related to SN38, which is a payload of Trodelvy. And again, this goes to the narrow therapeutic index that we've seen with these first-generation ADCs. There are a few other ADCs that are in development, namely DB1305. Also, Merck is progressing with a TROP2 ADC in clinic. Both these ADCs have already hit maximum tolerated dose as they go through their dose escalation, again, going back to that narrow therapeutic index. So what does this tell us? This tells us that TROP2 is a validated target in oncology. We've seen approvals with benefit over standard-of-care chemotherapy in several tumor types. However, these ADCs are limited in their development because of that narrow therapeutic index. So it leaves the opportunity for a best-in-class potential ADC with a superior ADC, which has better stability and also better characteristics, which brings us to some of the features and differentiating features of BHV-1510. So this is a novel, highly differentiated next-generation TROP2 ADC. It is ideally positioned to look for areas that are fast-to-market. It is also ideally positioned to be a partner of choice in combination with other anti-cancer agents, but namely PD-1s. It's important to highlight PD-1s or anti-PD-1s because these agents have shown remarkable efficacy across the board in several tumor types. But importantly, in a lot of epithelial tumors, they are making their way into frontline therapy. So an ADC, a TROP2 ADC that can synergize and be partnered from an efficacy and a safety perspective with a PD-1 will be the partner of choice. This ADC is fully optimized end-to-end. It has a highly stable linker payload. And I'll show some data in the subsequent slides to show that. Another important feature which differentiates this ADC from the first-generation ADCs is that it uses an enzymatic site-specific conjugation. So what that means is we get a very homogeneous product with a drug-to-antibody ratio, or DAR, of four. But this gives us a homogeneous DAR of 4, different from the first-generation ADCs where you actually have non-specific conjugation and you get very heterogeneous products. So even when you have ADCs with DARs of 8, it's actually a range anywhere from 0 to 8. So this is an important differentiating factor as well. I'll show some data in subsequent slides that shows us that we've actually seen synergy with an anti-PD-1. We've also seen a phenomenon called immunogenic cell death with this payload, differentiating it from the other payloads and the other ADCs that are in clinic. The last piece of data that we'll share is a very differentiated safety profile in our preclinical GLP-tox studies. So Datopotamab deruxtecan or DS1062 has a liability of interstitial lung disease. Sacituzumab govitecan, which is Merck's TROP2 ADC, and also Trodelvy have liabilities in terms of neutropenia, and for Trodelvy, also diarrhea. Again, these are narrow therapeutic index, and all these toxicities are related to their specific payloads. The breaking news, and probably the most important thing I want to highlight on this slide, is that we're really excited and proud to announce that the first patient on this trial has been dosed. I would also like to share that we've entered into a clinical supply agreement with Regeneron. So we will be starting combinations of BHV-1510 with their anti-PD-1 Libtayo. So over the next few slides, I want to share some of the data that we've seen, the points that I highlighted. So when we look at BHV-1510 in combination with a PD-1, we looked at this in syngeneic mouse models. These are immunocompetent models that are frequently used to look at the efficacy of anti-PD-1s. If I draw your attention to the green box there, that is looking at BHV-1510 in combination with an anti-PD-1. In the same experiment, we compared it to DS-1062 or datopotamab in combination with a PD-1. Remarkably, what we see is 5 of the 6 mice had complete regression with BHV-1510, as opposed to none of the mice having regression with datopotamab or DS-1062. This, again, goes and shows us synergy with a PD-1. I would say this is the first time we've seen such synergy with a TOPO-1 payload. The other thing I'll highlight is clinical data that has evolved with datopotamab plus pembrolizumab specifically in non-small cell lung cancer, is along these same lines as well. We have not seen synergy in those trials that they've reported out. Namely, the non-small cell lung cancer data in combination with pembrolizumab thus far is undifferentiated from historical control. The other thing I'd want to add is PD-1s have their own low rates of pneumonitis or ILD. Datopotamab, given its specific toxicity of ILD, when they're doing the combination therapy, they are seeing increasing rates of ILD. Some of their arms have reported ILD as high as 22%, including fatal events. These are some of the liabilities. Again, this shows us that this is a differentiated product. It's a differentiated TROP2 ADC that has potential to be developed in monotherapy, but notably in combination with PD-1 with better efficacy because of the synergy and also a differentiated safety profile. We also looked at several other attributes of BHV-1510 and also compared it to some of the benchmark ADCs. What you see on top there is a phenomenon called bystander effect. We see superior bystander effect specifically when we compare it to DS1062. Now, bystander effect can be an important feature for ADCs. The reason for this is that solid tumors are heterogeneous. Some cells can have very high levels of your target, but the neighboring cells may have no target expression or very low target expression. Having this phenomenon of bystander effect where you can kill cells, adjacent cells with low levels of target, is a very attractive feature. This is limited to the tumor itself. We have demonstrated better bystander effect specifically when you compare it to Datopotamab. We've also looked at immunogenic cell death and looked at immunogenic cell death comparing to the payloads of DXD and SN38, which are the specific payloads, again, for Datopotamab and Trodelvy. Now, when cancer cells die, a couple of mechanisms can take over in that phenomenon or in that cell death. One of them is called immunogenic cell death. This is important, but this actually revs up or activates the immune system and causes an anti-tumor response. And this will also be important when we look at combinations and synergy with an anti-PD-1. The other thing I'd like to highlight here is the stability. We've talked a lot about having stable constructs and how a narrow therapeutic index with the first-generation ADCs is a liability because of the labile constructs and the weak linker payload chemistry. So here, we looked at the stability of this construct in various species of plasma, including human plasma. And we compared it, for example, to DS1062 as well. What you see here, remarkably, is that there is 100% retention of the DAR regardless of the species of plasma with Biohaven 1510. And over time, you do see degradation of the DAR down to about 60% degradation of the DAR for DS1062, also known as Datopotamab. The last thing I want to show on this slide is that, again, given the stability of this construct, what we see is high delivery of that payload to the tumor. So when we compare the concentration of actual payload or toxin in the plasma versus tumor, it is about 70-fold higher in the tumor, again, showing that delivery of payload to the tumor cells very specifically and a highly stable construct in the plasma. The TROP2 ADCs that are in clinic, for the most part, utilize topoisomerase I inhibitors, which are really well-known chemotherapy agents and have been around oncology for a long time. This ADC also has a TOPO-1 payload. We call it TOPO-IX. However, this has many differentiating features which are superior to the other TOPO-1 payloads. Some we've already talked about, but I'll highlight again. So we have better in vitro cytotoxicity as a payload and also in conjunction with the corresponding ADC. We have better in vivo efficacy in various cell lines and various models that we've looked at. We've seen superior immunogenic cell death. And we've also seen superior bystander activity. The other thing I'd highlight here is that one of the mechanisms of DXD resistance is thought to be because it is a substrate for one of these efflux pumps. Notably, TOPO-1 is not a substrate for that efflux pump. And we think this would be an attractive feature to overcome some of the resistance mechanisms that have been seen with other TOPO-1 payloads like DXD. Now, switching over to the CINOPK data or the preclinical non-clinical safety data. Again, this is a very differentiated profile that we've seen. The data that you see there for DS1062 and for Merck's TROP2 ADC is taken from their respective publications. The first thing I want to highlight here is that the data that we've seen in these GLP monkey studies translates really well and predictably into the human studies. So as an example, for DS1062, they saw severe lung toxicity at doses of 30 milligrams or higher. And that is their main liability in clinic as well. For Merck's TROP2 ADC, they saw hematological toxicities across all doses. That is really their main toxicity in clinic. This data reliably translates to human data. When we switch over to Biohaven 1510, conversely, what we see here is a very differentiated profile. We dosed monkeys up to doses of 60 mg/kg. We really did not see any severe toxicity. Notably, no change in hematological or chemistry parameters and really no major organ toxicity and, importantly, no lung toxicity. This gives us confidence that this is a differentiated safety profile. The other thing with this data that's notable is that if we look at the corresponding data in terms of the toxicokinetic profile or the TK data, it really overlays well with this. What we've seen is really robust exposure of the ADC in these monkeys. What I want to highlight here is if you follow the two curves of the ADC and the total antibody, there's 100% overlap of the curves and almost minimal to no free payload that's detectable. That really goes to explain this differentiated safety profile and, again, goes to show a very stable construct in the plasma. Comparatively, the HNSTD, or the highest severe non-toxic dose for Datopotamab, is 10, but for us, it's 60. Really, it's 60 because that's the highest dose that we had tested. I would really call it the NOAEL. What this tells us is, again, a wider therapeutic index also gives us confidence that we can start at a robust dose in clinic. With that, we come to the clinical trial and the clinical trial design that we've put forward. So like most oncology studies, first-in-human studies, this trial will evaluate patients that have advanced disease or metastatic disease and that have failed what's considered standard of care therapy. We will go through a dose escalation design. We've designed this with a starting dose of 2 milligrams per kilogram, given the favorable safety profile that we saw. Just for comparison, because of their narrow therapeutic index, Datopotamab started their trial at a dose of 0.27 milligrams per kilogram. They dose escalated up to 8 milligrams per kilogram where they saw fatal ILD toxicity and then had to drop the dose. We are starting at a dose of 2 milligrams. We will use what's called a BOIN design or a Bayesian optimal design to dose escalate up to 12 milligrams per kilogram. Again, like most first-in-human studies, the primary endpoint for this study is safety. We will look at efficacy data as well in terms of response rate and PFS and also look at the corresponding PK data. We expect to have, as I mentioned, the monotherapy arm has already started. We've initiated. We've dosed our first patient. We expect to have early monotherapy safety data and also initiate the trial with the combination of Libtayo or PD-1 in the second half of this year. Okay. With that, I will hand it over to Gene. Thank you. Excuse me. I'm very excited to present our internal antibody-drug conjugate program, the so-called MATE conjugation. And this continues the theme that Dr. Khokhar just emphasized of optimizing stability of ADCs. We believe that our internal conjugation platform is a potential overall best way to conjugate payload linkers to antibodies. The protocol involves a single-step site-specific conjugation to wild-type antibody, the antibodies IgG1, 2, or 4. The antibodies don't require any kind of engineering of unnatural amino acids or extensive manipulation. The result is a very stable amide bond. I think let me just yep, highlight it here. To a chemist and I'm a chemist. A forewarning, there'll be some chemistry in this presentation. A stable amide bond to a single lysine on both heavy chains of IgG1, 2, and 4. So we achieve a DAR of 2 with a monovalent linker. This, to a medicinal chemist, is a very, very stable amide bond, a stable linkage that should not be reversible in vivo. This is in contrast to the industry standard maleimide conjugation. This has been used since the '90s. Actually, since I was at Bristol-Myers Squibb working on ADCs, we actually used this conjugation. It wasn't realized at the time that it can reverse. It's been noted over and over again recently that in stability studies in plasma and in vivo, you do see reversal. There have been some studies in animals that have actually quantified there's a reversal. This leads to essentially regeneration of the payload linker. This can conjugate to glutathione. It can conjugate to human serum albumin. It essentially represents a highly cytotoxic payload that can get elsewhere in the body in an untargeted fashion. The industry has been going toward site-specific conjugation technologies. Most of those involve either engineering unnatural amino acids, enzymatic removal of sugars, et cetera. Most of these protocols generate what is a very lipophilic linkage to the ADC, which results from what's called a strain-induced click chemistry. Again, to a chemist, this is a very large lipophilic sort of greasy group. This is undesirable if you can avoid it. We do avoid it with our conjugation. We believe that we should be able to improve the therapeutic index of ADCs, improve safety, less payload getting away from the ADC in circulation, improved efficacy, more payload getting to the tumor. We've filed IP globally on the overall procedure as well as specific ADCs and conjugation reagents. This approach is stable, site-specific, and very simple. This is how simple it is. This is just some chemistry. What we have is that the conjugation reagent is shown in the bottom left. It is a cyclic peptide that is directed toward a specific region of IgG1, 2, and 4. It's been modified to include what's called an active ester in chemistry. That is attached to the linker payload. All of these reagents can be manufactured in a straightforward fashion. What happens is that the FC3 peptide binds to a very specific part of the FC portion of the IgG. It brings the active ester in close proximity to a specific lysine, lysine 248. That displaces the active ester, resulting in that very simple amide bond. We get the ADC showing in the bottom right. The directing/leaving group is washed out. We've shown that this chemistry works on scale. We've actually scaled this up to 1.1 kilograms of GLP material and in very good yield. Also, we've demonstrated the breadth of this technology. We've made further 11 ADCs using this technology. The proof of concept ADC that we chose to make with this is BHV-1500, which is essentially a version of Adcetris. It's a CD30 ADC. It's a very validated target. The chemistry is highlighted. The chemistry is highlighted up here. I highlight in green and red here the only real differences. So we use the same payload linker as Adcetris. And we use the same antibody. We conjugate—we conjugate in a slightly different part of the antibody. But really, the only significant difference is the conjugation chemistry. And I want to emphasize that our ADC, BHV-1500, has a DAR of 2, whereas Adcetris has a DAR of 3.8, so almost 4. So we carry almost about half of the number of payload molecules. Now, we made this ADC. We showed it had very good plasma stability. We showed it had good in vitro cytotoxicity. And then we took it into a mouse tumor xenograft study at a CARPAS-299 cell model. And this was dosed head-to-head compared to Adcetris. It was dosed 4 times over a month weekly, 4 doses at 3 doses, 3 milligrams per kilogram, 1 milligram per kilogram, and 0.3 milligrams per kilogram. And you can see that at 3 milligrams per kilogram, both agents flatline the tumors. At 1 milligram per kilogram, it's not easy to see in this graph because the lines look very similar. But we do show an advantage. We get 8 out of 10 remissions in this model, whereas Adcetris showed 4 of 8 complete remissions. So at 1 milligram per kilogram, we do show an advantage. At 0.3 milligrams per kilogram, Adcetris is essentially inactive. It's almost like the saline control. Whereas our ADC at 0.3 milligrams per kilogram does show significant anti-tumor activity. And you can see it in that line there. This is more apparent in the survival curves in this model where, again, Adcetris, all of the animals are dead during the study. Whereas at 0.3 milligrams per kilogram, we have a significant number of animals that live to the end of the study. We took BHV-1500. Again, I want to emphasize that we carry half the number of payload molecules. So we also took this ADC, BHV-1500, and compared it head-to-head in a cynomolgus monkey PK study. We looked at two doses of our ADC, 3 milligrams per kilogram and 6 milligrams per kilogram. We compared it to 3 milligrams per kilogram of Adcetris. We see a significant improvement in stability over the, I think, this is a two-week study. So you can see that this is measuring total ADC. So you can see that Adcetris is losing payload over time, whereas we are much, much more stable in the course of this study. Oops. So based on that very good-looking data, we chose a list of further ADCs to make. We chose a diverse set of both marketed ADCs and ADCs that are in late development. And we made all of these ADCs without any trouble using our conjugation method. We've gotten good stability data, plasma stability data on all of the ADCs that we've looked at so far. Pretty much all of them have shown very good in vitro cytotoxicity data. And we're continuing to get data on these ADCs. This is a little bit of the plasma stability data. And again, I won't go through this in detail. But you can see our ADCs are essentially the line at the top that essentially shows no change. This is looking at loss of payload in human plasma over time. We have some more data. It's looking very good. We're getting some animal data on a set of these further 11 ADCs. It is also looking very favorable. Oops. Keep hitting the back button. In summary, we're very excited about the future of ADCs at Biohaven. As you know, Biohaven has a very good record of promoting novel science, but especially bringing very promising-looking compounds to the clinic quickly. We have industry-leading expertise now in oncology, in development, in chemistry, in discovery in CMC. We have a very flexible platform that we can use to optimize the profiles that we need. We have a sustainable portfolio of different ADC programs that we've already made. We're progressing those. Most excitingly, we've initiated phase 1 study of BHV-1510. We have another one behind it. That's it. Okay. Well, thank you so much, Brian, Nushmia, and Gene. I think maybe time for one or two quick questions. And then we'll take a break. There's coughing things outside. So I don't know, Matt, if we have any questions that we want to use or maybe if there are no questions, then we'll start in the break. Oh, here's a question. Just differentiate your program from the people who walk non-natural amino acids through the antibody backbone. What do you like about it? About your approach relative to Sutro and people like that? Well, so back at BMS, I worked in ADCs twice. The second time, we were actually working with a company I probably shouldn't name it, but the site-specific conjugation technology that involved engineering unnatural amino acids. Honestly, we often had great difficulty in getting good expression levels of unnatural amino acids. I mean, it's mostly a cost thing, I think. Again, sometimes it's just very difficult. I'll just add. Those large, bulky groups can lead to problems with aggregation of the ADC. They can impact stability. And so they do have features that can impact the clinical profile as well as the cost of goods issues and other issues that Gene mentions. Yeah. I've actually spoken to people. Sutro uses that strain-induced chemistry. One other scientist did say that, yeah, they often do have CMC issues, so aggregation issues with ADCs usually. It's just unnecessary, really. I I mean, as a chemist, because the payload already is much more lipophilic than the protein surface. Okay, folks. Thank you. We're going to go ahead and get started with the myostatin session. I'm going to go ahead and introduce, once again, our panel, which has both, again, internal experts, external experts. So I'll start with our external experts. First, we have Dr. Greenway, who was kind enough to fly in. His flight was delayed. And so he just made it. So we're very pleased to have him. He's director of the Pennington Biomedical Research Center in Baton Rouge and really an expert in obesity science and clinical. And he's been involved in obesity for all of the recently approved medications and historically for many of the approved medications over the last 20 years. Really pleased to have him here. He's going to be speaking today about some of the unmet needs in obesity. We have Dr. Barry Byrne from the University of Florida. Barry Byrne is the associate chair for pediatrics. He's a neuromuscular disease specialist. He'll be talking about our SMA program. Then for our internal experts, I'm going to start on the end. We've got Bruce Car, who's our Chief Scientific Officer. Bruce will be speaking about the innovative mechanism of action of taldefgrobep alfa. We have Lindsay Lair, who you met earlier. Lindsay is Vice President here at Biohaven. Then we have Peter Ackerman, who you've also met, who's VP here at Biohaven. So with that, I'll hand it over to Bruce to get started. There you go, Bruce. Thank you very much. This has been a fun time for us studying the mechanism of action of T-alpha. This, with the knowledge of biology in the area of how the myostatin and the TGF family ligands work, has been, since the last 10 years, emerged as an entirely new discipline. How T-alpha plays into that is extremely important. Illustrating that in this slide, muscle, fat, and the other tissues on this slide form a highly integrated reciprocal inductive cytokine network. That network perfectly allows a precise pharmacologic intervention of T-alpha in controlling both fat and muscle homeostasis. Very specifically, low muscle mass is associated with a range of age-related declines, cognitive decline, and increase in all-cause mortality, whereas the opposite is the case of high muscle mass, where these things are reversed. High adipose mass is actually associated with very high levels of these myokines that reduce muscle mass and increase fat mass. Within this broad context, I'll show in subsequent slides how T-alpha exerts its mechanism of action. A lot of this work has actually been demonstrated recently within our laboratories in Cambridge, U.S. Okay. If you look at that middle lane there, you can see these TGF beta family ligands. These are the things that we're calling myokines. Myostatin, Activin, GDF11, there's three of them. They all work through the Activin R2 receptor. They signal through this receptor. The signaling actually causes muscles to stop growing, ultimately leading to muscle wasting and increases lipid storage. You might say, "Well, that's kind of a crazy idea." That's actually what these receptors do. They're negative regulators. They control mechanisms. Each of these cytokines is expressed at much higher levels in an obese situation. So how does taldefgrobep alfa work? So taldefgrobep alfa binds myostatin and GDF11 tightly. And that taldefgrobep alfa myostatin complex also binds the activin receptor. So it works through two mechanisms. One is it inhibits these ligands. The other is it inhibits signal transduction through the receptor. The net effect of all of this is to slow down this signal transduction. SMAD phosphorylation is how we measure this. And the reverse is the case of what these myokines do in a situation in an obese individual or an individual with muscular atrophy or muscular dystrophy. You see muscle hypertrophy, increased lipolysis, so increased breakdown of fat, and thermogenesis. So it's really the ideal mechanism to interfere with the negative regulation of that activin R2 receptor. And I mentioned the complex very briefly. Myostatin and GDF11 bind T-alpha so tightly. That's a low picomolar affinity. This is less than 10 picomolar affinity that when they bind, they kind of stick around for a very long time. And that complex, and here you can see in the bottom left-hand panel at a 45 mg dose, the myostatin complex reaches very significant levels in plasma, well in excess of myostatin, Activin, and GDF11, and is actually able to then inhibit their interactions with the receptor. So if you think about the mechanism of action here, you've got a direct effect of T-alpha on myostatin and GDF11, just takes them out of the picture. And then the complex continues this activity, inhibiting the Activin receptor with a bit of a lighter touch and preventing the Activin signal transduction. So why is that important? So we've studied this extensively in vitro systems, in animals, in knockout animals, and in the clinic as well. The role of inhibiting myostatin in muscle growth and of the ActRIIB receptor in muscle growth is extremely well known. I won't reiterate that here. Of course, we interrupt this mechanism. We promote muscle growth. We've shown that preclinically and clinically, also shown that in vitro. But what about fat? This is a new area of biology that's only been emerging over recent years. What do these myokines, what do these TGF beta family ligands do to adipose cells? In the graph on the left, you can see combo there is a mixture of these three myokines. T-alpha takes them all down to a normal level in terms of signal transduction. So the graph on the left shows that T-alpha very quickly, very completely ablates the signal transduction of these myokines in fat cells, right? The graph in the middle is really the money slide. What does that mean? So myostatin and, to a lesser extent, the other ligands actually increase fat storage in cells. And I've already mentioned that these myokines are at high levels in individuals with overweight. So myostatin in those situations and other ligands are driving an increase in fat. T-alpha, and this is a direct experiment. This was just done in the last couple of weeks in Cambridge, as a matter of fact, reduced those levels of fat so significantly that you only see the small droplets to the right-hand side. And that's quantified in the graph below where that large blue line is reduced to the cross-hatched line. An important angle to all of this is we're not only reducing the fat storage, we're helping to break it down. So mitochondria are the little engines that we have in all cells except red cells that drive the basal metabolic rate. So a very important aspect that's new for T-alpha biology is that it increases the mitochondrial content in adipocytes. That means fat cells are better able to metabolize fat and reduce it into energy. So this is a really important thing. So what does this look like compared to the competition? This is a very complex signal transduction. I'll explain a little bit about. Scholar Rock and Roche have made specifically target the prolatent myostatin. They only inhibit that. So essentially, they only take out one part of what's important in regulating muscle growth. The other parts, GDF11 and the receptor, are extremely important as well. Inhibiting all three, in a number of experiments in the literature, have shown additive effects on muscle growth. So while this mechanism one would hypothesize is very safe, it's also possibly not completely efficacious. So that's the Scholar Rock and Roche approach. On the right-hand side is the Versanis molecule, Bimagrumab, that Lilly licensed. It binds with an extremely tight affinity directly to that Activin receptor. In fact, it's got an hours-long off rate. And it's dosed at an extremely high dose, 30 milligrams per kilogram every 3 months. And it's basically stuck there. And it stops anything at all interacting with that receptor. And these different pharmacologies, you'll see in a subsequent slide, translate into very different clinical profiles. And T-alpha, I already mentioned, binds GDF11 and myostatin. So it takes those ligands out, very much like the Scholar Rock Roche approach, just for myostatin. But that complex is quite stable and long-lived. And it interacts reversibly with the Activin receptor, blocking Activin signal transduction as well. Important that it's reversible. So how do these molecules stack up? Scholar Rock and Roche, likely quite safe, perhaps less efficacious, delivered as an IV infusion. Bimagrumab, extremely potent, blocks all of those ligands but has very significant effects. A little less than half of all patients have severe muscle spasms, fatigue, and diarrhea. And there are also some significant reproductive effects through this mechanism mediated through follicle-stimulating hormone. This also requires an IV infusion. Taldefgrobep alfa has a favorable subcutaneous dosing protocol and volume. I've already explained the mechanism of action. And the side effect profile, as this stacks up against Bimagrumab, is extremely favorable, but likely with the efficacy of Bimagrumab in accessing both the receptor and the ligands. So we've basically threaded a needle amongst the properties of the three different ligands in the receptor to come up with the best efficacy and the best safety of these three different mechanisms. So I'd like to hand this on. Thank you. I think I'll try and sort of set the stage with obesity. The obesity is a global public health crisis. Obesity isn't just too much weight. It's really too much fat. It's the fat that's driving the cardiovascular risk. It's estimated by 2030, there'll be 1 billion people in the world with obesity. People in the United States, 50%, will be part of that number. Obesity predisposes to cancer, diabetes, and cardiovascular disease, which costs the healthcare system $175 billion annually. Because of those associations, it's thought that about 20% of the deaths related to people who die between the ages of 40 and 85 in the US are associated with obesity. We need to do better. The medications have really taken off in recent times. We used to just use diet and exercise. Then we had Orlistat, which gave around a 5% weight loss, combining old medications. The short-acting GLP-1 agonist, liraglutide, gave about a 10% weight loss. But then semaglutide came along. It's a long-acting GLP-1 agonist and gave a 15% weight loss. When the GIP receptor was added to that, it increased it to 20%. By adding glucagon to that and with a triple form, you get up to a 25% weight loss, which is getting close to the surgical weight loss of between 25%-35%. Despite this tremendous efficacy, there's certain drawbacks. There's three in particular that I wanted to mention. The first has to do with the fact that GLP-1 agonists have an excess loss of lean body mass. So if you take a DEXA, you can measure fat well. You can also measure fat-free mass. Fat-free mass with diet shouldn't exceed 25%. It goes up to 40% with semaglutide. People lose about 5% of their muscle mass every decade after the age of 30. Muscle's important because it uses a lot of carbohydrate. Losing your muscle mass increases insulin resistance. It also is associated with bone loss and loss of muscles associated with cognitive impairment. In a meta-analysis of 16 different studies, they showed that the loss of muscle mass increases your death rate by 50%. The second thing that is a disadvantage to the GLP-1 agonist has been an accelerated bone loss. It affects the cortical bone, the hip, and the spine. It predisposes to osteoporosis. People who are 75 years old and older, and women, are at 4-5 times risk of developing fractures of their hips and pelvis. But when you look at Myostatin, Myostatin actually plays a role with that because it inhibits the osteoblasts, which make bone, and stimulates the osteoclasts, which break it down. Oh. I'm trying to change the. Are we? It's the top one. Is it the top one? Yeah. Oh, I thought it was the middle one. Okay. Pardon me. The other disadvantage is that GLP-1 agonists, when they're stopped, cause a rapid increase in fat loss. As you can see by the graph down in the right-hand corner, a year's worth of weight loss with the GLP-1 agonist, two-thirds of it was brought back as fat primarily in the year following. And people that are obese have a tendency to stop their medications periodically, either because they think they can do it on their own or they have side effects from the medication or they just feel it's too expensive and they can't afford it. So then when people keep losing and gaining weight, they tend to lose muscle and then regain it as fat and predispose later on to sarcopenic obesity. The other issue is that we measure fat by measuring weight or BMI, and it doesn't measure fat. We need better determinants of measuring fat. DEXA can measure fat accurately. Fat-free mass measured by DEXA actually is usually about 50% of it is muscle. There's also organs and body water that's measured in that so that it increases the variability. In order to actually image muscle, you need imaging studies. There's potential for reversing this by the myostatin inhibitors. If one looks at the Semaglutide, Semaglutide gives like a 15% weight loss, but it gives only a 19% loss of body fat and about a 10% loss of fat-free mass. Whereas Bimagrumab gives 6.5% body weight loss, but it gives more fat loss and actually an increase in fat-free mass. In conclusion, obesity is a disease of excess body fat, not excess weight. GLP-1 agonists can decrease muscle mass and increase the loss of bone and cause an excess of fat when the drug is stopped. The newer therapies like Taldefgrobep will help to reverse those, increase muscle mass, and increase the fat loss, and result in a healthier metabolic profile. We also need to have better measures of body fat to use in the clinic, which we really don't have at the present time. Thank you. You want to pass this off to. Thank you. Sure. Thanks, Dr. Greenway. Good afternoon. Over the next 4 or 5 slides, I'd like to talk to you about what makes us most excited about Taldefgrobep for the treatment of obesity. A couple of key points on this slide. It starts with the differentiated mode of action that Bruce talked about. Taldefgrobep's unique ability to bind certain TGF beta ligands that are important in inhibiting skeletal muscle growth, most notably myostatin, is important. But also the ability of that bound myostatin to bind to Activin II receptors and to serve as a competitive antagonist to other ligands that bind through Activin II, including Activin A. We believe that both of these are important to optimize skeletal muscle growth and to drive the metabolic change that we're hoping to see. Another important point is the fact that we've been in the clinic with Taldefgrobep. To this point, we've dosed Taldefgrobep in over 500 clinical trial participants. We have an established safety profile that's very favorable and appropriate for an indication in the treatment of obesity. Point number three is that we're very committed to pursuing an indication in the treatment of obesity with Taldefgrobep. And then point number four is the ongoing phase III program in spinal muscular atrophy. Dr. Byrne and Dr. Lehrer will talk more about that in subsequent slides as part of the panel discussion. I'd like to start with non-clinical data that have recently become available to us. This is a study that was conducted in kind of the standard diet-induced obese mouse model, C57 Black 6 mice, six weeks of age, put them on a high-fat sugar and high-fat and sugar diet for about 13 or 14 weeks, and then initiate therapy. The therapies include a placebo arm, which is marked by the black color, a Taldefgrobep monotherapy arm, which is marked in green, two different standard Semaglutide arms, both in the blue colors, and then the combination of Taldefgrobep with Semaglutide, both doses, in both orange and red. The high-dose Semaglutide and Taldefgrobep is represented in red. And again, the green color is Taldefgrobep monotherapy. From left to right, we're looking at change in total body weight in the animals. In the middle, it's change in total body fat mass. And on the right, it's change in lean mass. And what we're seeing is exactly what you'd hope to see with an investigational agent for obesity. We're seeing absolute decline in total body weight through four weeks with Taldefgrobep alone. And when you put Taldefgrobep with Semaglutide, you're maximizing the amount of total body weight loss that's realized. Importantly, Dr. Greenway has talked about this. The primary pathologic tissue in obesity is not overall mass. It's fat mass. And we believe it's very important that we can show meaningful reductions in fat. And we can demonstrate that here in this DIO study. Again, the green is taldefgrobep alone. You're seeing significant reductions with taldefgrobep alone and then even more when it's combined with taldefgrobep or with semaglutide. Finally, on the right, we're all used to lean mass loss with GLP-1s, including semaglutide. When you add taldefgrobep to the semaglutide in this DIO model, you're getting a net positive, an increase in lean mass in this study. And certainly, we're seeing significant increases in the taldefgrobep monotherapy group. Moving on to some clinical data. These are data from our phase I multi-ascending dose study that was conducted in healthy adults. These are important pharmacodynamic endpoints. On the left, you see free myostatin levels across varied doses of Taldefgrobep. You see significant declines through the dosing period, which ended at day 29. You stopped the drug, and the free myostatin levels returned back to baseline. On the right, you're seeing the effect or the accumulation of the complex. This is measuring Taldefgrobep bound to myostatin. The phenomenon that we're seeing here that's really important is that you're getting dose responsiveness, an increase or accumulation of complex throughout the dosing period. Beyond that, you're actually continuing elevated levels of complex that last for 30+ days prior to returning to baseline. We're seeing this in terms of the physiologic change in the clinic, and we'll look at that in the next slide. Again, these are data from our multi-ascending dose study, again, studied in healthy participants. The blue are the Taldefgrobep-treated adults, and the gray is placebo. We go from total fat mass. On the left panel, in the middle, we're looking at intramuscular fat change as measured by MRI. And on the right, it's change in lean mass. And what we see are a couple of important points. Number one is that you're seeing significant changes across body composition at day 29, by the time we finish dosing at day 29. What's unique to Taldefgrobep is that you're actually seeing almost a doubling of that body composition change at the last time point that we assessed, which was day 57, 28 days beyond the last dosing. And this is the effect of the complex that remains active in being a competitive inhibitor to multiple ligands signaling through Activin II. And so we see this both in fat mass and in lean mass by DEXA. We also see it in intramuscular fat. Intramuscular fat is a significant problem in individuals living with obesity. As many of you know, people living with obesity have increased fat mass overall, but the quality of that muscle increase muscle mass overall, skeletal muscle overall, but the quality of that muscle is very poor. There's a lot of intramuscular fat there. So clearing that intramuscular fat is an important measure for us, and we're able to demonstrate that. What's also important is that we believe that our differentiated mode of action allows for a preferred safety profile. Here, we're comparing ourselves and what we saw in our MAD study, Taldefgrobep MAD study, relative to what has been reported in the Bimagrumab program. Most notably, we aren't seeing kind of the muscle-related adverse events that have caused some problems there, including things like muscle fatigue, muscle spasm, etc., and some of the more difficult-to-manage GI-related issues, including diarrhea. Obviously, we're not looking to compound the AE profile of the GLP-1s. They have a lot of GI-related side effects. What also we're not seeing is we're not seeing the hormonal-associated AEs, the acne and these sorts of things that some other TGF beta inhibitors have seen. So where's this all leading? It's leading to us getting into the clinic and exploring the potential for Taldefgrobep to treat obesity in a representative population, a population living with overweight and obesity, and looking at it in an innovative way. So looking at the effects of Taldefgrobep as monotherapy, but also looking at it in combination with a GLP-1 agonist, and then looking at what happens with SWITCH. So looking at whether we start with Taldefgrobep and then bring on a GLP-1, seeing if we can protect lean muscle in that way, or starting with the combination and peeling off the GLP-1 and see if we can prevent that weight rebound that is common with interruption of those drugs. So with that, I'll stop, and we'll go on to Dr. Byrne. Thank you so much. Thanks very much, Peter. Thank you all. Thanks, Biohaven, for having me join this R&D day, which has really been fascinating to hear all that's going on at Biohaven. So I'm going to talk to you about the unmet need in spinal muscular atrophy, which is a condition that was recognized at the end of the 19th century. It wasn't until 100 years later that the gene for this condition was found, and then specific therapies were begun to be developed. It wasn't until 2016 that there was a strategy for treating SMA. Even though there are now three approved products in this condition, none of them target the effects of these mutations on muscle. This is a disease of motor units. Both the lower motor neuron and the muscle is affected. Even though we have standard-of-care therapy now because of newborn screening initiated in 2019, now this is up to every state to implement newborn screening. But as of a month ago, all states in the U.S. are screening infants for spinal muscular atrophy. There's an opportunity to treat presymptomatic patients with the more really disease-modifying therapy using gene replacement of Zolgensma, which was approved in 2019. Now that we have 4 years of experience of treating patients with Zolgensma, we are starting to see the unmet need that still persists in this patient population. There are about 25,000 patients in the U.S. with SMA across the age spectrum and disease severity. I just want to emphasize that that aspect, as we learn more about what the new natural history of this disease is, is an opportunity for improving the outcome in this patient population. Because T-alpha directly targets muscle, there's an opportunity to further improve muscle function in patients who may have received one of the disease-modifying therapies that is now standard of care following identification of the biolytic mutations in the SMA or SMN gene that causes SMA. Because muscle is targeted by T-alpha, it can augment the effects of salvaging motor neurons that are treated directly by upregulating the SMN2 gene, which is a duplicative gene that makes up for the mutations in SMN1. The RESILIENT trial has been initiated and been ongoing now as a phase III study to evaluate the efficacy and safety of this adjunctive therapy on top of the standard of care for SMN upregulating therapies like Zolgensma. RESILIENT has now really done well with enrollment. It's a huge study in the neuromuscular space. To have over 200 patients enrolled in a study is a great accomplishment. There's a 2-to-1 randomization to the active group, to the placebo group, and then after 48 weeks, an open-label extension. One of the unique things about this study is it has a broad age of inclusion, which is important to be inclusive in this patient population. The patients, really, this notion of how this might work is really to just further emphasize the residual unmet need in these patients and that by including patients that are both ambulant and non-ambulant, there's an opportunity to see how this patient population could further improve with a muscle-targeted therapy. You can see that the majority of patients are young, under 12. This is due to the fact that SMA is the most common cause of a fatal respiratory insufficiency in infancy. And so, before these disease-modifying therapies were available, we didn't have many patients survive. 90% actually didn't live to see their second birthday. So, a small portion of those who are teenagers or young adults who have the less severe form of SMA. And then, the majority, because of those delayed treatments or unavailability of treatment in the milder patients, are non-ambulatory. So, the RESILIENT study enrolled both of these populations, young and old, ambulant and non-ambulant. Of course, that represents a special challenge in terms of primary outcome measures because it requires a tool that measures the clinical outcomes across this broad age spectrum. So the motor function measure is a 32-point scale, which actually works very well in SMA and is designed to capture small differences in functional outcome that do not have either ceiling or floor effects that have been problematic in some of the other outcome measures that have been used. Additionally, there's that opportunity to study the modified Hammersmith functional score and the revised Hammersmith score that is subject to floor effects in just children that are over two. So if we compare the current therapies that are available for SMA, it was mentioned earlier, both the Scholar Rock and Roche programs. And importantly, the RESILIENT study includes a broad age category and broad functional capabilities. You can see that the two competitive studies either enroll only non-ambulatory or only ambulatory patients. Across all the SMA subtypes, they're therefore included in the RESILIENT study and requires that they receive standard of care, allowing for any of the current standard-of-care therapies to be used to qualify. If we see that the assessment, as time goes on through this open-label extension period, now the study being fully enrolled, all patients in the placebo or delayed treatment arm will cross over. And that ultimately, you'll, I think, start to see the benefits, particularly in addressing muscle function. Because there are no other therapies in SMA that target muscle, this is really another opportunity for T-alpha to have an impact in a condition with very severe outcomes and be transformative, ultimately, in the care of patients with SMA. Now we have time for the panel. Thank you, Barry. So maybe just in the interest of time, I'll hand it over to Chris Raymond from Piper to lead our moderator discussion panel. Maybe we can do 15-20 minutes-ish to keep on time? You bet. Okay. Not a problem. Oh, and we have a timer here. That's great. Yeah. Okay. But that one's 30 minutes, so we'll. Right. Okay. Well, first of all, thanks to Biohaven for inviting me to host a panel. I'm very honored to be up here with such an esteemed group of thought leaders, internal and external, to Biohaven. But I got to say, I'm not sure what's worse than having to do a panel before lunch, like poor Tyler, or one right before the degrader discussion. So we'll move along here. So I guess maybe first and foremost, Dr. Greenway, I wanted to talk a little bit about just you touched on the clinical need. When we talk to investors, I think there's maybe still a bit of a disconnect between what investors think is important and necessary clinically and sort of your view and the clinician's view. We often get, "Why do we need polypharmacy in this space? They can just work out." So that logic, obviously, applied to GLP-1s. Maybe they just eat less. So maybe just talk a little bit more. You mentioned sort of the need here, but in terms of the population, which populations would muscle preservation, in your views, most benefit? Is it the elderly, or is there some subpopulation that goes beyond age? Well, I think that the elderly already have a problem or developing problem that way. So yeah, I think that they'd probably benefit the most. I think that what you mentioned about, "Why don't you just push away from the table and walk around the block?" is just a leftover from what people used to think of obesity as. And that was just that it was bad habits. Then the psychologist said that in order to change bad habits, you only need up to 12 weeks to take the training wheels off your bicycle. And it wasn't until 1985 that it was really decided that it was really a chronic physiologically controlled disease. And so it's difficult for that perception to change. It's changing, but slowly. And I think it's but I think it is a challenge still for the field. Maybe just if we dream a little bit about once these agents are on the market, talk a little bit about how you see the overall benefit of these agents. You mentioned the overall metabolic health of the patient, but are you also thinking about potentially lowering the dose of GLP-1s, for example, as a benefit in reducing some of the tolerability issues associated with this class? Well, from a medical point of view, I think that these drugs like T-alpha are certainly much better in terms of addressing things. People are so focused upon weight at this point, and these drugs are so incredibly popular. I think that it's going to probably start with adding a drug like this to the GLP-1 agonist. But there's at least 20% of those people who can't really tolerate a lot of those GI side effects on the long term. And I think that eventually, this is going to come into its own. Great. And then maybe just another key question we get, and maybe this is for either Dr. Ackerman or Dr. Car, is the regulatory, the appropriate regulatory endpoint. We've talked to other thought leaders in the space who've postulated maybe fracture rates, especially in elderly, would be appropriate. But this seems to be a key question that keeps coming up. Any sort of thoughts there in terms of what an approvable endpoint might be? Yeah, thanks for that. I think that I think we're at an inflection point in terms of how we measure the effectiveness of anti-obesity medications. We have draft guidance from 2007 that are telling us that the end-all and be-all to chronic weight management is dropping mass, is the amount of mass drop. I think that that's, as we've gotten into an era, thankfully, where people are able to lose significant amounts of weight and achieve significant benefits and long-term outcomes based on things like the SELECT trial. We need better measures for the quality of the weight loss that's occurring. I think regulators are coming on board with this, and I think you will see change coming in response to this. The only other thing I'll say is that we've had interactions with the agency, and a lot of people are around this very point. We still believe there's established data around that 5% total body weight loss and its ability to drive positive long-term outcomes. That shouldn't be lost. The final piece and so I don't think that's going away necessarily, but I think there's other measures that are important to estimating the overall benefit of anti-obesity medicines. The final piece is that I can't stress enough that fat's the primary issue. When we're talking about whether it's anti-myostatins or any anti-obesity medications, we should be talking more about fat. I think that anti-myostatins have the ability to drive significant amounts of fat loss, and that changes significantly long-term outcomes for people on overall health. Excellent. I failed to mention also, just even though this is a truncated panel, this is meant to be participative. If you guys have any questions, just raise your hands. Maybe Matt up there with the QR code. Nothing. All right. Okay. Maybe pivot back to Dr. Greenway in your practice. As we think about maybe looking around the corner of the developmental landscape, just would love to get your thoughts on how lean muscle preserving agents could be incorporated more broadly into the treatment paradigm. In terms of the administration profile, would we need to see something that's co-administered, perhaps, with incretins? We've also got the onset potentially of oral agents. Just any thoughts there in terms of how you see things sort of evolving over time as innovation continues? Well, my observation has been that with chronic diseases, the time when you get the drugs that work most effectively and most safely is when you have drugs that act on the endpoint. We saw that with angiotensin-converting enzyme inhibitors with hypertension. They're well tolerated, and there isn't much chance for trickle-down effect. I think that although it's been concentrated on muscle with this, I think that fat's involved in this as well. I think that this is one of the trends in migrating to looking at the endpoint for what you want to treat, and that's fat tissue. Maybe pivoting back to the Biohaven folks, another key question that we get often is around the specific mechanism in targeting active myostatin versus pro or latent myostatin, which is what Scholar Rock and Roche are doing. Just maybe talk about the risk-reward profile as you view it and why you decided to sort of ride the horse that you're riding here. Well, I think that it has to do with the safety issues of the thing. In dealing with the Bimagrumab, it causes these cramps and acne and other things that this doesn't do. So I think it's a better mousetrap. When these molecules were first discovered, too, it wasn't sufficient information, really, to understand what the desirable targets were and the less desirable targets. At the time that Scholar Rock and Roche were targeting myostatin, it was thought there might be significant liabilities around Activin and GDF-11, which, through clinical trial study and preclinical studies, knockout mice, and so on, really proved to be untrue. They basically forewent the benefits of interfering with Activin signal transduction and GDF-11 and chose to have a pure-play, safest state-of-the-art inhibitor at that time. Ultimately, knowledge progressed in the field, and it turned out they were really missing a significant piece of the story, particularly with respect to muscle growth by focusing on those mechanisms. I think they went with the science of the day, and we moved with the science of the times as well. Okay. Great. So we're about halfway through our time. Maybe it's a good point to sort of pivot to SMA then, Dr. Byrne. You presented a decent amount of data or update on the RESILIENT trial undergoing. Maybe just a point-blank question. Maybe define success. What do you need to see in terms of feeling that that was a successful study? Yeah, it was. Obviously, as a phase 3 study, it's a challenge in such a big study to show significant changes in motor functional outcomes across a broad age spectrum. But we do see that the new natural history of the treated disease does represent some domains that are not effectively treated that represent other aspects of weakness that are associated with SMA. So even stabilization of the MFM would be a win. I think there's a goal to actually show improvement. And as the study continues, we'll see if that bears out. But it is still a population that has made significant advances in treatment in the past 8 years, but there's really more to be achieved. So excited about it. Then Scholar Rock has had their phase 2 data. Their phase 3 will read out, I think, in the fourth quarter. Just maybe talk about maybe a two-part question. How does sort of the phase 2 data that they've already presented sort of de-risk taldefgrobep alfa, I guess, in your view? And then I've got a follow-up question on their phase 3. I don't know if you want to contribute to that, Lindsay. About the competitive landscape might be appropriate for you to comment on. Sure. No, that's a great question. Thank you for asking. There's a lot of preclinical data, obviously, to show that adding a myostatin inhibitor on top of an SMN up-regulator can improve muscle mass and function. The phase 2 data from Scholar Rock also really speaks to that as well and kind of reinforces that there's evidence in the field that this approach could be successful. Certainly, that needs to be borne out in phase 3 studies. We're really excited that our study is going to complete its double-blind phase. And top-line results will read out at the second half of this year. Great. And then I know you've kind of addressed sort of the population differences, but maybe just talk about the underpinning decision to enroll this broad patient population with ambulatory, non-ambulatory patients in the age group that you did. Maybe talk a little bit about that. Absolutely. So when we went to this, we talked to a lot of our esteemed thought leaders. It's really clear that there is a huge unmet need in this space sort of across the SMA population. Our goal was really to be very patient-centric in our approach and try to fill the unmet need across the SMA population. The times have changed with the advent of the SMN up-regulators. Patients are achieving milestones they would not otherwise have achieved. Dr. Byrne spoke about the newborn screening, which is exciting, that even in the last couple of years or so has now become 100% in the U.S. So early diagnosis, early treatment allows patients to achieve milestones. So someone who was an SMA type 1 before, early treatment may present differently, may look differently now. So it makes sense to really focus on the functional capabilities of the actual person sitting in front of you. We designed a trial with that in mind to really look at the functional capabilities and measure the functional capabilities and to try to hit the broad age range as well. As long as there's functional muscle, there's reason to believe that they could respond to a myostatin inhibitor like Taldefgrobep on top of the standard of care. Okay. And I think oh, yeah. I was going to add one other hidden aspect of the notion of early treatment. So the goal is, with newborn screening, when the patient's identified in the first few days of life, to implement one of these really disease-modifying therapies that have been transformative for SMA in the first 2 weeks of life. But amazingly, still, there's not a complete acceptance of the urgency because there's a mixture of type of 2-copy and 3-copy patients. There may be maternal antibody that's interfering with the eligibility for Zolgensma. So in some regions, the delay for treatment actually leaves patients without achieving the maximum benefit of an SMN up-regulating therapy. So that's another, I think, that we're grappling with across treatment centers to try to improve access. Some of those are logistical. Some are just related to the nature of the treatment of a gene therapy product. All right. I've got time for one more question. It's kind of a loaded question, and it's for the Biohaven folks. So maybe relative to sort of the how is SMA sort of differentiated from DMD, I guess maybe I get this from investors a lot. Why did Taldefgrobep alfa fail in DMD? And why do you believe it's going to work in SMN? That's an excellent question. DMD, also a very devastating disease. And at the time the study was going on, there were no disease-modifying therapies, and it seemed like the right place to go. And the pathophysiology with DMD is a totally different pathophysiology, right? You get the replacement of muscle with fibrotic tissue and whatnot. And these children were treated with high-dose steroids, which can also impact changes in muscle. And there was no disease-modifying therapy. So SMA, totally different, right? The muscle's intact. There's disease-modifying treatments, as we were talking about. These participants in our SMA trial are on standard-of-care therapies and not on high-dose steroids. And so there's reason to believe that this is the right disease as long as they're on the stable regimen of care. Adding a myostatin inhibitor on top of that, we've seen that in the preclinical studies, even with Taldefgrobep, that Taldefgrobep can make muscle function and strength improved in those models. We're really excited about that for the SMA population, again, totally different than the Duchenne population. Awesome. Okay. Well, tons more questions, but no more time. So thank you. Thank you very much. Thank you, Chris. Thank you, everyone. I'd like to invite our panelists and speakers for our final session, the degrader session, to join us. Okay. David, you ready? Yeah. Okay. It's my pleasure to introduce the speakers and panel members for our final session of the day, the degrader session. So without further ado, let me start and introduce our external experts who will be either speaking or on the panel. We have Dr. David Spiegel, who will be a speaker. I think many of you know Dr. Spiegel. Dr. Spiegel is Professor of Chemistry and Pharmacology at Yale. He is the progenitor of the degrader technology and really sort of the initial person who came up with the idea for using the ASGPR receptor and liver to degrade proteins, to piggyback on the natural protein degradation system, and use that therapeutically. And so we're very pleased to have David here to speak and talk about the platform. We're also pleased to have two other external experts, Dr. Dennis Moledina. Dr. Moledina is also an assistant professor here at Yale. He is a nephrologist. He's a physician scientist. And really, he runs a kidney bank here. He also is very interested in other translational activities focused on rare kidney diseases. So we're lucky to have him here. And then we've got Dr. Chip Howard, James Howard from UNC. He's Professor of Neurology, and he previously was the head of the neuromuscular division. He's a legend in neuromuscular disease, particularly in the area of myasthenia gravis, and has been senior investigator on major international trials in MG that have happened over the last several years. So pleasure to welcome you. And last, on our side, the internal expert who will be presenting is Dr. Bruce Carr, who is our Chief Scientific Officer. So with that, let me hand it over to David to give sort of the vision of the degrader. Here you go. Thanks. Thanks so much, Irfan. Yeah. This is a pleasure to be here and talk to everybody about the degrader program. We've spoken about this, I know, with several of you in the past. Just to overview, the way that degraders work is they are bifunctional molecules. That means they're two-headed. One side binds to a target, a degradation target, as shown here in the top sort of row, a top bar. And the other side binds to the acyl-glycoprotein receptor, abbreviated ASGPR, which is expressed at high levels on the surface of liver cells. And so what happens is we administer the MODEs. That's step one, as you can see, either subcutaneously or by intravenous injection. And then once in the body, the MODE interacts noncovalently with its degradation target, probably first, in particular, as shown here. It's an antibody target. But of course, we talk about this. We are going after other types of targets as well. One side then, so it binds to the antibody target, then the other side binds to the liver. That mediates endocytosis trafficking into the endolysosomal compartment and then degradation of the protein target. We can control precisely the degree of target degradation on the basis of dose and also by virtue of how we optimize these molecules based on their affinity properties for each side of the interaction and also through the linker length. So there's a lot of parameters. We have control over. We'll talk about this in a moment. Let me just see if I can switch to the next slide here. Oh, wait. What did I do? No. Okay. Thank you. So here's a close-up of the structure of mode degraders. They are, again, as I mentioned, bifunctional, two-headed molecules. One side binds to ASGPR, the asialoglycoprotein receptor. The other side binds to the target, and then they're separated by a linker. You can see this sort of bifunctionality enables very rapid and modular design. We can mix and match different components to get to optimal properties very rapidly. That's exactly what we've done at Biohaven to arrive at our lead candidates we're going to be talking about a little bit later. What we found in the studies that we've done so far is that the compounds that we've developed, in particular, BHV-1300, which targets total IgG, IgG isotypes one, two, and four, efficiently can remove those proteins. It has a very rapid onset. It is selective, or the modality can be quite selective, but in this case, it's selective. What's really exciting is that because of the mechanism, we can co-administer with Fc-containing biologics, unlike FcRn inhibitors. Also, this modularity allows us to accelerate IND timelines substantially. The development of these compounds, or I should say the discovery phase in this process, is quite rapid. Just to give a little overview of comparison or the competitive landscape comparison with other technologies, the IgG degraders, that is, the pan-IgG degraders, the BHV-1300 and related compounds, the autoantibody degraders, had these incredibly rapid cycle times, I think, far faster than conventional small molecules as well as biologics. They work very rapidly because of the mechanism. They bring proteins to the liver very quickly. We see effects within hours, whereas it takes several days for the FcRn inhibitors, for example, to function. They can mediate very deep levels of lowering, equivalent to the FcRn inhibitors and other modalities. Importantly, we can co-administer with standard of care. Again, because the point here is somewhat subtle, our molecule is gone very rapidly, but the pharmacodynamic impact of the molecule persists long after it's gone. So we are no longer operating in the occupancy paradigm for these molecules. The drug doesn't need to be there to work. And so I like to say that we're liberated from the tyranny of occupancy, right? The drug can be gone but still function. And so the beauty of that is we've degraded the IgG, the natural IgG. The drug goes away. Then we can administer the standard of care agent, the biologic, without any expectation that the presence of our drug is going to impact that. And then finally, when we talk about the autoantibody-specific degraders, we can achieve the level of efficacy without any immunosuppression. So the platform currently is positioned nicely for sort of rollout toward the future. Again, as I mentioned, the platform, very rapid cycle times, very straightforward optimization. We're now developing the first-generation pan-IgG degraders that can remove IgG 1, 2, and 4 isotypes of IgG antibody. We can co-administer with biologics. And then we're also in the process of rolling out the second-generation antigen-specific autoantibody degraders, which can essentially incorporate autoepitope mimics that drive the degradation of only those antibodies that are causing disease. These are not immunosuppressive because you leave the majority of the IgG repertoire intact and can be developed for a whole range of different autoimmune diseases for which, in some cases, the autoantibody epitope is identified, and then in other cases, even when the epitope is not known. And so this platform then is poised for all kinds of advances moving forward. We're in the process now of working toward catalytic degraders, degraders for whom one equivalent of the molecule can mediate multiple degradation events, degraders that can function for short half-life proteins rather than long half-life proteins like antibodies, degraders that function through other mechanisms other than ASGPR. There are lots of endocytic receptors that we're looking to exploit, although we believe ASGPR has a lot of advantages, degraders that can function to degrade cell surface receptors. We're also working to accelerate and optimize our modular design strategies, optimize the manufacturing and scale of these compounds, and then finally, working to make orally available degraders because, again, these degraders are built on a synthetic sort of small molecule-based platform. So with that, I'd like to turn the platform over to Dr. Bruce Car. Thank you very much, Dr. Spiegel. It's a pleasure to be able to talk to you about one of my favorite topics, which is degraders, and it's really a privilege. Vlad communicated something of the excitement that we're all feeling at the moment about this presentation today, but also about this drug platform that we've built at Biohaven. I'd just like to acknowledge Dr. Spiegel and his team, past, present, and future collaborations that have really been very important in the genesis of many of the ideas: the Biohaven discovery team, non-clinical development, clinical development, the CMC groups already mentioned, really a tremendous teamwork event that have collectively created a platform and an engine for creating a large repertoire of medicines. I'll be talking about nine of those today. BHV-1300, of course, is the medicine du jour, but the repertoire of these medicines actually very significantly bridge gaps in the standard of care to tremendous improvements going forward. I hope we communicate some of that well today. Where are these diseases coming from? There are actually new diseases in this space that can be treated by degrader platforms are unfolding on an almost daily basis. In fact, the term idiopathic, which in pathology means unknown origin or unknown etiology, these diseases are being increasingly deorphaned. The cause for these diseases is being found to be an autoantibody. A proportion of diseases such as schizophrenia, autism, for example, a variety of encephalitides, including LGI1 encephalitis, long COVID, even aspects of aging with autoantibodies directed against adipose tissue, fibromyalgia, the more severe 50% of fibromyalgia patients, a disease ascribed to schizophrenia, sorry, to psychosomatic causes, has in fact been shown due to autoantibodies directed at epicytes in the dorsal root ganglia. So continually, new autoantibodies are being discovered, and there's an abundant literature that the titers of those autoantibodies directly correlate with both morbidity and mortality of all of these diseases. So these are antibody diseases. We're parlaying these antibodies to our advantages. So what do we do? So we take that autoantibody. The antibody itself is something of a diagnostic. We measure the levels of this antibody as the levels of this antibody disappear from the circulation with the degraders. We've got a pharmacodynamic biomarker. We use this antibody to find exactly what it's binding. If we know exactly what it's binding, we can make a degrader. To Dave's point, with our modular system, we can make a degrader. In the very best-case scenarios, go from discovery, conception, to IND within 12 months. We have to be a bit lucky. With the profound reversal of the effects due to these autoantibodies and I'll just take one example, LGI1 encephalitis. This is a disruption within synaptic terminal in the brain due to an autoantibody. You take that autoantibody out, and theoretically, you stop the incessant seizures that these patients have. You've got something you can measure quickly, seizures. The ability to move autoantibodies very quickly to proof of concept and then early registrations is afforded by this platform. We're really quite excited by it. I'd like to talk to you about our portfolio today. Including BHV-1300, for which the IND is approved, and I'm going to share some data with you shortly, we plan to have four INDs this year coming out of this platform for around. different diseases, 1310, 1400, 1600. I'll describe them separately later on. We're unveiling a group of four and actually a fifth, potential IND for 2025, new targets for which the degrader technology is exquisitely well-suited. We'll be talking about how we selected and found those drug candidates and even some proof of concept that we've established for these and the tremendous excitement around a new portfolio for which now we have nine shots on goal. Behind all of these, we're able to design these and thoughtfully select molecules. We have a detailed backup strategy. For all of these molecules, we have backups that may be employed across various price points if there are many indications and as contingency in case of attrition. And when you look at that number, you say, "Oh, will I multiply this by 8 or 9 diseases?" There's several of these that represent pipelines within an asset. So actually, there's a much larger number of indications that these 9 assets could potentially treat. So before we move into the BHV-1300 clinical trial update, I'd like to explain to you some of the more recent data that we've generated with this that really underpin some of the real advantages that you'll see for BHV-1300 in patients and for degraders generally, that they're uniquely differentiated in ways that FcRns, immolipidase, and other immunosuppressive therapies cannot do. Really important, removing immune complexes. There are no other mechanisms for directly removing immune complexes. Immunosuppressants can maybe decrease the components and decrease the probability that these complexes will form, but they don't remove them. So what is an immune complex? An immune complex is when one antibody binds another antibody and antibody binds another protein. Or, for example, an antibody binding a drug can be an immune complex. There are a variety of shapes and sizes. But when things go awry, when you have a pathogenic immune complex and it settles out in the synovium of a joint capsule, in a blood vessel wall, in the glomerulus of a kidney, you develop strong complement-mediated and cell-mediated inflammation and immune-driven events in those tissues that eventually lead to organ dysfunction and in some cases, death. So these immune complexes are really bad. No pharmacology has directly addressed them in the past, short of actually filtering them out in the blood, which is really not very effective. And we have a mechanism, and we've shown in the graph on the left, that we can take out the very largest of immune complexes. This is not rheumatoid factor. Rheumatoid factor is an IgM, IgG immune complex. This is simply to show that we can take out an immune complex of over 1 million daltons through the ASGPR system in hepatocytes. This is really cool and impressive. In the histology in the middle, you can see in the top panel is a vehicle dose. And if you look carefully, the yellow dots are basically filled with the immune complexes that have been internalized from the sinusoids of the liver. So we're able to remove these extremely large immune complexes and then just as quickly degrade them in the liver. And this is then shown graphically on the right-hand side. So this is a tremendous advantage, a differentiating advantage. No other drug platform today is able to do this. Another very important feature, I think, is the depth of the lowering and the speed of the lowering. So taking it completely out of the system very quickly is able to not only deplete what's intravascular but also deplete what's extravascular. You can see on the top left here is the depletion of pathogenic antibody that's really swimming around muscle fibers. And what you can see blown up just next to that, you can see here, is a neuromuscular junction. And the pink stuff surrounding that, just imagine these are autoantibodies against the acetylcholine receptor driving myasthenia gravis completely cleared out by a drug like BHV-1300. So really, a tremendous advantage. And in the kidney, this was even a little bit of a surprise to us. This is when pathogenic antibodies enter the kidney, they diffuse into the glomerular mesangium. Immune complexes diffuse into the mesangial matrix, the glomerular basement membrane. They get stuck. They cause all sorts of things, completely removed by a pan-IgG degrader. So just to emphasize how well we can deplete pathogenic antibody from extravascular tissues and remove immune complexes. Dave already mentioned the advantage that we have in being able to co-administer based on very rapid clearance. And we've shown this data before on the left-hand side where we're able to co-administer with Humira but not affect the exposure of Humira at all if that gap is only 12 hours. It's actually reasonably effective. The green line there is 2 hours. And at 6 hours, we're completely safe. 12 hours, you're absolutely certain. But what else happens with virtually all biologics, whether they contain FCs or not, whether they're antibodies or whether they're proteins, is you develop a level of immunogenicity. So what happens then is the body recognizes the biologic as a foreign protein, makes an antibody to it, and then a proportion of patients and in drugs like Humira and Remicade clear out a significant amount of that drug through the clearance mechanisms of antibodies. So the drugs just basically get filtered out from the spleen and stop working. These are called antidrug antibodies. So the picture in the middle here shows BHV-1300 very effectively removing antidrug antibody to Humira, adalimumab, in a monkey. And these are enormous titers. These are much higher titers than would typically appear in humans dosed with Humira. And very importantly, with the removal of the antidrug antibody, the function of TNF, ability to be neutralized by Humira, is restored. So this is a boost on top of taking out damaging proteins and so on, damaging antibodies like ACTAs and rheumatoid factor and so on in the context of a disease like rheumatoid arthritis, being actually able to help other medicines in this regard by removing antidrug antibodies. And I'll show you an angle for a very important medicine based around just this philosophy later on. Okay. And the last of these differentiating characteristics is something that I thought I'd share with you. The pharmacology of degraders is not terribly well understood. And generally, when you think of administering a dose of a compound and administering another, it's just to maintain those levels exactly over a period of time. Or if you want to get more bang for your buck and you administer it again, generally, it's at a cost of some toxicity. But with the degraders, if we administer the same amount of drug twice and in this case, you can see we've administered two sub-efficacious doses of pan-IgG degrader. And essentially, for the first sub-efficacious dose, the first dose, we take it down 30%. And for the second dose, down 60%. And if we increase the dose to 30 mg/kg, you see it going from 40% to almost 80%. So this is part of the pharmacology of these degraders so that they can be given consecutively and almost double the efficacy. So a huge advantage. So this is why we're all here. We're really excited to hear the results of this. I'd like to say, as these data were starting to unfold, and this is an open-label study, we've been looking very closely at the data, the confidence that this has given us for investing in the entire portfolio and to bring forward the repertoire of medicines beyond 1300 has been absolutely secured by the data that we've seen. So we've got great confidence that this platform is going to be transformational as a new platform for creating medicines. So the preliminary results from the ongoing first-in-human study have been completed with 4 cohorts involved in that. The fourth of these cohorts actually was subject to a very intense analysis yesterday. There were enough days to have good data. We're all sitting not far from here in Church Street looking at all the data and preparing it for presentation today. So summarizing this overall, we led into the clinical trial with the classical IND into toxicology species from which we drew both pharmacology and toxicity. We understood that the drug was very, very safe. And from the monkey, we were able to see very good pharmacology. We were able to make projections from the monkey in terms of pharmacokinetics, drug clearance, IgG, and so on, and then set up the doses in the clinical trial quite precisely. So we initiated the single-dose study. We completed the fourth cohort. And the data here take the fourth cohort out to 96 hours after that injection. This was a study for which these are standardized doses. They're non-weight-adjusted doses. So you should be aware of that. And I'm going to show this in two different ways. This is with averages and the next with median, which is more appropriate for such a non-weight-adjusted dosing regimen. So we saw dose-dependent sustained lowering throughout the follow-up period for particularly the latter of the two cohorts. The little pink stars that you can see throughout these are what we built in terms of our understanding of the preclinical pharmacology as projections for human lowering. And they were absolutely spot on. So based on this modeling, we anticipate achieving an 80% IgG reduction when this phase 1 is complete. And in fact, within those next two cohorts, we expect to see levels of lowering of that degree. And I'm going to illustrate that on the following slide. Is it advancing? This has stopped working. Sorry. Can we change the battery? No, so get back. All right. right. That's working. All right. We're good. Okay. This slide illustrates the same data as the previous slide with medians instead of average. You can see the very clear dose-response relationship for the fixed or non-weight-based dosing of all the cohorts down to a reduction level of about 43% or exactly 43% at the highest dose of the cohorts that we've shown here. These data, I've already mentioned, represent median values. This gives us great confidence in terms of our modeling as to where we can go in future, not only modeling what the next two SAD doses are going to be but also the multiple ascending doses. We have a very clear idea based on the pharmacology where this will go from here. So what is the context of all of this? We have a very safe drug. It was designed not to interact with any other immunoglobulins. We saw no meaningful changes in IgM, IgA, or IgE, the other subclasses. To understand how this compared to the FcRn mechanism where there's a mechanism-based hypoalbuminemia and resultant hypercholesterolemia, we, of course, as part of the normal safety labs, evaluated albumin and LDL cholesterol and saw no meaningful changes in those as well. So these are really very much expected results based on this mechanism. So far, the profiling has been exquisite. No SAEs or severe AEs. And those AEs that were mild were not related to drug and resolved spontaneously. This is very closely recapitulating what we saw in the non-clinical toxicology studies, which were quite clean to extremely high doses. So very thorough testing there and recapitulating the absence of ECG findings in the animals. We saw none in humans. We didn't expect to see any despite this being a small molecule. We saw, in fact, no clinically significant drug-related lab changes. But there have been very many questions about the pharmacology being mediated through the ASGPR, most of which is expressed in the liver. We looked at any evidence of hepatotoxicity, of course, extremely carefully in our animal studies. We saw none at very large margins of the highest exposures that we would evaluate in humans. So the drug clearly had a low potential for hepatotoxicity. We examined it. We saw no hepatotoxicity. We saw no clinically significant changes in LFTs. And that is, we saw no changes in alanine aminotransferase, aspartate aminotransferase, in conjugated and unconjugated bilirubin. None of those parameters shifted. This is actually not surprising. Alnylam have helped us by targeting the ASGPR receptor on the liver for a number of different drugs where it's important to deliver siRNA to the liver and haven't seen evidence of hepatotoxicity. We haven't seen evidence non-clinically. So this, I think, is not surprising at all that the profile to date has been very clean. And of course, looking very carefully for hypersensitivity, infusion reactions, those sorts of things, also nothing in that regard. So summarizing the efficacy, I'll go through this slowly. The dose dependence of the response was very clear. The rapidity of onset was particularly important. And I'd like to reiterate that in 96 hours, having a 43% reduction is something that took more than 10 days for most of the FcRns. In fact, our projected single-dose lowering of 70%-80% is something that takes multiple doses of the FcRns, in fact, 4 doses, in which case that nadir is only reached between 21-25 days. This has already told us we've got a rapid onset of action. We have the potential based on our modeling, which we've nailed so far, to achieve those low levels for efficacy. We're really excited about this. The doses that we're seeing, the behavior is still absolutely consistent and compatible with subcutaneous administration of this drug. Of course, we're doing all of the necessary GLP work to set up further clinical trials with subcutaneous administration. That's work that's ongoing. In terms of selectivity, no meaningful reduction of other immunoglobulin isotypes or impact on the elements of mechanism-based adverse events that have been seen with the FcRns. Not surprisingly, overall, very safe and well-tolerated, no noticeable events at all with infusion of the drug and no hepatotoxicity or clinically meaningful changes in transaminases, bilirubin conjugated or not. So I'll just pause there for a second to let all of that settle in. This has given us a super amount of excitement. And it's the first of our INDs for this year in degraders. Seeing these results basically told us that these other targets are going to be a slam dunk because in terms of going after pan-IgG degraders, that is the largest target burden of any degrader one could imagine. There's the highest concentrations. There's the highest molarity. You would theoretically need the highest concentration of the drug. We do it with modest concentrations. But any one of these, on average, is one-fiftieth of the dose of 1300. So for example, this next to BHV-1600 is a next-generation mode. And all of the subsequent drugs I'll be speaking about are next-generation modes. This one's targeting beta-1 adrenergic receptor antibodies. Those autoantibody levels are about 2% of the IgG levels overall. So this is a totally cool target. This is innovation that Dr. Spiegel got us kicked off with. And the IND is on track for the second half of this year. How does it work? So it's most unfortunate but agonistic autoantibodies. These are antibodies that actually mimic the action of adrenaline and noradrenaline on the receptor, beta-1AR, that's on the myocardium and cause its incessant activity. That means rapid heart rate, increased cardiac contractility, increased output but without stopping. So if you just imagined running a mile as fast as you could and then another and another and another, eventually, and this is what happens in animal models and in people, you cause myocardial injury that results in dilated cardiomyopathy, the incessant activation of the heart. The treatments for this, plasmapheresis, beta-blockers, are entirely inadequate to the treatment of this disease. Okay. So what I'm illustrating here is some of the most beautiful drug discovery that we've been done so far. And just for this particular drug, so this particular drug, so I don't show it multiple times, there's a number of different parameters we evaluate. So what I'm showing here, this little mountain, it basically shows that we have a stable ternary complex. That means when the whole thing binds together, it stays together with a very high affinity. You can see this is by SPR that we have a tight binding of the drug. Then we go in vitro and see that this tight binding, high affinity complex is taken up in cells. So if all of that's intact - and of course, we do that with six or seven different drugs - we then test this in vivo. And you can see on the left-hand side the extremely rapid removal of an anti-beta-1AR autoantibody in mice that's been injected into mice and then very rapidly removed by the drug. So this is a drug working exactly as it should. Oops. Stopped working again. Can you help me advance again? Oh, there we go. We're up. So we're good. So the question is, do we reverse the disease pathology that results in dilated cardiomyopathy? This shows this exquisitely well in an experiment in rats where the autoantibody to the beta-1AR, so mimicking adrenaline and noradrenaline and just driving that heart, the green boxes at the top are the effects on blood pressure. So we saw these effects on blood pressure, heart rate, contractility. And you can see the bottom two lines, it's basically the pink, BHV-1600, reducing the pathology induced by the autoantibody really back to baseline levels. So this is exactly what we hope to see in patients, a complete normalization of cardiac function by removing the autoantibody. So how does one assess this in a clinical trial? And I'll be showing cameos of how this might appear for these early drug candidates moving forward. For a disease like beta-1AR, the prevalence of these autoantibodies is really quite low. So we can get an but there is some. So we can get an understanding in single-ascending dose study in healthy volunteers, what the removal of those autoantibodies looked like, just very standard trial. But then moving to a proof of concept, there are a number of diseases within the family of dilated cardiomyopathies that have an extremely high incidence of these autoantibodies, as high as 90%. So recruiting those antibody-positive patients should be readily doable. The endpoints associated with those changes in cardiac function are readily accessible, those hemodynamic endpoints, and biomarkers that are associated with those endpoints and outcomes, things like proBNP, looking at ejection fractions and so on, six-minute walk tests. Overall survival, of course, becomes very important in the longer term and composite outcome endpoints. So a registrational strategy around biomarkers that are likely to change very quickly, with the most important biomarker of all of those being the removal of that specific autoantibody, can be harnessed to an accelerated registration, we believe, for programs like BHV-1600 and other programs. So there's two parts to this that you'll see in each one of those. The first is the biomarker. You remove the antibody. You go down. The second is a functional change that can be precisely measured that's associated with disease progression and that we show a reversal of that. These two things together parlay into a substrate for rapid registration. Okay. Okay. So BHV-1400 is another specific degrader that is on track for its IND this year. So this is a biologic degrader. This degrader was designed specifically to remove galactose-deficient IgA1, very specifically this species, and leave behind the rest of IgA1, IgA2, all the other subclasses and isotypes. This has been designed. This is on track for development. Importantly, the biomarker of the presence of the antibody itself, on the left-hand side there, you can see that the titers of this autoantibody directly correlate with morbidity and mortality within this disease population. Changes in that biomarker will be very important. Registrationally, some Biohaven work in the next column in IgAN patients and showing that the levels of those antibodies are 3- to many-fold higher, on average about 3- to 5-fold higher than healthies. Interestingly, there are a number of other diseases and renal diseases where these autoantibodies are increased as well that we could potentially address. This is an immune complex disease. I talked to you about the importance of the only drug class that's able to actively remove these immune complexes directly. What happens is that autoantibodies, IgG or IgA, recognize this Gd-IgA1 - so that's the bad actor - and these complexes settle out in the renal mesangium, in the glomerular mesangium. What this does is create a lot of inflammation in the kidneys and ultimately leads to renal failure and death or the need for transplantation. So we believe by taking out Gd-IgA1, we can avoid that progression entirely. And in fact, we showed that. Here in these slides, you can see the middle panel is the histology where all of the red is the autoantibodies - sorry, the Gd-IgA1 - floating around in the sinusoids. And on the right side, it's been all sucked into the hepatocytes by administering BHV-1400. That quantified nicely on the bar graph to the left. This is the same for these immune complexes. These are the immune complexes that actively drive the disease. You can see the uptake in cells in the graphic on the left, the colocalization with the LAMP lysosomal marker, in the right of the two middle histologic shots, so really clear internalization, which is then quantified in the bar graph on the right. You basically take out those immune complexes very nicely from the circulation and degrade them quickly in the liver. Getting rid of these immune complexes is the one thing that will most quickly slow down the rate of decline of renal function in these IGAN patients. That's an advantage that none of the other potential competitive mechanisms, which are all immunosuppressants basically, would have. So we believe there's potential for superior efficacy because we're targeting the upstream pathology. We can measure both Gd-IgA1 reduction, but we can also measure reduction in the immune complexes. We believe based on recent advances in this field where we're starting to see remission with APRIL/BLyS inhibitors, which take something like 13 weeks to decrease IgA and Gd-IgA1 to levels where you start to see improvements in glomerular function. And if we're able to do this within a day or two, that stabilization and improvement in eGFR, the glomerular filtration rate, should be very significant in providing us with early proof of concept and provide then a segue for subsequent studies for approval in which we would look both at eGFR and the urine protein-creatinine ratios that are used in the registrational trials for IgAN. So we believe a much more rapid onset of action can parlay into early registrational strategy for this particular molecule. We're very excited about this. This one's going forward fast. I mentioned there's other diseases. Chronic transplant rejection can be due to these IgA antibodies. Docs reach for something. If you're told in 13 weeks your IgA are lower, well, you're kind of out of luck. This is a potential for a rescue mechanism for such cases as well. One weekend, I was sort of texting backwards and forwards to this lab, and what do you think about this? This actually evolved into a research program. I had mentioned earlier the importance of antidrug antibodies in disease. Type 1 diabetes. Type 1 diabetes, insulin-dependent type 1 diabetes, occurs in 300 million people worldwide. 0.5% of the U.S. population are type 1 diabetics. That's close to 1.75 million people in the U.S. are dependent on insulin given once or twice daily for their survival. In a proportion of those patients, they develop autoantibodies to the insulin. It's a foreign protein, just giving you an idea of where this can go. But what we're really excited about with this disease is the ability to potentially halt, produce remissions in, or even cure type 1 diabetes as it's developing in children and adults. In 100% of children, up to about the age of 20 from infants that develop type 1 diabetes, the very first thing that happens is they develop an autoantibody to proinsulin. That autoantibody binds the beta cells in the pancreas, causes inflammation of those beta cells, an inflammatory process called insulitis, and ultimately complete loss of those beta cells. That's the bottom right panel here that you see. When you're in that situation, you basically need exogenously administered insulin once or twice a day. We believe we can make, and we don't just believe, we have made a degrader that can remove proinsulin and potentially halt this process. This is one type of a disease. If you're a type 1 diabetic and you've received insulin for 20 years and you start to have higher and higher levels of blood glucose and you have higher and higher levels of insulin administered and no effect and you develop ketoacidotic crises and other metabolic crises, coma, and more and more insulin not working, you switch to another sort of insulin. It works for a little while. Then it stops working. This is what happens when one develops autoantibodies to insulin. And this is not infrequent. In 300 million worldwide, 1.5 million in the U.S., a large number of patients go down this path. And in a very facile way, we're able to make degraders that remove those autoantibodies that recognize insulin and remove those from the circulation and after doing so, restore normal glucose homeostasis. So that's the hypothesis. And we've shown that that, in fact, can be done. So very similar design to previous. We assessed the binding affinity of a number of molecules. I'm just showing you the example of insulin here. We've done this for insulin, proinsulin, insulin peptides that are important as well. So there's actually three drugs that we have within this family. I'm just showing you a single example. We looked at cellular autoantibody uptake, showed it was really solid, as good as any of the drugs we've seen, and then in the mouse model showed that we could effectively remove—that's the right panel—effectively remove this anti-insulin antibody from the mice. And we had to do some very elegant protein design here to make sure what we were doing to capture the insulin didn't do anything silly like bind the insulin receptor or IGF1R and so on. So a lot of thoughtful protein engineering and small molecule engineering and manufacturing engineering had to go into creating something like this. Ultimately, it was successful. And this is a totally cool experiment that actually happened about 500 yards from here. Anna Bunin is one of our young scientists. She was working on this. And she noted that insulin autoantibodies occur in non-insulin diabetic - sorry, non-obese diabetic - mice. And the structure of mouse insulin and human insulin is actually close enough that those autoantibodies cross-react. So we said, "Well," Anna said, "maybe it'll cure these mice." We're not in the business of treating mice, but this was a great potential proof of concept for pharmacology. So a very complex assay was developed to find those mice that had the highest titers of autoantibodies. We selected those. You can see there, we grabbed 5 or 6 mice with the highest titers, then gave the—this is a highly optimized drug, 65/26—and lo and behold, it removed about 80% of the insulin—sorry, the insulin autoantibody—from those mice. So this basically proved the concept of what we could do. And then we would hope this would then restore normal glucose homeostasis to exogenous insulin. So a potential cure for those patients unable to take insulin anymore because we can take the autoantibodies out. So how does one do a clinical trial for something like this? Of course, type 1 diabetes is very well studied. In this case, the presence of those autoantibodies in fully insulin-dependent chronic patients allows us to do a single ascending dose study in those insulin-dependent patients and let us look at the lowering of those autoantibodies. So we get direct pharmacodynamics of the disease antibody in the SAD study. Then we're able to segue that quickly to a proof of concept study. There are two different sorts of proof of concept studies. One is just simply restoring insulin efficacy when one is completely dependent. And that's just measuring blood glucose and hemoglobin A1c and so on and so forth. There's a variety of different parameters of normoglycemia or insulin homeostasis that lend themselves directly to registrational program there. In type 1 diabetes, this is about diagnosing, and these happen and are diagnosed readily, 300,000 new cases of this occur in the U.S. over every five-year period. So we ought to be able to find some of these patients pretty soon. Parents are very aware of when children first begin to develop glycosuria. It's something that mothers are incredibly sensitive to picking up on and generally follows a viral infection. Sometimes it's not understood why. But if we're able to find these patients as they're first expressing the proinsulin autoantibodies and we can take that proinsulin out, we can potentially slow down, cause remissions, even potentially cures in some of these patients with endpoints such as directly lowering those autoantibodies, looking at C-peptide levels, the AUC for exogenous insulin requirements, glycemic control, and so on. So how one would conduct such a proof of concept study, I think, is very straightforward. And leveraging that then to a registrational paradigm is quite straightforward as well. So we think just a weekend texting between the lab and myself may have come up with something that could be a transformational medicine. This is just an example of one of many of the different sorts of cool things we can do with this particular technology. This is another pretty cool one. I mentioned earlier on that a number of these drugs that we were creating potentially form an asset within a pipeline. I'd just like to give a shout-out to the protein engineers and the small molecule chemists that designed this structure. It is really cool and neat. A number of years from now, when you all get to see it, you'll realize how cool and neat it is as well. This is the highest level of science to produce an IgG4 specific degrader. So you already know 1300 and 1310 take out IgG1, 2, and 4. This just takes out IgG4. As it happens, there's a series—okay. Go. Could you? I think the battery's dying gradually. There's a series of different IgG4-mediated diseases and some IgG4-associated diseases. Let me just right-click on this. Okay. Thank you. Hey. Developed very similar to the others. You can see on the left-hand table affinity for IgG4, high affinity, completely missed one, two, and three. So it did exactly what it should do. It was taken up very well in cells. And then we gave this to mice that it had been administered human IVIG, so carried all of the different isotypes. Just the IgG4 was very quickly eliminated from those mice. So this is something that could go forward very, very quickly and be a drug for IgG4-mediated diseases. And there are many IgG4-mediated diseases. You can see here, there's no rhyme or reason as to why they would occur. They tend to be skin and central nervous system diseases that are the most important amongst them. Also, very importantly, many antidrug antibodies are IgG4 specific. We feel you can take out IgG4 very safely. So it's the one of all of the four immunoglobulin subclasses for which if you take it out, you don't expect to see any immunosuppression at all. Okay. And these are three examples of those diseases. There's a subset of myasthenia gravis that only indirectly acts through the acetylcholine receptor. We're able to remove the IgG4 that does that. For pemphigus vulgaris, pemphigus foliaceus, pemphigus erythematosus, there's a family of proteins called desmogleins that hold keratinocytes, their skin cells, together. And when the antibodies bind these, skin cells dehisce and they form blisters. And neutrophils influx into that space and you see clefts throughout the skin. That's specifically IgG4. Who knows why? But we believe we have a mechanism to potentially fix that. And I mentioned this as an example very early on. Anti-LGI1 encephalitis, there's actually quite a bit of research going on in the Yale Medical School on this particular disease where this autoantibody disrupts an interaction between ADAM22 and LGI1 in the synaptic cleft and results in really an extreme seizure situation and poor neurocognitive - really profound neurocognitive effects - for which if you're able to specifically take out this autoantibody and not affect any other part of the immune system, should have a registrational path just based on we took out the autoantibody and we stopped the seizures. And you can imagine that that cascades very quickly into an early registration. Okay. So we already talked about one renal target. That's IgAN. There is a specific autoantibody that's directed against PLA2R. This is an autoantibody that's present in the middle largest circular panel. It's present on the surface of a renal cell called a podocyte. It means a foot cell. What happens is autoantibodies bind these cells, kill them basically. It's most unfortunate. What that cell does is it maintains the glomerular basement membrane. When your glomerular basement membrane is gone, you filter protein into the urine and you ultimately die of renal failure. It's a very, very specific autoantibody. The disease levels, morbidity and mortality, profoundly correlate with the presence of this autoantibody. We know how to specifically take out just that autoantibody and have made a degrader to do so. Same things. High affinity, removes from cells, works well in mice, and we're off to the races. So each one of these three that I've described - or four that I've described - actually have a very sort of similar paradigm. And we've moved into evaluating these for tolerability and toxicity. And we're projecting 2025 INDs for all of these three. And the last but not least of these, and interestingly, Dr. Barry Byrne in our last session, who was talking about the difficulty of poor uptake of gene therapy, things such as Zolgensma, when there are autoantibodies against AV9, you're only able to administer a gene therapy agent once. 50% of all of us carry autoantibodies to AV9. So not everybody's even substrate for gene therapy. And what very frequently happens is you have a dose of gene therapy, the expression is maybe 30% of what you would like, and the disease is really not reversed to the extent that would be ideal. So being able to top up the expression in gene therapy is critical, but you can't do it because after one dose of gene therapy, there are lifelong high-level autoantibodies directed against AAV9. And so as we would - and this also happened in the lab - that it's not 500 meters from here. We looked to find the recognition sequences on the capsids of AAV9 that are recognized and could be turned then into a potential degrader. We showed we could internalize these autoantibodies against AAV9 with this structure. And then on the next slide - and I'll see if this is working. There we are. Here we basically showed that by removing those autoantibodies, we restored the ability of the AAV9 virus to infect those cells and then potentially allow a second generation of gene therapy to occur. And in talking with Dr. Burns last night, it was very clear this isn't a one-shot trek. If the adsorption of the virus into cells is something that happens over a couple of weeks, and actually keeping down a high-production antibody for a couple of weeks really takes a bespoke medicine like I've described here to work. So we believe by giving something like this ahead of gene therapy, we can render this platform that's had such promise that really only works a bit in 50% of patients a much better disease platform as well. Okay. So this is my last slide. I've presented the excitement that we have bringing BHV-1300 forward, the clinical data that confirms for us after the fourth cohort, still a relatively low dose, that we're able to achieve 43% lowering and that that lowering was sustained as we expected it to be. That told us that all subsequent proteins - all subsequent programs using this technology - are quite doable if we can find that right binder from the library and that right ASGPR ligand that David talked about earlier. So we have our pan-IgG platform. We have very smart backups beyond these two. We have the next generation antigen-specific targets that I talked about today. And in addition to this, we're able to remove ADAs. We can target pathogenic proteins. They don't have to be autoantibodies. We can target transplant rejection. We can work with combination therapies. This is a drug therapy that's tailored very well to improving standard of cares, replacing standard of cares. A great drug like Humira that isn't working so well, we can improve its action. And at the end of the day, this path is leading to transformational immunotherapies for a variety of different diseases. I talked about improving the standard of care. What we really hope is this will lead to remissions and in some cases cures for a proportion of these patients. Thanks very much for your attention. Thank you, Bruce. Truly transformational. I want to welcome Tessa Romero, who's going to be moderating our final session. Welcome, Tessa, from JP Morgan. Thanks, everyone. Thank you so much for having me to round out the day in what is hopefully a comprehensive 25-minute discussion. We learned a lot about the progress with the MODE platform at Biohaven over the course of these prepared remarks. I'd love to kind of bring in the clinician perspective to start here. Dr. Howard, thank you so much for being here. Are you able to share a little bit of background on yourself? And do you think the pan-IgG mechanism could be advantageous in the context of the current treatment paradigm in MG? In other words, which aspects are you looking to see improved upon? Thanks for the opportunity to speak with all of you. Myasthenia is undergoing a transformational change in terms of our treatment. Yet we're still burdened by the burden of treatment and the burden of disease. In the pan-IgG program, I see a nice fit for those individuals who acutely deteriorate into either myasthenic crisis defined by ventilatory assistance or rapidly declining for a number of reasons - infection, drugs, etc. We know that over the course of a lifetime, up to 30% of myasthenics have at least one crisis. And in our REGAIN trial with eculizumab, nearly 25% of the patient population had crisis within two years of study entry. So it's not an infrequent situation, to say nothing of the acute exacerbation. None of our therapies to date work fast enough. These individuals end up being hospitalized, undergoing therapeutic apheresis for intravenous immunoglobulin, and some places in the world, high-dose corticosteroids to turn this around. So I see a beautiful niche if one could have instantaneous, essentially, cleavage of an immunoglobulin molecule and clearing it. We know from FcRn therapeutics, we know from therapeutic apheresis, the clearance of antibody reflects clinical improvement. And so this is where I see this working out. Yeah. So maybe just you could remind us how well correlated the degree of autoantibody reduction is with clinical outcomes across various neuromuscular and neurological diseases or other autoimmune diseases. For myasthenia, it's a little bit difficult. The absolute value of antibody bears no correlation to the severity of disease. I have patients with 1,000 nanomolar who are in remission and 0.02 nanomolar who are on a Ventolin. Yet within an individual, the clearance of an antibody, that lowering of antibody clearly correlates with improvement with interventions like FcRn therapeutic apheresis. Not so much if we use our SOCs of methotrexate, mycophenolate, azathioprine, etc. Dr. Moledina, I'm not sure if thank you also for being here. I'd love if you could also give a little bit about your background and how do you think about the MODE approach being used for specific autoantibody degradation and its applicability? What do you see as the potential advantages and any possible drawbacks? Thank you for the opportunity to come and talk to all of you. So I am a nephrologist here at the Yale School of Medicine. I see patients with kidney diseases at the Yale New Haven Hospital. I'm also a physician scientist, and much of my work is around translating some of these basic findings into bringing them to our patients. And in that role, I've enrolled participants in this Yale Kidney Biobank that Dr. Qureshi had mentioned with rare kidney diseases. The first kidney drug that we heard about was targeting IgA nephropathy. There's a lot of excitement around IgA nephropathy. A number of newer drugs are either in the pipeline or have recently been approved for IgA nephropathy. IgA nephropathy is the most common form of what we call primary glomerulonephritis, primary diseases affecting the glomerulus. And yet it's fairly rare, about 4,000 or 5,000 patients a year in the US. Most of the drugs that are either approved or are in the pipeline or are part of standard of care are immunosuppressants. These are basically drugs that suppress the immune system broadly. There are problems with that - risk of infections, bone disease, etc. So we reserve them for participants who are at high risk of bad outcomes with IgA. So there are a whole bunch of people with IgA nephropathy who we don't treat because we're afraid of the complications that the immunosuppressants will cause. Moreover, these immunosuppressants don't actually target the problem with IgA nephropathy, which is, A, the galactose-deficient IgA that we heard about, and the antibody to that galactose-deficient IgA. So this degrader that specifically targets the problem with IgA nephropathy, I think it's quite exciting, first of all, that we're targeting the underlying pathogenesis. There's a lot of work done on what causes IgA. But it's one of the first drugs that would actually treat the underlying pathogenesis. So I think avoiding immunosuppression, like in Dr. Spiegel's talk, where there's very little immunosuppression with these degraders, that's great. I think targeting the underlying pathogenesis, being able to target patients who are who we don't wait months and years before we start treating them, I think all of that is quite exciting for IgA nephropathy. Similarly, for membranous, the other disease that we heard about, membranous nephropathy with the anti-PLA2R, again, we know that that's strongly tied to most primary membranous, the PLA2R. There's no drug that actually targets the underlying problem with membranous. And so being able to target it specifically with these degraders, I think, is quite exciting. Same thing as IgA in membranous. We, again, wait and wait and wait until the patients develop bad enough kidney disease that we start treating them with these immunosuppressants. So it'd be great if we could start treating some of these patients earlier with safer drugs. Okay. I think I would love to just dive right in on kind of the clinical experience that we've seen so far. And really, for any of the panelists, please do chime in if there's a comment you wish to make, as well as our KOLs. To start us off here, how well is the non-clinical data translating into the clinic across key parameters - PK, PD, dose requirements? Are there any complexities in the biology in terms of translation that are important for us to keep in mind or might be translating less well? We have looked at this really carefully, and we actually built an enormous store of knowledge in this unique pharmacology from our preclinical work. One of the things I didn't talk about, actually, that I could have was the preclinical translation of the pharmacokinetics. That was also absolutely accurate. We didn't show that, but there's nearly 100% correlation with where we expected to be in terms of human exposure from extrapolating from mice and monkeys and allometry and clearance mechanisms and so on. The projection of how we would then model the human dose and the pharmacology, which is, again, quite complex, it took PBPK type of modeling, not the classic sort of modeling that one would do. Actually, independently, Dr. Spiegel, myself, David Perman in the Cambridge Laboratories, we all modeled this, and we came up with something very similar. And those lines that you saw correlating with the median and the average declines down to 43% were spot on. So actually, you could say, "Weathering difficulties? Well, we're pretty pleased. It actually turned out exactly as it should be," which means further planning of what comes next in the next couple of cohorts or even how to plan how we administer the dose in the MAD studies is something that we can almost take to the bank in terms of our modeling. I don't know if you want to add to that, David? Yeah. I mean, again, it's always entertaining, I think, for me to hear people's impressions of the science because from our perspective, we've been blown away. Situations where the modeling fits the data so well are incredibly rare. And you look at those and it's not like we went back and modified these models. These models These models predicted these PD results. So that was really exciting. And I think it's a really remarkable place to be to see the science working so well. So when you ask about difficulties, it strikes me that we're sort of waiting for them to come. Also, I should point out the liver toxicity is something that people have brought up a lot. And it's interesting. We see very clean—I mean, extremely clean—tox results, again, speaking to the ability of this mechanism to tolerate the capacity that we're delivering to it. And IgG is the highest bar, as Bruce said in his talk. It's high concentration. We want to remove it quickly. And we're achieving. We're hitting on all those cylinders without impacting the liver, without impacting any other endpoints. Again, so we're kind of over the moon with this. There are a lot of learnings going into making these not hepatotoxic, though. So we're harnessing the most potent uptake mechanism in the liver. If you put something that was a little bit hepatotoxic on that molecule, it would be magnified. So we're aware of that. And of course, we design these molecules not to have magnified toxicity of any type at all because it is very important. It is markedly concentrated within the liver. But David knew that going on in. So I think in the SAD study, Bruce, you talked about how you were using flat dosing versus something more weight-based. Can you discuss the dosing approach? And as we think about these 4 dosing cohorts you've shown us today, how do these 4 doses relate to one another and relative to what you'd anticipate for your dose 5 and your dose 6? So it was a very standard paradigm in terms of selecting doses. Ultimately, we gave as much as we possibly could to animals. We're really seeing not very much at all. So we couldn't stop things at the top end other than to the extent of lowering. So that defined how far we would go. And you always start at a sub-pharmacologic dose. So we chose a dose that we projected would be 4%-5% lowering. And that's exactly what we saw. So that's how we bracketed from the bottom to the top. Not doing this on a weight basis is a very standardized way of conducting such studies. One expects at very low levels of pharmacology for the weight to make a difference. But as you move up to higher levels of lowering, any differences in weight will become much smaller. And obviously, that lends to the convenience of medicines where a single dose of a medicine is given to all patients at some point in the future. So that's the obvious thing to do. The other part of the doses that were selected were doses that very specifically would inform us whether a particular dose could be used for a subcutaneous formulation. So the doses were also selected on the basis of that. And so far, everything is projecting towards being able to give a subcutaneous dose, which we expect to have really identical pharmacokinetics and pharmacodynamics from the preclinical modeling in humans. Next question. Whenever you're saying sub-Q format, are we talking about a 1-2 mL format of sub-Q? Yeah. Exactly. If we're autoinjectors, our goal is to not have as the FcRns, they currently are healthcare administered and yeah. They're high-trigger. Yeah. Ours is projected to be an autoinjector. Okay. Yeah. The solubility of these small molecules is about 10 times that of an antibody. So it gives us a lot more flexibility with how we can formulate them and use them in injectors and so on. I think it will give clinicians and Dr. Howard, I think we're at myasthenia right now, to be able to administer something like that quickly as opposed to even not having to send somebody to an infusion center or a hospital, right, that there could be a big advantage to that. And so that's, I think, it becomes very critical. Yeah. So you won't have to use something like an additive or whatnot to do the dissolve-in solutions? No. No. Very nicely behaved. Yeah. F%, we've probably reported this for these compounds, is 100% sub-Q, so should be a very, very appropriate mode of administration. What's the limitation, typically, for milligrams that you can dissolve into a 1-2 mL whenever you've tested different types of formulations? And generally, in autoinjectors, you want to be below 2.5 or 2.3 mL. So you're somewhere in that mL range. But we don't say what the solubility or the molecular weight of our compound is. So you can do up to 2,000 mg. We won't comment on that for competitive reasons. But we're absolutely confident that the projections that we have from the cohort 4 let us use these subcutaneous devices readily within that framework of volume that you just mentioned. Are you guys capturing IgG levels beyond 96 hours? Yes. When Yes. When Yes. When you present that later? Yeah. Yeah. We're taking that out to 35, 36 days with each cohort. It was at 96 hours for the cohort 4 yesterday. That's as far as we made it with that cohort. We are seeing sustained lowering, if that's where your question's going. Yeah. Okay. It's what we'd expect. The beauty of this mechanism is there's a short exposure of the drug once. It's all gone within about 24 hours. Then IgG returns based on the regeneration rate of IgG. So you only expect 4% of it to return every day. So that can define your dosage interval as to what you want it to be. Yeah. I think you just said your drug's around about 24 hours. And then do you see smooth declines for 24 hours and then? Absolutely. Or Or does it vary fast? It's very, very fast. The free drug is almost gone immediately. And the drug binds antibody. And then the drug bound to antibody then disappears over a slightly longer period. So the rate-limiting step is the receptor? Actually, there's not a rate-limiting effect. When you say rate-limiting, you mean rate-limiting for what? I mean, the slope of the curve. Is that how fast the receptor can pull these if you make complexes instantaneously, the decline of antibody, is that limited by how quickly? You're talking about the PD curve. Yeah. The clearance is actually driven by human blood flow liver blood flow, sorry. So it's just as it's captured going through the liver, it's the chance interaction with one of the 2 million receptors per hepatocyte. So there's nothing limiting. It's just how long hepatic blood flow takes to eliminate that volume of the drug load. So is the data that we saw today, was it IV or is it sub-Q? IV. IV. And what's the ultimate goal for the frequency of administration? We'll tell you when we're done with our MAD. Remember, we're still in the very early stages of the SAD. So once we complete that and we have our MAD data, we'll be able to accurately project. I mean, we already modeled out it. We've said it'll be infrequent dosing, whether it's biweekly or monthly. That's to be determined based on the MAD. We may choose to vary it with different diseases. If we choose to peg a level of lowering lower for a particular disease, we might make it weekly instead of biweekly. It's just inherent within the system. There's a lot of flexibility, so. And it'll depend on the indication. If it's acute exacerbation myasthenia, you might treat an acute episode, right, and not need to treat chronically. Transplant rejection would be more acute as well. Then there's obviously chronic administration where you will have to keep it down to a certain level. But that will vary depending on each disease. Yeah. Another important point here is that the depth of IgG lowering is something we have exquisite control over, unlike the FcRns, right? We can go down very, very deep. We may not need that for every condition. So the frequency of the dosing will vary based on what the needs are. And there's different ways to get there. As Bruce's data showed, you can get there with higher single doses or lower multiple doses, right? And so there's many different ways to really titrate down what level of reduction you want of IgG. Do you want 40%, 50%, 70%? And you need to maintain that. And then we'll have different paradigms depending on the disease indication. Talk about just some of the variability we see in the dose cohorts and why as you go to higher doses, that variability's not tightening or just like you kind of spoke, right? It's really a function of how the plots are represented. What we're seeing is as we move up doses, an increasing proportion of those individuals with sustained lowering. So we are seeing that. And we expect you expect more variability, actually, at in-low doses always when you do pharmacology. And as you get to very high doses, it sort of forces the whole cohort to be similar. So it's simply a function of the low doses that we've administered up to cohort 4 that I think contributes to any of that variability. Of course, in the very lowest dose, there'll be where we have an average of 5%, there'll be some responders and non-responders. Everyone's a responder in the cohort 4 and cohort 3. In terms of the chart that you showed, I don't see the patient numbers there. Is it the full data from the 8 patients for all those 4 cohorts? Or are we looking at just a partial picture? Oh, no. It's all the patients in the cohort. So we have 6-8 patients. We have some placebo. Some active. We haven't said exactly the number of placebo versus active. But we do have a placebo. They're placebo-adjusted. So if placebo changed, it's always a change from placebo. It's all the patients. In the 4 dose, what percentage of that is of the target dose? Target dose? We're flat dosing. So there's no dose adjustment? No. Compared to the 6 dose that you would get to, what percentage? We haven't disclosed what the difference between the dose cohorts are. We'll wait till we have all the phase 1 data and then present all the data. But we haven't yet said what number we're going to. Based on what we've seen, we were saying before 6-8 dose cohorts. We now think we'll only need 6. And so based on the data, we're going to have less SAD cohorts and then start the MAD. Flat. Can you just comment on what switched you over to say you only need 6? Was it the cohort 4 data that came in? Well, that reinforced it because now we have 4 data points. To Bruce's point, we didn't know, "Is it going to completely match up with our primates and others, rabbit and mouse modeling?" So now that we've seen everything match up perfectly with 4 data points, we have confidence we will hit our target of greater than 70% in the phase 1. And that made us confident that we didn't need as much dose change. Can I just follow up on your question and your answer? So as you think about moving into these higher cohorts, what would be an acceptable amount of variability at those levels? I think when you say variability, the current variability, I think, is within SAD. If you look at other people's FcRn data, there's no difference in variability. In fact, if you look at Immunovant data, they don't actually put error bars in some of their studies because obviously, you're dealing with few subjects. So with 6 subjects, you would have tolerance for a wider variability, especially when you're not weight-adjusting it. So there's nothing unusual about the variability there. In fact, in the fourth cohort, as Bruce was saying, as you go up in dose, you expect less variability. Exactly what we saw, the majority of patients actually, by the median that's reported of 43% reductions, most are having that high level of reduction. What's really great as you go across dose groups, you actually see variance and variability go down with your drug group. And in drug development, when you look at the placebo arm, you have a large range of variability. And then your treatment group, that variability goes away. That's a great indication you have a real drug effect. That's exactly what we're seeing. So if you looked at our placebo arm data, most of the placebo is actually going up, as is reported with the FcRns. You get increases above baseline, slight increases, not a lot. I mean, maybe because the IV is in the patient and you're having immune reaction related to that. But it's been highly consistent with everything we've seen to the publications with the FcRns today. The only difference is we're seeing effects at five days. And with the FcRns, you see much later. Five days earlier. Yeah. We're seeing it earlier. Yeah. We're seeing it earlier as expected, right? Exactly. And we also see a lot less variability in subcutaneous dosing. The FcRns, they're antibodies. And they have anywhere between 50%-70% oral bioavailability. So in the efgartigimod trials, they're very good trials, you see a significant proportion of the patients, given the drug subcutaneously, that have inadequate exposures, really, for efficacy. We don't expect to see that at all with sub-Q. It will faithfully recapitulate IV. There are the patient characteristics that were. So you said with the degree of reduction, I mean, were higher BMIs less efficacious or anything else? Since we got the data less than 24 hours ago, I can say we haven't done every analysis. But I think there are some trends with body weight, so lower individuals. And so that's all consistent with drug effects, typically, as concentration goes up. But it takes about 7 days to get the PK back. So we really can't comment on PK data because we don't have it back yet on the last cohort. But to date, it does look like it's correlated to body size or body volume, which makes sense because we haven't adjusted for weight, right? Are you compromising any efficacy by sparing IgG3 for them? It's interesting. People think when we say sparing that we're not hitting at all. That's not accurate. And from the data we're seeing, we are reducing IgG3. It's just to a much lower degree, which was what our goal was because we wanted to keep some immune function intact so that you had a lower rate of infections, especially if you're going to be on this chronically. So it's a relative sparing. But you still are reducing some IgG3, just leaving enough intact for an immune response. I don't know if Bruce, you want to ask. We've actually looked at this very carefully at all possible diseases that could be mediated in part by IgG3 for which we would be potentially less relevant. There's actually very few. Out of the hundreds of different autoantibody-driven diseases, we only found two or three in which IgG3 contributes to the pathogenicity. There are autoantibodies in several diseases. For example, myasthenia gravis has IgG1 and IgG3. But the concentrations of IgG3 are so low, they're actually thought - and this is in many publications - thought not to contribute to the pathogenicity. There's only a couple. Autoimmune hemolytic anemia is one for which IgG3 may be important by itself and we'll be paying a lot of attention to that. We think, actually, there's hardly any diseases because of the low concentration, less than 5% of circulating IgG, for which the pathogenicity might be significant. What's great about a modular platform, you've heard from Bruce, is we can degrade one, two, three, or four. So if it comes to be that there's a disease where you want to keep three intact, you can. Or if you want to remove it solely, you could, right? We could develop a separate degrader. And I think, as Dr. Moledina commented on renal diseases, this is the potential where you could pluck out a positive agent, right? And we haven't seen that before. With the current therapies, we're using these broad immunosuppressives. And so the modular aspects of this technology, really, I think, are what's making it special and differentiate from others. So just to confirm, so it was by design that you're sparing IgG3? Correct. Yes. Okay. And why it was by design is that that is the IgG with the most potent effector functions. So it's small but mighty. So we believe by leaving this small amount of IgG3 behind, we'll disproportionately preserve host defense in the immune system while taking out the mostly pathogenic IgG1 and 2, which are the main targets. You mentioned that there aren't that many disadvantages lowering IgG4. But are there any special precautions that you might need to take with patients in clinical trials that might be tough to replicate if this product makes it to the real world? Not that I'm aware of. There's actually relatively little biology information that allows us to understand what happens when you take IgG4 out of the system. It has more anti-inflammatory than pro-inflammatory properties. People don't understand that. But it is also only about 5% of circulating antibodies. So it's possibly, from a quantitative aspect, not important. That is something that one would discover. But in the IgG4-mediated diseases, those diseases are so absolutely profound that, as always, the benefit-risk of the medicine considers fatal seizure and neurocognitive decline versus whatever the drug happens to do, that'll be a physician's judgment. But so far, we're not aware of something that would be limiting. I'm sure you don't really have that much of an analysis on cohort 4. But across cohorts 1 through 3, have you looked at any baseline characteristics and how they can be impacting variability of IgG response, like baseline IgG levels, weight, anything that kind of helps? We're looking at all of those sorts of things. But we're not sharing them just yet. They're absolutely the right questions to be asking. Yeah. I think it's fair to say there's multiple factors that will play a role. But most of that will get washed out as you get the higher doses. So, lower doses are probably more relevant, those things. But the goal is to get to the therapeutic exposures where, once you degrade the target, those other factors are less meaningful. What are the chances of getting anti-drug antibodies for these degraders? Zero. Yeah. For a small molecule degrader like this, it's really gone very, very quickly. And so I would say zero. For a degrader that is a biologic, an antibody-based degrader, we have those too. It's the same as any other biologic. So I would bet with David that we won't see any immunogenicity. We saw none in the animal studies. If you're going to see, you'd often see lowering exposures in monkeys over time. But we saw just as good immunoglobulin lowering in monkeys in the first week as in the fourth week. So it's. Is the data you're showing, is that best response or that's everyone at 6 hours? And is it fair to say there's different periods where patients have different levels of IgG? You spoke a little softly. Everyone. It's not best response. It's not best response. Speaking to this issue of immunogenicity, just a couple of additional points. First of all, small molecules, of course, we're not worried about immunogenicity unless you're talking about penicillin or something that makes a covalent bond to a protein. These molecules don't do that. And then there's also an additional feature of targeting ASGPR, which is likely tolerogenic. And there are even strategies for inducing tolerance by conjugating proteins to ASGPR. So we think, if anything, our protein bioconjugates will be, when they're ASGPR ligand conjugated, will be less immunogenic than your average protein. Yeah. I think maybe a final question from me. Wow. What an active panel. Appreciated all the thoughtful questions. I didn't have to do much work. So I think a nice way to close this might be to open it back up to our KOLs. And just if you can comment on, is there anything you still think you need to see before moving this platform approach into the clinic in patient populations? And what are the key questions that you have about the approach overall as we see more and more data, which should unfold over time in the coming years? Can I go first? Yeah. I I go first? Yeah. I I go first? Yeah. I think, as I said before, I'm very excited by the fact that we're targeting the underlying pathogenesis of IgA nephropathy and membranous nephropathy as we've understood over the years. No drug has targeted these before, specifically. And so these are hypotheses. They're very credible hypotheses but would need trials to show us that targeting these things actually improves patient outcomes. Once we show that, I think the uptake among physicians, among patients, will be really high because this is how we understand the disease. If we're targeting the underlying problem, I think the uptake will be quite good. That's exactly the data we need to see, is that targeting these problems actually fixes our patients. That's, I think, the next step, really, to try it out. Howard? Yeah. I fully agree. The next step is in trials, I think, and IgG has a clear need. Where I'm excited is if we can selectively target specific antibodies themselves. That, to me, would be a huge game changer if you could go in and target near the immunogenic region in the acetylcholine receptor antibody attack and selectively improve that. Then this becomes a beautiful pharmacotherapy to control disease. Dr. Howard, Dr. Moledina, Dr. Spiegel, and Bruce, thank you so much for having me.