Thanks everybody for joining us. My name is Laura Chico. I'm the Senior Biotechnology Analyst here at Wedbush. With our next panel, we will be discussing novel approaches in neuro. I'm not only extremely excited about our speaker lineup today, but also a little awestruck by the collective years of experience that our panelists bring together here. After confirming with everybody, it works out to be about 95 years of experience on our panel today. Pretty amazing. Going around the Zoom room, though, we have Aptinyx CEO, Norbert Riedel, CEO of Denali Therapeutics, Ryan Watts, Sage Therapeutics CEO, Barry Greene, and Karuna Therapeutics CEO, Steve Paul. Gentlemen, thanks to all of you for joining us today. Just to set the stage, we are going to have each of our panelists start off by giving a brief introduction to themselves and their companies. I suspect everybody in the audience is probably already largely familiar with everybody. We'll then transition to our panel discussion. We'd love to have some audience participation, though, so there should be a chat box at the bottom of your screen if you'd like to participate and launch a question in there. With that, Barry Greene, why don't you start us off on an introduction? Laura, thanks for having me. I'm also equally excited to be on the panel, and I've got my pen ready to take notes and learn a lot. I joined Sage late last year, December, as the CEO, after 17+ years at Amylin and Millennium before that, and plus big pharma experience before that. The reason I joined Sage is because I was very interested in getting into brain health and saw a phenomenal team with a deep background in really tremendous science, specifically in brain circuitry, NMDA, and GABA, with a really organic pipeline that should provide sustainable product for years to come. I was actually very impressed with the team and Steve Paul being the scientific founder. The first drug ever approved for postpartum depression based upon Sage science and Sage methodology. I'm excited to be here, and I'm looking forward to more discussions on the panel. Okay, wonderful. Thanks, Barry. Steve, how about over to you? Well, thanks, Laura, again for pulling this session together. It should be very interesting. I have had a long history in neuroscience, almost 40 years. I'm 75 years of your 95 years at least of experience. I began my life as a pure academic. I was a branch lab chief at the National Institute of Mental Health, the NIH institute that was focusing on major mental disorders such as schizophrenia and depression, anxiety disorders, et cetera. Still huge unmet medical needs. I left NIH and joined Eli Lilly shortly after Lilly launched Prozac and was really developing a number of very interesting drugs in the psych space, Zyprexa and Cymbalta being two of their blockbuster drugs that I helped to oversee the development of. Became head of drug discovery there, over all of our therapeutic areas and then head of R&D, then retired from Lilly in 2010 and wanted to go back to my early work in neuroscience and particularly in psychiatry. I helped to co-found Sage, and Barry's given you that introduction. More recently, I've been at Karuna, developing what we believe will be the first in a new generation of antipsychotic drugs to treat this horrific disease we call schizophrenia. It works completely differently than the traditional first and second generation antipsychotic drugs. It does not block dopamine receptors, serotonin receptors. It's a muscarinic receptor agonist M1M4 preferring, we've demonstrated some very unique and hopefully important clinical benefits in this disease, schizophrenia. We're going to stay focused on that. We're going to work on dementia-related psychosis, a psychosis that occurs in upwards of 50% of patients with Alzheimer's disease. Then go into some other areas in terms of cognition and maybe get back to mood disorders as well. That's wonderful. Thanks, Steve. Ryan, why don't we transition over to you? Great. I'm very grateful to be here on this panel, especially with these individuals. My background is in developmental neuroscience, but have focused primarily on neurodegeneration over the last two decades. Spent time at Genentech and then founded Denali with several other co-founders in 2015. The focus of Denali is really to defeat degeneration. We know that this is almost an unachievable goal, but I think our sort of area of expertise is engineering medicines that cross the blood-brain barrier. Focusing on really crossing the blood-brain barrier, focusing on genetic targets. I'm really looking forward also to, like Barry said, taking notes and listening, and it's great to see everyone else on the panel that I've known for years. Thank you. Wonderful. Thank you, Ryan. Norbert, last but not least. Why don't I turn it over to you? Yeah. First of all, Laura, thank you for having me on the panel. It's terrific to be on this panel with Barry, Ryan, and Steve. That's really a wonderful experience. I started with my team, Aptinyx, in August of 2015, so that's exactly six years ago, by spinning it out of its predecessor company, Naurex. We were at the start a preclinical stage company. Now we are a full-fledged clinical stage development company with three highly innovative compounds in 4 CNS indications. The focus is entirely CNS. Our focus is entirely on the NMDA receptor. That is because the receptor is a very well-researched and well-known pivotal regulator of normal brain biology and associated with numerous diseases of the brain, of course. We have spent together with the academic work, 25+ years functionally and molecularly characterizing the receptor. We designed by molecular design, novel, highly specific compounds that are targeting the receptor, and they act by modulating that receptor. That is a very new mechanism of action, a new set of compounds around which we have exceptionally strong IP. I am very excited at that stage of development, as well as the information and data we have so far gathered, both by doing extensive preclinical work and, of course, doing phase I and phase II studies that led to the current stage of our pipeline. Thanks again. Thank you, Norbert. Okay. Really looking forward to jumping into the discussion. Just one last reminder for our audience. If you'd like to shoot over a question, feel free to put one in the chat box at any point. In preparing for this panel discussion, I found an article that's about 12 years old now, talking about the impending crisis in CNS drug discovery, basically, the need for more regulatory tools, establishing proof of concept earlier in development, finding new biomarkers, trial design. A lot of the things that are starting to become more prominent in our discussion. I guess, as we look back at the past 10 years, I'd ask each of you, what do you see as some of the most meaningful shifts that have occurred in the space that have helped lead to more success in the CNS drug development space? Perhaps, Norbert, we could start off with you. Yeah, thank you. I think I have a pretty good sense of that because in the early 1990s, when I was in large pharma, we had a strong focus on CNS, Alzheimer's disease and schizophrenia. Look, I would start it by saying, in those days, I think the industry was very much chemistry driven with a much, much, much more rudimentary understanding of biology as well as molecular biology. Today, I think we have a really good understanding of molecular mechanisms of disease and the biology of disease. I speak for us as Aptinyx, but I think it's similar for the colleagues I have on the panel, that we look to characterize our compounds by doing very extensive pre-clinical work, where that pre-clinical work can be somewhat predictive of a human disease. Again, we do it at the molecular level. We, for example, use crystal structures of the receptor to design and map our compounds and identify where they are and how they interact. That's a level of resolution that just didn't exist in those days. We do very, very thorough target engagement studies in healthy volunteers. We look for pre-specified and validated biomarkers. There are some of those, for example, in fibromyalgia patients, there is clear data that in those patients you have pain perception and pain processing issues that relate to hyperconnectivity in certain brain regions. You can measure that by fMRI. You can also measure glutamate levels using spectroscopy. We characterize our compounds long before we put them into a patient population. Only when we actually do those signal finding studies, do we then conduct patient studies. Those studies are initially also exploratory and signal finding for us to get a sense of informing the right patient population, de-risking the next study. Then we go from there into more definitive, call it registration supportive or pivotal design studies. That has sort of characterized the path we've been on. I think we take advantage of the molecular biology in general, the molecular pharmacology, and as many sort of data points that are related to biomarkers and early readouts that we can use to try to be successful in what I think all my colleagues will agree to be an exceptionally difficult space to be in, namely the CNS diseases that we are trying to find therapies for, where there hasn't been really a meaningful contribution, in some cases, in 20+ years. That's the path we are on. I think we have better information today than we did 10 years ago, and we use that information to the max to get to the right answers. Okay. Thank you. Steve, I might pivot over towards you and, psychiatry, neuropsychiatry, there's not a ton in the way of success that we've actually seen materialize in the last decade or so. I'm wondering if you could just spend a moment. Yeah. Neuropsychiatry differ from other CNS areas, and what progress has been made in the last 10 years? Yeah. Where is the unmet need still? Well, let me start with the unmet medical need, then I'll get back to the progress. We're not adequately treating mood disorders, depression. We're not adequately treating schizophrenia. We're not adequately treating major anxiety disorders, PTSD. The drugs we have are okay, and they do save lives, and they're very useful. I don't want to diminish the role they've played. Patients with these diseases are often left still symptomatic. Very few remissions occur in any of these disease, particularly in the more chronic diseases like schizophrenia, where we have drugs that can reduce psychosis, psychotic symptoms, hallucinations, and delusions, but don't touch the other symptoms of the syndrome that we call schizophrenia, negative symptoms and cognitive symptoms in particular. Progress has been slow in psychiatry versus neurology. I'll lump them into those two big buckets. Because of the real lack of tractable biology. We don't have a lot of genes that we know cause mood disorders or schizophrenia. In contrast, virtually all of the heritable inherited neurological disorders from Alzheimer's disease, Huntington's disease, ALS, frontotemporal dementia, many of the epilepsies, pain, we have tractable biology. We have mutations in gene that either cause disease in simple Mendelian forms or actually increase or decrease your risk for causing a disease. This really helps validate targets and really allows us to focus on biology that's likely to result in a therapeutic effect, if you will, if we craft a drug. Now, in psychiatry, we've been left really with some biology that's tractable. I think the work that Barry's been doing at Sage on allopregnanolone and postpartum depression is a very elegant example of tractable biology that will hopefully lead to a very effective drug for treating PPD as well as major depressive disorder. We're really relegated to some observations in the clinic, like we've made with xanomeline over the years, and then being able to exploit those by understanding at a very molecular, very detailed level. As Norbert said, we know a lot about targets in the brains. We just don't have a lot of validated targets. NMDA receptors and GABA receptors and in our case, muscarinic receptors, which are G protein-coupled receptors. The others are what we call ionotropic receptors. We have really intricate understanding of how signal transduction occurs in the brain, but finding those validated targets so that we prosecute them, we have a high probability that we'll come up with drugs that actually work, has been a bit of a challenge, certainly in psychiatry. That's why we've leaned towards human validation, as we've done with xanomeline. The observation we made on its antipsychotic properties in DRP were serendipitous, clearly, but now we understand how the drug works. We understand which receptors have to be tickled, if you will, and how to do that. Because these structures are known, we have three-dimensional crystal structures and cryo-EM structures, and we can model these things. We can craft even better drugs. It's been a bit of a tougher road in psychiatry, for sure, but I think we're seeing a new generation of drugs. Medically and commercially, these have always been big markets. If you go back 20 years, if you look at the SSRIs, the SNRIs, the atypical antipsychotics, arguably among the most successful commercial markets that have been. What we need is new biology, new targets, and new drugs, and I think you're seeing that emerge right now. That's helpful, Steve. Thank you. Ryan, neurodegeneration, what would you highlight in this space in the last 10 years as most impactful? I'd say I think the most meaningful shifts in the last 10 years are the following three things. The first are the genetics, as Steve already highlighted, what we call the degenogenes. These are the oncogenes of neurodegeneration. The second, and an area that we're focused on is being able to engineer brain delivery, getting small molecules, kinase inhibitors, large molecules, ASOs across the blood-brain barrier with systemic delivery. I think the third, which we'll talk about in detail, are advances on biomarkers, both fluid-based, but also imaging-based biomarkers. I think all of this, if you look at the last 10 years, the success around monogenic diseases has laid the foundation for these more polygenic or more complex diseases. We actually have spanning that. We're working on monogenic diseases where there's a much more clear understanding, but also using that as really the springboard to treat these more complex diseases such as Alzheimer's and Parkinson's. That makes sense. Barry Greene, any last comments you'd like to throw in here as well? I think, as you mentioned, you took over recently as the Sage CEO, interest in the CNS health space, so I'd be curious of your thoughts here. I think that Ryan, Steve, and Norbert laid out exactly the rationale for why I felt that the brain health and CNS is primed for real innovation over the next 10- 20 years. It feels to me like oncology felt back in 2000, and I think Steve said it well. There were lots of drugs made, people's lives were being saved, but they were toxic, and the side effects were horrific, and people were still dying at rapid paces. Now we have people alive with very specific drugs. I think that's where CNS and brain health is going. Very specific drugs, deep understanding of biology. Norbert laid out all the tools we have, and throw imaging, as Ryan said, into that mix. It really feels like we're on the cusp of big breakthroughs across brain health and CNS, just like we saw with oncology over the last 20 years. Okay. Very helpful. One theme that really stands out here is there's just tremendous unmet need, a lot of perhaps past failures. I guess I'd ask the question to each of you as you're thinking about your own pipelines, maybe help us walk through some of the parameters or considerations that you're making to guide candidate selection, whether it be in terms of a target or a disease area. Ryan, maybe we can start with you first. Obviously, a lot going on, both on the target side and the disease perspective, but let's maybe focus on the small molecule kinase efforts and how do you apply that? How do you drill down and focus the efforts? Great. Thanks, Laura. I guess the question is where to begin. For us, it's always the genetics. It's the degenogenes. Let's just use, I think, one example, as you mentioned, the kinase inhibitors. LRRK2 is a kinase which is mutated in about 3% of Parkinson's disease. It's sort of the perfect drug target in the sense that the mutations are kinase activating, and therefore, we're looking for kinase inhibitors that we can discover that can cross the blood-brain barrier and show robust target engagement. In addition to that, we've developed a number of biomarkers that look at first the target itself. The subsequent pathway, and then broadly sort of patient phenotyping. In an ideal scenario, you can pick patients based on genetics. Let's go back to LRRK2. These hyper-activating mutations basically disrupt lysosomal function. What's really fascinating is if you look at the genetics in Parkinson's disease, many of the genes point to endolysosomal function. It's not just LRRK2, it's GBA, it's GALC, it's a number of genes that are linking the biology of lysosome. In fact, in some ways, I think Parkinson's disease is a mild lysosomal storage disease. The idea then is to improve lysosomal function by using these tools. For us, when we think about going after a target, let's start with the degenogenes as the major point there. Much better if it's a monogenic disease like IDS and Hunter syndrome. There are many diseases where we can start with the genetic insights. Okay. Norbert, perhaps the question, this might be relevant for you because NMDA, you're pursuing multiple indications, fibromyalgia, PTSD. There's a pretty diverse range of indications I would note there. walk us through kind of the selection rationale that you use. For us, Laura, as I mentioned earlier, it's really a focus on the NMDA receptor and the knowledge of 25+ years of extensive research, both in academic labs as well as in industry labs. The receptor is extremely well known to be involved in plasticity, neuroplasticity, and plasticity changes. It's also known to be involved in cognition, memory, and learning paradigms. For us, the question then becomes where does the science take us as we explore the compounds we have in how they engage as modulators with the NMDA receptor? Because there are various subtypes of the receptor, they have a spatial distribution, they have a temporal distribution, and that's our focus. It's really important and helpful to understand that if you think about chronic pain like fibromyalgia, it is actually a dysfunction of plasticity or a change in plasticity. If you think about PTSD, it's the inability of a PTSD sufferer to basically extinguish a fear that they become exposed to and to basically consolidate that fear extinction. The approach we take is to address plasticity changes, to look at our NMDA receptor modulators as modulators that have positive allosteric effects on the receptor. It's known that in areas of PTSD in the prefrontal cortex, there is a hypofunction of NMDA receptor activity. Again, we connect the biology with the molecular target, with the specificity of our compounds in a disease in which we believe the mechanism should be relevant to alleviate the symptoms that we see. The benefit we also have with NMDA receptor modulation is it addresses not just plasticity, it addresses mood, it addresses sleep, all well characterized. It's a much more holistic approach to the underlying complexity of these diseases, because all chronic pain, all PTSD, all cognition impairment in one way or another associates with mood and sleep disturbances. We have seen really, really excellent results. It all comes down to selecting indications based on where our science takes us. Of course, it is helpful to know that they are highly underserved with respect to the medical need, that they are also highly attractive from a commercial point of view. We arrive there not by jumping into an indication for those reasons, but by taking us there steadily, by exploring mechanism of action-related biology that suggests that we are on the right track when we focus on fibromyalgia or PTSD or cognitive impairment in Parkinson's or in Lewy body disease. Those are our core areas of focus, and the results are so far extremely supportive with the superb safety profile we've seen in our compounds. Okay. Barry Greene, your NMDA platform is also at kind of an interesting point where you have some data in Huntington's, and you'll be pursuing that. How do you weigh pursuing the other indications that are arguably quite different in terms of the commercial footprint? Yeah. If I take a step back, Laura, Norbert described very well what he's doing in NMDA, we're doing something very similar with GABA, designing a whole series of molecules that dial up or dial down these receptor pathways and then match those with clear biology that we've already articulated. When we bring things into the clinic, what we're doing is looking for big effect sizes really early on. Probe studies, not a whole lot of patients, but big effect sizes. As we'll talk about probably later, clinical studies in these areas are very, very hard because of the placebo effect. If you don't have a big effect size with your drug, your ability to differentiate when you get into later-stage clinical trials becomes challenging. Those big effect sizes in probe studies is a key aspect of our strategy. What I'd add to what Norbert said, because he articulated sort of following the science, the biology very, very well. I'm sure Ryan, Steve will say similar things. The other thing that we do is we really think about the unmet need from the patient's perspective. Let me take depression as an example. If you interview 100 of Steve's peers, I think what you'll hear from almost all of them is it's a well-served population, and I have cheap drugs to treat these people. If you talk to the patients, they won't say that. First of all, the drugs aren't cheap because they're paying out of pocket for generics, so they're not cheap to them. The patients don't feel they're well-served. The average person's on a depression drug For seven weeks, it's like the three new depression drugs on a regular basis. These are chronic medicines that cause weight gain and sexual dysfunction in a majority of patients. It's interesting that we have a field that thinks it's well-served, but the patients don't. We look at those disconnects with big effect sizes, and we talk about support. We believe that we have a drug, zuranolone, that could be a paradigm shift in depression. We have the data that suggests that patients get better in as little as two or three days. It's quite remarkable. When you go to the physicians, they're not ready to make that paradigm because they've been taught, like Steve and others, to treat these diseases chronically. We also look for the disconnects between what patients believe and what the opinion leaders believe and try to create that paradigm shift. It's not easy, but ultimately, that's where real innovation will occur. Okay. Steve, I think you really highlighted the unmet need in schizophrenia, but clearly there could be broad applications for the muscarinic family here. How do you think about expanding out to other indications for KarXT? We look for things where the mechanisms of action of KarXT, which is xanomeline, which is the active pharmaceutical ingredient, is a M1M4 preferring. There's five muscarinic receptors, and this is a potent agonist of those two. Almost any situation in which you sort of have hyperdopaminergic function, it looks as if this mechanism is operational. We're looking at, obviously, schizophrenia, acute psychosis. We're also interested in the pro-cognitive effects of the M1 agonist activity of our molecule. Things like Parkinson's psychosis, levodopa-induced psychosis should respond. Something called levodopa-induced dyskinesia, which is again, due to overstimulation of dopamine receptors by either levodopa or dopamine agonist. There are a number of different conditions that we're looking at. The question, and my not-so-great answer, reminds me to make another point here, and that is that in even neurodegenerative disorders where clearly the best way to treat a disease is to prevent it from happening in the first place, and where you have a lot of genetics that can help you focus on certain tractable biology, as I indicated earlier. There's still a lot of people with Alzheimer's disease and Parkinson's disease that are symptomatic, and disease-modifying treatments, while they hopefully will slow the progression of the disease, maybe even prevent it someday, still leave a lot of patients with a lot of symptoms. As Norbert Riedel was saying, maybe an NMDA receptor PAM or the work that Barry Greene is doing at Sage with SAGE-718, which is an NMDA receptor PAM using different chemical matter, maybe these will help improve the symptoms that patients with Parkinson's or, in terms of cognition, Alzheimer's disease have. When they come to your clinic, you want to give them a drug that slows the progression. What you're telling them is that you're going to be less worse off two years from now with our drug than you would be if you were on placebo or didn't have a drug. The patients are saying, "Well, Doc, I'm having problems now. I can't remember now where my keys are. I can't function." We do need better drugs for symptomatic treatments as well in light of how many patients we're talking about here. Hopefully, we'll get to a point where we can completely eradicate a disease like Alzheimer's disease. I think in fairness, with 6 million patients roughly with Alzheimer's today in the U.S. and maybe 10 million with mild cognitive impairment still impaired, I think we're going to need all kinds of different drugs to treat these patients effectively. Yeah. Well said, Steve. We could tackle one other area, and that is heterogeneity across different disease areas within CNS. It's already a challenging space, layer that into the mix, it makes it even more complicated. We have the COVID pandemic too, also kind of influenced background noise. I'll pose this question to the group. If you put on your clinical development hats, what are the most important takeaways from your own clinical trial experiences thus far in terms of how do you mitigate risk? How do you tackle the issue of perhaps a changing placebo response in a complicated environment such as today? Barry, we'll start with you. There is the recent Waterfall experience, I guess I'd ask it more broadly. How have you approached risk in some of the other landscape studies as well? Yeah. Our industry's been trying to figure out placebo effects in these CNS for 50+ years, and it's a challenge. If you look at the literature, in fact, placebo-controlled studies in depression, placebo groups have improved about 1 point a year for the last 30 years. It is getting harder and harder because of expectation bias. It's a very challenging thing. First of all, you have to have a drug with a big effect size. If you've got a marginal effect size, you'll just never beat placebo. We all know that when you take a depressed person out of their home, particularly in a COVID situation, and you give them good medical care for several weeks, they're going to get better. I think it's actually a biologic phenomenon that they're getting better. They're not as better as if they have an important drug on board. Make sure you have a big effect size. This is well-controlled, large enough studies. Make sure that your endpoints are right. You need a number of studies. I think Steve mentioned PROZAC earlier. I think PROZAC had seven failed phase III studies and three or four successful ones. To get a drug approved. You really have to do a lot of studies here and make sure you're looking at it. The other thing that I was very impressed that Sage had done, frankly, before I got here, but I've had a chance to see the data, is asking patients. Patient-reported outcomes is huge here because it's clear that the physician's bias is much higher than the patient bias. In previous work that Sage had done, particularly with zuranolone, physicians were reporting patients getting better, both placebo and drug. When you look at the patient-reported outcomes, placebo patients did not feel better, but the drug patients did. It's looking at it both from a physician evaluation, but very importantly, from a patient evaluation. At the end of the day, this is a theme of mine obviously, the patients can tell you how they're feeling, and they actually know, versus someone evaluating how they're feeling. Interesting. Ryan, you're tackling right now Hunter syndrome, and there hasn't really been a lot of success there with current standard of care. I'm just kind of wondering how do you approach the heterogeneity issue and risk in that kind of disease setting where there's not a lot of benchmarks to follow? Great question, Laura. I think for us, we're a little bit more fortunate because it is monogenic disease. There is still a fair amount of heterogeneity, both in the mutations but also in the clinical manifestation. About 70% of Hunter patients develop what would be considered neuronopathic, so neurological disease. Interestingly, the attenuated patients also have some signs of neurologic symptoms. It's much more attenuated. For us, we have to follow the clinical data. We have to understand that it's a monogenic disease, loss of an enzyme. The enzyme has a specific substrate. What we've seen, at least in the periphery, is that reduction in the substrate correlates generally with clinical benefit. Now we're making that leap that in the nervous system we have the same correlation. That being said, we also need to treat early these young boys before they fall off the developmental curve, and you'll be able to see, I think, much more robust effects. Also, since it's a developmental disorder, the idea is you're not just stopping regression, but in fact, some of these boys, as we saw in our just small cohort of five patients, can improve, which is not necessarily what you'd expect in some of these more chronic heterogeneous diseases. Okay. Norbert, anything you might add in terms of your risk mitigation strategy? I would maybe like to make 2 points. One, to follow up on the comment that Barry made on placebo effect. I think that placebo effect will be part of how we do drug development, in particular neuropsychiatry. I do believe that the placebo effect can be mitigated. We, for example, use in our chronic pain studies, blinded algorithms and handheld devices so that patients basically have sufficient but also insufficient information as to what they should be looking for and how they should be recording their results. There is the whole question of how do you, in a psychiatric study, train sites in not actually developing a therapeutic relationship with patient, but a research relationship. That entails a rigorous training in how you do measurements. In our fibromyalgia study, that is now a phase IIb study in 300 patients, we would not have started that study had we not shown in 22 patients prior to that we get a very consistent fMRI signal in looking at hyperconnectivity and glutamate levels, looking at spectroscopy. To minimize that heterogeneity by starting with a population of patients that all have a similar response was very helpful. Then, Laura, I would just add one item that we all need to be paying attention to, and that is not patient variability, but measurement variability. I am concerned about having sites that are not properly trained, as I mentioned. I'm concerned about doing studies outside the U.S. where the standards are very different, where you have so much measurement variability that it becomes hard in Barry's paradigm to basically make a sufficient distinction between a placebo response and a treatment response. It's complex, but I think it has to do with managing heterogeneity, managing how we engage with sites, and managing how many sites and how many studies or how many arms in a study we conduct to basically try to do the best job we can to mitigate that placebo and heterogeneity problem we have, because the problem is not going to go away unless we do it for Ryan's approach by having monogenic disease paradigms that are specifically targeted. That doesn't necessarily serve very broad patient populations. It's super helpful, but perhaps not as broadly applicable as some of the other areas we work in. Okay. Steve, rounding this out here, you've progressed from the phase II EMERGENT-1 study to a multiple phase III programs right now, and I think the variance you had in EMERGENT-1 was actually quite impressive. I'm curious, how do you ensure that you don't lose that fidelity when you go from phase- to- phase? Okay. Well, let me just, Laura, repeat a few things that Barry and Norbert said. First of all, it always helps to start with a drug that really works well, at least in a subgroup of patients, because we're not able to fully delineate that very easily, particularly in psych disorders that are highly polygenic leading to the endpoint of either, in our case, schizophrenia, psychosis, or let's say, depression. Starting with a drug that really works, and undoubtedly it's likely that our drug works very well in a subgroup, but when we put everybody together, the effect is robust enough that, in fact, we can see a good drug placebo difference. Keeping the placebo response low is also very important. As you point out in our phase II study, EMERGENT-1, we had a very moderate, very modest placebo response. Our clinical team, our CMO, and our head of Clinical Operations worked very hard to design a study, importantly, to execute a study that enrolled only appropriate patients. Believe it or not, there are some simple things, like getting people who really have the disease, that do not have inflated ratings at baseline, which can occur with the certain incentives that occur in place that have built-in checks and balances and oversight on good data. As the trial actually progresses, we can actually blindly eliminate sites that are giving us bad data or increase enrollment at sites where there's no issues around inter-rater reliability or remote raters and things like that. A lot of things that you can do. Keeping a large placebo group in terms of the size of the placebo group sounds simple, but if you have a four-arm study and three of them are active drug, there's an expectation bias that you're going to be in an active drug trial. You're going to be on an active drug. We have a one-to-one randomization, a very large placebo group, fewer numbers of sites, which believe it or not, variability from site to site, The more sites, the more patients, you can get a lot of noise into your study. There's some real simple things. Eventually, I think we're going to have biomarkers. There's some really exciting, encouraging data using pharmaco-EEG, for example, to discern which patients respond to a given agent. There's some pretty good clinical data, it has to be reproduced, that suggests that we may be able to stratify patient populations by people we predict might respond to a given agent or a given mechanism. It's still too early to tell. We feel good about where we are at Karuna because we have a drug that's shown good results in now three phase II studies. We're executing and sticking with our knitting, so to speak, and have designed several studies that are very similar to what we did in phase II. Okay. That is really great perspective from everybody, so thank you. I'd like to delve into the regulatory front a little bit here, and I think there's obviously been some controversy on the CNS side of the house at FDA. All of you have interacted in various capacities with the agency, and I think obviously this has been a pretty challenging environment. I guess the one question I'd like to pose to each of you are, one, has the agency changed the bar for approval? Do you think it's becoming more challenging or less challenging to get a drug approved? two, FDA's budget has really not kept pace with the breadth of their responsibilities. This last year, we had 53 drugs approved. About 10 years ago, it was 21 drugs approved per year. A big increase in terms of the bandwidth that they're needing there, but not really keeping pace in terms of budget. First question, has the bar changed and is it more challenging? Second question, what needs to change at FDA to keep it running efficiently? Steve, let's start with you. You've had a great perspective here, has the bar changed? I don't think so. I've been encouraged by our interactions with the divisions that basically oversee the development of drugs in our space. I have to say, I think they've done a good job of understanding the unmet medical need and the potential value that our drugs bring to that. I don't think it's changed. Nor do I think some of the controversies really relate to events moving forward. I think you're probably referring as much to the aducanumab decision as anything at this point, which we could probably debate for hours. I think they're doing a good job. Now, I will say it's gotten a lot more complicated, hasn't it? You mentioned numbers, but the kind of chemical matter that's being developed now. It used to be all small molecules, and then we became fixated and focused on biologics, mainly monoclonal antibodies. Today, it's gotten much more complicated. I'm sure it's probably not easy to keep up with the advances in technology, and I'm sure that's been a bit of a challenge. I have never heard that from anybody at the FDA, but I can't imagine that's not an issue. We need to do everything we can to ensure that we get really good people there, and there's some very good ones there with good common sense. Keeping up on that technology curve, because it's really exponential. You think of antisense oligonucleotides, RNAi, some of the work that Barry's done at Alnylam, some of Ryan's extremely innovative approaches to getting proteins across the blood-brain barrier. These are all much more complicated than they were before. I'm encouraged by the cycle times. Yes, you don't hear the whole story when something happens and there's a blip, somebody's put on hold or they get a complete response letter. You just don't hear the details. In our interactions, we've been very impressed with the folks that we're talking to. Okay. Ryan, Steve just kind of mentioned some of the innovation piece there, I'd ask you the question, even if the bar has changed, does that really even impact your strategy? I agree with Steve. We've also enjoyed our interactions with the FDA and have been impressed by the FDA. We work with several divisions because we're working in rare disease as well as neurological disease. I would hope that the bar changes. I think that oncology should be the example. I think the idea of accelerated approvals based on robust biomarkers and definitive genetics would be a great path for us to take in neurodegenerative diseases, in part because the diseases we're treating are so difficult and their duration is so long. We need to be a little bit visionary here. The current circumstances aside, which are more complex for different reasons, it's actually pretty visionary to imagine you have this genetic lesion that's causative in a population, in a mutation, like we're talking specifically about APP. We discovered a mutation in previous life before Denali that protects people from Alzheimer's disease that basically reduces Abeta. You have this really strong genetic understanding and then you have the associated biomarker. Obviously, we all know what the missing link there is that correlation of biomarker to robust clinical benefit. Maybe if you look across programs, that would change. I like that the FDA can imagine this as biomarkers for accelerated approval, obviously do no harm. That's going to be key. I do think, Laura, your point is well taken, is that in the last decade, the incredible advances in biomedical research, the tools we have, the different modalities, as Steve laid out, the FDA needs to quadruple its resources to be able to handle this explosion in invention. I think that's going to be really challenging. We, of course, are working, for example, crossing the blood-brain barrier and using biomarkers as a driving force for that. I think that it's good to imagine an FDA that is trying the best we can to provide benefit for patients and using biomarkers and genetics as the underpinning. I do think I agree with Steve, that there isn't this completely frame shift. It's not like everything can get approved now on an accelerated biomarker. I do hope, controversy aside, that we start behaving a little bit more like oncology. These patients are an incredible unmet need. We need to be forward-thinking about approval of some of these medicines to provide a benefit, especially if you have to prevent the disease. You're going to have to treat one decade or two decades before the disease actually hits in terms of cognitive deficits. I'm with Steve. I think our interactions have been really good with the FDA. I do think bandwidth is an issue. Yeah. Okay. Barry, real quick, you've actually been able to leverage specific pathways like breakthrough therapy designation. I'm just kind of curious what your thoughts are here. Well, I was going to say exactly what Ryan said. When I was at Millennium, was able to get VELCADE approved on accelerated approval on phase II and repeated phase III, first-ever drug approved in multiple myeloma. Richard Pazdur was young and very creative back then. If I fast-forward, our interactions have been great. The neuro division's done a phenomenal job, and I feel actually for people at FDA who get dragged before Congress and questioned about everything they're doing. I certainly wouldn't want to do that every day. There's a certain scrutiny that they're under. With all that, they've done a phenomenal job. I'm not going to say anything the FDA doesn't know. There needs to be a modernization of FDA, and there needs to be a progression of oncology-like thinking across other divisions. Ryan said that. Well, I agree. We need to leverage biomarkers, novel phase III designs, so we're not conducting 15, 20 phase IIIs across thousands of patients to get a drug that we all know works across the finish line. It's frustrating when you have a drug that you know works, and it still takes two or three years to get that drug across the finish line because of all the other things that go into a regulatory filing. That can be modernized. Okay. Norbert, any final comments here? No, you're probably for in the sake of keeping our time, I think it was all said, right? I do believe that Barry Greene just mentioned we have a lot of drugs that work, and the hurdles can be high because, in particular, our space, placebo responses are a really, I would say, complicating factor, right? It is much more straightforward to measure whether or not a tumor will reduce in size or goes away, or whether your blood pressure is lower if you take a drug, compared to looking at whether or not your depression symptoms are alleviated. From that point of view, I think really having a bit more of a sort of like receptive environment at the agency to look at it from a point of view, is your drug active? Most importantly, is your compound safe? All these diseases are chronic long-term diseases that require not acute intervention, but chronic intervention. For me, it's also a matter of show me that your drug is really superbly safe, and then I can probably get a bit more over the hump of how do you best show that it's efficacious, and maybe there we can maybe suggest some change as to how we go about endpoints that in some time, in some ways of looking at it, may not be the most adequate endpoint any longer in addressing the disease paradigm. Okay. Laura, just to add that the FDA is pretty savvy. The divisions know the nuances of the space they're working in. If a failed phase III trial caused a psych drug not to ever get approved, we'd have no psych drugs, okay? In other words, they understand they need to see compelling evidence that your drug works and is safe. There's no doubt about that. Occasionally, even for an active drug, because of placebo responses, other sorts of things. The final comment I'd make is just think about what they've been going through with COVID and all the work that they're doing there. To be able to manage all the things that this panel is doing with them and deal with COVID vaccines and diagnostics, that's a heck of a lot of work. Well said, Steve. I think they have more than a full plate right now, unfortunately. One theme that keeps coming up across all of you guys is biomarkers. When we look ahead rather than looking back, when we look ahead and think about tools that might help us push forward towards reaching successful milestones, biomarkers and genetics stand out there. Steve, I think it was your earlier comment, you mentioned CNS neuropsychiatry specifically has really been lagging versus other therapeutic areas. There have been some really good initiatives that have led the way here. I guess I consider genetic screening in epilepsy, Alzheimer's Genome Project, Bluefield Project in FTD. There are definitely some great examples there, but what is most notable in terms of biomarker advancements for either monitoring therapeutic intervention or patient selection? I guess, what do you see as areas that we need to work on the most? Maybe Ryan, let's start off with you. Lots of great case studies on perhaps how you're using these, but how do you perhaps see NFL as an example, how does that evolve from here? I think the simplest way for us to view biomarkers, although certainly not linear, is starting with biomarkers of target engagement, just really definitively understanding dose and its relationship to target engagement. Second to that would then be modulating the pathway. I think the third, where we could see the most advances but with the most uncertainty, is around patient phenotyping. I'd put neurofilament in that category as sort of a broad marker. It's essentially a structural marker of the neuron. If a neuron dies, you would expect neurofilament levels to go up. It's clear that using neurofilament as a predictive biomarker is not as simple as we would all like it to be, and as we progress this work, especially in Hunter syndrome, we've realized it's more complex. I think one really simple example is that in diseases like SMA, neurofilament and tau is also a cytoskeletal marker, a biomarker. Both of them are elevated. In a lot of diseases where neurofilament is elevated, in fact, almost all that we've observed, we've seen tau elevated, but it has to be elevated to a certain level. We, in Hunter syndrome, see neurofilament very heterogeneous, so some patients elevated, some not, and 0 elevation in tau. We've seen no elevation in tau. It's not as simple as like, "Hey, this is just a marker of neurodegeneration." As we lay the foundation in new areas like Hunter syndrome, we're going to have to be willing to blaze these new trails with the uncertainty. I think we can't hinge on a single biomarker. If we're looking at neurofilament, we should be looking at tau, we should be looking at other markers of, let's say, neuroinflammation. Our hope is that in totality, you might be able to see these changes in biomarkers. Start with the target engagement, go to the pathway, and then ultimately these patient-associated biomarkers. Maybe one last point. There also appears not to be a correlation necessarily between neurofilament reduction and clinical benefit. That's been shown in Batten disease, where clinical benefit actually preceded neurofilament decline, which took upwards of two to three years later. I think these sort of uncertainties around, let's say, neurofilament and tau or other biomarkers, it's not going to be linear. It's not going to be like, here's one biomarker, you change that's going to result in some form of accelerated approval. Okay. Barry, any thoughts here? You're welcome to take whatever area you want to focus on given the pipeline breadth. Anything we need to do in terms of patient identification, monitoring? What biomarkers do you envision kind of advancing going forward? First of all, I agree with everything Ryan said and the way he laid it out. Before the call, we were joking, tell me what the biomarkers are and tell me what they mean, and I'll happily use them. Many times it's incumbent upon us to figure out the biomarkers ourselves. For example, SAGE-718, which Steve mentioned, is an allosteric modulator of NMDA. The team here discovered a natural endogenous chemical that actually modulated NMDA, that it observed with these lower levels in Huntington's patients, 24-hydroxycholesterol. In the early phases, we're using that as a biomarker. This is what Ryan said. You can't take that too far because that's not necessarily the right biomarker in Parkinson's disease or Huntington's disease. Use the biomarker to come up with a biologic hypothesis. We confirm that in Huntington's patients. It turns out that modulating the pathway the way we have actually works well in Parkinson's patients as well, and we're testing in Alzheimer's patients. We use this to find the molecule, the pathway, the first patient group to start with. We're starting, as you know, Laura, with Huntington's disease, because it's an orphan disease, and we as an industry haven't developed a molecule that can improve cognition, memory, and learning ever. How do you design that Phase III? Well, designing it around an orphan disease gives us some of that pioneering regulatory flexibility to come up with memory panels that we can fulfill Huntington's disease, and if it works, apply that then to Parkinson's, and if the data supports, Alzheimer's. It's a series of markers along the way used for certain points that then gets other "biomarkers" for regulatory purposes. Okay. Norbert, PTSD would be a fabulous indication to apply biomarkers to for patient selection, but anything there that you see as very promising looking ahead? Look, I would say just in general, I look at biomarkers very much in line with the comments that were made, either to validate target engagement or to help stratification of patients for clinical studies. I do not look at them today as endpoints that could be used for regulatory approval. That might come, but that's not the emphasis we have. I mentioned earlier on fibromyalgia, fMRI and preKIMAT as validated biomarkers. You asked about PTSD. I would say, neuroendocrine markers, neuroinflammatory markers, actually even markers of white matter integrity are markers we look at and are interested in because they are known to be connected to the glutamatergic system. That is, of course, the system that we influence through NMDA receptor modulation in biology. It will truly, to come back to it's for the purpose of enhancing our confidence level through target engagement and patient stratification that we are giving the compound the highest probability of success to deliver a positive result in the clinical study. That's the emphasis on why biomarkers would matter, as long as the biomarkers can deliver that as a contribution to the study. Okay. Steve, schizophrenia, are there any biomarker initiatives that you think are going to bear fruit in the next decade here? Or what should we be watching for? Well, let me just say that biomarkers can be used, as Ryan mentioned, for several purposes. One is to assure that your drug is engaging its target and maybe evoking some physiological event that might underlie its therapeutic benefit. I think there are markers that are emerging for that in all the psychiatric disorders. To use a biomarker, though, and to qualify one for clinically stratifying patients like we do for breast cancer or lung cancer, you'd never treat a patient with breast cancer or lung cancer without knowing what kind of breast cancer and lung cancer they have, is going to take a long time. To qualify, it requires clinical data. Whether neurofilament light chain works in this disease or that disease, it's just going to require a lot more work, and we shouldn't prematurely assume that it's going to work in any given disease without getting the clinical data that we need. How many years did it take us to qualify LDL cholesterol, which I think is a pretty good biomarker for hypercholesterolemia and treating it? It's just going to take a long time. The brain is a little bit more complicated. There's not a lot of things we can measure in serum, plasma, or CSF, but our knowledge is increasing. I think ultimately it's going to require clinical data to know whether these markers can be useful in next generation molecules, almost. Okay. With our last couple of minutes, this is going to be our lightning round. I'd love to go around and ask each of you what you're most excited about over the next 10 years in the space. What might be the biggest success points or development wins that we might see unfold over the next decade? Ryan, how about we start with you? Yeah. I mean, never a more exciting time than now to be in the neuro space. I think what we're most excited by are the various modalities and ability to modulate protein and gene function. I can't imagine a more exciting time to be in neurodegeneration research and development. Wonderful. Barry, how about yourself? Yeah, I agree. I guess I put my body where my mind was. I think this is going to be remarkable for the next 10, 20 years in terms of the innovation. I'm excited by a paradigm shift in depression. I'm excited by coming up with an essential tremor drug where there hasn't been innovation in 50 years, the first ever drug to improve memory and learning in neurodegenerative diseases. I mean, if you think about what the four companies are doing, and clearly not everything we're doing is going to work, but there'll be tremendous innovation for the benefit of patients for years to come, and that's really exciting. Norbert, how about you? I would echo those comments. I am incredibly passionate about what we do as a group of companies in CNS. I think we are so much more informed with respect to biology, molecular biology, and molecular pharmacology. I think if we can make a contribution to addressing the enormous unmet need, I can only hope that all of us will be successful because with tens, if not hundreds of millions of patients in really dire conditions, there will be plenty of room for new treatment modalities. The fact that we work on different approaches to address very similar diseases gives me enhanced confidence that we will be successful. It will not be a matter of if, it will only be a matter of when. I think we are on a really encouraging path. I'm very pumped up about what we do. Steve, any final thoughts? Well, just to elaborate. For the major mental illnesses, major psychiatric disorders, one could state that we probably don't have drugs today that are all that much better than the very first drugs introduced some 50 years ago in the U.S. We need to do much better. These patients are suffering. Suicide is a big problem. Patients with schizophrenia live on average 15 years less long than those that don't have schizophrenia. Disabling illnesses, depression is the number one cause of disability worldwide, number one DALY, if you will. We've got a lot of work ahead of us, and I think all of us share an optimism that we're starting to see drugs that really are going to change the way we treat these diseases. These are potential game changers. We have nothing really in neurodegeneration, and hopefully the work that Ryan's doing will create, and other companies will solve that problem. That's a huge problem as well. Just across the board, I think science is winning, and we hope to be a part of that. Well, thank you gentlemen, all for your participation today. Really excited to see what unfolds for each of you in the coming months and years ahead. With that, I think we have to conclude our panel. Thank you everybody for participating. Enjoy the day. Thank you, Laura. Great to see everyone. Yeah, guys. Thanks a lot. Bye-bye.