Sage Therapeutics, Inc. (SAGE) Earnings Call Transcript & Summary

Good morning, everyone, and thanks for joining the Second Annual Sage FutureCast call. I'm Jeff Boyle, Head of Investor Relations at Sage. Before we begin, I encourage everyone to go to the investor and media section of our website at sagerx.com, where you can find the press release related to today's call as well as slides that we will discuss. I'd like to point out that we will be making forward-looking statements, which are based on our current expectations and beliefs. These statements are subject to certain risks and uncertainties, and our actual results may differ materially. Please consult the risk factors discussed in our SEC filings for additional details. We will begin the call with prepared remarks by Dr. Jim Doherty, our Chief Research Officer; Dr. Rob Lasser, Vice President, Late Stage Development; Dr. Helen Colquhoun, Vice President, Early Development; Dr. Mike Quirk, Vice President, Pharmacology; Dr. Aaron Koenig, Medical Director, Early Development; Dr. Steve Kanes, our Chief Medical Officer; and then finally, Dr. Jeff Jonas, our Chief Executive Officer. I will now turn the call over to Jim.

Good morning, everyone, and thank you for joining us for Sage's Second Annual FutureCast call. Last year, we explored Sage's approach to translational science, which is fundamental to our approach for developing new therapeutics for central nervous system disorders. We also discussed how thinking differently about interrogating the role of our target receptors in brain networks. We think differently about treating brain disorders, but we also think differently about how we approach the discovery and development of new medicines across multiple disease states. The breadth and depth of our pipeline reflects this approach with differentiated assets and the potential to help millions of people. And I'm happy to share more of that approach this year. Our R&D strategy is both proactive and predictive. That means we proactively follow the science with our discovery efforts and lead with human data to predictively approach drug development by addressing unmet medical need in patients who need new approaches. We believe this may lead to improved chances of success for a number of programs and areas across our depression, neurology and neuropsychiatric franchises. As Dr. Jonas has said many times over the years, there really has been a paucity of innovation in the treatment of central nervous system disorders for quite some time. This is in part because the human brain is very complex. And despite many advances in our understanding of central nervous system function, using preclinical studies and animal models, we still don't have a full picture of how brain function goes wrong in disease. Therefore, we focus on understanding how modifying key brain receptors like GABAA and NMDA receptors impacts brain function at the network level. That principle guides our preclinical studies in modeling the potential impact of our molecules on network function and experimental systems, and it also guides our focus on translatable endpoints in clinical development that we can use to understand how our candidate drugs may influence the human brain. The questions we ask in our discovery and development programs are thus designed to be proactive and predictive. For example, how does the GABA receptor PAM or NMDA modulator work? Well, if it works the way we expect, it should have defined and mappable effects on brain circuits, allowing us to select the correct molecule and also to identify which patients may benefit from these effects. Or can we use human subjects to give us increasing confidence. By leading with human data, we strive to solve for real human problems like impaired executive function, essential tremor and other diseases, addressing the things that matter most to patients. All this is underpinned by our translational and back translational approach. Everything we learn from a program can be applied to other programs, leveraging learnings to define the path forward, while driving for seamless execution. And with a wealth of chemical equity, more than 6,000 molecular structures in our library, we can be very selective on what we pursue. Today, you'll hear from the Sage scientists who are driving this development process. We'll start with SAGE-718, the lead asset in our NMDA platform that we believe offers great potential in conditions where executive function and cognitive deficits significantly impair patients' ability to lead meaningfully independent lives. We'll then talk about our next-generation GABAA receptor PAMs, SAGE-324, for the treatment of neurological disorders and how GABAergic deficits associated with essential tremor led us to consider whether GABAA receptor PAMs could be effective therapies for the disorder and how the properties of SAGE-324 could be ideally suited to address the needs of essential tremor patients. You'll also hear about our thinking behind the fundamentals of zuranolone, as a potential, as-needed treatment for major depression. And hear from a clinician who will discuss the value, if successfully developed, of both the rapid action of zuranolone in major depressive disorder and its lasting effects beyond dosing. And finally, we will tell you about our focus on following the science and back translation with brexanolone that led us to investigate the potential of brexanolone in COVID-19-related acute respiratory distress syndrome, or ARDS, where we've initiated a Phase III pivotal study based on a sound scientific hypothesis. We believe this is how drug development should be done. R&D at Sage requires the fortitude for deploying our methodology consistently over time. We're flexible enough to follow the science for our molecules, but we also set out to address particular unmet needs. And by doing that, I believe we have one of the most interesting pipelines in the industry. And on that note, I'll turn it over to Mike Quirk, Vice President and Program Lead for our NMDA platform. Mike?

Thanks, Jim. Good morning. I'm Mike Quirk, and I lead the pharmacology and translational science department at Sage. I came to Sage over 6 years ago, enthused by Sage's approach to translational science and drug development and by the opportunity to lead Sage's emerging NMDA platform. NMDA receptors constitute one of the major classes of excitatory receptors in the brain and play a fundamental role in regulating brain function. Over 40 years of scientific research has detailed the importance of NMDA receptors in learning, memory and complex cognitive processing, including executive functioning. However, despite the importance of NMDA receptors, drug developers have been historically challenged to deliver medicines that safely and effectively target this critical receptor. Since Sage's founding, we have been driven by the vision that the best approach to modulating NMDA receptors is to leverage the brain's endogenous modulatory machinery. Emerging research, including contributions from Sage's scientists and collaborators, has shown that a specific brain metabolite 24S-hydroxycholesterol can both modulate NMDA receptors and serve as a potential biomarker to identify patient populations of interest. Driven by these novel insights, scientists at Sage have spent the last 8 years creating a portfolio of drug-like molecules, targeting NMDA receptors. Importantly, we have been able to dial in the degree of modulation with compounds ranging the gamut from strong positive modulators to full negative modulators. Amongst this portfolio of molecules, 2 compounds, SAGE-718 and SAGE-904, have entered clinical testing. Today, we will discuss SAGE-718, a molecule specifically designed to be a strong positive allosteric modulator with good oral drug properties. Much like our approach to GABA modulators, we see SAGE-718 as the first asset in a broad neuropsychiatry franchise. Specifically, if we look at how disorders of cognition and behavior impact individuals in all age groups, we believe that Sage's portfolio of NMDA receptor modulators has the potential to provide the substrate and the flexibility for treating a range of Neurodevelopmental and Neurodegenerative disorders. I'll now turn over to Dr. Aaron Koenig to walk through some of our early clinical strategy with SAGE-718.

Thanks, Mike. My name is Aaron Koenig and I'm a Senior Medical Director in early clinical development at Sage. Given the broad spectrum of potential indications for our NMDA molecules, we thought it would be important to first, level set and ensure that everyone understands what we're talking about when we use the term cognition, and in particular, how we at Sage are thinking about disorders of cognition differently than others have in the past. So what is cognition? In a few words, it's the sum of all of our mental abilities. That's a fairly abstract definition, so to help explain, we've broken it down into its parts. In these 6 circles, we've highlighted the key cognitive domains as they're defined in the DSM-5. On the right, complex attention describes our ability to focus on a task of hand and do so, while tuning out everything else. Problems with attention are the primary deficit in ADHD, a fairly common disorder of adolescents and early adulthood. You're probably also familiar with learning and memory, which is the ability to take in information, file it away and then retrieve it at the appropriate time and place. This domain gets a lot of attention because it's the primary deficit in Alzheimer's disease. And while you need all of these domains working together to function normally, it's really executive functioning that's the conductor of this orchestra. Executive functioning is what controls our ability to plan, make decisions, confront challenges and navigate unfamiliar situations. In fact, executive functioning touches on so many aspects of what we do on a daily basis, organize, problem solve, multitask, skills that we need to do things like prepare meals, drive, even get dressed in the morning. Without executive functioning, it becomes really hard to operate independently. Executive disorders are seen in different conditions, but it's really the primary cognitive deficit in 3 diseases: Huntington's, which I'll talk about more in a minute; Parkinson's, which is typically thought of as a movement disorder, but really has a very strong cognitive component; and frontal temporal dementia, which is a form of rapidly progressing dementia. There are also other conditions in which executive changes play a smaller, but still important role in patient disability, and we think represent an important area of unmet need. And these include, among others, Alzheimer's disease, ADHD and Autism. Here's an example from a patient that we spoke with, is experiencing changes in executive function. It's a patient with Huntington's disease, though the narrative could be just as applicable to a number of other conditions. This 40-year-old has worked at the same job for over 4 years and now is starting to make mistakes that she never has before. She's having trouble at home managing her finances, needing to rely on her husband more and more to get by. And then this third comment, when something unexpected happens, it really throws her off. She describes herself as becoming a mess, needing to rely on her husband to help her get back on track. This difficulty with flexibility, dealing with the unexpected is a core deficit of executive dysfunction. That was a patient with Huntington's, which is an inherited Neurodegenerative disorder characterized most prominently by abnormal movements, so-called Chorea. While the disease is typically staged by the severity of these abnormal movements, labeled here in dark gray as the manifest period, it's appreciated more and more that there are other symptoms like cognitive dysfunction that precede the prototypical motor syndrome. Starting at a young age, even in their mid-20s, patients begin to show behavioral changes and problems with higher [indiscernible] cognitive abilities. And when you think about the life stage of these patients, adults in their 20s and 30s, working, perhaps even to start to raise a family, you can imagine that these cognitive problems can have very serious consequences. Leveraging biomarker data is really the first step in our early development journey at Sage, and that's how it began for the Huntington's story in our lead NMDA PAM, SAGE-718. SAGE-718 is a synthetic neurosteroid that shares a similar molecular pharmacologic profile with the naturally occurring molecule 24S-hydroxycholesterol, which itself is an endogenous modulator of the NMDA receptor. Using data from the naturalistic follow-up study TRACK-HD, we correlated 24S levels with measures of cognition, psychiatric symptoms and motor symptoms in patients with Huntington's. And we found that levels of 24S correlated with cognitive performance and in particular, executive performance as measured by tests like the Stroop, Symbol Digit and Trails B, but notably did not track with other aspects of the disease, including motor and psychiatric symptoms. Based on these and other findings, we arrived at our initial therapeutic hypothesis. By giving an NMDA PAM acting similarly to naturally occurring 24S, we might be able to correct aberrant NMDA activity seen in early HD and perhaps also improve cognitive symptoms. To test this hypothesis, we recruited a sample of patients with mild Huntington's and ran them through one of our signature open-label studies. The table at the bottom shows that these are patients with objective cognitive impairment and overall, mild functional impairment when measured by the TFC, or total functional capacity, which is the most common measure of functioning used in HD. We treated these patients with 1 milligram of open-label SAGE-718 daily for 14 days and measured their cognitive performance every other day, and we saw some very interesting things. First, when we looked simple cognitive processes like attention, measured by the detection task on the left, we didn't see any movement over the course of 14 days. Patients finished right where they started. But when we gave these patients more difficult tasks, here on the right, the 2 back task, which is a measure of executive performance, we saw improvement that kicks in robustly on day 8 and is maintained through day 14 of treatment. And this is consistent with the signal on the 2 back that we saw in our healthy volunteer, experimental mess and study. And we talked about at last year's FutureCast. And like we said then, there's nothing out there in the market that has a profile like this, the potential to augment that key cognitive domain of executive functioning. The Huntington's story is really interesting, but we also want to make sure that we keep an eye on the bigger picture. Where else is there unmet need? And how broadly can we apply these early findings? As I mentioned, we see prominent executive deficits in Parkinson's and to a lesser, but still significant extent in early Alzheimer's. This offers us an opportunity, both to see how an NMDA PAM impacts executive performance across different conditions, but also to learn about the drug's effects on other cognitive domains, for example, the core memory deficits seen in Alzheimer's. We already completed the open-label HD study and have just opened the Parkinson's trial, named the PARADIGM Study for enrollment. We also hope to start recruiting for our open-label Alzheimer's trial sometime in the near future. Conducting a series of small open-label signal finding studies isn't new for Sage, in fact, it's the same approach that we took when we first started developing our GABA PAMs. And it will help us understand the full opportunity offered by our novel class of NMDAergic agents, ultimately, so that we can make the right portfolio level decisions as we plan for later stages of development. Be it in the Neurodegenerative disorders, like we spoke about today; Neurodevelopmental disorders, like Autism and ADHD; or the challenging area of cognitive recovery and rehabilitation. Sage's NMDA platform offers a host of opportunities to help patients across the lifespan, and we remain energized by early clinical signals that we hope to build on over the course of the coming year. Now I'll turn it over to Helen Colquhoun, to talk about SAGE-324.

Good morning. My name is Helen Colquhoun, and I'm Vice President of Early Development at Sage. I'm also the program leader for SAGE-324, the lead asset in our neurology franchise. We are currently in Phase 2 of our program in essential tremor. Essential tremor is the most common in movement disorder. It is estimated that over 6 million people in the U.S. live with essential tremor. This prevalence is generally held to be an underestimate due to under and misdiagnosis and is predicted to increase as the population ages. Despite the high prevalence, at any one time, only about 20% of patients are actively seeking treatment. Essential tremor is a Neurodegenerative condition. It is associated with GABAergic deficits in the cerebellum that result in unwanted abnormal movements, typically of the hands, arms, legs, head and/or voice. The inability to execute normal function as well as the stigma that is associated with these abnormal involuntary movements can have profound impacts on the living of those living with the essential tremor. The GABAergic deficits associated with essential tremor led Sage to consider whether GABA PAMs could be effective therapies for the disorder. Last year, we presented data on the reduction in tremor seen in studies of patients with essential tremor treated with brexanolone and zuranolone that gave us confidence that we should progress SAGE-324 in this underserved patient population that has witnessed almost no innovation for over 50 years. It is important to understand that essential tremor is a serious and progressive condition that often steadily deteriorates over time, leading to social isolation and loss of independence. The abilities to write and perform fine motor tasks are commonly impacted and activities related to self-care, such as cooking and getting dressed, get increasingly difficult for patients with more and more reliance on assistance from caregivers. As essential tremor progresses, significant physical and psychosocial impairment and disability results. The burden of disability is a key concern for employers and private and public payers, including Medicaid. Improving tremor has the potential to also reduce the associated morbidities, improving functioning and well-being. Almost half of patients living with essential tremor report that the tremor interferes with their work. When you listen to patients, you realize that the tremor can affect nearly every aspect of day-to-day living and can make the simplest tasks difficult, if not impossible. We learned that tremor can lead patients to feel frustrated and embarrassed, while others are left significantly disabled and unable to really care for themselves. Rather than me give you examples of how, for instance, a moderately severe voice tremor can lead to significant communication difficulties, loss of self-esteem and loss of employment, let's take a moment to hear from some people living with the essential tremor about how it affects them. [Presentation]

We thank this international Essential Tremor Foundation for allowing us to share these patients’ vignettes with you. I'm sure that you understand how motivating the patient voice is in driving our work to develop an effective and well-tolerated treatment for essential tremor. The factors that determine whether and when a patient seeks treatment are varied and complicated. However, as patients progress from mild to moderate to severe symptoms with a corresponding impact on their ability to perform day-to-day activities and engage socially, they are more likely to present for assessment to the treatment. Research shows that eventually 60% to 70% of patients with essential tremor will seek treatment. Initially, pharmacotherapy is prescribed. Prescribing habits vary, perhaps because there is little evidence to support the use of one drug or another. Approximately 60% receive beta-blockers as first-line therapy, most of which is propranolol. Around 25% are treated with anticonvulsants, the most common of which is primidone. And approximately 15% receive a variety of other medications, such as benzodiazepines. Propranolol is the only approved pharmacological treatment for essential tremor in the U.S. Its approval in 1967 was based on a 2-week study in 9 patients. Data requirements for approval have certainly evolved since then. Not everyone can take propranolol, such as those with asthma, heart block, heart failure, emphysema or diabetes. Of those treated with propranolol, approximately 40% report that their tremor is at least halved, 30% report a partial or suboptimal response and 30% discontinue, either due to intolerable side effects or lack of efficacy. Treatment then evolves in partial or nonresponders with the addition of or substitution with multiple off-label therapies, ultimately leading in severe cases to surgical options. Essential tremor is associated with a substantial mental health impact. 43% of patients living with essential tremor report depression. Patients describe embarrassment and stigma associated with the tremor, which further contributes to the spiral down towards social isolation and mental health issues. Assessing the effects of drugs on the associated physical and psychological impairment and disability as well as on the tremor itself is of keen interest to patients, employers and payers. Reliance on caregivers is significant, providing a societal burden that is difficult to quantify. Most of the caregivers are spouse, but almost 1/3 are adult children. Now let me walk you through how we are progressing the SAGE-324 program in essential tremor, leveraging our learnings from other programs and listening to stakeholders, particularly patients. You've seen the data on this slide before, but as a reminder, this was a single-dose study of SAGE-324 in 6 patients with essential tremor, and we measured tremor amplitude with the Kinesia accelerometer as well as the clinician-assessed TETRAS performance scale. SAGE-324 was well tolerated in this study. The top figure shows the immune reduction over time in the combined upper limb tremor amplitude as measured by accelerometry. The bottom figure shows the plasma concentration of SAGE-324 over that same time scale. What you see is a striking PK/PD relationship. As the class of a concentrations of SAGE-324 increase after the single dose, the tremor amplitude decreases. As the plasma concentrations diminish, the tremor amplitude increases, although not quite back to baseline. Our hypothesis is that with repeat dosing and an optimized PK profile, we can maintain plasma concentrations above an effective threshold for all or most of the dose interval and that this will translate into sustained tremor symptom control. SAGE-324 was selected as the lead compound in our neurology franchise, partly because of the predicted good oral bioavailability and long plasma half-life in humans. These 2 attributes have now been confirmed in our early clinical studies. As you see from the PK simulations in the figure, plasma concentrations rise from the first dose to steady state. There is not a much excursion between the peak and trough plasma concentrations at steady state. And if, as an example, in this simulation, the optimally effective plasma concentration was 600-nanograms per ml, after approximately 2-weeks dosing, patients would maintain plasma concentrations above that threshold throughout the daytime dosing interval. Utilizing our understanding of the PK/PD relationship, our organizational expertise in formulating this class of compounds and lessons learned from zuranolone, we hope to be able to exploit the pharmacokinetic characteristics, I've described to tune the exposure to maximize clinical benefit throughout the day. We're currently enrolling in our KINETIC TRIAL, which is a double-blind, placebo-controlled study to evaluate the efficacy, safety and tolerability of a daily 60-milligram dose of SAGE-324 in patients with moderate-to-severe essential tremor, our target population. The data from this study will be used to model the doses for our planned dose-ranging study, which will be designed to define an exposure range and profile with the goal of balancing efficacy and tolerability for patients. The Sage team is planning to develop a comprehensive evidence package for the SAGE-324 clinical development program that, if successful, meets the requirements of all the key stakeholders, such as patients, physicians, FDA and payers. To maximize the effectiveness of this effort, we listen to patients, commission research with clinicians and payers and engage with FDA early in development to learn about which factors will be most meaningful for approval, future utilization and access to the drug, should it be approved. We also pull through learnings from other programs. An example of this is how we understand the utility and behavior of endpoints utilized in previous studies in essential tremor. We incorporate a broad range of endpoints that will service the needs of our stakeholders from the technological, such as accelerometry; to the traditional, such as the TETRAS performance scale; to the innovative, such as exit interviews with patients in early clinical studies. The SAGE-324 team is adopting innovative strategies and tactics to drive the program for essential tremor forward, with the goal of developing a safe and effective medicine for this serious progressive condition that is currently underserved by available therapies. The execution of the KINETIC TRIAL is on track, despite the COVID pandemic. And plans for next steps in the development of SAGE-324 are mature and being implemented on schedule. I will now hand over to Dr. Rob Lasser to talk about zuranolone.

Good morning. My name is Rob Lasser. I'm an adult and geriatric psychiatrist who leads late-stage development at Sage. On behalf of myself and the Sage team, I want to begin by thanking the doctors and their staff and the people with depression in their families who've agreed to participate in the zuranolone clinical development program. We've recently announced that following discussions with the FDA, we've identified 3 unique opportunities, which offer multiple pathways to develop zuranolone as a potential treatment for people experiencing episodes of postpartum or major depression. As a reminder, the vision for zuranolone rests on its potential to work rapidly and continue providing benefit beyond the period of dosing. Most current treatment options for MDD, target therapeutic mechanisms that block monoaminergic uptake in the brain to support symptomatic improvement. This older mechanism can result in lengthy delays to symptomatic relief, often up to 6 weeks, with that relief frequently being incomplete as many people experienced a partial response to treatment. New mechanisms have emerged over the past decade with recently approved antidepressants, impacting the more fundamental mechanisms of GABA and NMDA. Some examples are ZULRESSO for PPD as well as esketamine for treatment-resistant depression. In focusing on the GABA pathway, we set out to think differently about defining wellness and people with depression and provoke a fundamental change in the way depression is treated. Our goal is to provide a rapid response and sustained effect beyond dosing in most people. We know a few things must be true for depression treatment to be used as needed when symptoms occur. First, a treatment must work quickly to be used as needed. As clinicians, we don't want people with depression to endure symptoms for a month or more, while waiting for a response or a re-response if their depression occurs. Second, and as needed treatment must provide a response that allows for a treatment-free interval or remission by restoring the balance in the brain. We can now see more easily, though, the variety of ways in which zuranolone, if we're successful in our development efforts, might be used in the real world, if approved. We can ask also maybe zuranolone is more of a pharmacologic tool than a simple antidepressant and what does that mean for clinicians? Well, based on our knowledge of MDD, we've identified several ways the target profile of zuranolone might be flexibly matched to different types of people with different types of MDD progression over time. This reflects the potential for 3 distinct indications with topline data expected in 2021 and multiple potential regulatory pathways for approval, if we're successful. First, zuranolone as a single day -- a single 14-day course of treatment in postpartum depression. Our ongoing SKYLARK study is evaluating the efficacy and safety of zuranolone at 50 milligrams in postpartum depression. In our earlier ROBIN study, women in the 30-milligram zuranolone treatment group showed an improvement in depressive symptoms by day 3 compared to those in the placebo group, and that difference was maintained through day 42. Second is the use of zuranolone in major depression under 2 paradigms, monotherapy and add-on therapy. You could think of the monotherapy target profile as being analogous to a person being a zuranolone patient. The typical zuranolone patient, if we're successful with this treatment approach, would experience a rapid early response, followed by maintenance of their therapeutic benefit through several months following the initial 14-day course of treatment. We recently reported on the 6-month follow-up cohort from the MOUNTAIN Study in MDD and in participating subjects with response after the 14-day treatment period, a large majority maintain this response throughout the 6-month follow-up regardless of arm. Out of those participants who were responders to zuranolone 30 milligrams, about 75% who responded to day 15 continued to maintain that response 6 months later. If approved for this treatment paradigm, should symptoms return, a repeat 14-day course would be used as needed. We do expect that the FDA will require additional long-term safety data to support a regulatory filing for this treatment approach. Now importantly, other people may also find a new therapeutic option because of the issue of partial response to a prior antidepressant. Well, why is this? We know that chronically used therapies, which dominate treatment today can produce a partial response, and that add-on therapy is often needed to move a person from a fading partial response with worsening symptoms back to stability in a rapid fashion. That partial response can leave people with persistent insomnia, anxiety, lack of motivation or a low mood, which limits their return to full brain health. This add-on treatment approach is one we continue to explore with our ongoing WATERFALL study, given that participants in the trial, as with our completed studies in depression, can be on stable doses of antidepressants. Of note, between 15% and 30% of people who enter our short-term trials like ROBIN or MOUNTAIN were on stable doses of antidepressants. Finally, we're exploring the use of zuranolone with a newly initiated standard antidepressant starting both medications simultaneously. In the CORAL study, we're evaluating the specific use of zuranolone for a rapid effect and a standard antidepressant for maintenance. This approach is not because zuranolone doesn't have the potential for maintenance of effect, but because we believe the co-initiation approach, if we're successful, gives us the potential for an expedited regulatory pathway to approval. Now each of the studies I've just described are part of our Landscape and Nest programs. So let's turn to the clinical program components briefly so we can connect our studies to the data they'll deliver with callouts to guide our focus. First, on the left-hand side of the slide in lilac, the approach to monotherapy in postpartum depression will be supported by the ROBIN and SKYLARK studies. We expect that about 1/4 of subjects will enter the studies on standard antidepressants. This will allow us to generate data on the benefit of zuranolone in either the monotherapy or add-on context. In the middle of the slide in orange, the approach in MDD will be supported by the MOUNTAIN and WATERFALL studies for short-term outcomes and SHORELINE and REDWOOD studies for long-term outcomes in safety. SHORELINE will provide more insight on how an as-needed approach to treatment works over the long term, including people who may have started an antidepressant at different points over time. This naturalistic study will first give us data to potentially support the development of zuranolone 30 milligrams and 50 milligrams as a monotherapy as well as add-on approach. On the right-hand side of the slide in blue, zuranolone as a rapid response therapy when co-initiated with a new antidepressant is being evaluated in the CORAL study. The idea here is to use the study -- to use zuranolone for a rapid response during the period where standard antidepressants often have a delay in onset to effect, while we use the standard antidepressant for a maintenance of effect. If the CORAL study is positive, we believe we could first seek regulatory approval for the use of zuranolone when co-initiated with a new antidepressant. And if we're successful, this would enable us to leverage the profile for physicians and patients to gain real-world experience with zuranolone for this use, and in turn, better understand its rapid-acting profile. If we're then able to generate data from the long-term placebo-controlled studies of zuranolone, we would have an sNDA seeking approval for repeated use over the long-term with zuranolone monotherapy and MDD as well as add-on therapy. If we're successful in these efforts and zuranolone is approved for all these uses, the clinical use of zuranolone would then be flexibly applied based on the profile of the person undergoing treatment. I want to share a little background as we approach the initial presentation of a data cut of cohort 1 from the SHORELINE study later this year, that is people initially treated with 30 milligrams of zuranolone. The SHORELINE, or MDD-303 study, was designed to provide a naturalistic approach to uncover real-world use patterns with a primary focus on the safety of long-term repeated use of zuranolone. People with MDD in the study received an initial 14-day course of zuranolone with responders being permitted to continue into full 1-year follow-up with access to zuranolone as well as all available antidepressants. Depression self-assessment was conducted every 2 weeks, with in-clinic visits every 2 months or as clinically needed. COVID has not significantly impacted the collection of clinical data in SHORELINE. Zuranolone is being used in SHORELINE in adult and elderly people with MDD in 3 primary ways. The first is monotherapy with repeated dosing as needed. The second, in conjunction with a chronically dosed antidepressant which was started prior to zuranolone, what we call partial response. Or third, in conjunction with a chronically dosed antidepressant, which has started after zuranolone course is completed for its maintenance properties. The evaluation of the use of zuranolone in conjunction with chronically dosed antidepressions in which the treatments resided simultaneously will be the objective of the CORAL study, MDD-305. To help us further understand the potential real-world applications of zuranolone, including the profiles of people with MDD who may benefit from zuranolone, if we're successful in our efforts, I'm pleased to welcome Dr. Greg Mattingly, one of the psychiatrist investigators in the SHORELINE study. Dr. Mattingly is an adult and pediatric psychiatrist and an associate clinical professor at Washington University in St. Louis, where he received his medical degree under a Fullbright Scholarship. As personal investigator in clinical trials for Midwestern Research Group and a founding partner of the St. Charles Psychiatric Associates, he's executed more than 200 clinical trials across multiple psychiatric disease states. He's also served in an advisory capacity for Sage, given his expertise in clinical psychiatry. Thanks for joining me, Dr. Mattingly.

Glad to be here, Rob.

As we've discussed and been talking, Sage is developing a number of different ways that zuranolone can be used. Can you comment about how you think people are hearing about and thinking about this new way to approach treatment?

I think people are hopeful, hopeful this may work in a way that is different, hopeful that this may work where a standard in their present hasn't been fully successful, and hopeful that this may allow patients to take an antidepressant when needed instead of forever.

And so when you think about different patients and you meet with them, are there some critical elements that you like to bring out in those conversations that can point to how a person would embrace a more rapid response therapy?

Certainly. Our traditional antidepressants have taken 4 to 8 weeks to work. For many of my patients, 4 to 8 weeks can feel like a lifetime. Imagine a mom, struggling to get up and take care of her kids. Imagine a dad, whose head is locked in a depression, as he struggles just to make it through the workday. Imagine a university student sitting in class who's afraid of losing a semester due to his head being stuck in a fall. Now imagine if this was your wife, if this was your husband or if this was your child. These are the patients I see each and every day. So the ability to alleviate the symptoms of depression in a matter of days to several weeks versus many weeks to several months could be a game changer for many of the patients I see whose lives are being damaged by a depressive episode. As a clinician, the flexibility to use zuranolone by itself or on top of another antidepressant could be a real advantage to a lot of my patients. If established, the ability to use zuranolone along with another antidepressant to potentially jump-start a response or even achieve remission within several weeks would be an added advantage for patients who've had a partial response or a prior response to their current or a standard antidepressant. In that case, the choice of which antidepressant would be based on what's been your best medicine in the past or is it worth trying a new antidepressant when we try zuranolone. If Sage is able to demonstrate the ability to use zuranolone as needed to treat depressive episodes, it may revolutionize the way we think of treating major depressive episodes. Rob, for this type of use, we'd rely on patient education and how to monitor their own symptoms, education for clinicians on how to detect a relapse and the importance of shared decision-making between patients and their clinicians to improve the lives of individuals who have struggle with depression. In this era, the good news is the wide adoption of telepsychiatry enhances our ability to communicate with our patients and bring information into their lives. The easy exchange of rating scales, such as the PHQ-9, where patients can monitor and measure their own symptoms, has been a major breakthrough.

Well, it’s great to see how things like telepsychiatry are really moving things forward. From the medication sphere as well, I'm wondering if there are particular patients that you think would benefit from this nonchronic approach.

And as needed, fast onset treatment would be a vital enforcement for patients who need to return to function to get their life back. A student, a busy working executive, a mother or a father trying to take care of their family, A fast onset as-needed treatment may also be of special interest to a vulnerable elderly patient, whose physical condition is quite often being impacted by an episode of depression.

So I think it's clear how zuranolone monotherapy could be used as needed as symptoms recur. I'm wondering, if you could talk a bit about this partial response. As I understand, it is one of the most difficult to treat or difficult issues within depression because it's such a large number of people.

Day in and day out in my clinic, I see people who are better, but not well. So for antidepressant partial responders, the potential to have a fast onset, as needed treatment could be a dramatic improvement compared to our current switching or augmentation strategies.

So when you think as a clinician, and all of these different ideas about how to treat people, how do you put them together when you're matching an outcome that you want to see for a patient with the drug that you might use and thinking particularly about how the flexibility of zuranolone may offer you a new opportunity.

Rob, if Sage's development is successful, it would give the clinicians like myself and antidepressant treatment that could potentially improve symptoms of depression in days to weeks and that you only need to take for 2 weeks when you're having a depressive episode. This will give our patients the flexibility to treat their depression when needed and an option to take this medicine either by itself or along with their current antidepressant.

I remember in practice that many patients, and particularly men, often would get worried about being trapped on these chronic antidepressant treatments. I'm wondering if you could comment on your thoughts on that.

Many of my patients, many of our patients are resistant and use the exact word of getting trapped, getting trapped on an antidepressant that has to be taken on a day in and day out basis and on a chronic basis. So the need to take a medicine on a daily basis, it's fraught with a lot of, not just compliance issues, but long-term side effects, such as weight gain, sexual dysfunction. And the fear that if I try to stop my medicine, I may have withdrawal side effects that can occur with a lot of our standard antidepressants.

Understood. Well, it's a lot to think about and a lot for us to think, going forward together Dr. Mattingly. I want to thank you for joining me today, and I look forward to working with you more as SHORELINE data comes out.

Thanks for inviting me, Rob.

So to close, I just want to finish with a reminder about where we are in the execution and planning of the Landscape and Nest programs. WATERFALL and SKYLARK are enrolling well, and the new 50-milligram cohort of SHORELINE, that's roughly 175 people, is now fully enrolled. The CORAL study is planned to commence enrollment before the year's end. We continue to hold on 2 trials in RAINFOREST and REDWOOD, pending the outcome of new data reading out in 2021. In SHORELINE, the zuranolone 50-milligram reports so far of the treatment being well tolerated with only 20% of subjects, down titrating to 40 milligrams. We continue to execute as planned for data releases in 2021, and I really want to thank you for your attention this morning. Now I'd like to turn things over to Dr. Stephen Kanes.

Thanks, Rob. And I also want to thank Dr. Mattingly for joining us to share his really interesting perspective. You've now heard about how we think differently about our traditional mechanisms of action and brain networks. Beyond that, also how we proactively follow the science with our discovery efforts and predictively approach drug development by leading with human data to solve real human unmet needs. Let's turn now to another way we're applying the same approach with another of our GABA modulators, brexanolone. We announced in our second quarter earnings call that we've received clearance from the U.S. FDA under the CTAP program, or a Coronavirus Treatment Acceleration Program, to initiate a Phase III study with brexanolone in patients with advanced COVID-19-related acute respiratory distressed syndrome, or ARDS. Before I get into the specifics around the study, I want to take a minute to share our rationale for this development and our initial approach to potentially expanding the utility of our platform. Now in the process, by doing this, we'll inform both the underlying mechanism of our drugs as well as how they might be developed in the future. We've talked a lot in the past about PPD with brexanolone, potential for major depressive disorder and additional brain health disorders with zuranolone, and we've also gone into some detail about the potential utility of SAGE-324 in essential tremor. These are all brain health disorders. They're all related to network activity and the potential for our GABAA positive modulators to impact network activity. But we know that GABAA receptors are not just located in the brain and they don't only modulate brain circuitry. These same receptors can be found around all over the body, but also very specifically in lung tissue and in respiratory smooth muscle that may potentially modify both lung vascular permeability as well as inflammation. Mechanisms related to GABA PAMs are what led us to consider exploring the use of brexanolone in COVID-related ARDS. This is a unique mechanism and it may very well be important in modifying the course of lung inflammation that requires the use of ventilators for ARDS in patients with COVID-19. So what's the pathophysiology of ARDS in COVID-19? And how might brexanolone will help? Moving from left to right, as we all know by now, COVID-19 is a respiratory coronavirus, so its main initial infectious site is the respiratory tract. Symptoms can range from mild or even asymptomatic to severe, and many of these more severe cases can lead to life-threatening pneumonia and acute respiratory distress syndrome. In addition to damage done to respiratory tissue by the direct infection of lung tissue, through uncontrolled inflammation, the so-called cytokine storm, excess fluid accumulates in the lungs in what's now called acute respiratory distress syndrome. This may occur -- in very severe cases, it can lead to complex treatment and has a high mortality rate, estimates of up to 50%. As we learn more about COVID and its impact on the respiratory system, we're seeing injury to the lung tissue that is epithelial damage, we're seeing impacts on the blood vessels or endothelial damage, fluid accumulation, ultimately, fibrosis and scarring. So how might brexanolone improve ARDS? First, through its effects on inflammation. GABAA receptors are located on key cellular components affected by the coronavirus in the lungs and in animal models have been shown to modulate inflammation. The cartoon on the left shows the general mechanism in a healthy person, and the cartoon on the right illustrates the same pathway in the setting of COVID-19. A harmful stimulus such as infection, mechanical damage or tumor causes tissue damage. The tissue damage releases what are known as DAMPs or DAMPs, damage associated molecular patterns. These are the alarm system of the cell, which alerts the immune system to react and potentially fight off the harmful stimulus. These DAMPs also activate the receptors of immune cells called TLRs, which cause this reaction to occur in excess, resulting in more tissue damage, more DAMPs and so forth in a feed forward accelerating cycle. This is the cytokine storm associated with severe COVID-19 acute respiratory distress syndrome. The figure on the right illustrates these effects in the specific setting of COVID. Published and internal research led to our hypothesis that by decreasing inflammation and increasing smooth muscle relaxation, brexanolone may reduce, in a unique way, the mechanisms that are driving ARDS in people with COVID 19, thereby improving the likelihood of ventilator-free survival. We, therefore, are conducting a study in ventilated patients with COVID-19-related ARDS in the ICU. This is another example of our predictive approach to drug development. The primary endpoint is the most clinically meaningful one, this is the percentage of subjects who are ventilator-free, 28 days after treatment. It's what patients want. They want to be off the ventilator and home. It's what physicians are looking for, the place that's the most difficult to achieve in these severe patients. But beyond the primary endpoint and equally important for us understanding the utility of this mechanism in COVID are the secondary endpoints that we're including. These are measures of respiratory function as measured by changes in lung compliance and improvements in ventilatory capacity. We'll be looking at the use of other anesthetics, potential improvements in the use of other anesthetics. And perhaps most importantly, we'll be looking at changes in serum biomarkers, both inflammation and inflammatory cytokines. We intend to enroll, up to 100 patients in the trial. And if successful, this study could be used for registration purposes, but perhaps most importantly, it's the right thing to do. We believe that based on our unique science and the fact that brexanolone has been extensively used in the ICU setting during its earliest development as a treatment for severe seizures, we are in a unique position to test brexanolone for an entirely new indication. In the face of an ongoing global health crisis, this is the right thing to do. And during this unprecedented pandemic, we believe it's our responsibility to investigate any potential solutions that may help to improve patients' outcomes or reduce the overall burden on the health care system. Beyond that, we'll be understanding our -- the medications, our mechanism of action, the potential utility of these mechanisms for respiratory function as well as its improvement in inflammation. Every bit of data that we obtain from such a trial will be useful both for patients in -- who are experiencing COVID ARDS, but also for our internal decision-making, as we think more deeply about our pipeline and how to move it forward. I look forward to updating you on our progress. I suppose that is just getting underway. And I'll turn it over to Jeff Jonas for closing remarks.

Thanks, Steve. And I want to take a moment to thank all of the folks at Sage for the tremendous science and hard work that have brought us to this point today, and I'd like to thank all of the audience for attending our FutureCast today. One of the question that comes up is how much confidence do we have in zuranolone, and I can say that zuranolone is really the end result of a lot of really great science and exploration. And the confidence we have in zuranolone and its potential is no more or less than the confidence we have in the rest of our multi-franchise brain health programs. We have multiple programs in the clinic. We're very excited about all of them. One of the things that we take pride in at Sage is our ability to move with urgency and in a way that's different than other companies, and I hope you've seen that today. We've begun with aggressive medicinal chemistry, a large chemical library and compounds, moving through translation rapidly and into clinical trials. This has led us to a position where we have multiple catalysts upcoming over the next 18 months. This has been a hard time for many companies, and I think, I'd like to, again, congratulate the folks at Sage for really working hard and with great urgency in areas that we all think are so important for patients in our attempts to advance brain health. So with that, I'd like to thank everybody again and open this up for questions and answers.

[Operator Instructions] Our first question comes from Cory Kasimov with JPMorgan.

I guess, 2 of them for you. I guess one thing I wanted to ask about, can you talk about the main causes of the down titration for patients getting 50 milligrams of zuranolone in the SHORELINE trial? And then I have one follow-up.

Cory, this is Jim. Thanks for the question. I think we'll ask Rob, if he wouldn't mind, to take that question on.

Sure. I hope folks can hear me. The down titrations have been associated with the same as we saw with 30 milligrams, patients doing the next morning, often the explanation for down titration.

Okay. Perfect. And then, Jeff, you mentioned, I believe in your last quarterly call, the potential that we see some of the 50-milligram data from SHORELINE in 2020. I just want to clarify, to the extent we do get any data from this, would it include both patients who were treated with 30 and then retreated with 50 as well as just those who start on 50 in the new cohort?

I'll take that, since it's a simple answer. Right now, we're still analyzing the data, but our hope is that we would have that sort of data. It's a little -- we're not quite up to the point of finalizing it. But certainly, we're hoping to see some re-treatment data as well as some initial treatment data and some safety data.

Our next question will come from Brian Abrahams with RBC Capital Markets.

Two for me, one on 217 and one on 324. First off, on 217, I guess, just a follow-up on the SHORELINE dose reductions, just wondering if you can maybe put that into context. It looked to us like that didn't seem too much higher than what you had seen with 30 milligrams. But just wondering, if there's any impact on what you're seeing there to your powering and approach for the ongoing Phase IIIs? And then I had 324 follow-up.

I can jump in quickly...

Sure, Brian. Go ahead, Rob.

Sure. So with respect to 50 milligrams, we now have over 250 people exposed. The safety outcomes are exactly what we expect. There's no loss of consciousness that's been reported. And I'm not going to go any further there. I'd just ask you all to stay tuned for our fourth quarter release.

Got it. That's really helpful. And then on 324, you've mentioned that the -- stated the ongoing KINETIC study would be -- that you plan to use that as a basis for additional dose-ranging work. And I'm just wondering if that stems from anything that you're seeing emerging in terms of activity or PK/PD from that study and some of the things you may be exploring in terms of dose frequency, dose level titration or alternative formulations for the subsequent study.

I think the answer to the question is that at this stage of the program, recall that we've learned quite a bit about the response to essential tremor from both the ZULRESSO and zuranolone programs. But as we're moving forward with the SAGE-324 and start thinking ahead to all the aspects of a registration program, if you think of Helen's presentation and really trying to understand the full benefit of SAGE-324 to tremor patients, we are thinking in terms of understanding the effects across multiple doses. And it's really dealt the evolution of the program as we continue to gain confidence in the mechanism in essential tremor patients. Helen, would you care to add anything to that?

Sure. Thanks for the question. We -- obviously, the study is ongoing at the moment, so we don't have any data. But you saw that closed PK/PD relationship. So we'll be able to use the data from the Phase II first study to select doses for the dose range next year. As to the dosing frequency, from the PK profile, you can see that we were on the money when we selected 324 for its long plasma half life. And we are expecting that we will be able to dose once daily at the doses that we select to put into that Phase II study.

Our next question will come from Gary Nachman with BMO Capital Markets.

First on zuranolone, there's a lot of different approaches you're looking at. Is it most likely the first indication would be simultaneous with other antidepressants and then followed by other indications for use as monotherapy? Is that the likely approach that we'll see?

Gary, the way I would think about it is, as you know, the Landscape program has 3 parallel pathways to potential approvals. And if any of them are successful, that would potentially be the constitute of regulatory filing. And so it's -- we -- as you heard from Rob, we're expecting results in 2021. So at that point, it's really going to be driven by the results from the studies, results from the CORAL study to drive a co-administration application, results from the SKYLARK study to drive a partum depression application and the results from WATERFALL and REDWOOD for episodic use. But I think we're -- although we will be driven by the data and look at the sequencing, I think the results from the CORAL study and the SKYLARK study are the ones are most likely to drive us in 2021.

Okay. And then if you use zuranolone together with other antidepressants, any concerns with drug-drug interactions, as you're exploring the higher 50-milligram dose in greater detail?

Well, at this point, of course, we do all our...

Yes, Jim, I can just add...

Yes, Rob.

Yes. As you know, in SHORELINE study, there's a large number of our antidepressions, we have not seen anything difficult or concerning at this point.

And recall, we won't restrict new antidepressants in the studies. And so we have a fair amount of data already with a variety of antidepressants from already completed studies.

Okay. And then just lastly on brexanolone for the COVID patients, would they take that with anything else potentially? Or is the study just looking at monotherapy? And what's the timeline for getting that data?

Yes. Thanks for the question. The goal actually is to add brexanolone on top of standard of care. And so we know the standard of care continues to evolve. But regardless of what patients are taking, this is actually a placebo-controlled Phase III trial. It's added on top of standard of care for those patients with COVID ARDS. In terms of time lines, what I can tell you right now is what we're starting this fall, a lot of it's going to depend upon what the trajectory of COVID is. We'll have a better -- as we always do, we'll have a better idea of what our time lines will be as we start the enrollment process.

Our next question comes from Paul Matteis with Stifel.

I have one tremor question and a couple of quick zuranolone ones. On 324, can you comment on the degree to which you'd characterize different aspects of the profile like sedation and also tachyphylaxis, understanding that those are 2 pretty significant drawbacks of chronic [ enzo dose ] in the tremor space? And then separately, on the SHORELINE study, just 2 quick questions. One, what are you seeing for long-term retention rate? A lot of the recent depression studies show something like 40% to 50% at a year. And two, were you surprised by the durability of response in the MOUNTAIN study? It seems like it was a little bit better than the implications of what you had said previously around this whole 1/3, 1/3, 1/3 type of thing. So just kind of curious if that MOUNTAIN 6-month data has created a new benchmark in your mind for what to expect as it relates to retreatment and durability.

Paul, thanks for the question. This is Jim. Let me answer the 324 question, I think, that's a quick one, and then I'll turn over to Rob to address your questions around the SHORELINE study for zuranolone. For 324, of course, with the mechanism of action that we have with molecular steroids one of the key features is that we do see a distinctive profile relative to benzodiazepines. And so we have data from both nonclinical as well as clinical studies suggesting that we don't see a lot of acute tachyphylaxis. Now of course, in the 324 program, the Phase I work we've done to date is mostly with short-term dosing. So that will be one of the values in the KINETIC study that's running at the moment is to understand what, if any, effects we're seeing around tachyphylaxis with 324. But based on the volume of data we have from the programs to date, that's not something we're expecting to see. And then, Rob, can you address Paul's questions on the SHORELINE study?

Sure. I can tell you, Paul, that the discontinuation rate is right in line with what we expect for other trials. And I'll also tell you that we really need to wait until the study is complete when we start thinking about repeat treatments and so on because it's used in many different ways. So we are seeing repeated use, of course, and that will be part of the collective data that we released in the fourth quarter.

Well, I'd just add one other thing, which is that the long-term data, and you had asked about it, it's not surprising at all. I mean we've been getting anecdotal reports around patients having long durable responses throughout the development program all the way back to little less. So -- and so this is our first real chance to start talking about those things with rigorous clinical data. But no, it's exactly the kind of information that we've been looking to establish in the program and why we think this is potentially such an important treatment for patients. So yes, no, it was not surprising at all. And actually, it's great to be able to talk about it with data to back this up.

I just wanted to ask, Jim, one more quick thing. How long do you think is it okay to dose, to derisk tachyphylaxis. How long does it take to manifest on a benzodiazepine?

So definitely with a benzodiazepine, it's within days. And so I think that's -- we would be expecting not for a very long course of dosing to be necessary to be seeing any kind of tachyphylaxis side effects. So a couple of days, would be the short answer.

Our next question will come from Ritu Baral with Cowen.

I did want to follow-up on Paul's question about the sedation related to the tremor program. Can you walk us through, I guess, what you're seeing around the therapeutic index, where sedation comes in relative to the plasma levels that you showed on that slide and weather we should be thinking about that in terms of like an absolute threshold effect -- I'm sorry, absolute threshold plasma level or a Cmax-driven event? And then I got a follow-up on CORAL.

Sure, Ritu. And I think the way we think about it, of course, is we -- the profile of 324 is distinct pharmacokinetically from the profile of 217, but the pharmacological profile is similar. So the types of sedation-related events, you would expect to see would be similar. But the doses in the times at which you see those effects are going to vary a little bit with 324. And the best data set to describe that is the results from the healthy volunteer Phase I studies with 324 that were conducted last year. And maybe, Helen, could you talk a little bit about the tolerability profile that we saw from the healthy volunteer studies in Phase I for 324?

Sure. Thanks for the question, Ritu. We did see some somnolence at the higher doses, as we expected to, in the Phase I studies. And so we're confident that, that is the tolerability sort of marker that we'll be looking at in our Phase II program. And remember, I went through how we're going to model the data and use that PK/PD relationship to select our doses for our dose range. And we're lucky that we are able to map out the balance between the efficacy and the tolerability. So we're keeping it front of mind. And in the end, the goal is to select a dose to move into our Phase III studies, and that is well tolerated, does not have intolerable sedation, but it's still active. And we're confident that we'll be able to move the program forward on that basis.

Great. And then on CORAL, do you -- have you finalized the protocol for concomitant SSRIs or SNRIs for this trial? Like what's allowed? And are you requiring a certain holiday for previous cycle of SSRIs or SNRIs.

Yes. I think Rob can give you the details of the playing around the protocol for CORAL.

Sure. We're just finalizing the protocol now, they're speaking back and forth with agency, but other than that things will progress. There hasn't been requirements for people to stop drug, but those are all maintained within the protocol and on some level proprietary. So folks will be starting new medications at the same time. But some of those details, of course, we'll review with you once the study results are available.

Is it fair to assume that any oral antidepressant would be allowed for concomitant start as part of CORAL?

Not any. As you can see with these trials in the past, there's been a collection or a representative basket of antidepressants, and that will be the same in this trial as well.

Our next question will come from Akash Tewari with Wolfe Research.

Just on this down titration, can you go over what the current protocol is for the upcoming Phase III trials on how you handle that downtime titration? Will those 40 -- let's just say, some patients get on the 40 mg arm, will they be allowed to be included in some type of primary analysis? And can you kind of walk -- is that 20% number that you saw in SHORELINE, in line with the expectations you had going into the study?

Quite simple, the answers are yes. I mean these are the expectations....

Yes. Akash, I think the way...

Go ahead, Jim. Sorry, do you -- the expectations that we have are roughly around 20%. There really hasn't been anything out of the ordinary. And when patients need a decrease in dose, they return to the clinic and get a decrease in dose and continue on to the trial to finish. So it's quite simple when you're decreasing the dose in the course of the trial, and those patients are counted in the overall analysis. So that's a very standard approach in these studies.

Yes. One way to think about it, Akash, is when we look at the outcomes from these trials, we're trying to understand what percentage of patients actually, say, cease treatment in any of the arms due to adverse events. That's the kind of data that we have at the end of a trial, and we really look to see what effect it has. I mean you can try to decipher that by simply looking at numbers that we relate around sedation or somnolence. Those really are, for the most part, mild or moderate and don't necessarily represent a reason for people to stop taking the medication. So that's really a part of the outcomes that are important for -- to follow, whether you're -- it goes back to Ritu's question before, if you're looking at 324, we'll get the data about what effect the adverse event profile has and specifically which ones, as a result of the placebo-controlled Phase II data. So that's the way we look at those data. Obviously, all drugs have an adverse event. We look to see how it effects the [indiscernible] do what it needs to do.

Right. And just a quick follow-up. It seems like based on your response, the reason for down titration is kind of mild-to-moderate somnolence. Is that fair?

Rob, you want to take that one?

Yes. Yes, it's fair. Yes. I think that's widely accurate.

Yes. Yes. I mean, it's not -- it wouldn't be the only reason, but of course, that would be the most common adverse event. And so, yes, of course.

Our next question will come from Yatin Suneja with Guggenheim.

Just a couple of questions on SHORELINE. So with regard to the 50 mg arm, are you enrolling patient with a certain cutoff for HAM-D, I think, 24 or above, if you can clarify? The other question I have is that in that study, you have -- or you have the ability to evaluate these 4 different treatment approaches within just one study. Given that there is no placebo arm, just help us understand how should we interpret those data when you're going to be presenting those? Is the trial big enough for you to detect a meaningful difference between the patient, let's say, who is on a complete monotherapy versus on a stable antidepressant versus somebody new who starts on an antidepressant later.

It's a big question. I'll answer...

Sure. A couple of questions in there. I think to the first question, there are...

Sorry for the delay. I can answer the second question first, which is it's a naturalistic study, and it's really a safety study of repeated dosing. So what we're going to see is real life. You're right, it's not a controlled trial, but it gives us a very robust picture of how the medication could be used in the real world. The answer to the first question, we have not changed any of the enrollment for SHORELINE when it comes to HAM-D. So it remains the same as it always has been, which is patients enter the study with a HAM-D of at least 22 and then they get retreated as needed. They have to, of course, surpass certain thresholds. But there's been no change there as well. The only changes to HAM-D have been in the WATERFALL and the SKYLARK studies.

Our next question will come from Andrew Tsai with Jefferies.

One question on 324. So my understanding is you're comparing that to placebo with efficacy being the change in upper limb score. So one, curious what a placebo response could show within the subscale. Two, what do the existing drugs, I guess, interval show specifically on this subscale? And then three, based on the Phase I data, I think you showed a 50% reduction in that upper limb score within 3 hours. So is it fair to assume we assume that type of magnitude over 28 days?

You want me to take those, Jim? So the -- it's a -- yes, it's a placebo-controlled trial based on the previous work that we did with brexanolone and zuranolone. Our prediction is that we'll be able to demonstrate a difference from placebo based on the placebo arms in those other studies. But of course, the study is ongoing at the moment. You had a question about the likely magnitude of effect, I think, of the current standard of care. And part of the issue is that there hasn't been very much research done. So what we do know is that clinicians feel that a minority of patients about -- well 40% of patients do see a reduction in that tremor of maybe approaching 50%. And it's difficult to translate that single-dose data into what we might see after chronic dosing with 324. But remember that after the single dose, the plasma concentrations are much lower than when you get to steady state. So our expectation is that at the same dose at steady state, we would have a greater effect on the tremor score. And we await the data, which will be out before the end of the first quarter next year.

Yes. I mean one bit of data that we've talked about in the past, and we've been working in clinical tremor for a really long time, is data that we collected in SAGE-217 early on. And there, we actually did repeated dose studies with 217 and showed over the course of a week that while there is a fairly tight PK/PD relationship when the drug is on and tremor improves and so forth, we also saw a day-by-day improvement throughout the day -- throughout the full week of dosing. So what that says to us, and I think this is what your question was is, what do we expect with chronic dosing? Do we expect that simply the effect that we've seen after a single dose is what you'll see chronically? That's what Helen was implying. But perhaps more importantly, do we see a larger effect after chronic dosing? It's exactly the kind of thing that we'll be able to assess after a month of treatment. Now recall, the 324 trial is a chronic dosing, in this case, it's a month of dosing, and we'll have a lot more to say about that. With regard for placebo response, well, every program is a placebo response, every drug has a placebo response. We know for certain that the effects that we've seen either with single dosage is at least as good as those who have been approved in the past. And, I'm sure, as I think, that our placebo response rates will be anything other than standard. So we use a lot of that kind of data to power our study, and we believe it's that case appropriately designed and really gives us the answer we're looking for.

Our next question will come from Laura Chico with Wedbush Securities.

I do have one on 718 that I want to get to. But really quickly on SHORELINE, I guess, I just want to clarify one thing. As I recall, the purpose of the 50-milligram dose was to increase exposure, not necessarily efficacy. So how should we be thinking about the differences in exposure level at 40 milligrams versus 50 mgs? Does this still give you sufficiently greater coverage versus the 30-milligram level? And then I do have a follow-up on 718.

Sure. I can answer that quickly...

Laura, this is Jim. Yes. So the short answer is, yes. Recall again that when you think about the population...

Sorry, go ahead, Jim.

No, I think that's the answer, Rob. So we're good.

All right. I've just got one on 718 then. I'm a bit intrigued by the NMDA platform and curious if you're seeing any signs of excitotoxicity since you're activating the NMDA receptors. And just help us understand what is the translation between the Huntington's setting and the Parkinson's setting. So as you roll out that additional Phase II data, what translatability do you see there?

Mike, can you speak to the plans in the 718 program?

Sure, Jim. Thanks. So in developing the NMDA platform, safety has actually been a very important consideration that we've had. And what I can say is based on the data that we have seen so far, we are encouraged and confident in SAGE-718's overall safety and adverse event profile, both preclinically and clinically. And I think even more critically, the totality of data that we see across all the studies that we've completed, the translational medicine studies, there are studies in Huntington's as well as the standard Phase I SAD and MAD, supports our overall thesis regarding the value of targeting the brain's endogenous modulatory machinery as a way of dialing in that appropriate benefit risk profile, especially in terms of being able to deliver a safe NMDA modulating compound. And it really does speak to our focus on translational science and sort of early clinical development. And maybe, Aaron, you can kind of touch upon what we're looking to see and how the signals that we're seeing in the Huntington's patients may translate and be extended in our Parkinson's study. Aaron?

Yes. Thanks, Mike. The Huntington's population is an orphan population -- Mike I'm going to hand back to you. I have a lot of feedback.

All right. I'll take that. Yes, in terms of the -- what we're expecting to see in Parkinson's versus Huntington's, what we're really looking for is the breadth and the distinctiveness of the cognitive profile. Again, if you look at Huntington's disease in to date, most domains that have been affected have been modest effects in memory and attention. And what we're looking for is the breadth and a distinctive cognitive profile. At minimum, I think we're going to be looking for a replication similarity between Huntington's patients and Parkinson's patients in terms of executive function. But we're also looking for the opportunity to see if we can expand that into other areas of learning and memory, where the ability to detect those signals may be stronger in Parkinson's patients or potentially other patients that we go into that Huntington's where they're really enriched on the executive function deficits.

Yes. I mean this really goes to our R&D approach. And Jim teed this up at the very beginning of the FutureCast. The thing we don't want to do is move into a specific indication until we've surveyed the real potential for an entirely new mechanism. And so the way you do that is look at the ones that are closest to the biology first, Huntington's. We look at related indications where there's a strong scientific thesis. And then you may ask the question, just like what we did with GABA, are there more -- is it -- are we seeing a more generalized phenomenon? So over the course of the additional open-label trials in cognitive disorders, we'll really be dialing in which are the best patient populations to go after and how best to do that. And I think it's a great example of that.

Our next question will come from Jay Olson with Oppenheimer.

I just wanted to follow up on 718. You've outlined a number of different options for the development of your NMDA platform. I'm especially interested in Neurodevelopment. You spoke about Autism, schizophrenia, ADHD. What are the gating factors for initiating studies in some of those Neurodevelopment indications?

Mike?

Yes, I can take that one. Yes, so I think to your point, we are actually really optimistic about the breadth that we -- and the opportunity we have to build not just a single asset in the NMDA space, but really a whole franchise around the NMDA portfolio. And that's driven by both the long-standing research and the long history that we have in understanding a receptor that's an important target, the diversity of the portfolio that we have in chemical assets really to dial in different compounds of different profiles. And again, when you talk about Neurodevelopmental, Neurodegenerative and sort of cognitive rehabilitation our belief is that we can really take the sort of early development plan of these open-label study, really trying to understand signals across different populations, much as we have done with the GABA historically into not just the Neurodegenerative, the Neurodevelopmental space. And as we learn about what the key features are about the types of property, the types of data that would be best presented for Neurodevelopment where you have potentially pediatric cases being tested versus neurodegenerative where you're different -- in a different age group, we can really dial in the most appropriate assets into our chemical equity, actually use the ability to both translate forward, but more importantly, back translate to understand what is the best profile across the portfolio to go after these indications.

Great. That's super helpful. And then if I could ask one zuranolone question. I think you described the REDWOOD study as halted, pending data in 2021. I was wondering, if you could comment on what that 2021 data is that you're looking for? And whether or not results from the REDWOOD study would be required to submit a regulatory filing for zuranolone?

Rob, you want to take that? Listen, we're having some technical things, as you can see. So I apologize for all of this stuff. We're doing our best here.

Yes. We've discussed really linking REDWOOD to WATERFALL. So to conduct a long-term study, it makes sense to have your short-term studies in hand. So that will be one of the important connections and links to REDWOOD.

Yes. The -- and thanks, Rob. I think the one thing to keep in mind is that this is a program that's being undertaken under breakthrough therapy designation. So as a result, we do have the opportunity to have an ongoing dialogue with the FDA around how best to move the program forward. As Rob said, we are working under the assumption that data from the long-term studies, SHORELINE and REDWOOD, would be needed or potentially needed for a filing. When we have the data from whatever our next trial is that reads out, we'll certainly engage the agency about the most efficient pathway to filing. So it's one of the reasons why we are holding off and really focusing on those efficacy trials. How that all fits together in terms of which trials we initiate, which trials we're going to need for a regulatory filing, all of that comes after our -- after we have more clarity around where and what is going to be our first pathway to submission. For -- if it's PPD or RRT, we know we don't need the long-term data. If it's MDD that we might want to take, it might. And so we're really looking to have those discussions in the context of SHORELINE readout, WATERFALL readout and all the other data that we'll have in the upcoming year.

Steve, one other point that might be worth noting, and I think it was Paul Matteis who asked this earlier, is the 6-month data showed durability that was, I think, at least certainly met, if not exceeded, what our expectations were. And that's the kind of data we'll need to develop the actual final protocol for REDWOOD, where initially, we were looking at every 2 months. Now we may have to rethink what kind of dosing interval might be needed. So to your point, I think there's a little bit more data we want to acquire.

Our next question will come from Neena Bitritto-Garg with Citi.

So I just wanted to ask about the SHORELINE 50 mg arm, again. So the 20% of patients who titrated down, I was just curious, is that amongst all of the patients that were taking the 50 mg dose or only patients in that 50 mg second cohort? And then my second question is, how quickly were patients titrating down? I mean, was it within the first few days or a little bit further down?

Rob, you want to take that?

Sure. The second answer, first, which is it really can vary. There's some patients who notice issues on day 2. There's some that notice it a little bit later. So it really can vary based on the patient. And any more detail, it's really helpful to wait until the fourth quarter. There are nuances around starting at 50 and 30, and I think it's best to give you most accurate information when we have that all together.

Our next question will come from Douglas Tsao with H.C. Wainwright.

Just following up again on the issue of down titration from the 50 to the 40. Just can you describe the process? I mean, was this a situation for patients who came in and complained about somnolence and were given the option? Or was it outlined early in the study that -- just the availability of the 40? I'm just curious, as you go through this, is there thought to potentially investigating the 40 because perhaps that sort of hits the sweet spot for a little bit more effect than the 30, but not so much of the adverse events and issue of somnolence?

Sure. I'll answer it. It's a quite simple process. As I mentioned before, patients are given 50 as they go out, when they come home. If they notice on day 2 or day 5 that it's a little bit more sedation, they go back to the clinic and they trade in for 40 milligrams. When it comes to which doses we would be looking at, I think we're a bit premature in that space. I think as WATERFALL shows us data and so on, we can talk more about what strengths might be what we're looking at for a filing.

Yes. Yes. I would just say that -- this is an approach that we've taken across our development program even going back to ZULRESSO. So we've always allowed for the ability to decrease the infusion rate for ZULRESSO or the dose, even when it was on 30. People's responsiveness and sensitivity to drugs vary, we all know that. And so as part of the approach to any of these trials, we allow for some level of dose adjustment without impacting either our statistics or ability to follow patients in the trial. So yes, this is sort of a standard approach. I mean, we are talking a bit about this to demonstrate the ways in which we may -- keep patients remain in the trial and have them complete to the degree to where that's possible. And again, as I said before, one of the things we look for in our placebo-controlled trials is what are the number of -- what are the rates of patients that -- the drop out from the trials and from the drug and placebo arms. Those are the ways that really get a sense of whether or not this has an impact on patients.

Yes. And I think it's important to contextualize this a little bit differently, which is down titration is routine. The patients are not dropping out. It's very well tolerated. And if you remember from MOUNTAIN, we had virtually adverse events similar to placebo. So we're very comfortable with the 50 milligrams. These are basically -- as Rob has said, the nature and the severity are very similar to what we thought -- saw with 30. And absent anything else, it looks to be more routine dose adjustment that patients would do in almost every trial.

Okay. Great. And then just would you have enough patients on the 40, enough patient experience to file for that approval on that strength?

Yes. You don't -- again, everyone's overthinking this. Even hypothetically, we had 2 doses, people routinely down titrate, whether -- whatever the dose in the label as you see this with SSRIs, with other kinds of drugs. So we certainly have enough data, we believe, for 30. We'll see -- the 50 looks like it's performing very, very well. As Rob has said, we have more than 200 people in higher doses. So what we're offering is flexibility for patients. And that's always been our intent. So this will -- you got to -- so it's really pretty -- I hate to say it, it's -- there's a lot of emphasis on this. It's actually, extremely unremarkable in terms of a trial.

Our next question will come from Dane Leone with Raymond James.

Congrats on the progress with the clinical efforts. Two questions for me. In terms of the NMDA modulation program for PD, it's been interesting because there's been good work done around the glu and 2D subunit and how it can interact with levodopa-induced dyskinesias. I wanted to ask, it seems like you're avoiding any modulation of the glu and 2D subunit. I want to confirm that. And ask more broadly, how do you think about your strategy for modulation, potentially impacting the potentiation of dopamine release from L dopa for these patients? And then the second question I had really simply was just to ask whether the 20% down titration to 40 milligrams for those patients in SHORELINE that started at 50, is that absolute? Or is there another subset of patients that were down titrated to 30 milligrams or lower in addition to that 20?

So I'll just take the first one, and then I'm going to go to -- I'll let Jim or Mike handle the second. The down titration is pretty benign. No one's -- we're not seeing dropouts. And we'll give you all that data when it comes up. But again, as I said earlier, these are mild -- these are patients who were simply feeling -- but by and large, having mild sedation in the morning and opt -- and just opting to lower to 40. It's really nothing more than that. So let me turn this then to Jim or Mike. I'll let you 2 fight this out over HAM on answering the other 2 questions or Aaron, whoever wants to go on that one?

Why don't I take it first? And then maybe I'll turn it over to Aaron for just a little bit of context of this, if he's back online. So in terms of the NR 2 selectivity of a molecule, in general, we don't think that the -- that SAGE-718 or any of the compounds that we've developed so far have a specific selectivity profile. Based on where the target of the receptor is in membrane, we believe that these are likely to be kind of nonselective at the different subunits. So -- and to date, we haven't seen much distinction between 2A, 2B, 2C and 2D. In terms of potential interactions with dopamine, as you alluded to, there's a wealth of literature of how these different systems interact with each other, and it's always something that we are mindful of in conducting our nonclinical work. And I think maybe I'll let Aaron talk to you a little bit about why we're focused on cognitive endpoints in the Parkinson's trial, it's not necessarily the only thing that we have the potential to look at. Aaron?

Thank you, Mike. I hope you can hear me now. So while we are focused primarily on cognition, we are also interested in understanding the motor impact of a drug like SAGE-718, also on certain psychiatric symptoms, and we take a very comprehensive approach. It's actually how we built out our whole Cognex early development platform for NMDA modulation. So what we're trying for is a very deep [ pipeline ] in these early stage [ parties ]. And I think this will ultimately set us up for success as we move into later stages of development.

Our next question will go from -- will come from Sumant Kulkarni with Canaccord.

I have a couple, both on zuranolone. We appear to be moving from chronic to episodic treatment of MDD as we see these developments in relatively rapid-acting treatments. But at what point should we start thinking about episodic treatments potentially morphing into prophylactic treatments for MDD? And is there any quantification you can provide on how long durability should be for an MDD treatment to be considered a preventive? And then I have a follow-up.

Rob, do you want to take that one?

Yes. It's a very speculative question, and I think it's a very interesting one. But I think that it's going to take a while for a shifting in clinical care. There have been standard patterns for decades. And there is fairly different types of patients. So be clear that not everybody has the depression that could be simply treated as needed. There will be people who need to be on chronic medication. That's the whole point of trying to create these multiple pathways to be able to match the product with the right patients.

Got it. And the second question is sort of corollary to that. In your clinical development programs so far, what have you found out and how easy or difficult it would to convince patients that they may not need to be on a chronic therapy to stay better?

Rob, can I comment...

It's a large question. Sure. It's more of a commercial question. Please go ahead.

Yes. I think what we're seeing is based on the enrollment figures in our clinical trials, that there's a great deal of acceptance of this kind of approach. It's extremely user-friendly, not only for patients, but for clinicians. And I think that's why you're seeing our clinical trials are enrolling, frankly, ahead of schedule, the ones that have already started. And if you think about this more broadly, especially nowadays, the idea of having to go to your physicians' office every 2 weeks to get checked versus being able to use telemedicine, which we are -- which Rob and the team are employing aggressively, we think is going to be an extremely user-friendly option for patients. So you made a really important point. They're different stroke -- different options for patients. But we think, especially in this world -- sorry, I'm trying to avoid getting into trouble. Especially in this world, the way it is today and for patient -- if you think about patients needing to get back to work, take care of the families, the ability to get 2 weeks and then go back to your life, we think, is going to be an extremely attractive option. And what we're seeing in the enrollment really suggests that this really is being accepted as a very user-friendly alternative.

Yes. I would say, it's also very patient focused. We know from our research that, in reality, patients don't take their antidepressants chronically. What they're looking for and I think during -- documenting these as well, we -- people -- patients are looking for a way to get better, they want to get better quickly and want to go on with their lives without the -- all of the baggage associated with chronic treatment. And that's what we're looking to establish. It's an entirely different way of thinking about it. Even in other programs that are being developed, often it is -- built into that is the notion of chronic or top-off therapies. So what we're thinking about here is playing directly into -- and Jeff just mentioned it, playing into other ways of following patients, treating them when they need it and also educating them to identify their own symptoms in ways that will allow them to take use of the medicine. So it's nothing short of, for appropriate patients, an entirely different way to treat it that's more like every other disease, where the treatments are reliable and where they provide treatment and relief when you need it.

Our next question will come from Ritu Baral with Cowen.

I wanted to ask -- sort of a corollary to the down titration and sedation. Does this mean that we should be thinking about any additional either next morning hangover effect studies differently that would be required for registration? And currently, are there any expectations for requirements for formal sleep studies around zuranolone? And then I've got another follow-up.

Yes. So this is Jeff, Ritu. A couple of things. We've already done high-dose next morning work in our Phase I, and it's been very benign. And again, look, I think we just have to -- look, if the FDA approves zuranolone, we expect that a range of doses will be approved. We think this is going to be a major commercial advantage. And again, this is not -- we're talking about 80% remaining at the 50-milligram dose. So to me, and if you look at clinical trial data, I think that really indicates that we are now in the exact right dose range we need to be to explore efficacy and safety. And the good news is that this is a completely consistent profile with what we've seen. There are no other unexpected effects. And again, hopefully, we'll see good efficacy, and if we do, I think we're going to have a very useful drug. And as we've talked about earlier, we're not seeing -- and again, to reemphasize, we've already done a lot of these studies already, liking, driving and things like that. We don't -- they're all part of NDA submission. We don't make those public, nor are we interested in providing a pathway for competitors. But -- so yes, we are very pleased with the way the 50 milligrams is working.

Got it. And then my follow-up is actually on your NMDA program. Can you talk about -- I mean, the 3 conditions that you spoke of, Huntington's, Alzheimer's, Parkinson's, can you talk about the nature of the cognitive dysfunction in each of those 3? They vary. And I'm wondering if that is -- if there's any remarkable difference in background NMDA dysfunction between the 3 programs and whether that's driven by disease pathology.

I'm going to turn that over to Aaron. Hopefully, he doesn't have feedback.

Yes. Thanks, Jeff. Thank for the question. There's a lot of overlap when you look at different variables in those conditions. What we've started to put is a very targeted approach. So in Huntington's, it's not only a disease that has very prominent executive deficits. If you go into Parkinson's, executive is the majority of the cognitive deficits, but then you start to get into other things like visual or spatial memory. And then we go to Alzheimer's, as we've sort of shown in the graph earlier, you're seeing primarily memory deficits, but you're also seeing executive deficits. And so by doing this sort of stepwise approach, a very disciplined approach in the different studies, I think we're going to be able to characterize really the full clinical sort of -- that would be amenable to treatment with 718. I'll hand it over to Mike maybe for a second just to talk a little bit about the role of NMDA in each of these conditions.

Yes. Thanks, Aaron. And yes, I think to follow-on to that point, as Aaron talked about, the clinical presentation of all those 3 indications really allows us to look at sort of the breadth and depth of opportunity. But in thinking through what populations made sense at this point in the development of SAGE-718, we also were laying in towards different types of mechanistic rationale. So for example, Huntington's really drove out of the work that we did looking at 24S as the endogenous modulatory system that might be disrupted. As you look at Parkinson's, there is a NMDA rationale, but it comes from a different perspective. And similarly, when you go into Alzheimer's as well, there's also a rationale around NMDA receptors and different types of impacts in terms of the circuitry that's disrupted in Alzheimer's disease and the high levels of NMDA receptor that is involved in those circuits. So we're looking both in terms of different clinical presentations, different symptom domains, but also different underlying mechanism rationales to really build up that overall profile and picture of not just 718, but where we can use the whole portfolio to go after cognitive dysfunction.

I'm showing no further questions in the queue at this time. I would like to turn the call back over to management for any closing remarks.

Well, I'm going to do the close. I want to thank everybody. First, I want to thank Jim for dealing with his technical difficulty. We're going to take a pool to help him get better internet at home now. I know. Sorry, Jim. He just heard that. Listen, I just want to thank everybody today for tuning in. I can't, say, complement everyone highly enough in the company for all the work they've done and moving everything forward the way they have during this COVID. I'm very, very excited about what we're seeing and what the pipeline is offering. I think the other point I want to emphasize, and this has been a theme, I think, the scientists on the call that I think have made, is we're really focusing on not only our chemical equities, but on translational work that's going to look -- we hope, will actually benefit patients and make a real difference. Not just me toos, not repurposing drugs, but really clinical effects that will really change the way patients live their lives. And it's not only in depression, but you can see that in essential tremor where we help people put on makeup or help meet their deals. And then into NMDA, where it's so exciting to think about exploring executive function, which is really the soul of how people operate on a day-to-day basis, and potentially impacting that is such an important -- potentially important advance. And then the work on COVID, which really exemplifies our early exploratory work, which has always been under the radar, but which I think is really important for the future of the company. So again, I want to thank everyone who attended today. I hope all of you are well. I want to thank all the scientists and the people at Sage for really persevering during COVID. And I hope all of you have a really, really good day. So thanks again, everybody.

Ladies and gentlemen, this concludes today's conference call. Thank you for your participation. You may now disconnect.