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

Good morning. Welcome to Sage Therapeutics' conference call to discuss top line results from the WATERFALL study. [Operator Instructions] This call is being webcast live on the Investors & Media section of Sage's website at sagerx.com. This call is the property of Sage Therapeutics, and recording, reproduction or transmission of this call without the express written consent of Sage Therapeutics is strictly prohibited. Please note that this call is being recorded. I would now like to introduce Jeff Boyle, Vice President of Investor Relations at Sage.

Good morning, and thank you for joining Sage Therapeutics' conference call to discuss top line results from the WATERFALL study with zuranolone in major depressive disorder. Before we begin, I encourage everyone to go to the Investor & 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 contain supplemental details. I'll point out that we'll be making forward-looking statements that 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 today's press release and in our SEC filings for additional details. We'll begin the call with prepared remarks by Barry Greene, our Chief Executive Officer, to provide an overview of this morning's announcement. Barry will then be joined by Steve Kanes, our Chief Medical Officer, who will review the findings in more detail. Jim Doherty, our Chief Research Officer; Al Robichaud, our Chief Scientific Officer; and Kimi Iguchi, our Chief Financial Officer, will join for the Q&A portion of the call. I'd now like to turn the call over to Barry. Barry?

Thanks, Jeff, and thank you, everyone, for joining us this morning. This is an important day for Sage, our collaborators, Biogen and Shionogi, and most importantly, the millions of people suffering from depression, also called major depressive disorder or MDD. As you may have already seen from our press release this morning, we have a very successful study, and we're pleased to report that zuranolone met the primary endpoint in the Phase III WATERFALL study, demonstrating a statistically significant reduction in HAM-D scores at day 15 at the end of the 14-day dosing period as well as a rapid onset of activity beginning at day 3 that was the earliest time point measured. And perhaps just as importantly, patients who responded to zuranolone at day 15 retained more than 85% of their improvement out to the last time point in the study, day 42. And just to be clear, that's 4 weeks after the last dose of drug. Before Steve reviews the data in detail, let me provide some context about the importance of these data and how zuranolone could represent a breakthrough in the current management of MDD. If approved, zuranolone would offer a new class of treatment for depression with the potential for patients to achieve a rapid response, in fact, just days, and accelerated sustained path to wellness. In other words, when you use zuranolone to get patients better, faster and keep them better longer, unlike anything on the landscape today. The current standard of care for people with depression is monoamine-based antidepressants. That hasn't changed in nearly 60 years. And while many patients have been helped, significant unmet need remains and, in fact, has increased dramatically. These treatments require chronic dosing and often take up to 6 to 8 weeks to show effect, if any. To add to that, unfortunately, in the real-world setting, current antidepressants have very low adherence with rates that are between 20% and 40% over 12 months. Additionally, the data in the real world show that patients switched to a different antidepressant just about every 7 weeks on average. In clinical trials to date, zuranolone has consistently demonstrated a significant and rapid positive effect on depression even after just 2 doses. And after a full 14-day treatment course, reductions in HAM-D scores have ranged from 12.6 to 17.8 across 4 pivotal trials in the Landscape and Nest programs. So collectively, the Landscape and Nest programs to date have shown a consistent and differentiated data set for zuranolone. That includes data showing rapid onset of activity, again, after the second dose, as measured by day 3; the potential to treat depression when patients need it, notably the pattern of response seen across demographics and WATERFALL study built on the profile seen in the SHORELINE study, where more than 70% of patients receiving a 30-milligram dose of zuranolone needed only 1 or 2, 2-week treatments over the course of a year to maintain wellness, suggesting that zuranolone, if approved, may offer a unique option for treating MDD with a significantly differentiated treatment approach and clearly, a differentiated benefit risk. And importantly, zuranolone has been generally well tolerated in more than 3,000 patients across multiple studies. The fact that more than 90% of those receiving zuranolone completed the WATERFALL study speaks to the tolerability profile. Based upon these WATERFALL study results and the profile we've seen from the Landscape and Nest programs to date, we believe zuranolone has the potential to offer a new approach to treating depression and to provide a differentiated tool for health care providers that may help patients take their lives back by getting faster and staying better longer. I'll remind you that zuranolone's breakthrough therapy designation from the FDA, and we intend to discuss next steps with the agency. Now just to remind you, we sat down with the agency and mapped out 3 distinct Phase IIIs, any of which, if positive, was a fileable event. We believe that that's what we have here with WATERFALL. And we'll sit down with the agency to map out next steps, and we'll share updates in the coming months. We also, as I mentioned, have 2 additional ongoing Phase III studies with zuranolone: CORAL for MDD and SKYLARK for PPD, which remain on track. We expect these studies, if positive, will further validate and expand the potential for zuranolone. We're also continuing enrollment in the 50-milligram cohort of the open-label SHORELINE study. Our work does not stop here. This is only the beginning for Sage. As the world continues to battle the COVID pandemic, we see that we're also in the midst of a brain health pandemic that includes, according to at least 1 study, a fourfold increase in the rate of depression symptoms in the U.S. alone. This unmet need is compounded as the pace of true innovation has not kept up with people's needs. For people suffering from depression, zuranolone may represent a novel approach to treatment with a hope that they will no longer be defined every single day by their depression. To end, we are working with our collaborators, patient advocates and others to move from brain health awareness to brain health action. And we continue to be motivated, passionate and urgent as we plan for the continued progression of our depression, neurology and neuropsych franchises over the year and in the long term. With that, let me call -- turn the call over to Steve for more details. Steve?

Thanks, Barry, and good morning, everyone. I'll start with background on zuranolone and MDD. Zuranolone is a novel positive allosteric modulator, or PAM, of the GABA A receptor that is in development as a treatment for both MDD and postpartum depression, or PPD. MDD is a brain health disorder commonly marked by symptoms of depressed mood, accompanied by changes in affect, cognition, sleep, appetite and functioning, which last 2 weeks or longer. And according to the World Health Organization, more than 264 million people worldwide are estimated to suffer from depression annually. As Barry referenced, rates of depression symptoms have been reported to have risen fourfold during the COVID-19 pandemic. Despite the fact that standard antidepressants are widely used for treatment, large-scale and epidemiologic studies have demonstrated that in the U.S., approximately 8% of adults or 19 million people experienced at least 1 episode of major depression each year. The need for innovation in new therapies is clear. Now let me turn to the top line results from the WATERFALL study. As you can see on Slide 4, the WATERFALL study is a double-blind, placebo-controlled pivotal Phase III study evaluating the efficacy and safety of zuranolone in adults with MDD. The study consisted of a 14-day treatment period and an additional 28th day follow-up period. The primary endpoint of the study was changed from baseline in the 17-item Hamilton Rating Scale for Depression, or HAM-D, total score at day 15. On the next slide are the inclusion and exclusion criteria for the trial. We ensured that patients in the trial had MDD using standard evaluation, scales rating instruments and with a HAM-D-17 total score of 24 or above at entry. We allowed patients who are on a stable antidepressant for at least 2 months prior to enrolling the trial. Moving to the endpoints of the trial on Slide 6. The primary endpoint is the change in baseline in HAM-D-17 total score at day 15, 24 hours after the last dose of zuranolone with both the key and other secondary endpoints listed here. The secondary endpoints were powered in a hierarchical manner, requiring a statistical significance in the prior endpoint to claim statistical significance in subsequent endpoints. And of course, we looked at safety. Moving on to the next slide, with the disposition and subject progression listed. In the study, 268 adults with MDD received zuranolone 50 milligrams and 269 received placebo once nightly for 2 weeks. It's important to note that only 9.7% of those receiving zuranolone during the study discontinued versus 12.6% for placebo. This speaks to the tolerability and effect of zuranolone, which I'll discuss shortly. On Slide 8, you'll see the man -- the mean HAM-D total scores at baseline were 26.8 for the zuranolone group and 26.9 for the placebo group, representing patients with moderate to severe MDD. Approximately 30% of patients in both the zuranolone and placebo arms maintained use of chronically dosed standard antidepressants throughout the trial. Turning to Slide 9, we can see that the primary endpoint was met at the end of 2 weeks of treatment with a p-value of 0.0141. Furthermore, there was a significant reduction in HAM-D scores in the zuranolone group at all previous time points beginning at day 3, following just 2 doses of zuranolone. Zuranolone provided rapid and sustained effects through the 2-week treatment period and in an additional 4-week follow-up. The ability of zuranolone to provide a significant benefit to patients at an early time point is a consistent feature we've seen across all trials with zuranolone. Slide 10 is a forest plot examining the responses of important subgroups from the primary endpoint at day 15. Core subgroups were not powered to show the subgroup level of statistical significance. But nonetheless, we include confidence interviews and p-values for your benefit. And importantly, you'll see that all of the subgroups favor zuranolone. On the next slide, you see the benefit patients who responded to zuranolone is maintained for the month after the last treatment in the trial or out through day 42. In the zuranolone group, patients retained -- retained 86.1% of the benefit seen at day 15 at the end of the study period. So both the primary endpoint data and these retention data support the durability of that effect after 2 weeks of dosing was completed and having been off drug for 4 weeks. That's a profile unlike current standard of care. Moving now to the Clinical Global Impression-Severity scale, Slide 12. The first key secondary endpoint, here, you can see that significance was achieved days 3, 8 and 12. And while we did not achieve statistical significance, we did see numerically better scores at day 15 in all subsequent time points. Using the MADRS scale, another key measure of depression, you can see on Slide 13, separation from placebo on day 8 and day 15. Importantly, at day 42, the zuranolone treatment group maintained more than 87% of its effect with this measure. People typically have a great deal of anxiety associated with MDD. And the benefit we saw in the zuranolone group on the HAM-A scale on Slide 14 was significant at day 8 and day 15. Note per protocol, we did not measure a day 3 or day 12 for either the MADRS or HAM-A to lessen the patient's study burden. Turning to Slide 15, we talked about CGI-S, severity, here represents CGI improvement, or CGI-I, measuring the physician perspective on whether the patient is much or very much improved. This is a key endpoint that speaks to the clinical meaningfulness of results. What you're seeing here on the x-axis is time, and the y-axis is the percent of subjects who were deemed improved or much improved by the commission. You can see improvement and significant separation favoring zuranolone at all time points from day 3 through day 21. And finally, on Slide 16, bringing all of the efficacy endpoints together, here, you can see the day 3 through day 15 treatment period outcomes across the major scales I've described. I hope you can appreciate not only the significance and consistency of zuranolone's performance across the scales as well as the clear benefit patients in the zuranolone group saw, including a rapid response and the ability to maintain that response beyond the dosing period. Now turning to safety. Beginning on Slide 17, here, you can see on the right-hand side the overview of treatment-emergent adverse events. The majority of TEAEs more than 95% occurred during the 2-week dosing period. The vast majority of AEs were mild to moderate, with only 2 serious adverse events in the zuranolone group and 2 in the placebo group. All of the serious adverse events were thought to be related to prior conditions. Importantly, there were no deaths and no loss of consciousness. There was also no weight gain, sexual dysfunction or euphoria, all of which can be associated with current standard of care antidepressants when they're taken chronically that can lead to discontinuation. To that end, importantly, very few patients discontinued the study because of adverse events, with 3.4% of those receiving zuranolone and 1.5% on placebo discontinuing because of AEs. The overall completion rate is consistent with the benefit effect seen. The full table of adverse events greater than 5% is listed on Slide 18. The rates are consistent with what we've seen with zuranolone in previous studies, and recall patients are tolerating these for 2 weeks of treatment versus chronically. Interestingly, events like diarrhea, dry mouth, nausea, which are very common with chronically used antidepressive medications, occurred at similar rates or higher rates in the placebo group compared with the zuranolone group. To round out the safety section on Slide 19, we saw no signal of increased suicidal ideation or behavior in either group. And with respect to withdrawal symptoms, there were no signals for withdrawal as assessed by the PWC-20 following the discontinuation of medication. And finally, there were no clinically meaningful differences in the safety profile of zuranolone when dosed in the study with an antidepressant that is used chronically. The final slide shows the key efficacy takeaways that reinforce our belief that zuranolone has the potential to be a transformational medicine for treating depression. We believe that, if approved, zuranolone may provide a unique treatment option for people suffering from MDD with the goal that it can be used when needed and has the potential to present a differentiated profile with the data we have on rapid onset and durable antidepressant effect. As a reminder, current treatment options require chronic dosing, may take up to 4 to 6 weeks to show an effect, if any, and often cause troubling side effects. We look forward to sharing full study results, including additional secondary endpoints at upcoming medical meetings and in peer-reviewed journals. I'd like to thank the participants in the WATERFALL study, their families, the trial investigators and our dedicated Sage team. Your support is invaluable as we pursue our vision of developing medicines that matter, so people can get better sooner. As Barry mentioned, we are just beginning here. We intend to speak with the FDA about next steps, and we're hard at work with our collaborators. Our medical affairs and patient advocacy teams are working with passion and urgency on scientific and medical education to support brain health action. We look forward to updating you as we advance these efforts in the upcoming months. With that, I'll turn the call over to Jeff to manage the Q&A session. Jeff?

Thanks, Steve. [Operator Instructions] And now I'll turn it over to the operator. Operator?

[Operator Instructions] Our first question comes from Paul Matteis with Stifel.

Congrats on the data. I wanted to talk about regulatory, I guess, and I hope you don't mind if it's two-part question. I guess for one, how would you characterize your confidence level that this is truly a positive study from a regulatory perspective, and that a lack of delta, small waning of effect of the drug at day 42 doesn't matter regulatory-wise? And then the second thing on regulatory, maybe you could just comment on what your base case is for how much safety data you'll need in terms of retreatment and long-term follow-up, and whether or not you have that right now or you're likely to need materially.

Paul, thanks for the congratulation, and thanks for the compound one question. I really appreciate it. It's really important -- yes. Really important question. So I mentioned this in the call, but let me remind you, we set out at the beginning of the year guiding that after meeting with the FDA, we designed 3 unique Phase IIIs that presented 3 unique opportunities for positive readouts, any of which, if hit, were a fileable outcome. With the WATERFALL positive result, a very statistically significant p-value at day 15, the primary endpoint, we believe that we have a fileable data. We intend on meeting with the agency and mapping out next steps. And just to be really clear, we have CORAL reading out at the end of the year, another MDD trial, which we think is important in the overall data landscape because that trial will provide data on zuranolone when prescribed concomitantly with an antidepressant. So if the agency believes that those data are important in a major review, we'll likely wait and file CORAL along with WATERFALL, so that we don't have 2 different clinical filings and have the potential to suffer a 3-month delay. So that's the overall strategy. You asked about safety and overall profile. So let me ask Steve to talk about the safety database. And then you also asked about the absolute effect, and Steve, someone who's treated these patients, can talk about that as well.

Yes. So yes, let me talk about that first. The overall effect that we're seeing is really extraordinary, Paul. With -- in this study, we're seeing a greater than 14-point drop; in others, we've seen 17. The profile of this drug is clear, and it's one that's very noticeable to patients. And if approved, will be an extraordinary new tool for physicians. With regard to safety, we have more than 3,500 patients treated in this -- with zuranolone. That's a really large safety database. It's large, and it's growing. We have ongoing placebo-controlled trials. So the profile that we're seeing has been consistent for quite some time, so we're confident in our safety database. The other thing I'd say is with regard to regulatory, the way that we're thinking about maintenance of efficacy, we've had this question before with ZULRESSO, what's important here is stability. And we certainly have demonstrated that across the entire program with more than 85% of patients maintaining benefit. So overall, we're looking forward to having those discussions with the FDA. But this is clearly fileable, and as Barry said, we're looking to think about the most efficient way to do that.

Our next question comes from Andrew Tsai with Jefferies.

A big congrats on the positive data readout. Congrats again. So can you talk a little bit about how you think physicians down the road will decide whether to prescribe the drug or not? I mean do they merely look at the delta versus placebo at day 15 and day 42? Or do they look at other aspects of 217 as well as other parts of the data? Or do they also look at SHORELINE, for example? Can you just talk a little bit about what physicians would care about down the road?

No, Andrew, thanks and thanks for the congratulatory note. Obviously, we're thrilled with these data. And importantly, to your question, we're very gratified with the overall profile of the Landscape and Nest programs. And it's the totality of the data that ultimately will be in our label and ultimately lead to our education and promotional activities. We know that while patients live with an underlying depression, as depressive episodes that cause them to go deep, we're going to help the world reimagine the treatment of depression with zuranolone. And once we educate, I believe that for many patients, physicians will reach for zuranolone either alone or concomitant with another antidepressant, which doesn't necessarily help get them out of the depressive symptoms but can help them, once better, stay well as we've seen with zuranolone. And Steve, why don't you -- again, why don't you talk about how you would think about this when your patients walked in the door?

Yes. Thanks for the question. It's really at -- at the end of the day, we know that in clinical trials, people talk about placebo and delta from placebo. For clinical treatment, it's how quickly do patients get better, how much better do they do and what happens after they stop therapy. And what's unique about this and absolutely different from everything that's out there is the ability to treat patients quickly, a 2-week course of therapy, and then know that if they need additional retreatment as we've seen in SHORELINE, they'll have reliable response again. That's something very, very different. It's going to be a very welcome tool for physicians in their toolbox if approved. So we talked to a lot of docs. The ones that we've had a chance to speak to really understand where we're going with this and recognize how important it will be for -- as a treatment option.

Our next question comes from Salveen Richter with Goldman Sachs.

Congratulations. In the context of the delta that you saw on HAM-D, can you put the placebo response in respect of what it's been seen prior?

Yes, absolutely. And Salveen, again, thank you, and thanks for the congratulatory note. So first of all, we believe the reduction in HAM-D scores in the WATERFALL are clinically meaningful. Steve will talk more about that. And just to emphasize, what's important here again is that patients got better rapidly and stayed better longer with maintenance of effect out to day 42, and that's what matters most to patients. Steve?

Yes. The consistency that we've seen in terms of the overall benefit for patients on zuranolone has been rock solid since our first study, Salveen. And so the delta from placebo, that's driven entirely by the variability that we see in placebo. And that's a constant challenge within depression trials. It's one of the reasons why we emphasize what it is that the drug does as opposed to trying to war game out what that delta from placebo is. So at the end of the day, as we said, the large drops in terms of symptoms as well as the maintenance of efficacy is what really matters to patients. And from a physician and from patient perspective, those are the things that are absolutely critical for when they would choose to use this and for which patients.

And Jim, maybe you can talk about having effectively conducted -- designed and conducted in a global pandemic, how placebo responded here relative to other studies.

Yes, absolutely, Barry. And let me take that opportunity to really thank the Sage team investigators and most especially, the people who participated in the WATERFALL trial. These are really remarkable results. I'm particularly proud of the team for delivering those compelling results during the COVID pandemic. And really, when -- the question was around placebo. What we do see is, and Steve said it, placebo response does vary. We do see a significant placebo response in the trial. And more importantly, as you saw in the figure from the primary endpoint, we're seeing improvements in placebo continuing, right? So we see what we've always seen a rapid response with zuranolone. And you can see on the -- if you look at the graph, we get that rapid response as early as day 3, and then that's maintained out through time. But then if you look at the placebo response, we do see additional placebo responding that's occurring all the way out through day 15.

Our next question comes from Ritu Baral with Cowen.

When we've asked our KOL consultants about interpreting this data, they've always directed us to focus on a couple -- actually, 3 measures that I don't think you guys have addressed this morning. Can you talk about what you saw on calculated effect size? And also, have you calculated remission and response? I see it's under additional secondaries and, therefore, nominal, but it sounds like it would be important to doctors both at day 14 and day 42.

Yes. Thanks, Ritu. Let me ask Jim to comment.

Yes. A couple of thoughts on some of the secondary endpoints, relapse and remission rates. First, they're really comparable across the entire Landscape program, and that's a consistent finding that we're seeing across the board with the drug. More specifically, those response and remission rates are really consistent with the pooled response and remission rates across the placebo-controlled Landscape and Nest studies, including both MDD and PPD. And as far as additional details around secondary endpoints, we're looking forward to sharing those additional data from the study in future scientific forums. But I think, look, the important point is, these are valuable markers, but the key point is that patients are getting better fast, and they're maintaining that improvement out through the study.

Our next question comes from Cory Kasimov with JPMorgan.

When you think about durability and maintenance of effect of 86% at day 42, how far would that effect have to fall before more physicians or patients would think about restarting therapies? Anything you can glean from SHORELINE on that front?

Yes. Let me ask Steve to address that.

Yes. So that's the study. So SHORELINE is our naturalistic study that actually looks to answer that question. And in that study, we followed patients out for a year after they've been initially treated. And what we've seen is something really important, Cory, which is that about 70% of patients require no more than 1 additional treatment in a year using standard diagnostic criteria. So that's really better to understand what the rate of retreatment might be than rather -- rather than at 30 days. So we looked for where our triggers of true MDD, where patients actually have recurrence of symptoms. And as I said, 50% of patients didn't require any additional therapy over the course of the year, and 70% required -- excuse me, 70 -- the additional 20% only required 1 additional treatment. So a really durable response and an important option for patients if approved.

And Cory, just to highlight, what's remarkable there, again, 70% of patients, that was a 30-milligram group, only required 1 or 2, 2-week courses. So that's 4 weeks of drug out of 52 weeks, being drug-free for the rest of the time. That's important to patients. And what's really critical and often missed here, not by physicians per se, is that they know they're off drug, and yet they're still maintaining that benefit. It's almost a reverse placebo effect, if you will.

Our next question comes from Yasmeen Rahimi with Piper Sandler.

Congratulations on the data. I guess I wanted to just drill down on the somnolence seen. Do you think that or have you completed driving tasks that the FDA may ask for? So if you could comment on that potential requirement. And then secondary, what percentage of patients had like sort of severe somnolence versus moderate somnolence during that 2-week period? If you could just comment on that, that would be helpful.

Yas, thank you, and thanks for the congratulatory note. I'll provide some context and then ask Jim to comment. So we're thrilled with the overall adverse event profile here. The fact that over 90% of patients finished the study is evidence of the tolerability profile, and it's far better than the estimated 20% to 40% adherence we see on today's standard of care.

Yes, absolutely. And as Barry said, you asked a question about somnolence. So the vast majority of those events, as Steve said during his presentation, were in the mild to moderate category. And again, I think as we think about the overall profile, importantly, there were no deaths. There were no loss of consciousness events. There was also no weight gain, no sexual dysfunction, no euphoria, all of which can be associated with standard of care antidepressants when they're taking chronically, which can lead to discontinuation. So as we're saying, we are very happy about the tolerability and safety profile that we're seeing, again, in the WATERFALL study, very consistent with the broader program.

Our next question comes from Tazeen Ahmad with Bank of America.

Congratulations from me as well on the top line. Can I just get some color about the results for the HAM-D? Specifically, can you talk about what particular domains of the HAM-D score showed statistical significant effects? And then maybe as a point of clarification, can you just talk about what was the maintained response rate for placebo? I think you showed the drug arm, but I was wondering about the placebo as well.

Absolutely. Steve, you want to talk about that?

Sure. We'll go into the details of the subgroups from this study, as Jim said, as we start to roll out more detail the dialysis of the data. What I can tell you is that what we've seen consistently across all of our trials is that all of the measures within the HAM-D have been in favor of drug. And that's been shown in trial after trial after trial. So we've been seeing a very consistent profile, and that's what we would expect to see when we dig more deeply into these data. The other piece of this is we've been looking at -- and some of the other data is related to SHORELINE and others. And again, we've been seeing a really consistent profile. And apologies, what was your -- what was the second part of your question?

Yes. Sure. What was the maintained response rate for placebo?

Yes. Yes. So the placebo was maintained. I mean we see placebo across all of our trials. And that's why our -- obviously, we didn't separate from placebo at day 42. A little less relevant because if you look in the real world, the placebo responses aren't what and what no treatment really refers to. And that's one of the challenges in this field, but really, the important part here is day 15 in the primary endpoint.

And then, Tazeen, just to round up, and Steve mentioned this, we will provide way more detail about all the HAM-D endpoints. But when we look at the totality of data across the Landscape and Nest programs, we are seeing with zuranolone, really, the big 3, the improvement in mood, anxiety and sleep. And that's what -- if you talk to your physician consultants, that's what they want to do for their patients. And we're seeing it time to time again.

Okay. Did that also hold true for patients that weren't on background SSRIs?

Yes. If you look at the WATERFALL plot or the plot that Steve mentioned, all of the subgroups responded about the same. So the pace -- the 2/3 or 70% of patients not on background antidepressants performed equally to those that were.

Our next question comes from Laura Chico with Wedbush Securities.

I guess I just wanted to follow up on the delta between zuranolone and placebo over time. There appears to be a bit of a narrowing on that HAM-D improvement, and that wasn't necessarily something that we saw in MOUNTAIN. And I guess, I'm just asking this in the context of potential COVID impact on the study. I might have missed this, but were -- could you comment on the proportion of patients that experienced a first-time depressive episode in the study and maybe contrast that versus what you saw in MOUNTAIN?

Yes. So you're right, this is a very unique study and that was designed and conducted in the course of COVID. And that may have had an impact in the placebo. Steve, do you want to take the color on that?

Yes. Well, first and foremost, we're -- we mentioned this during the call, this is an extraordinary trial in that we were able to both execute and conduct and have a positive study in the midst of COVID, something that initially wasn't -- people even questioned whether it would be possible to do. So really excited about that. We did not specifically call out patients with COVID. We can tell you that there weren't many patients early on that were affected by COVID. But perhaps more importantly, we wouldn't expect expression to be any different in patients that -- during a pandemic, you're not talking about depression due to COVID, the depression during a pandemic than they were due to any other stressors. So overall, we're not looking to see any differences in the responsiveness of patients to the drug. And likewise, during the follow-up period, we know that there are going to be some patients that have a slow return of symptoms. And those are the kinds of data that we really are looking to SHORELINE to see in terms of potential for need for subsequent retreatment. So overall, the profile that we've seen here, Laura, is very consistent with what we were expecting and we think represents an important new option for patients.

Our next question comes from Sumant Kulkarni with Canaccord.

Given the consistency of data theme and the relatively rapid onset of zuranolone in your discussions with your partners, when you meet with the FDA or in real-world usage, what's the preference on positioning zuranolone purely as an acute or episodic treatment for MDD? And how much of a challenge would that be to change this paradigm, given patients are used to chronic treatments and doctors as well? I'm only asking because even 4 treatments per year or 14 days each is much better than the standard of care right now.

Yes, Sumant, thanks for that observation. And you're exactly right. I mean, the idea here is to ensure that when -- if a physician believes their depression patient needs zuranolone, they get zuranolone, whether that's first line, whether it's after they've been on an SSRI or concomitant to an antidepressant. And we intend, with the overall Landscape and Nest data, to have all of those data out there. I mean I'd ask you -- and you've mentioned this, if you walked into your physician's office and they offered you something that may get you better in 6 to 8 weeks, if at all, with significant side effects or profile like zuranolone, I'm confident you would pick zuranolone.

Our next question comes from Gary Nachman with BMO Capital Markets.

Congrats on hitting the endpoint. Patients came in at a reasonably high HAM-D of 27. So are you surprised you didn't get a bigger treatment effect with the 50-milligram dose with the 14-point reduction in the HAM-D? Specifically, relative to what you saw with some of the lower doses in previous studies and maybe just review those baselines and treatment effects, I'm curious how the FDA is going to look at treatment effect across the doses in the different studies and if they'll be looking for improvement there.

I'll -- let me start with context, and then Jim or Steve can dive in there. Look, we're thrilled with the effect size. I'll remind you that we increased the HAM-D inclusion/exclusion criteria for just that purpose. People were less well in this trial. And the data across the program, Landscape and Nest program, and this is what's important to the agency. We have a positive study, and the data in every trial is consistent with significant reductions of HAM-D, statistically significant effects at day 3, consistent effects at day 3 and durable effects out to day 42. We mentioned this already, but we saw in SHORELINE at a lower dose that 70% of patients only required 1 or 2, 2-week treatment. It's just a remarkable profile for a new class of drug like this that can get better -- get patients better, faster and keep them better longer.

Yes. I'll just say the consistency that we've seen across all of our trials, including MOUNTAIN, has been remarkable. We've had now -- 3 out of 4 clinical studies have been not only statistically significant but clinically meaningful improvement with very few -- very limited dosing. I mean who doesn't want to get better within 3 days, quite frankly? So the profile itself is extraordinarily consistent, and the overall change from baseline, which is in the mid- to high teens in every study, that represents a level of consistency that's rarely seen. So when we talk to the FDA, when we start thinking about filing, it really is about the totality of the data across all of the studies that speak to the overall benefit risk profile. And we think that these data continue to support what we've known about this drug for quite some time: rapid, very large effects and durable effects that we can use to treat patients episodically.

Our next question comes from Jay Olson with Oppenheimer.

Can you talk about why the efficacy of zuranolone seems more pronounced on MADRS compared to HAM-D and then describe some of the key differences between those 2 scales?

Yes. Jay, great question. And in fact, patients performed a bit better on MADRS. And Steve, why don't you talk about the difference in scales and sort of how a physician thinks about these?

Yes. They're both used. They're both regulatory endpoints. HAM-D has perhaps a little bit more emphasis on sleep and anxiety; MADRS a little bit less. But overall, they track very closely. So when we include as our primary HAM-D, that tracks back to the way that we've sort of designed this overall program. We include MADRS almost as a checkpoint. It allows us to understand our results as they come out and make sure that everything is working the way we expect. They're on a different numerical scale. The numbers themselves are a little bit different. But that said, across all of our trials, we've always seen our HAM-D-17 as well as MADRS scores tracking very, very closely in terms of overall change as most significant. So those are there to help us make sure that all of our data remains consistent across the board.

Yes, Jay, and just to add to that, maybe to reiterate the point, it's a good question. We, of course, look at MADRS across all of our trials. And as Steve said, there are some slight differences in what their -- what the items are measuring. But really, the main message for us is there's a consistency, not just in the HAM-D scores, but that same consistently shows up very clearly in the MADRS scores as well. So what you're seeing is an absolute drop at day 15 of 17.5 points on the MADRS scale. So -- and the same maintenance of response that we're talking about for HAM-D also appearing in MADRS. So of course, 2 different scales. So it's the same message in general. But I think that's really an important point is that it doesn't matter which scale we're talking about. We're seeing clear and consistent antidepressant response rapidly and that response being maintained with whichever scale you're looking at.

Our next question comes from Brian Abrahams with RBC Capital Markets.

Realizing that the numbers are small when we cut the subgroups, but it did look like you saw a slightly lower treatment effect for patients on concurrent antidepressants. I think the HAM-D delta was about 1.2. Just curious how that impacts your view on the powering of the CORAL study. Might you consider any amendment to that design or upsizing to maximize that probability of success when we see the data year-end?

Yes. Let me ask Jim to take that.

Yes. So what we've seen here again is what we've seen consistently in the past. It really -- we get comparable results in both the groups that are or are not on concomitant ADTs. I wouldn't go too far into some analyses, as you say, they're not necessarily powered for that. But I think, of course, we -- as we always do, we're going to take learnings from WATERFALL for the entire program and apply that moving forward. So it's just always something we look at.

And this will also be part of our agency regulatory discussion. They're also very interested -- as evidenced by the alignment on 3 unique Phase III trials to get a path for approval, they're very interested in doing the right study to demonstrate patient benefit.

Our next question comes from Neena Bitritto-Garg with Citi.

So just a question about the baseline HAM-D scores. Can you talk a little bit about the distribution of those scores at both the placebo and the 50-mg arms? I know in MOUNTAIN, there was a little bit of a dumbbell shape, just given that some patients were kind of up-scored. So there's a little bit of a heavier tail kind of on the lower end. Can you talk a little bit about what you saw here in this study?

Thanks, Neena. Steve, do you want to take that?

Sure, Barry. Yes. The data here are very clean, and it speaks to what we've talked about for quite a bit. We've taken the learnings from MOUNTAIN and applied them directly to our design of WATERFALL. So we're very confident in the results and are really proud of the way that everybody has performed during what would otherwise be a relatively challenging time. So no, this -- some of those technical challenges that we had run into are not ever in that tier.

Our next question comes from Matthew Harrison with Morgan Stanley.

I was just wondering if you could just comment again on some of the regulatory considerations. And I guess, the question would be, given that this is the first study using the 50-mg dose, just wondering how the regulators may think about that and needing 2 studies with 50-mg dose versus obviously having studies with multiple different doses.

Yes. Matt, thanks for the question. Let me take that. So look, that -- again, I'll mention, we sat down with the agency and, with concordance with the agency, mapped out 3 placebo-controlled studies. Any one of which if hit was fileable. So we now have WATERFALL, which hit, we believe the data are fileable. SHORELINE, in fact, was the first trial where we used the 50-milligram and have substantial data on the 50-milligram. As I've commented, Steve commented, the safety profile is actually remarkable in terms of how clean it is across the whole Landscape program, whether 30 or 50 milligrams. The top adverse event we see is somnolence in this patient population. Being sleepy at night and getting a good night sleep is actually a benefit. And what we want to see is mood, anxiety and sleep improvement. We see all of that across the totality of data. So the real technical question when we sit down with the agency is their view on CORAL being a major review or not. If they believe it's a major review, which I suspect they will, because it's an important study, then we'll file all the clinical data in 1 section, telling 1 clinical story with WATERFALL, with CORAL, with SHORELINE, with a positive Phase II we've already had, including the PPD data. And that will be the totality of filing. And then we expect to be talking to the agency with that totality of data about MDD and PPD. Hope that's clear, Matt.

Our next question comes from Tim Lugo with William Blair.

And given the issues around dose in the past, can you just maybe talk about the BMI data? It looks like there is maybe some sort of issue between the 18.5 to 24.9 kilogram patients versus the heavier patients, and then it may be reversed a bit in the over 30. Nothing but significant, but can you just discuss if there is any -- if you're seeing anything there?

Yes. Let me set some context. And then maybe when I kick it to Steve, he can talk about different dose options actually being available. But Tim, as you said, all of those subgroups-favored drug, the sizes were not sufficient to make any major conclusions in all subgroups irrespective of BMI benefit. Steve, do you want to talk about multiple dose options for patients and physicians?

Yes. Absolutely. Tim, for this -- for our psychiatry drugs, we know that we need more than 1 dose option. We studied 30. We have 2 studies that were positive at 30. We are here, a study that's positive at 40. And we've allowed for some level of dose adjustments based on tolerability. We think this covers the entire sort of spectrum. When we sort of get down to looking specifically at dose recommendations, we don't view weight as a major consideration, although patients often need to have their doses adjusted. So the package that we're putting together, if we're approved, will cover a variety of doses along with instructions on how best to optimize benefited risks. So that's standard for depression treatments. And is what we'll have, assuming that we're able to move -- to get approval for zuranolone.

Next question comes from Marc Goodman with SVB Leerink.

Steve, what is the effect size for standard of care antidepressants? And what was the effect size for this product at day 50 on the endpoint? And just a clarification, just for -- I know it's a second question, but I apologize. But it sounds like CORAL is important. What if the CORAL study is not stat sig? How does that work in? Or you still feel confident that you still have a fileable package?

Yes. Well, I'll take -- let me take CORAL and effect size, and then Steve can talk about standard of care. So obviously, a positive CORAL will be beneficial. But CORAL will be important data nonetheless and enriches the safety database, if nothing else, and shows how zuranolone performs, zuranolone plus a concomitant antidepressant. So I'll remind you that we have 3 unique Phase IIIs. Any one of which, if positive, is fileable. So a positive CORAL would be great, but it's not needed for approval. We believe that what we see in WATERFALL is what we need for a drug to be approved. We will talk about further endpoints later on as we present these data in medical meetings and publications. We wanted to get enough out to give itself to the data, but we want to make sure that we can publish these data as well. Steve, do you want to talk about standard of care effect size?

Yes. What we know based on the best meta analysis is that on HAM-D-17, typically, you see about 1.3 to 1.5 difference from placebo. What's different about what we're seeing here is the large effects. In other words, we're seeing large differences from baseline, and we're seeing it very early on. And that's been true across the board for this program. So the metric of just trying to understand difference from placebo, it could be misleading. What you need to look at is how large an effect was there for patients, would they notice it. And when you're talking about a 14-point difference, that's notable to patients, for loved ones, it affects things like, as we've seen in other -- in prior studies, quality of life and other measures of activity of daily living. And so those large effects are what's different. Placebo will do what placebo does, but we're seeing some very dramatic and very different effects over a very short period of time. And it just bears repeating that we're talking about these effects being durable after only 2 weeks of therapy. Most patients are taking antidepressants for 6 months, a year or even longer, experiencing the side effects and potentially not demonstrating any real benefit. So we think this is a really important option for patients, and it's one of the reasons why we're so excited about the program and the continued, really, I would say, highly replicated data that we've seen across all of our studies.

Our next question comes from Danielle Brill with Raymond James.

I'm just curious what proportion of patients enrolled in the study were newly diagnosed or experiencing their first depressive episode. And then what would the forest plot data by prior history of MDD look like?

Thanks, Danielle. Steve, you want to take that?

Yes. What we're seeing is about 20% of the patients are on their first dose, but what we need to understand is that some patients -- and that's by episode. So what some patients are doing is they're coming in and having monotherapy. And that means that they've either started an antidepressant and stopped it on their own or were -- have not for this episode taking medication. And we've seen that across all of our studies. About 1/3 of the patients are on underlying antidepressants and haven't received benefit. Regardless of whether they had underlying antidepressants or not, we've seen very similar results across the board, and we've talked about those results quite a bit this morning. So regardless of whether it's a first episode, on an underlying antidepressant or monotherapy, receiving very consistent results, and that's been true across the entire program.

Our next question comes from Douglas Tsao with H.C. Wainwright.

Just curious, given these results when you've -- obviously, you've addressed it from a regulatory standpoint, but just curious when you think about clinically and in the marketplace, which dose of zuranolone will be most commonly used when you think about this? I mean it doesn't look like there is that much of an incremental benefit for the 50-milligram, and there was some caused in somnolence. So do you think now that the 50-milligram or do you think that there'll be some patients who need the 50, but most patients will ultimately end up on the 30-milligram?

Yes, Doug, that's a really important question. And obviously, we'll see how things play out when we file, assuming approval and a successful launch. It's very important -- as Steve mentioned, it's very important to have multiple choices out there for physicians and their patients. And over time, I think physicians and patients will figure it out. We will certainly start with the 50-milligram dose. That's what these data are based on. That's where we saw the positive results. And if some patients need to down titrate because, in 2 weeks, they can't tolerate certain side effects, then they'll have that option. And in certain patients that might be lighter over time, and therefore, a lower dose gives them enough blood plasma concentration, blank concentration, that's to be figured out over time. But we plan on having multiple doses available, and I think 50 will be the dose of choice. Very importantly, what we're going to get approved here is the drug, not specifically a dose, but we'll get the drug approved.

This concludes the question and answer session. I would now like to turn the call back over to Barry Greene for closing remarks.

Thanks, operator, and thanks, everyone, for joining us this morning. Let me also echo what Steve said on the call. We really want to thank the patients, the investigators, the entire community for conducting an unprecedented study in the midst of a global COVID pandemic and with these absolutely remarkable, incredible results. As you can tell, we're pleased with the successful WATERFALL study. And I'm excited by the opportunity we have to make a difference in the lives of millions of patients who suffer. I'm absolutely convinced that we're going to help millions of people -- patients and save countless lives. Today's announcement represents an important step towards our mission to make medicines that matter, so people can get better sooner. And it's an important milestone in our goal of becoming a leader in brain health and a top-tier biopharmaceutical company. I look forward to sharing more on the next steps for the zuranolone program in the coming months. And thanks again. Have a great day.

This concludes today's conference call. Thank you for participating. You may now disconnect.