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

Well, great. Good afternoon, everybody. Thanks for joining us for the next session. I'm Matthew Harrison, one of the biotech analysts here at Morgan Stanley. Very pleased to have SAGE with me for this session. Just quickly before we get started, I need to read a disclaimer statement. Please note that this webcast is for Morgan Stanley's clients and appropriate Morgan Stanley employees only. This webcast is not for members of the press. If you are a member of the press, please disconnect and reach out separately. For important disclosures, please see the Morgan Stanley Research disclosure website at morganstanley.com/researchdisclosures. And if you have any questions, please reach out to your Morgan Stanley sales representative. So with that, really pleased to have Mike Cloonan and Jim Doherty from SAGE with me. Mike, I'm going to turn it over to you to make some opening remarks and then we'll jump right into Q&A.

Yes. Thanks, Matthew. Thanks for having us, and good afternoon, everyone. I'll just say a few words and turn it back to Matthew, so we can get into the Q&A. But just to set the stage of where SAGE is at this point, we are pleased with our progress year-to-date and how we're executing in the face of the pandemic. And I think, hopefully, many of you got a chance to dial into our FutureCast last week when we went through some of the pipeline in our neurology franchise in 324 and our neuropsych franchise with SAGE-718. I'm sure Matthew is going to get into some of that today. But I think you got to see the depth of our pipeline and that we've got a lots going on over the next several months and into 2021 that we're excited about, including the continued execution of zuranolone in SAGE-217. We're on track with our trials. We've made a lot of progress that we'll get into today. And we have a very clear path on -- forward with zuranolone, including 3 different pathways to filing, which we'll talk about today as well. This is all about execution for SAGE in 2020, right? Coming off the restructuring we had in April, the team has rebounded really nicely, and we're really proud of what the SAGE employees have done in the face of a pandemic and social injustice and everything that's going on in the world to execute on the trials, the way we have has been really impressive and we're excited to talk to you about that today. And probably now more than ever, when you think about the disease areas that Sage is focused on, mental health specifically, right, the work that we're doing, extremely important with what we see is happening in the world. So we're looking forward to talking through all the different programs we have and exciting developments that's setting up what is a very exciting end of 2020 and even transformational 2021. Oh, you're on mute.

All right. I'm still on mute. Thanks for that, Mike. So I think maybe just a good place to start because we're going to talk about a lot of different trial names here. Can you just remind everybody across the zuranolone program, which are the studies and what you're studying in each of those studies? And then I think we can hit some of the key data points.

Yes. And I'll take it, and Jim, feel free to jump in, too. So maybe think about -- there's the 3 paths that I mentioned before, right? There's 3 distinct pathways for zuranolone right now, which is our PPD path, which is the SKYLARK study, which has initiated the dosing. We have our rapid response therapy, or RRT, indication for a major depressive disorder, which is the CORAL study, which we'll be starting this year. And then we have the episodic treatment path, which is on the backs of the WATERFALL study, which has started this year. It's over 50% enrolled. We'll have data in the first half of 2021 from WATERFALL. And we've also said we will need some long-term safety data for that episodic treatment paradigm. So that was the REDWOOD study that we paused. So that's sort of the overall landscape program. Probably the one of the ones worth mentioning that we may talk about today is SHORELINE, which is the open-label retreatment study that we have said we'll have data by the end of this year in SHORELINE in the 30 milligrams and possibly the 50-milligram dose as well that we started up this year. So that is the composite of the overall program that we call Landscape that gives you the 3 defined paths to get to filing for both PPD and MDD.

Okay. Perfect. That's great. So I guess, first, maybe one of the questions I get a lot is, SHORELINE, what do we really learn this year from SHORELINE? How does that inform any of the data sets that we're going to get next year?

Yes. Jim, you want to take that one?

Yes, let me take one. Absolutely, SHORELINE is an important part of the Landscape Program, as Mike was saying. The study itself is an open-label naturalistic study, the largest study of its type for the treatment of depression to our understanding. And really, it plays an important role in really helping to understand how zuranolone will impact the treatment paradigm for depression. Zuranolone represents a novel approach and a novel mechanism of action in the treatment of depression. And we understand that the onus is on us to really articulate for patients and for providers and for payers, what the true opportunity is in zuranolone. So thinking about the SHORELINE study, what it's doing is it's giving us information around how the drug can actually be used in a clinical setting. And so by design, the study has a first phase of 14 days of dosing with zuranolone, similar to what we've done in the other studies in the Landscape Program. But importantly, for those folks who have responded, they stayed in the study, and they stayed in the study for up to 1 year's time with the goal of, as long as they have -- they're maintaining their response, they're just monitored. But then when symptoms recur, they have the opportunity to be treated again. And so we get a lot of information around retreatment, both in terms of how often do patients need to be retreated and when they're retreated, what kind of response do they have to a second or potentially a third treatment with zuranolone. So we're going to get an awful lot of information from the study.

And then as we think about the 50 mg dose and the people that you've treated there, how informative is the effect on those patients as we think about what we might expect from WATERFALL? I think we're going to be the first one to read out.

Yes. And of course, we have adjusted dosing to 50 milligrams for multiple studies. So Mike laid a bunch of them out just to reiterate. So we're dosing at 50 milligrams in the SKYLARK study that's enrolling. We're dosing at 50 milligrams in the WATERFALL study that's enrolling. But yes, we did also modify the protocol in the SHORELINE study, which is currently enrolling in. So there are a number of 50-milligram patients that have enrolled into the SHORELINE study as well. And, yes, I think we have a number of questions around going from 30 milligrams to 50 milligrams. I think the important questions are around tolerability and what we're seeing when it comes to tolerability. And then, of course, we think that the range of dosing between 30 and 50 gives us a really good way of targeting a range of doses for efficacy. So we'll get both pieces of information from 50-milligram dosing in SHORELINE.

Okay. And I think at the FutureCast, you guys said that you haven't seen any events of loss of consciousness with the 50-milligram dose yet. Maybe just update us on what we know about safety of that dose, given some of the concerns around those kinds of events?

Yes, absolutely, that is correct. We have not seen loss of consciousness in -- frankly, in the zuranolone program. So at this point, over 2,000 patients dosed, and that does include the data to date with the 50-milligram cohort. And we do know that with the mechanism of action, sedation related events are part of the tolerability profile, and, yes, that goes all the way back to the Phase I study with the -- with zuranolone. So I think it's important to be looking at things like the sedative type response and both the total number of patients who report a sedative type effects and then also the magnitude of effect. So it really is a continuum.

And for those kinds of events, I mean, what kind of regulatory scrutiny do you expect around sedative-like events and -- I mean, do you have a view on -- is there a threshold at some point where there's more concern from the regulatory agencies versus not?

Well, I think sedative-type adverse events are a common feature in CNS products. So I don't think that's a particular surprise. And I think there's a fair amount of experience with understanding sedation-related adverse events, and there are definite ways to deal with that, I think. Where it becomes more of a topic for discussion is around that -- the more severe events, and really, that's why the question around the loss of consciousness happens. So -- but I think that's a distinct issue than the sort of mild to moderate sedative -- sedation-related events that you can see at the concentrations we're talking about.

Okay. Okay, perfect. And then just maybe one of the last things that I think came up on FutureCast was around some of the dose titration events where people moved from 50 down to 40. And I guess it seemed to cause a little bit of controversy or just a little bit of concern from people. But I guess my real question is, I'm sure you've gone back, you've remodeled this from a PK standpoint, probably pretty intensively. And so as you guys think about, let's call it, average trough level, how do we think about what that number needs to be to maintain the effect? And how does some variability on the dose level for certain patients impact that?

Right. Well, yes, I mean we definitely think in terms of what kind of plasma levels we're targeting for the drug and we do think that's important for efficacy. So the fact that there's down-titration study, not a surprise to us at all, that's a provision that are built into studies. And what you're trying to do is to afford a population of patients to hit a general target range. Some people will metabolize a little bit more, some people will metabolize a little bit less. And so that gives you the flexibility to sort of be in the appropriate range. When we think about it, we know from the studies already completed with 30 milligrams from both the ROBIN study as well as the Phase II study in MDD that an exposure range based on 30 milligrams can get us into that efficacy range, going to 50 milligrams keeps us in that range. And so we're very comfortable that titrating between 50 to 40 keeps us into the target range that we're looking for, for efficacy.

And Matthew, maybe just helpful, too, because a lot of what you're asking about the tolerability. To me, when you think about the overall profile, right, of zuranolone, how we're thinking through this, right? The one thing we've seen throughout the studies is the rapid-acting nature, right? And that's something that absolutely has been -- resonates with all the key stakeholders, whether it's the physicians, the payers, the patients, et cetera, right? So the tolerability comes into play as well as we're kind of talking through here. But when you look at all the things that we've got in that profile between the efficacy, the rapid-acting nature within days, the tolerability and then durability, right, we did show that through the 6-month follow-up study to MOUNTAIN and then we'll get some additional data through the SHORELINE. Those are the things that Landscape Program is really bringing together as we look at this from sort of a holistic product profile perspective. We really think we have an opportunity to create a differentiated molecule, both within PPD, sort of that 1 path that I mentioned before, but also both in the RRT indication, which is the co-initiation with antidepressant and then also episodic as well. So when we look at the whole package, with all the data that's kind of coming together, we're really excited about the potential profile here.

Okay. Perfect. That's very helpful. Maybe if we could move on to WATERFALL. I think a lot of people are focused on that study. And maybe, I guess, one of the questions I get a lot also is what are the kinds of patients that you're enrolling in the study? And I think the reason people are asking is, has COVID changed the dynamic of the kinds of patients that you're seeing? And is there any risk associated with that?

Yes. So let me take that one. So I think certainly the impact of COVID on mental health has been substantial. I'm sure people have seen the results from the recent Boston University study, suggesting the prevalence of depression might be up 2 to 3 fold. So it certainly has had a major impact on mental health and on depression. When it comes to the clinical trial, this is where the protocol and the trial design elements are really very important. So our entry criteria are rigorous enough that we're confident that we're recruiting patients with a comparable level of depressive symptoms, both in what we've been doing previously and also consistent within the trial. So although the depressive symptoms are higher across the board, entry into the trial is very much dictated around those entry criteria, especially around the HAM-D.

Okay. Perfect. And then I guess, second question is, can you just talk about how you're taking various endpoints now in a COVID environment? Are these over Zoom? How -- follow up those kinds of things. I mean, any of those issues that you're facing in this kind of environment?

Yes. So we spent an awful lot of time as we're getting the Landscape studies going, talking about these kinds of issues and trying to project ahead around how we are going to handle those kinds of interactions. And really, the principal has been as much as possible for regular in-person connections and discussions with physicians and data points. So as much as possible, we're trying to keep it to those face-to-face meetings. But as a fallback position, in some cases, we are using telepresence and telemedicine as a way of getting information if we can't collect it face-to-face.

Okay. But so overall, from a data continuity and data clarity standpoint, it sounds like you feel pretty comfortable with how you're collecting things.

Exactly right.

Okay. Okay. Great. Maybe then if we could walk through -- so as you outlined, right, there's 3 sort of paths to approval here. Maybe we could walk through each of those and how that works and how that plays out. So I guess, since we're talking about WATERFALL, why don't we start there? So WATERFALL data comes in the first half of '21. But as you said, you're going to need some amount of data from REDWOOD as well. And I believe REDWOOD is still paused. So could you help us play out how REDWOOD plays an impact here and how you're thinking about getting that data from REDWOOD?

Yes, I'll start, and Jim can jump in. But -- yes, you're right, Matthew. At the end, we paused REDWOOD back when the MOUNTAIN study readout happened, right? And it gave us an opportunity to pause that study and really take the lessons learned that we had from MOUNTAIN, including the dose and the HAM-Ds and everything else that we learned from the study and make sure we incorporate those into our next studies. And then it was a portfolio decision. When you think about what we had to go down in terms of the 3 paths that I mentioned before, we had the SKYLARK study in PPD, we had the CORAL study in rapid response, and then we had the WATERFALL study. We really had to look at our resource and say, where are we going to place the resources to make sure we're maximizing the program. And that's -- that was the other reason why we paused REDWOOD at that time. So our strategy was, let's pursue those 3 paths. WATERFALL is going to be the first to read out, as you said, but as we've had our corresponds with the FDA, the feedback has been, you're going to need some level of data cut from REDWOOD. So until we sort of have that data from WATERFALL, right, we'll make the plan, our strategy will be, okay, let's look at REDWOOD, let's think about what the data shows us in total from the Landscape Program because again, we're going to have SHORELINE. We're going to have the 6-month data from MOUNTAIN. We'll have the MOUNTAIN study. We'll have WATERFALL. There's always an opportunity, again, for us to reengage with the FDA and have conversations about the full data sets that we're aligned on what that path forward is, right? So hard to say exactly when we would sort of reinitiate REDWOOD and what that time line is, because you really want to have all the data in hand and then have some discussions with the FDA around what that path looks like. And then if you want to get into the other paths, like so RRT and PPD, the beauty of those 2 individual paths are that it's a single study and a pathway to filing, right? So in PPD, if the SKYLARK study is positive, we can file off that and have an indication in PPD. If the CORAL study, which is the antidepressant -- co-initiated with antidepressants and zuranolone, if that study is positive, we can file off that, right? And so maybe worth just a second for me to elaborate on that -- how those come together, right? So if you think about MDD, you may have this question, but we actually think that's very synergistic between the CORAL study and WATERFALL. If the path plays out the way we think it could, the RRT indication could come first in MDD as we're waiting for REDWOOD. But the beauty is you're going to give patients and physicians and, hey, there's real-world experience with zuranolone, even though it's co-initiated with antidepressant. They'll see the rapid-acting nature within days of zuranolone. They get a sense for the tolerability profile. Building up that real-world experience is really important in advance of when you get the episodic treatment paradigm. Because it will actually help, I think, seed the market in terms of getting experience with the product on the market at that time. We'll be able to publish retreatment data from both SHORELINE and REDWOOD during that period of time. And when you get the episodic treatment indication, again, we think that by that point, they already have a lot of comfort with the profile of zuranolone, which might make that change less onerous to them.

Okay. Got it. That's helpful. So I mean, is it fair to say that you're not going to make a decision on REDWOOD until you at least have some data from WATERFALL?

I think that's the plan right now. I think that's what we -- we made this path, right, to pause it intentionally, right? Take the learnings from both MOUNTAIN, but also WATERFALL, SHORELINE, the six-month data, et cetera. So again -- and the portfolio perspective, right? We want to make sure we're disciplined in our resource allocation. So I think that's the plan as we sit today.

And I think there's an opportunity too in that since it's -- as a breakthrough program, one of the features that we've had is a fair amount of running dialogue with the FDA. And of course, that led to the 3 pathways that Mike was describing. But it also, as you think about data that's coming in, Mike referenced the 6-month data from the MOUNTAIN study. We'll have data from the SHORELINE study as well as ultimately, data from the WATERFALL study. And we will just continue the dialogue with the agency, right? I mean that -- and that's one of the opportunities with a complex program like this is as data continues to come in, we continue to have ongoing discussions.

And then just I guess 1 more question related to WATERFALL. How much does positive WATERFALL data, since that comes first, derisk your view on CORAL?

And Jim, do you want to take that one?

Yes. So I think the program is intended that multiple studies do support each other. So I do think that efficacy from the WATERFALL study is definitely good for the overall program, I think. But they've also build that CORAL study where we're doing co-administration of an antidepressant with zuranolone does also stand on its own, and that is an opportunity to take that pathway forward. So I think that it definitely is beneficial when we see good responses from any of the studies. But there's also a level of independence built in that -- each of these trials alone allows us a pathway to a regulatory approval, if successful.

Okay. Okay. And I guess just 1 last question before maybe we move on to 324 and 718. Any DDIs or anything to worry about in terms of co-administration with SSRIs or other antidepressive agents?

Not that we see, right? So we do all of the standard of work looking at DDI effects. That's part of the zuranolone program overall. And recall, too, that we have not restricted the use of antidepressants in the studies that we've completed, and we see somewhere between 20% and 30% of subjects in those trials have been treated with antidepressants and we don't see any obvious effects. And so we're relatively comfortable that co-administration is not going to give us a different profile than we've seen today.

Okay. Okay, great. 324, you have the KINETIC study here, maybe to sort of remind people around the thesis in essential tremor and what we're going to get from KINETIC.

Yes, absolutely. So the thesis for the SAGE-324 in essential tremor is around this mechanism of action. And it's a different part of the brain that's involved, but it really does come back to the same concept that we think that this MOA can have really beneficial effects for patients that need it. And we have clinical data from earlier studies with first ZULRESSO but then separately with zuranolone showing that, as predicted, these GABA PAMs can reduce tremor behavior. And so we are -- we've applied that learnings to SAGE-324. The molecule is distinct from SAGE-217 but has a comparable pharmacology. So all of that comes together to potentially treat a patient population where -- there hasn't been any innovation in the treatment of essential tremor for a long time. The first and only approval for the treatment of essential tremor is propranolol and that was in 1967. And we're talking about a disorder where there are almost about 6 million in the U.S. alone. So there are quite a lot of people suffering with essential tremor. The KINETIC study is a placebo-controlled double-blind study with SAGE-324, building on the learnings that we have taken from our Phase I work, intended to demonstrate reduction in tremor activity over a 28-day dosing period. So that study has been enrolling for a couple of months now, and we are looking for results from the study either in the fourth quarter of '20 year or first quarter of '21.

Okay. Perfect. So maybe 2 comments. Safety profile versus propranolol. Any key differentiators here that are important to highlight?

Yes, definitely different. I mean so propranolol definitely cardiac effects, and there are a number of patients who have various cardiac conditions who can't tolerate propranolol. GI effects propranolol. Whereas what we're seeing with 324 is pretty consistent with what we've seen across all of our GABA PAMs and really, it is the sedation-related side effects that are the key elements in tolerability profile.

Okay. And then on the efficacy side, you guys highlighted a range of endpoints there. Which endpoints do you think are most relevant from a regulatory standpoint and from a clinical standpoint?

Right. So we have -- we are using TETRAS, which is a physician rating scale where they engage with the patient and measure the magnitude of the tremor. And that scale is pretty widely used. As we talk to KOLs, the -- it's the TETRAS scale as well as people use accelerometers. So we've actually included both endpoints in our work to date. And this is where you see one of the values of Sage's approach in back translating across programs. So between the ZULRESSO and zuranolone trials, we got a good look at the TETRAS and accelerometer end points in our hands. What we saw is pretty good identity between the two. So we see reactions in both endpoints. And so we continue to use both of them moving forward. The TETRAS remains the primary endpoint of the study and the accelerometers are key secondary. Since it's been quite a while since there has been an approval in this space, one of the next key stages in developing any therapy for essential tremor will be to have some regulatory discussions around key end points. So we believe that we've got the right endpoints to be measuring tremor. As we talk to KOLs and patients, it's interesting. There's a lot of comorbidity in anxiety and mood effects. And so that is also something that we're paying attention to in our trials and we will be measuring. So the plan is that we're collecting all the information that we think we need to have a robust discussion with the agency around what a pivotal program can look like.

Okay. Okay. Great. Maybe in the last couple of minutes, we could turn to 718. And I guess just talk broadly about the mechanism here and how you think about that sort of broadly across a range of diseases?

Yes, absolutely. And I think -- let me tell you about the mechanism, and then maybe Mike can talk a little bit about what we see as the opportunity for different patient populations. But yes, SAGE-718 is Sage's most mature asset in our second molecular platform. So unlike the other drugs we've been talking about, 718 is an NMDA receptor positive modulator. And it really is the other side of the coin where GABA receptor modulation dampens down activity, NMDA receptor activity increases excitability. Important for things like learning and memory, and then also importantly, executive function. So what we think here is we've got an opportunity with a brand-new mechanism of action. This is the most mature asset that we believe has been tested for this approach. And also importantly, similar to what we've done on the GABA receptor side of things, we've built off a template of an endogenous mechanism. So there is a natural mechanism to potentiate NMDA receptor that we think that we're building off of with SAGE-718. I mentioned that because we think that's important when it comes to tolerability. That's often a question around NMDA receptor modulation. We have been moving 718 through Phase I studies. As we have done in other programs, we add endpoints for activity into those. And so we've included a battery of cognitive tests and even in our Phase I work. What we've seen is that an improvement of executive function as measured by a number of cognitive tasks. So this is important because that matches the science of what you might predict you'd see with NMDA modulation. But I think it's also important in that we're not seeing just a generic effect across the board. There are a number of reaction time tasks or things like that, that don't involve NMDA receptor function as much and they're not moving. So we do think that there's opportunity to improve executive function. We have studied in a small open-label cohort in Huntington's disease. And what we saw there, with the same cognitive battery, it was a similar profile where we see improvements in executive function in Huntington's patients. Importantly, that they start from a lower baseline as healthy volunteers. And so that demonstrated a fairly substantial improvement in performance. So given the data we have, Huntington's disease is a good potential population to take forward. But given the nature of NMDA receptor in brain function, before we go any further, we really want to take the opportunity to apply the same trial methodology to some other patient populations. So at the moment, we are running a study in Parkinson's disease, so looking for cognitive improvement in PD patients. And we're about to commence a study of similar design in Alzheimer's patients. The goal really then is to understand the breadth of the opportunity here, but also which populations might be the likeliest to respond. Mike, do you want to talk at all about what the opportunity could be amongst these patient population?

Yes. And I'll just build on what you said, Jim. I think Jim hit it on the head, is we've got an opportunity, right? There is the differentiation of this molecule itself and the pathway, right? We're really excited about the potential here. And we -- and as Jim said, we were exploring it in Huntington's disease. We've got that data set. And now we've got the Parkinson's disease and Alzheimer's, the cognitive dysfunction aspect. And so the whole goal of this is to really step back once we have that data and take a portfolio approach, right? Because you can think about rare diseases versus broadly treated diseases in terms of Parkinson's and Alzheimer's, what is that next step that we're going to place on this program. It's all going to depend on what the data set shows that we can make the right portfolio and resource allocation decision going forward. But we think we've got multiple pathways for this molecule and for NMDA in general. And it's really exciting, right, because the level of differentiation is something new and different than what we've seen to date.

Okay. Great. Well, we're just almost at time. I think that's probably a great place to end it. Thanks for the time today and I appreciate you being here.

Thanks, Matthew. Thanks for having us.

Thanks, Matthew. We enjoyed the conversation.

Thanks.