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

Thank you, everyone, for joining us here at the Cowen Health Care Conference and the Sage Therapeutics fireside chat. With us today for the conversation, we have SVP of Clinical Development, Chris Silber; Mike Cloonan, Chief Business Officer; and Kimi Iguchi, CFO. I will turn it over to Kimi for a few opening remarks, and then we'll jump into the Q&A.

Great. Well, thank you, Ritu. Thanks very much for -- we're delighted to be here. Again, my name is Kimi Iguchi. I'm the Chief Financial Officer. And just to start, a little housekeeping, we will be making forward-looking statements. So you can certainly access our SEC filings on our website. But just for those of you who might not be as familiar with Sage, we were -- we're about 9 years old. And when we started, it was really all about brain disorders and thinking about how we can help people with brain disorders. And at the time we've started the company, there was not a lot going on in the field. There was clearly a lot of people that weren't engaging, if you will. So we knew that it was going to be difficult, and we knew that we had to do it differently. And so we've always said we were going to take an innovative approach to how we thought about it. And I'm glad to say that we've made progress to date. We have 3 franchises that we're working for us. Our depression franchise is our most advanced with our first approved product in postpartum depression. That's the first approved product ever in postpartum depression. And that's our product ZULRESSO. And then following that, we have zuranolone, which is an oral program, that's also being studied in postpartum depression and major depressive disorder. So a very interesting depression franchise that's moving along. We also have our neurology franchise, which is our SAGE-324, is a leader in that, and we'll be starting a Phase II study in essential tremor with SAGE-324 this year. We also have our neuropsychiatry franchise, which is exciting. It's our first NMDA platform, an NMDA-positive allosteric modulator. And we will be starting 2 open-label Phase II studies in cognitive disorders, and we're looking at a number of cognitive disorders, and Chris will probably talk more about the breadth of how we're thinking about that. So that's where we've gone to, to date. We have $1 billion of cash to really continue to execute on those 3 franchises. So we're in a good financial position. We did have our earnings call last week. A couple of highlights from that earnings call was with regards to zuranolone and what we've always said is that we were going to have a meeting with the agency, with the FDA, in the first quarter, and then we would absorb that, come up with our plan, and we would communicate that plan. We don't want to speculate about what those scenarios might look like, but I know how important it is, and we're doing everything we can to rapidly get that information out as soon as this company has had the right amount of time to think that through. The other important thing we talked about was the ZULRESSO launch. We continue to see progress there on the long-term indicator. And we talked about the fact that we think that we'll have a modest growth in the revenue over the next couple of quarters. And then we're going to try to -- really try to improve the number of sites and then the volume of treatments going on at particular sites. And lastly, we talked about the fact that we were going to have a very focused approach in 2020. We're going to continue to be disciplined as we have in the past. We're going to be rigorous about how we think about prioritization of the pipeline and across the commercial and R&D. So that's something that we're going to be doing very hard and be very focused on in 2020. So with that, I'll turn it back to Ritu for Q&A.

Yes. I think that investors -- right now, investor focus is slowly, maybe not slowly, but very, very much on the amendment with the Phase III program for zuranolone, just call it 217 for now. And so I wanted to ask right now, the implication on your calls that maybe you had your FDA meeting, you're waiting for minutes. How are you thinking about the amendment to RAINFOREST and REDWOOD? At JPMorgan, there were some topics, some potential levers thrown out, dose timing, entry criteria. Are you still thinking about those as potential levers? Are there other things under consideration?

There were an enormous amount of data coming in from the MOUNTAIN study. We have been and pulling through all of those analyses and elected to take a pause. In REDWOOD and RAINFOREST, those 2 studies, as components of the overall Landscape Program, had very particular targets in terms of what they were looking at. In the case of REDWOOD, looking at preventing relapse. In the case of RAINFOREST, looking at comorbid insomnia with mood disorders. As we've indicated previously, some of the key learnings coming out of the MOUNTAIN Study really confirmed the overall profile of early and robust onset of effect. In terms of dose, we think that one of the most important learnings coming out of the MOUNTAIN Study was looking at the fixed-dose 20-milligram arm that, that does not appear to be an efficacious dose. That, together with the overall side effect profile that was observed not only in the MOUNTAIN study, but in our earlier positive studies in PPD and MDD, leads us to conclude that we've got room to go on dose. As well, given the number of patients that have now been exposed across trial, more than 2,000 patients, we've been able to look at exposure/response data that also props up the -- this notion, this hypothesis, that higher doses and higher exposures may contribute to a greater efficacy. Now as that opposed [indiscernible]

Do you think -- Chris, do you think you need a higher dose in MDD versus PPD for some reason? Your Phase III PPD trial worked very cleanly.

Importantly, we've got positive data, both in MDD and PPD at the 30-milligram dose. We think that there are a variety of factors that contributed to the results that we saw in MOUNTAIN. But again, very importantly, at the doses that were utilized across both those 2 completed studies earlier than on the MOUNTAIN study itself, what we saw replicated in those studies was an early onset of effect, an onset of the order of days relative to 6 or 8 weeks seen with SSRIs and SNRIs. So we think that the 30-milligram dose showed its stuff in the first 2 studies, the first 2 completed studies, and as well confirmed the overall hypothesis of its profile. To the question of higher doses, we think that there is potential opportunity, both in PPD and in MDD, to explore that. Clearly, we are going to be putting together that information from the available trials, together with the exposure/response data that I mentioned in the context of our discussion with the FDA and the overall program as to what we do and in what sequence.

Are there going to be any concerns if you do decide to take a higher dose forward for a potential PPD application? Just going to hypothetical side effects, the whole loss of consciousness or sedation issue that was brought up with ZULRESSO? Have you thought about that at all?

Again, looking at the overall side effect profile seen with zuranolone kind of across populations including exposures within our clinical pharmacology program at doses higher than 30, we think that we now have a more than adequate basis to be able to explore those higher doses. Again, both in MDD and in PPD, we have not seen loss of consciousness anywhere in the program as well single-digit adverse events occurring across the overall program, give us adequate comfort to be able to move up for both indications.

As far as RAINFOREST, this is your placebo-controlled acute study. Is there any aspect of the insomnia monitoring that was part of the study that interfered with capture of depression scores? Do you have all the key secondary endpoints, CGI, et cetera, that MOUNTAIN had?

So for the RAINFOREST study, we do have a number of key secondary outcome measures, including HAM-D score, clinical global impression of severity score that's contained in there. But very importantly, that study was designed to be able to look at this question of comorbid insomnia and mood disorders. All of that said, that propped up the reason to pause that study, take a look at what we had in terms of available evidence our interactions with the FDA to decide a path forward.

Is MADRS -- are you tracking MADRS in RAINFOREST as well?

I would need to confirm that's it's MADRS. But for sure, we've got HAM-D and relevant other self-report measures, including THQ in that study.

And going back to why you started it in addition to addressing the comorbid insomnia, is there any necessary sleep architecture or sedation safety that is critical to NDA that FDA asked for that is being specifically generated by RAINFOREST?

While I would not comment on the specifics of the interactions with FDA or the nature of that discussion, we were particularly encouraged by the side effect profile associated with zuranolone as well an earlier experimental medicine study that showed improvements in overall sleep architecture...

Those are Phase I/II [indiscernible]

Yes, correct. At doses, again, to this notion of dose, at doses that include the doses that have been used elsewhere in the program, as well single doses of 45 milligrams there. That was the genesis of that aspect of the program. But I think, very importantly, where we are now is taking a look at that study together with the overall suite of studies that we have included in landscape that will illuminate potential scenarios and potential path forward.

If RAINFOREST, for whatever reason, doesn't work, is there the potential for the randomized withdrawal study, REDWOOD? Could that count as a pivotal given that withdrawal design? I mean, you don't have a completely blinded acute phase to that. But on the other hand, do you see, like, for instance, there's a DRP withdrawal study that's going in as a pivotal. Is that a possibility? Is that something you're bringing up with the FDA?

Again, while I'll not comment on specific elements of specific scenarios, the 302 or REDWOOD Study is designed as a relapse prevention study. It is a placebo-controlled study. And there certainly is regulatory precedent for that kind of work to be counted as supportive and/or pivotal. Now again all of that said, we're going to take a careful look at the available evidence, the input glean from our interactions with FDA and make a portfolio judgment of which of the scenarios, which trials, in what sequence we undertake.

Are you having parallel discussions with EMA right now on the path forward about changes to these trials and what you're going to need for filing there?

I'll take that one. This is Mike Cloonan. So just maybe the European strategy. So maybe just talk a little bit about that. If you may recall, what we've said is that we're taking a portfolio approach to Europe, in general. When we look at ZULRESSO, when we look at 217, we've got some additional correspondence that's happened with the regulators there, and it gives us an opportunity to step back, which...

Recently or last year?

So it's ongoing, right, we've had ongoing discussions with them. And so for us, it gives us an opportunity to step back and say which assets do we think is the best to enter Europe with and also indications, as that's kind of your question on how we think about the indications that we would go in with. And I think even more importantly, even though the asset themselves or our prioritization and the resource allocation decision, Europe is also -- falls into our prioritization process. So as Kimi [indiscernible] we'll be very disciplined in how we allocate our resources in 2020 and beyond, Europe falls into that. So it's not just as specific in Europe, but Europe as a geography will be a portfolio prioritization process for us.

It doesn't sound like it's a priority in the first half of 2020.

Yes. So at this point, we don't expect to file in Europe in 2020.

Okay. And then are there any -- so SHORELINE has finished enrollment. SHORELINE is open-label to retreatment study.

Correct.

That has finished enrollment. Are there still any changes that you can make to how the data is analyzed, even though it's fully enrolled?

So as you've indicated, SHORELINE has completed the enrollment -- completed front-end enrollment in the fall of last year. Importantly, 303 gives us patterns of retreatment following an initial response period following patients for up to [indiscernible] carefully tracking their symptomatology, and providing the opportunity for a retreatment if such is needed. There are a variety of potential amendments to be included in that SHORELINE Study, including, as I mentioned earlier, the opportunity to enroll an additional cohort in addition to the cohort that was completely enrolled toward the latter part of last year, as well as looking at what retreatment dosing might be. So we're taking to look at all of those possibilities as to what we could include in SHORELINE.

Chris, have you looked at the SHORELINE data, open label, have you looked at it closely enough to date such that it might inform you about how -- what you might want to do with REDWOOD? Are there learnings that you could bring to REDWOOD? Or have you not looked at that data closely enough?

So one of the things associated with the SHORELINE study is it is an open-label study and affords opportunities to be able to inform aspects of the program. What I would say is I'm not going to comment on specific data out of the SHORELINE Study. But we think it provides complementary information to that of the design of REDWOOD. So REDWOOD undertakes in a relapse prevention design a fixed schedule on the administration of drug, independent of symptoms. We think that the important thing about SHORELINE is that it provides additional complementary information, really driven by patient's symptoms as to whether retreatment is needed. So it provides very, very useful information on retreatment patterns.

And I think it's also -- you may remember the retreatment phase also are going to help our pricing strategy, right. This is something we've talked about in the past that between both SHORELINE and REDWOOD, it gives us an opportunity to assess how often patients get retreated and then how we will ultimately determine the per course of treatment costs and price for zuranolone.

So you guys reiterated on your call recently that you're going to have SHORELINE data presented in 2020. Can you give any more granularity on that? Again, first half, second half? Or is there a certain threshold of treatment duration and a certain number of patients that you're waiting for?

At this point, we have no further guidance in terms of timing to be able to deliver results related to SHORELINE. But what I would say is if we conclude that going higher dose is a possibility, again having available information at the appropriate time to be able to share where we are in terms of a fully enrolled cohort and at least the possibility of any additional cohorts, again, its open-label design affords us some latitude on what we're able to deliver when we elect to.

But one of the questions that I've got a lot from clients is, why not -- why go with rapid retreatment or relatively rapid retreatment? Why not go for a longer duration of treatment? Is it something about preclinical tox coverage? Is it something about the mechanism that scares you to treat, say, 4 weeks rather than 2 weeks? How should we think about that?

Let me answer that first quickly [indiscernible] and then you can -- so first, which I think is one of the things that we like about the 2-week treatment is the differentiation, right? So when we think about what's best for the patient and what's best for the market, what we saw in our clinical studies is the 2-week rapid effect could be very differentiating in this market. There's still a high unmet need that exists in MDD and PPD. And so when you think about from a patient perspective, it's really important for us to continue to pursue that path. But Chris can speak to some of the scientific questions that you're raising.

So from a tox coverage perspective, we've got a number of alternatives that could be taken. We're actually quite comfortable that the 2-week pulse delivers positive results. Again, remembering the positive 2 studies that we already have in PPD and MDD, the durability of effect that's seen after that 2-week pulse, we're quite comfortable with 2-week interval of administration.

But was that defined by a toxicity seen in preclinical? Or is that defined by PK/PD efficacy data?

I would say that it was defined by some of the earliest, most positive result and robust findings that we saw, first in open-label studies. Again, thinking about our overall approach, starting with small open-label studies that was then born out in initial placebo-controlled studies that got us very excited about the possibility of episodic treatment as a new paradigm that we're pursuing.

So if I made you -- this is totally not a fair question, but I'm going to ask it anyway. If I had to make you a portion blame, like by percentage numbers to what went wrong in MOUNTAIN between what we discussed, the wrong dose, great inflation, as I call it, upon enrollment, compliance at certain problem sites or timing of dosage could affect that sort of thing. Between those 4, how would you sort of proportion the blame? Is it even? Or is it -- do you think it's driven by 1 or 2?

Well, you've obviously identified among the top 4 factors that we've evaluated and assessed. Obviously, there's an interplay among those 4 and additional factors that exist. I'd be reluctant to offer precise percentages associated with one or the other. What I would say is, in looking at those factors, in understanding the enormous amount of data coming out of the MOUNTAIN Study, we're going to be pursuing both in design, conduct and operational elements the addressing of all of those potentially contributing factors.

And then preliminary thoughts on your treatment-resistant depression study. Design, timing and there's a lot going on in the TRD space. So where do you think 217 could fit?

Well, what I would say about TRD as well, other indications for the overall 217 program. We're going to take a careful look at, again, the available data, our conversations and potential clinical development and regulatory path forward, both for PPD and MDD, as we've been saying, as an initial set of paths forward. And then we'll take a very disciplined approach including looking at TRD or potentially GAD further down the track.

Okay. In our remaining time, I want to split the time between talking about the ZULRESSO launch and specifics around operationalizing that launch as well as I would love to review some of the pipeline programs because that's going to be really important in 2020. Maybe Mike, could you talk a little bit about some of the challenges that you guys have encountered getting patients onto the drug you've got, I think, like 175 certified centers but only 29 are treatment ready and have actually infused a patient. You've said -- you thought it was going to be 6 months, now it's 9 months. What are the main things that you are seeing? And what are you doing to address them?

Yes, Ritu, let me take you up a level in terms -- or let me give you the background to the ZULRESSO launch, and then I'll talk to you about the specifics there. And I think similar to what Chris is talking about, every time we do a launch, like you do a clinical trial, there are learnings, right? And so that's what we're seeing with ZULRESSO, and I'll talk to that. But if you think about where we are, so first product ever approved and launched for postpartum depression, a 60-hour infusion that is most likely going to be predominantly in the hospital-based setting, and we knew it was going to take time, right. We've said this all along that it was going to take time to really activate the sites of care. And what we've seen is ZULRESSO falls at the intersection of women's health and mental health. And if I'm being honest, I'd would say the sense of urgency that exists for those 2 disease states may not have the same heightened urgency that you might see in other areas like oncology or rare diseases. So that's upon us to build that sense of urgency. But what I can say is we've seen progress. We're making progress across the board. All of our key metrics that we reported in Q4 saw growth. So demand, physicians that are referring, the number of sites of care that are treating, and our payer coverage has been very favorable as well. But not dissimilar to what Ritu is pointing out is that there have been some barriers that we've seen that we are overcoming and the lessons learned will help us overcome those barriers, and one of which is just you've talked about is the REM certification. So that's one of the things that we've seen. We have 175 sites that have been REM certified. That, in and of itself, has not been a challenge. What we're seeing and one of the areas that we want to overcome is how do we further accelerate the operationalizing of ZULRESSO in the health care setting, both large institutions and small sites as well. And really what that's coming down to is this is a whole new service line that's being introduced into the hospital base system, right, a 60-hour infusion for postpartum depression. So some of the complexity that the sites are working through is where in the hospital should it be administered. We've seen various kind of best practices, if you will, that we can share with other organizations and other sites of care. So that's one area. Reimbursement is another. So reimbursement is clearly an important criteria for many of the sites to gain comfort in treating with ZULRESSO. And we knew this going in. We're very pleased with the coverage we have. Over 80% of payers have been statable coverage for ZULRESSO this early in the launch that's exceeding our expectations, but what we have learned is that sites want direct experience with reimbursement in their sites of care. We share our best practices across the different sites, tell them what's working, what hasn't.

So they haven't actually gotten any pushback. They're just so going slowly until they've confirmed there is none?

So again, separate reimbursement and coverage are 2 different things. So coverage is saying they're willing to provide access to patients for ZULRESSO. Then there's the reimbursement where they're going to pay the sites of care for the reimbursement. And there's a lag that takes place between when you treat a patient and when you've actually received payment from the payer. That can be months that it could lag. And so what we've seen, you probably heard us talk on the earnings call, there is this revenue lag between site activation and revenue based on the fact that we're seeing at some of the sites, we'll treat a few patients, they'll pause to assess reimbursement. We sort of see the start/stop aspect of the sites. But what we can do is pull through the learnings, get sites experience the direct reimbursement experiences to build confidence, so that's what we're working on. And one of the things I think is important to note is we did get granted a C-code designation that went into effect in January that, that creates more transparency for reimbursement in the outpatient setting. If sites are willing to treat in the outpatient setting, there will be published reimbursement rates that create a lot of transparency for sites. So we think that will be an indicator of future success for these sites. So as Kimi said early on, all of that sums up to making progress with the launch, breaking down the barriers, taking the learnings and leveraging them and continue to make progress. And what we've said is modest growth over the next couple of quarters where we see an increase in the growth rate in the second half of the year as we break down some of those barriers as we see hospitals come on board and the patient volume through site starts to increase.

We have 3 minutes to go with the whole rest of your pipeline, so there's plenty of time. Yes, I think instead of rushing you, I'm going to ask you fun, fair question. 324 in essential tremor versus 718 in Huntington's cognition in the other cognitive disorders, which one do you think you -- which one would you guys consider more promising, either from the commercial perspective or an unmet need perspective?

So I'll start. And yes, it's a great question. It's hard to choose between your children, as we always say, that old adage, but I think what we -- it always comes down to a portfolio of prioritization with us, which is when you look at what we're doing with both 324 and 718, it's not different than 217. Both of those assets have what we think a lot of breadth and depth across multiple indications, but we're going to develop them in a very disciplined and prioritized approach. So we're starting in 324 with essential tremor, and Chris can elaborate on that. That gives us the most data that we have across multiple assets in that indication. And so we will start the Phase II study this year, as Kimi mentioned, and then we got other indications like epilepsy, Parkinson's disease and others that we may have an opportunity to continue to expand that neurology franchise and that will be a portfolio based decision based on the data we'll get from the 324. And with 718, again, I can let Chris elaborate on the plan. Multiple indications, doing what Sage always does, stopping to evaluate what makes the most sense to go forward. We've got Huntington study and data that we're very encouraged by, but we also want to look at some open-label studies in cognition that may give us an opportunity to really determine which path that we want to go on a disciplined approach with 718. But Chris, I don't know if there's anything you want to elaborate on?

Sure. No, with respect to 324, a compound that is next in the series of GABA-positive allosteric modulators, I think this is a terrific example internally within Sage of leveraging learning. We were able to explore with earlier compounds, both brexanolone and 217, the space of essential tremor, have a good appreciation of efficacy of the platform and pharmacology. 324 represents, because of its PK profile, a unique asset for essential tremor. We'll be pursuing a Phase II placebo-controlled study as up next. For 718, as the first in the series of NMDA-positive allosteric modulators, we're really excited about this pharmacology showing some initial effects at cognitive improvement within the HD population. Some of those learnings coming out of that HD experience give us a reason to explore some of a return to our roots and looking at open-label studies in some other population, including Alzheimer's and Parkinson's, to be able to explore the full breadth of this pharmacology.

Great. I think we're at time, and appreciate all you guys coming out, having a discussion with us. Thanks, everyone.

Thanks, everyone.

Thanks.