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

Hi, everyone. Welcome to the Cowen Healthcare Conference 2021 Fireside Chat with Sage. With us today from Sage is the new CEO, Barry Greene. Barry, thank you for -- thank you for joining us today.

I'm going to -- I'm just going to kick it off with a strategy question, and you can take this opportunity to elaborate a little bit if you want, if you think that there's something not covered in our discussion outline, but I'm sure you've gotten this question a number of times for the last couple of months. Why Sage? I mean I've heard from third parties that your phone was ringing off the hook the second, your your planned move from Alnylam was announced. So what did you see in this opportunity, not just as a Board member, which was the original plan, but in the CEO role. What withdrew to Sage at a time where it was kind of in the doldrums from where it had been at one point?

Yes, Ritu, and thanks for the question, and thanks for Cowen, for having us here today. It's exciting to talk in this fireside chat. Yes. So look, when my plant announcement occurred kind of in the late summer, as you said, a number of people reached out. And part of the planned exit was to really think about and spend some time on what I should be doing next, how I can contribute kind of to global health next. And I was very intrigued with CNS and brain health. As you know, Alnylam has a fantastic partnership with Regeneron, really uncovering a lot in the CNS area. And I was amazed to learn just how far CNS had come over the last few years. And it felt to me like CNS, today, is at a place where oncology was 15, 20 years ago, where, looking forward, we really have an opportunity to transform brain health. And particularly at a time, unfortunately, due to COVID and the pandemic, where things like depression are increasing 300%, 400% year-on-year because of the lockdown. So I was looking for a company with great science, who develop real products that have major patient impact, and Sage was among the top of that list. So when Jeff and -- Jeff Jonas and Kevin Star contact me kind of within minutes of my release going out, asking me to get involved, first as a consultant, then a Board member, it led to the opportunity to be the CEO of Sage and carry Sage forward to the next 10 or 15 years. But it really was grounded on phenomenal science, real products with, I hope, demonstrable impact on patient health to be transformative for global health.

Got it. And one of the things that people have asked me is, did Barry get to weigh in on the Biogen deal? Why did the Biogen deal happen? When it did? I mean, obviously, this is an incredibly transformational time, potentially transformational time for both companies. And people wonder, did you derisk too cheaply? Was this done on your watch, so to speak? Did you buy into this deal? Maybe you could walk us through the history of the deal, and how you sort of evaluated it.

Yes. So as many deals occur in the biotech, big biotech or pharma sphere, not just Sage, but all companies, companies talk throughout course of years. And Sage certainly, has been talking to a number of companies, really intrigued by the big market opportunities that Sage is pursuing. What Biogen felt was -- and it's true, Sage has a phenomenally strong team, has achieved a lot in the last 9.5 years, particularly on the R&D side. And they were very intrigued with the science. So there were lots of conversations back and forth. I got involved in the summertime as I started consulting to Sage. And it was important that the deal would be a robust deal and have the 50-50 deal that we landed on. So I'm thrilled with the deal. I'm thrilled with Biogen as a partner. The last thing you want to do when you're launching a drug with a profile, for example, that zuranolone has, where we really have to reimagine the treatment of depression. You don't want a company that's done depression for the last 20 years. We have to think about depression going forward, not in the rearview mirror, and Biogen was like-minded in terms of reimagining depression. Certainly, the balance sheet strengthening allowed us to immediately say, we're going to expand and accelerate the pipeline, something that Sage has wanted to do for a number of years. And I know we'll get to it later. But in addition to zuranolone and 324, which we'll do in parallel -- more indications in parallel with Biogen, we have a weary on SAGE-718, but we can now think about aspects of how do we accelerate that, now that we have the balance sheet to do so.

Got it. So it's fair to assume as a takeaway from what you've said that the deal -- the deal you negotiated was sweetened over time. It was just a better and better deal for Sage as you...

Yes. So I think the deal, as landed, got to a point where -- it's a great deal for Biogen. Clearly, it's a phenomenal deal for Sage. We've got the right partner, the balance sheet strengthening. And yes, why now, Ritu? If you -- and I clearly -- you don't need to -- but if you take sort of the net present value, being able to expand, accelerate in December of last year rather than waiting until third quarter or fourth quarter 2021, that's potentially billions of dollars of value we're going to create by accelerating and expanding our pipeline. And not -- needless to say, we're going to help a lot more patients globally now with a partner like Biogen.

Got it. So obviously, the near-term investor focus is the WATERFALL data that's coming. You guys mentioned that you have closed enrollment that you're still randomizing patients, but that you've got clear sight into first half data. We've published that we think the data is May-ish depending on, obviously, how long it takes to clean up the data, et cetera. The question I get most frequently is, why do you think WATERFALL will work when MOUNTAIN did not? And we know Jeff Jonas' answers, but you obviously did your diligence before you signed your contract. And as you saw what happened in MOUNTAIN versus how WATERFALL is being conducted, of the different drivers, dose, food effect compliance baseline, what really struck you as this is probably the left turn in Albuquerque?

Yes. So if we take a step back and just think about the data generated for zuranolone thus far, zuranolone has been tested in 3,000 subject/patients, and the data have been remarkably consistent. You see a rapid onset of action. As soon as 2 or 3 days, people are starting to feel better. You see a durable effect. And the evidence that durability is the SHORELINE data, the 30-milligram data, even not the 50, where there was about a 50% remission rate, and 70% of patients only need 1 or 2, 2-week doses in the course of the year. So just think about that. That's pretty remarkable that people are on drug only for 4 weeks and off drug the rest of the year. That allows people not to be defined by taking their depression drug. And the safety and tolerability has been consistent across all that. So we clearly have a drug that's well tolerated, and that works. The question is designing the study properly that demonstrates that. And you're probably aware that many depression studies, across many drugs, multiple Phase IIIs have failed for various [ reasons ]. So now let's talk back about MOUNTAIN, and I think MOUNTAIN provided a tremendous amount of data, including learning to design WATERFALL and CORAL going forward. First and foremost, when you looked at the scatter data from from MOUNTAIN, the patients even at 20 milligrams, that had enough blood concentration because they took it with a meal at night, it worked. And more, obviously, in the 30-milligram work, particularly those that took it with dinner at night. So when you get the right concentration of drug, the drug clearly works. And I kind of looked -- I had a chance to see those data and have those data explained to me before kind of I signed on. Now let's look at the specific lessons learned, applied to WATERFALL. So its dose, so it's 50-milligram versus 30 milligram. Its inclusion/exclusion criteria, HAM-D-24 or above. Its better instructions, take with your dinner at night. And those sets of activity clearly set up to create a stronger study. The other thing the team did smartly is took a hard look at the sites and how the sites performed and really only are working with those sites that are the best-performing sites, meaning they follow the inclusion/exclusion, they give good instructions, they've got good and clean data. So obviously, we'll need the readout, know for sure, but lots of confidence in the go-forward studies, both studies with zuranolone.

As we think about the sort of time course of WATERFALL's conduct, were there any COVID accommodations that you had to make? Were there any patients you had to screen out? One of our previous panels mentioned that oftentimes you can have sort of new psychiatric symptomatology with the stress brought on by the pandemic, and it could be hard to tell what is depressive disorder versus -- no, I'm forgetting the exact term, but one of the terms related to PTSD. So it can get squishy for somebody reacting to an enormous sort of black swan life stress. So how did that -- what did you have to do to avoid that sort of interference with -- in WATERFALL?

So the study was designed by the team in an excellent way, understanding that COVID was occurring. So COVID sort of was on the scene as we're imaging the study. So we -- this has come as a surprise and pivot came as an explicit part of what we were studying. The other aspect here that we're proving out, Ritu, and we'll prove this with more and more data is, while depression is a chronic condition, you're not necessarily chronically depressed. It's created by depressive episodes that's sending people into a state of depression, actually rewiring brain circuitry. So whether -- if you're depressed because of genetics, you've had a baby, or because of COVID, you're depressed. And MDD is a specific diagnosis that requires a prior history, and a clear documentation that patients are, in fact, MDD patients. So the chance that someone has kind of general anxiety disorder due to...

Adjustment disorder was what I was thinking.

Right. Exactly. Due to COVID at this point, those aren't the patients. We're getting really sick MDD patients in this study. So very little concern of that. In fact, patients that might have had a previous depressive history and are, in fact, depressed because of COVID, that's a patient group that, I hope, we're able to help, and we're -- who certainly are going to need that help with depression going up 300% to 400% year-on-year due to lots of things, including the pandemic.

Got it. Any concerns about the safety and tolerability of the 50-milligram dose since the WATERFALL dose based on what you've seen from the open-label SHORELINE study, whether you're talking about the patients who went from 30 to 50, or the patients who've been on 50 with retreatment? Yes. And so any signal there? And when might we get -- when might we get more SHORELINE data sort of illustrate the 30 to 50 and 50?

Well, the short answer is sort of in the 50, nothing concerning to date. The longer part is, the 30-milligram readout will occur in the first half of this year, the 50-milligram readout will occur in the second half. Now clearly, there's some 50-milligram data in the first half readout. And to the extent that there's anything new to share, and I don't know that until we look at the data, we'll certainly provide an update if there's anything. What we've seen thus far in patients treated is actually improved efficacy, with a similar tolerability profile. We've seen the same kinds of adverse events. They've been mild to moderate. We have seen a few more adverse events in the 50 as expected, but nothing concerning. And the key adverse event we see that concern people is the adverse event of somnolence or sleepiness. And what's interesting is what you worry about is not necessarily the sleepiness at nighttime, when you take your pill with dinner because sleeping well is actually a benefit to a patient that's depressed. In fact, many depressed patients require sleep aids to get and stay asleep. So it's possible here, and we'll try to prove this at some point in the future, that zuranolone is actually helping sleep, rewiring sleep architecture in the brain, which is a beneficial impact to a depressed patient. What you worry about is real sleepiness next morning some people get with benadryl or ambien or other things like that. That's not how this drug works. So we might see some of that. But keep in mind that all of this is in the context of a 2-week course of treatment. This is not something you have to live with 365 days a year. So even in that small percent of patients that might have some level of adverse event, can you get through it for 2 weeks. And the data we have thus far on dropouts and -- suggest that these adverse events aren't so significant to drive [ people ] to discontinue drug because that's really [ something ] to worry about.

This is in SHORELINE. This is in WATERFALL. This is in CORAL. This isn't SKYLARK.

Everywhere we look. Yes.

Okay. Got it. As we think about somnolence versus sedation, Barry, what's the difference? And like what degree of somnolence is acceptable? Like is it just about the next morning hangover? Or is anything between loss of consciousness and next morning hangover acceptable? Or is there a certain level that starts messing with the market model?

Yes. So we're not seeing loss of consciousness. That's really an extreme. So it really is more or less sleepiness. And for most patients at sleepiness, the night they're taking the drug, you see very few patients flow over to the next morning. And we'll see how many that is in WATERFALL. So the key here, again, is compliance. Are patients staying adherent to the drug for 2 weeks, are they getting better, faster and staying better? And we're -- so far, compliance looks really good. So as you know, in any drug, the adverse events you worry about are adverse events that drive lack of compliance or drive lack of adherence or patients have to discontinue the drug. And at least so far for zuranolone, we're not seeing a profile that's changing adherence or driving discontinuations. And that's really what you have to look for.

So we'll talk about SHORELINE versus REDWOOD for a second. Once upon a time, you needed REDWOOD forced retreatment study to establish the safety of data every 2 weeks. Your argument has been, since SHORELINE suggests retreatment is much, much, much less frequent and you're capturing that data, forced retreatment every 6 weeks is unnecessary. Have you been able to speak to FDA about this sort of general theory? I understand you don't have anything in writing, but have you socialized them to this idea? And is it like -- best case scenario, would you still need to do like a tiny REDWOOD, just like 20 patients redosed at that every 6 weeks kind of thing, just on safety, no efficacy measures?

Right. So I'll remind you that zuranolone has breakthrough designation, which enables a more frequent and consistent dialogue with the agency. That's why we got breakthrough. And the agency also recognized that zuranolone, which is different than 35 years of antidepressants, has a fundamentally different benefit risk profile, again, because of the 2-week profile, better adherence, less -- and higher levels of efficacy quickly. So it really -- and we've shown this -- these graphs. It really separates in terms of benefit and benefit risk. So those are really important characteristics. When we -- the team met with the agency early on -- or remind you, we describe that we designed along with the agency adhered 3 distinct sort of opportunities to get zuranolone to market. The opportunities that WATERFALL represents, CORAL represents in terms of rapid release, and then obviously, our PPD opportunity as well to get zuranolone to the market. So the plan has always been, let's get the WATERFALL data. We'll take the totality of data. We'll sit down with the agency and figure out what the right timing and filing strategy is. As we stated clearly, I'll state again, when REDWOOD was designed, it was answering the question of frequent dosing throughout the course of the year and what happens to that patient from an efficacy and safety perspective. And as you said, Ritu, and I agree, based upon the data to date, including the SHORELINE data, that's no longer a question that we think needs to be answered because that's not going to be happening in the real world. 70% of patients...

Right. Like 3%, 4 % at max would need that. But wouldn't you still need something just to make FDA feel better?

Well, look, that's a discussion with the agency. Our view is that REDWOOD has design doesn't make sense. Other small data, a post-marketing commitment that might make sense. That's something that we and the agency will talk about. But I think we're all on the same page that the current state of depression is growing. The unmet need is growing. The current treatments do not offer solutions for most patients. We need something new. We need something new on the market quickly. And while we can help some patients in clinical studies, the way to help the most number of patients is to get the drug commercial.

Do you think that the probability of success for WATERFALL is materially different than CORAL? I mean they're different -- they're a little bit different, but a little bit not. So how do you think of that?

Yes. I remind, WATERFALL, we're delivering zuranolone as a single agent. For CORAL, we're delivering it at the same time as a newly prescribed SSRI. So...

But some patients in WATERFALL will be on background SSRIs, chronic background SSRIs. Yes.

That's correct. That's correct. So hard to differentiate the fine line of probably success. It's our hope, as designed, that both studies result in statistic significance. And if so, they're both important studies.

So first-world problems, CORAL and WATERFALL both work, and you're looking at RRT versus episodic. And you're trying to build that -- those market models. How would you think about finding a price that would make sense for both indications? But -- and if you -- if only one of them move forward, would there be a price differential between the 2 indications?

Yes. Ritu, it's too early to talk price specifically, but what I can say is that, in the hope they both read out positively. What we're offering is different approaches for patients and physicians to treat MDD. They're not really distinct patient population, per se. It's patient -- until we have more data, it's patient's and physician's choice. So there'll be a number of physicians that, with SHORELINE and WATERFALL, are convinced that zuranolone alone is right for their patients. There'll be others who are uncomfortable with that yet and want to prescribe both. And hopefully, we'll have information based upon physician and patient choice. What I can tell you is that our plan is to be very proactive with payers and launch zuranolone under the context of a proactive value-based agreement. So that if payers are concerned with certain things, we can start solving for them in a value-based agreement, providing some level of certainty and, in certain levels of protection without going to that classic rebate model.

Got it. That time of the talk where you have to pick a favorite child. I'm not going to ask you to pick a favorite drug, but it might be worse. I'm going to ask you to pick a favorite next indication for zuranolone. So we've got generalized anxiety disorder, bipolar depression, probably things I'm not even thinking of. What are you most excited about for either zuranolone or one of the related compounds in other indications...

Well, first of all, what I'll say about zuranolone is we're still working through the development, clinical development plan with Biogen as our partner, I need to do that before I can comment. I think we want to make sure that we answer all the questions we need to answer in MDD and PPD so that we do launch that the data gaps that we have today are filled with other studies. So we'll really go deep in MDD and PPD to make sure we can answer your question. Because it would be a shame to have major segments of patients not get drug because levels of treating physicians were uncertain about certain aspects of data. So we want to make sure that we win in MDD and PPD. And by winning I don't mean beating anybody, I mean winning for the benefit of patients, getting every patient who can benefit on drug. And then we'll work out. Do we go after general anxiety disorder, seasonal anxiety disorder next, we've got to work that out with Biogen. In terms of excitement, I've mentioned this before, I'm incredibly excited by SAGE-718. It's a first-in-class an MDA drug that is studying cognition-specific executive function, which has really not been done before. So it's an opportunity for Sage, along with all the other pathways we're blazing, to blaze a whole new pathway in neurodegenerative diseases to study potential improvement in executive function, which is incredibly exciting.

Right. I wanted to actually get to 718 in just a second, but let's hit 324. We've only about 6 minutes left. 324 and the ongoing Phase II study in essential tremor, you've set the bar at 30% to 50% reduction of tremor. But you already said -- I think, you said on the earnings call that you basically pick doses for a maximal, maximal effect. How confident are you that if you hit 50%, but the sedation profile requires tweaking that you can maintain that 50% while still impacting the -- or improving essentially the sedation or somnolent aspect of the drug?

Right. So I'll remind you that the KINETIC study was designed as a Phase IIa to pick the maximum dose of 324, and to test whether we see that 30% to 50% improvement in tremor is measured by upper limb score from a patient's baseline that's sustained over the course of the study. So that's number one what we're looking to see. The other part on the safety is to make sure that the adverse event profile is as expected, and we don't see wholly new adverse events, some new things pop up. If that's the case, and we have the efficacy and we have a safety profile that's consistent with a drug class like this, we're highly confident that with dose, dose frequency, dose levels, formulation that will have a path forward. And it's really important. We want to lean into this on the earnings call. The profile that -- and I don't know the data. So the profile that we exit phase -- we exit KINETIC with is not the Phase III profile or the commercial profile. There's work to do between best readout and the product that goes into Phase III. And in the course of small molecule drug development, that's how this works. So I'm looking for efficacy, as we described. I'm looking for an adverse event profile that's as expected with nothing new. If we have that, that's a big win and allows our next step, which is a Phase IIb.

Got it. And what would that Phase IIb look like essentially? I mean suppose you've gotten a satisfactory profile. Would you still want to take multiple doses forward just given the kinetics of the compound and the disease? And yes. And what are what are the other considerations for a Phase IIb tremor study?

Well, look, if you asked the question, if we get KINETIC, and we're pleasantly surprised that the maximum dose presides efficacy with the tolerability profile that is acceptable to move forward, then you move forward. What we're trying to say is, again, don't know the data. We pick the maximum dose to not run into the mistakes. You see some companies run into where they pick a low dose and then they see nothing in the Phase IIa and then what they can do. So we did pick that maximum dose so that we would move to the IIb. We think we moved to IIb. And I don't know what that looks like, Ritu, until we see the data. You can imagine picking multiple doses or multiple doses and frequency or nighttime dose with a true-up the next day. There's a lot of possibilities. You could see extended-release formulations happening. So there's a world of possibilities for IIb, and we'll look at the data, and we'll share our thinking as soon as we have the data and that level of thinking.

Got it. Just a quick minute on 718, your NMDA modulator for cognitive impairment. Good Huntington's data. It is the Alzheimer's study that is ongoing, the Parkinson's cognitive impairment should read out shortly. Many investors are -- they're hesitant to give you value for this program just because they're afraid of the same challenges that have hit Alzheimer's dementia programs. Like how should we think about your approach to derisking a cognition program?

Yes. So look, when you think about cognitive studies in the context of ongoing dementia, it's complicated, as we've seen. And patients are deteriorating rapidly. There's comorbidities, there's behavioral aspects, there's motor impairments. What we're doing here is fundamentally different. We're looking at improvement in cognitive performance or executive function. As that's what we're studying, by definition, we're working kind of earlier in the disease [indiscernible] some of these co-founding impairments occur. So let's take Huntington's, for example. You might have somebody diagnosed with Huntington's disease at 25, which is terrible. They might not get motor impairment for another 10, 15, maybe 20 years, but they tend to get cognitive impairment sooner in their disease. So if you think about helping that person stay self-sufficient for the next 15 to 20 years, that's remarkable. And then we can think about Parkinson's and Alzheimer's sort of earlier in the disease, that maintains and improves executive function. Really, we're talking about is we're talking about people doing things that you and I do every day. We make a shopping list, we go to the store, we buy [indiscernible], we come home, we put stuff away, and we remember where we put it. That's a remarkable achievement in degenerative disease. And if we can maintain people's ability for independence, that's a huge win.

Okay. And then just as far as the rest of your pipeline, you unveiled a slew of additional candidates with potential indications on your last earnings call. Which one sort of strikes you as particularly promising?

I don't want to name a specific program. But if you step back, what Sage has done is understood endogenous receptors for both GABA and NMDA and dialed in, allosterically, different levels of impact for both of those positively, neutrally and negatively. And that allows us to fine-tune drugs for a variety of indications. So as we move these medicines along, what Sage does and will continue to do is small studies and patient groups, we think where it works, see big effect sizes like we did with 718 and Huntington's and allow human data to guide the indication selection rather than de novo pick the indication before we have data.

Got it. Great. Well, we're at time. Barry, thank you so much for your time today. Great talking to you, and look forward to talking in, hopefully, May-ish with positive WATERFALL data.

Thanks, Ritu.

Take care. Bye.