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

Good day, everyone. I'm Sumant Kulkarni, a senior biotechnology analyst at Canaccord Genuity, and I'm pleased to welcome you all to our growth conference this year. Hope all of you are staying safe and enjoying your summer so far. I'm pleased to have Sage Therapeutics with us at a very interesting time for the company. For Sage, we have CEO, Barry Greene. We'll be keeping this very interactive. So please feel free to send questions to me using your web-based interface or e-mail me at [email protected]. And with that, I'll turn it over to Barry, or we can just go straight to Q&A, depending on how he'd like to go.

Yes. Sumant, thanks for having us, and I'd like to thank the organizers for Canaccord for having us here. It's very -- as you said, it's a very exciting time for Sage. We have had a very catalyst-rich year. We see ourselves as leaders in brain health and believe that we could be a top-tier biopharmaceutical company in the next 5 to 10 years. As you're well aware, the deep work that Sage has done in brain circuitry, specifically around NMDA and GABA, have led to the first-ever drug approved for postpartum depression. We've got zuranolone on the horizon, which I'm sure we're going to talk about, and a very significant pipeline, behind that, of drugs that have been invented by Sage. So I believe that we're well on our way to serving millions and millions of patients with brain health disorders, and I'm happy to talk to you about that more.

Sure. So I'll just start off with a big picture question. You've been at Sage for some time now, first as a director, now as a CEO. You just had a recent stock purchase. Could you just give us some of your thought process behind that and how you're thinking about the stock for now?

Yes, Sumant. So as you said, I was able to join Sage late last year when I exited Alnylam after almost 18 years there. And I really wanted to do something focused on CNS and brain health. As I looked around the industry and felt -- I felt that the next area prime for major innovation and major breakthroughs was, in fact, brain health. And I likened it to what I saw in oncology when I got involved in the -- in 2000, 2001. We had decades of really important good medicines, but not necessarily the breakthroughs for patients. We have the same thing going on in brain health. We've had good medicines for 35, 50 years, but not great medicines, not medicines that are transformative in the lives of patients, things like depression, essential tremor. We've had nothing for cognition. So as I looked around the industry and looked at deep science, great biology, strong team and pipeline, I felt Sage was well positioned to be transformative in the landscape of brain health disorders. So that's why I joined first as a consultant, then the Board, then the CEO. And in terms of stock purchase, as you know, we have very small open windows. And I felt that based upon the great news we've had with some of the negative stock sentiment, it was sort of a no-brainer, given where I think we're going to go over the next 5 to 10 years.

And then I've got some questions on zuranolone here, specific ones. First, just to clarify, is it still fair to assume that Sage and Biogen are mostly interested in treating MDD episodically with the product, and that's changing the paradigm?

Yes. Look, if you step back and think about really the last 35 to 50 years of antidepressants, what psychiatrists have been trained is if you diagnose someone with MDD, you want them to be on an antidepressant, checking in, in 6 months. If they're in really good shape, wean them off. If they're not, you'd want to keep them on their antidepressant really for life. And that's just not the way kind of the modern drugs that we're developing work. We're almost resetting the brain circuitry to a normal state. So the way zuranolone works, as you know, is someone diagnosed with MDD or PPD take this every night for 2 weeks. Starting after even the first dose, but measured clearly after the second dose, day 3, we see a rapid improvement or a decline of the depressive symptoms, remove their anxiety and sleep. And then after 2 weeks, they're back to, if you will, a normative state. So it's more of a treat as needed. And the analogy here is that if you have a lower respiratory tract infection and you take a Z-Pak and I don't, you're going to be better very quickly. We both might be better in 3, 4, 5 months, but you're going to be better quickly and stay better. And if we get a reinfection, then we can both take that drug. Same kind of thing here. Most patients get normal again after 2 weeks. And then if they or their physicians see their mood start to darken, their anxiety start to increase, their sleep patterns start to worsen, they may be headed into another depressive state, and that could be time to re-treat. As you know, we've got a really good data set with SHORELINE that showed in the 30-milligram dose that 70% of patients only required 1 or 2 2-week treatments in the course of the year. A little over 10% require a third, about 10% a fourth and slightly under 10% for a fifth. So if you step back and say, would I take -- as a patient, "Would I rather take my pill every single day, reminding me of my depression? Or should I treat it when I start heading into that depressive state?" In the data set that I just articulated, even in the worst case, I'd argue that 10 weeks of taking a drug is certainly better than 52 weeks of taking a drug. And that's what we're hearing from patients as well.

Sure. And you alluded to this a bit, the Z-Pak analogy. What would you say to people who might perhaps be paying too much deep thought on day 42 because, as we know, 42 is not the answer to life and everything?

Well, so I think, unfortunately, some of what happened was a big misinterpretation with -- for the WATERFALL study for day 42. So day 42 is simply the last data cut or the data read of the study. And importantly, what we saw is that for those patients that responded at day 15, they held that response out to day 42. Numerically, that was about 85%, 86% of the overall response. But what's really happening is out of the data set, some patients are getting worse again. As we saw in SHORELINE, a small percent get worse. But most of those patients at day 42 were as well or better than they were at day 15. And clearly, we have an overperforming placebo that visually throws off the graph. But the important part is they were just as good at day 42 as they were at 15. And reflect back to the SHORELINE data, we know that, in a kind of real world-like study, patients get better and they stay better longer. And what's really interesting about SHORELINE, and it's unique in the study of depression, is it's almost a reverse placebo effect. So I'm a patient taking my drug every day. After 2 weeks, I now know I'm not taking a drug. So I'm on nothing, and yet, I'm not rebounding or seeking another drug or worried that I need to get back. My anxiety is not increasing. My mood is not worsening. My sleep pattern is not worsening. So it's really good evidence when you look at the overall landscape that the drug works. It works really well.

On the SHORELINE data, is it fair to assume a dose dependence on the cuts of people that might need re-treatment over time as you go to a higher dose?

Yes. That's a great question. So what we've been able to present so far is the totality of the 30-milligram data, and I articulated that. We also presented the 50-milligram data interim analysis, which showed that we had a slightly higher response rate, 80% response versus 70% response on SHORELINE. So we don't know yet if that 80% response requires less re-treatment down the -- I suspect the data will be about the same. And frankly, if we can get 60% to 70% of patients only requiring 1 or 2 2-week treatments in the course of a year, that's a pretty spectacular data set.

And then moving on to CORAL, which a lot of people are focused on. What would you think would be an ideal result in CORAL, given we've seen zuranolone work relatively rapidly so far?

Yes. So what CORAL -- and if I step back, what we did at Sage is last year sat down with the agency, and this is after the MOUNTAIN study, which just barely missed stat sig, and all appreciated that we have a drug, zuranolone, that works. What we need to do is we need to prove that it works in yet another randomized controlled Phase III study. And if you think about a drug like Prozac, Prozac failed 7 Phase IIIs to get -- out of 11 studies to get approval. So failed studies in depression is rare. And I was on a panel with Steve Paul yesterday who made the statement that if we judged every depression drug by a failed Phase III, we'd have 0 on the market today. So this is part of what we're all dealing with. Fortunately, we've had mostly positive studies. So we sat down with the agency, we said, what's the right path to approval? We designed 2 studies for MDD: WATERFALL, which we read out positively; CORAL, which is coming up; and then another PPD study, SKYLARK. Any one of which, if positive, provided the efficacy data set to file on. So we saw that positive study out of WATERFALL. CORAL was designed to ask the question, if we give a concomitant zuranolone with an antidepressant, do we see the same kind of results that we've seen with 3,600 patients treated? And then obviously, SKYLARK was answering a PPD question. Again, in case the 2 MDD studies didn't work, we could back off to just the PPD label. So what we expect to see out of CORAL is what we've seen across the board. We expect to see a rapid response at day 3, 8, 12 and 15. We know that antidepressants don't work that rapidly, and there's thousands of patients of literature that say -- that suggest that. We might see some interesting new side effects that aren't zuranolone side effects but are side effects due to antidepressants, things like weight gain, sexual dysfunction, which we don't see. So that will be very interesting because both arms will be on that antidepressant. And I expect the drug to work as the drug worked, rapid response at day 3, 8, 12 and 15. What we don't know is how placebo will respond. And we know that placebo rates have been improving over time in depression studies. But I will say this, we are looking at all the studies we've done, including WATERFALL. We're applying those studies to CORAL. And if we believe that there needs to be changes in CORAL, earlier time points, bigger N, we'll make sure that we work with the agency to make changes because we're not going to run a study that we know is going to fail. We're going to run -- we want to run a study that has a level of success that we can utilize that study to help patients.

So that brings me to my next question. You have WATERFALL. So you have a fileable product as you've said. Some safety data, my understanding, may be necessary from CORAL, but maybe not all the data for a fileable package. But how important do you think the CORAL data would be in terms of potential real-world usage and commercial implications of that?

Yes. So what we believe we have today is we have the efficacy data for a fileable package now that we have that next complete positive Phase III with WATERFALL. So we have the efficacy package. We've said -- we said at the beginning of the year, and we sort of have tried to repeat this, we chose to rerun the pharmacology studies at 50 milligrams. We had the 30-milligram complete, but we're rerunning them at 50. Those studies will be done kind of at the end of the year. Why? We didn't want to leave any room for interpretation about any boxes that didn't get checked as we filed for the NDA. This is a large indication, millions of patients. So we want to make sure that we provide the highest quality data package. We expect no surprises with the repeat pharmacology studies. And then because it's in zuranolone's best interest because of the clean safety profile, our intention is to take a blinded look at all safety on ongoing clinical studies as part of the package. Now we expect CORAL to be complete as we've guided. So with a complete CORAL, both efficacy and safety, that will get integrated into the package. But we've said historically, and we confirmed this with the agency, that unless we see some strange pattern, it doesn't work on day 3, 8, 12 or 15, that statistical significance at day 15 is not necessarily a requirement for filing. We have the efficacy data we need for filing. This is yet another study.

And we know there's been a lot of unpredictability with the FDA recently. So -- and I'm not sure this is a fair statement over time, it might always be unpredictable. But when was your last discussion with the agency to ensure that you are completely on board with the regulatory requirements? And I'm assuming the last discussion was the one that resulted in not requiring REDWOOD and RAINFOREST? Is that a fair assumption?

Well, so we have -- given the breakthrough status that we have with zuranolone and given the way breakthrough works, there's ongoing discussions with the agency. Some of those discussions are more strategic, which leads to decisions to not run certain trials, like you mentioned. Some of the more tactical, in terms of dates and time and sort of administratively, do we do a rolling submission? Do we do a full submission? The kind of formal meeting where we do the official Type B NDA meeting, that's coming up in kind of weeks, not months. And once we have that formal meeting with sort of everybody present that locks in the data sets, all the administrative steps, as we've said before, we'll come out in a Reg FD format with Biogen and clarify what we believe we and the agency have agreed for the path forward. I do expect it to be as we've already articulated.

Got it. Next is, again, a real-world type question. So if zuranolone is approved, how does a physician reconcile what might be an episodic treatment versus the potentially "safer way" of using iconic treatment to make sure patients are always on a standard of care? And does that mean that episodic treatments will mostly see use as an adjunct therapy?

Yes. Sumant, what you just said is -- it really sets up the paradigm shift that we need to create. And I kind of mentioned this before. What you're taught, even today in psychiatry school, is that depression is always there underlying. It's a chronic disease. And put someone on antidepressant, check in, in 6 months. If they're better, wean them off of their antidepressant until the next time they have a depressive episode. If they're not better, keep them on antidepressants for life, which was the standard of care. We've got something different and modern and innovative, sort of a take it as you need it, when you need it to get yourself better. And again, we likened it to -- slightly different, we likened it to a Z-Pak. So what we have to do is educate physicians that there is an underlying chance of depression, but they're really sparked by episodic effect -- events. COVID hits. I've had a baby. I lose my job. I lose a loved one. And my mood, my anxiety and my sleep start to change. They all worsen. And when I'm heading down that path, I'm heading to a depressive episode. That's when you treat and get them better with zuranolone. So it's a different way of thinking about it versus I must have a drug on board. And go back to SHORELINE again. I mean if you ask the patients, they won't tell you that their standard of care is a better, safer way to treat because of the weight gain, which means I need concomitant metformin, or I've got sexual dysfunction or I've got euphoria or all the other side effects that come along with someone that takes an antidepressant. And the data are pretty clear. Now the average length of use for an -- standard antidepressant is about 7 weeks, and patients cycle through 2 to 3 antidepressants in the course of a year. I wouldn't call that a safe way to treat.

Right. And I kind of see this as the migraine market in reverse, where that went from like acute as-needed treatment to more prevention chronic type aspects, and now you have a blurring of the lines with an older product or 2 coming. So that's why I look at the MDD market sort of in reverse. So we'll see how that plays out, hopefully, in the real world. So what are your latest thoughts on how an episodic treatment might or should be priced? And what do you think real-world durability could look like? We've seen some SHORELINE data. We know that. But -- and how would all this change if the PPD market comes along?

Yes. So let me sort of go back to a comment you made, Sumant, because I think it's important. We believe that the data that we'll have at launch will allow a patient and a physician to ask, "Do I want this by itself?" Or many patients will be on -- already on an underlying antidepression that hasn't given them the bad side effects so it's fairly stable. Or this will be an add-on. Or they'll decide, "I'm going to give you zuranolone to get you better faster. But just in case, because I might not see you, you're traveling, let me add an SSRI on top of that. Not that, that's going to get you better very rapidly, but that might help to prevent the next episodic episode." And there's interesting data that suggest that all 3 approaches to treating might be important for that patient. So I think we'll have a data set for a patient and physician to add, alone or do both at the same time, depending on that patient's case. So it's not an either/or, it's really a patient/physician decision. And that's a great place to be when you're launching a drug, to give that kind of optionality to patients and physicians. In terms of pricing, it's too early to talk about the price. We do believe that we have the data for an indication for both PPD and MDD. We'll confirm that with the agency. And we'll think about this as sort of a branded drug that might look at chronic cost of a year but in a shorter time frame. We've got data out of SHORELINE that suggest how many 2-week treatments that a patient might need in a year. And then what's really important is we intend on launching zuranolone with a proactive value-based agreement, which enables a level of dialogue with payers that's different than simply a contracting, rebating approach, which I think would be the wrong approach to use in this case. Although a proactive value-based agreement allows us to talk about health policies, how many times they might re-treatment, what the epidemiology their patient group might look like. So it broadens the conversation so that if a physician writes the drug, the patient gets the drug. That's what we want to see.

So Barry, from your discussion just now, it's clear that you've put a lot of thought into the commercial aspect, but your partner, Biogen, has a lot of things going on right now. How confident are you that the asset will get the appropriate attention it deserves, especially when Biogen will also be learning along the way when it comes to targeting primary care? Are other partners thinking of this as initially especially focused with the pricing dynamics that come along with that?

Yes. So I can tell you that the Biogen team, from the top, Michel, Alisha, have all been very involved. Their regulatory and clinical team are frankly outstanding, and they've already partnered with our team on a series of regulatory interfaces we've had thus far. And any kind of partnership starts, and this started through late last year, there are start-up time and start-up costs, if you will. We're sort of over that hump and really working very, very, very well as a team. So I'm thrilled with Biogen, not in terms of the balance sheet strengthening that happened for us, but also their thought capital and how they're interfacing as a partner. We've seen really great thinking and great work, not just for zuranolone, but also for 324, which is coming down the pipe. So I'd say it's in great shape. Now interestingly enough, Sumant, the patients that they're treating today, MS patients and others, have a significant amount of comorbidity with depression. So they sort of have a natural playground of call points. We already have patients that are suffering with depression. We're going to expand into psychiatry together. And as you mentioned, many of these patients get treated with large primary care practices. But you know we live in a world of almost perfect information. So we know who those practices are. We know what kind of drugs they're treating, how many patients they have. So we'll use the right targeting techniques to build out the right call interfaces to get to the practices that we need to get to.

Got it. How do you expect to position yourself versus competing approaches from maybe a practice and the potential use of psychedelics in MDD or treatment-resistant depression?

Yes. I think we have to separate out psychedelics versus sort of modern drugs. So I mean psychedelics are very interesting. Particularly for people who went to school in the '70s and '80s, there's a certain allure there. You got to think about psychedelics more as sort of a business model. "I'm going to bring my patient in, in a well-controlled environment. I'm going to give them psychedelics to enhance my therapy of them." And it's an overall therapeutic business model, which is very different than, "I've diagnosed you. Here's a drug. I expect you to be better in a couple of days, let me know. Go on my portal, let me know, and let's check in, in a month to make sure that -- or a month or 2 months to make sure that you stayed better." It really is a portable go-where-you-want approach for zuranolone. And as I see drugs on the market and the competitive landscape, I don't see anything that's an oral drug that works rapidly that -- and would sustain benefit out for months and months and months. There's no really competition to zuranolone that I see out there.

Yes. And do you expect to target other types of depression with zuranolone, for example, depression and bipolar disorder?

We do. It's just a matter of -- it's a matter of how and when. So what we've agreed with Biogen is that the paradigm shift that we have to create for MDD and PPD is so significant. We've talked about that. And we will. This is not an insurmountable mind shift set. And it really will be driven by patients wanting the new modern approach rather than the generic approach that sometimes work or sometimes doesn't. So we're going to do all the work we need to do in MDD and PPD and kind of win there, if you will. And then as we get more experience with the drug, we'll think about things like GAD and SAD. The good news is that if you want to think about sort of the hat trick of what you want for an antidepressant, improvement in mood, anxiety and sleep, we have all that. So that gives us great flexibility for other future studies.

Got it. Now moving to SAGE-324 now. How are you thinking about the next steps there? And any refinements that might need to be made as the product progresses to the next stage of trials?

Yes. So we designed -- so first of all, SAGE-324 is initially designed to treat essential tremor. You're aware there has not been innovations in essential tremor for over 15 years. The standard that's used today is a beta blocker, but they're not used very often because it doesn't have much of a positive effect and it can have a deleterious effect, unless you're treating the heart as well. So we designed a long-acting GABA PAM to improve essential tremor. What we showed everybody in January was with a single dose of 324, when blood concentrations got to the right level, we saw a decrease in essential tremor, as measured by the upper limb TETRAS essential tremor score. Upper limb is really where the unmet need comes from. Lower limb, voice are problems, but upper limb really makes people less dependent. The next question asked was, can we see that effect durably over time? So if we give this every single day, does the drug continue to work? Or do we see different side effects? Do we see tachyphylaxis? So when we presented the KINETIC study, we showed a 30% to 46% reduction in essential tremor depending on patient population, with no surprises on side effect. We give a very high dose so we expected high levels of somnolence and sedation. Now what we're doing is a dose response, dose escalation over, of course, longer than a month. So at the end of this, we've simulated what a Phase III might look like, and we're hoping that in time frames longer than a month, given at multiple doses, we have a dose and frequency to move forward. And things like evening doses, lower doses over the course of time, based upon the data we're seeing, we think, solve that benefit risk equation and give us a chronic treatment. So that's the question we're answering right now.

Got it. And we just have about a minute left so I'm going to squeeze in 2 questions. First, on SAGE-718, could you give us a sense of what you expect to learn or see in the upcoming LUMINARY study? And the last question, big picture. You've been CEO for some time now, hopped on ahead of a very important data set. But behind that now, there's a lot of things ongoing. So where do you see Sage in the next 3 to 5 years? And how does your significant cash balance help you to achieve that, specifically in terms of how many products might be in pivotals before you may need to replenish?

Right. As I said, we started the year with an expand and accelerate theme, and we've continued that. The Biogen alliance allowed the strengthening of the balance sheet, with continued strengthening coming because 50% of zuranolone and 324 are paid for. So we can do a lot for a long period of time, given the ongoing milestone payments. I'm thrilled with the data for 718. And I am very excited to have an NMDA positive allosteric modulator to treat cognitive -- motor-cognitive impairment across neurodegenerative diseases. We've got great data for Huntington's and Parkinson's. I believe our first approved indication will be in Huntington's, which gives us phenomenal optionality to globalize. And we're going to expand the rest of the pipeline along the way. And we have a long runway to go and generating multiple data sets.

Okay. That about wraps it up. So thank you, Barry, and thanks, everyone, for tuning in and listening to the replay.

Thanks, Sumant. It's fun.