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

Good morning, everyone. I'm Brian Abrahams, senior biotech analyst at RBC Capital Markets. Our next presenting company is Sage, featuring their CEO, Barry Greene. Barry, thanks so much for joining us.

Brian, thanks for having us, and I'd like to thank RBC for organizing this today.

Our pleasure.

So maybe let's kick things off on the most imminent data catalysts coming up, the WATERFALL data for zuranolone in major depressive disorder. Can you maybe start by talking about your latest views on how some of the adjustments and changes that you made in things like dosing, inclusion criteria and conduct relative to the first -- the MOUNTAIN study improve your expected probability of success?

Absolutely, Brian. And just to take a quick step back, I'll remind you that zuranolone is being studied as part of the overall landscape program, of which WATERFALL is 1 of 3 distinct opportunities we have for a positive Phase III readout and for submission for a potential NDA. So we've got 2 in MDD and 1 more in PPD. Any one of those Phase IIIs, should they be positive, gives us an opportunity to file and have an important medicine on the market for MDD and PPD patients, depending on success of the trials. So specifically with WATERFALL, we're very pleased with the conduct of the trial. I'll remind you that our team designed the WATERFALL study, understanding that we were in the midst of a pandemic, with COVID specifically in mind. Unfortunately, for the world, major depressive disorders have gone up three- to fourfold pre-pandemic to now. And in fact, adolescent and adult suicidality has nearly doubled. So the pandemic has certainly exacerbated the brain health pandemic we're also in the midst of. And therefore, we were able to upsize WATERFALL twice. We learned from previous studies and used the 50-milligram dose. We've learned from previous studies and leveraged the sites that were most effective. We increased the HAM-D scores for the inclusion/exclusion criteria and believe that in the first half of this year, we'll have a robust readout for WATERFALL. I'm highly confident, based upon the data to date in kind of over [indiscernible] [ 3,000 ] with a very consistent profile with zuranolone, that we should have a very robust readout.

Great. And you mentioned COVID having an impact on the rates of depression across the population. And that's, of course, enabled you to upsize the study. But how do you control for any potential COVID effects? For instance, the potential for greater numbers of first episode depression patients who may respond differently to drug or the drug and/or placebo, or any other potential variability that COVID could introduce relative to the way you had initially powered and planned the study?

Yes. So Brian, as I mentioned, the team designed WATERFALL, and actually the Landscape Program, understanding that we were in the midst of a pandemic due to COVID. What we believe, the hypothesis at Sage, and this has, frankly, been proved out with SHORELINE, is that while people have a depressive undertones, the actual bouts of depression are caused by specific trigger events. Those trigger events could be the loss of a loved one, the loss of a job. It could be being trapped due to COVID. So the MDD that we see that may be driven or exacerbated by the COVID lockdown is not that different than MDD we see with other kind of trigger events. And therefore, we believe that the results we have should be the same, whether in the midst of a COVID pandemic or not, or whether it's a trigger event due to having a baby or, as we talked about here, due to COVID. And now let's look at the SHORELINE data. I mean SHORELINE was conducted in the midst of the COVID pandemic. The 30-milligram final readout was consistent with the 30-milligram readout in the fall. We saw that 70% of patients treated only required 1 to 2 2-week treatments. So think about that. At max, 4 weeks of treatment in 52 weeks to get yourself in a normative state. And in fact, we saw better efficacy with a 50-milligram dose and 80% response rate with no really additional safety liabilities. So I take great confidence in the SHORELINE data being a good read-through for WATERFALL.

Got it. And then speaking of SHORELINE, you recently announced expansion of the study. Can you talk about the rationale? And maybe, I guess, how much you're learning in addition to what you just mentioned about -- with respect to COVID. What else are you learning about the data there and how that study plays into the regulatory strategy?

Yes. So we have agreement with the agency that SHORELINE will serve as part of this overall safety database. So we're in pretty good shape now. The reason for reopening SHORELINE, number one, is, as we talked about, the rates of depression are up three- to fourfold from pre-pandemic. So giving more patients the opportunity to potentially benefit by zuranolone was certainly in the cards. We wanted to study that patients could roll over from CORAL. And while we're in the midst of conducting these Phase III studies, it was not appropriate to open expanded access program. It's something we plan on doing later when the studies are closed. But right now, the way to offer zuranolone to more patients in need is by reopening SHORELINE and enlarging the safety database, which is already sufficient, but enlarging it will only be helpful in the context of a regulatory filing.

Got it. That makes sense. And at the beginning of our discussion, you mentioned that WATERFALL is 1 of 3 distinct opportunities for a positive Phase III readout. What's your latest view based on the progress of these studies and your regulatory feedback as to how these 3 different Phase III studies play into the overall regulatory strategy? What if -- clearly, if they're all positive, that -- there's an answer; if they're all negative, that provides a clear answer as well. But what if the studies are mixed? What's the extent of data or number of positive studies or patients within positive studies that you would expect to need for approval of zuranolone?

Yes. So if we step back and think about studies in depression over the last 35 years, and Brian, I know you're aware of this, actually, most Phase III studies in depression have failed over the last 35 years. Prozac, which is a very important drug in the landscape, multibillion-dollar drug, had 7 out of 11 failed Phase III studies. So it needed -- it had 4 positive studies and again, it became a very important drug. And that's because of the variability. We're not looking at biomarkers. And these are advances we're going to make, I believe, over the next decade. But right now, it's physician assessment. So these are tricky trials to run. I think what we've done is taken the learning over the last 35 years, certainly, the learnings for zuranolone, and design very robust studies with WATERFALL, with SKYLARK and with CORAL. So I'm of the belief and the hope that we're going to go 3 for 3 here. But as you said, should we not, any one of these studies, should they be positive -- and the basis, we already have a positive PPD study with ROBIN. So any one of these studies forms the basis of approval. Again, hopeful that we go 3 for 3 here. We certainly designed the studies to do so. Zuranolone's benefit risk is very different than any other antidepressive seen to date. And by that measure, we should have a robust effect size. So the plan, should WATERFALL readout positive, is to sit down with the agency, talk about time lines, talk about the outstanding studies. If the agency believes that the CORAL study is going to be part of a major review, then likely, we'll wait for CORAL and file the whole clinical package together. If the agency does not believe, based upon WATERFALL, that CORAL will be a major review issue, then we go ahead and file. But the thing we don't want, Brian, is we don't want to file, have a readout of CORAL become a major review and then have to announce a 3-month extension. Not that it's a big deal, but it gets people a little bit wonky. So we kind of want to avoid that.

Got it. With the SHORELINE data in hand, what do you think is going to be important to demonstrate out of the WATERFALL study out to day 42 beyond the 14-day endpoint? What are your expectations there?

Yes. So I want to be really clear on this. If we hit the primary endpoint, that is a statistical significant result at day 15, we have a drug. We have a positive study. We have agreement with the agency that that's what they're looking for in the context of the overall data, including, as you mentioned, SHORELINE and a positive ROBIN study. Now clearly, it'd be great if we were able to announce that we hit all primary and secondary endpoints, but the study even upsized is not powered for statistical significance at day 42. That would take a much, much larger study. So what we're looking for at day 42 is maintenance of effect. Is day 42 relatively where day 15 is or within 50% of day 15? So what we don't want to see, just to put it that way, is we don't want to see complete loss of effect out to day 42. We want to see around 50% of the effect we saw on day 15. But again, we're clearly not looking for statistical significant -- stat sig at day 15 is a positive study. And given the differential benefit-risk of zuranolone, and by benefit-risk, I'm talking about lack of discontinuations and remission rates, much different than all other antidepressants, we have a very important medicine in the depressive landscape.

Got it. And ahead of these study readouts with these studies -- with the data right around the corner, what degree of market prep work have you and your partner, Biogen, done? And what kind of commercial infrastructure do you think will be required to launch zuranolone?

Yes. So we, clearly, with the SHORELINE data, have gotten experience in a number of psychiatry and neurology. So we've started the market prep on the health care provider, patient advocacy, payer front. But just to be clear, this is going to be a very big lift for us and Biogen. The world is used to treating depression chronically with a pill that a patient has to take every single day. So coming out with a 2-week treatment that potentially is rewiring depressive brains back to normative states is a paradigm shift, a major paradigm shift in the landscape. I can tell you, and this is gratifying, that we have a chance to educate health care providers, given the SHORELINE data. People are becoming believers that, in fact, their patients are being triggered, they're reaching a depressive state, we can help them recover from that depressive state with 2 weeks' worth of treatment. And the other thing I'd add, Brian, which is really important. If you think about SHORELINE, so a patient gets drug for 2 weeks and is off drug. So it's an open-label study. That patient knows that day 15 and on, they're not on drug and yet still report a normative state behavior. That's almost a reverse placebo effect. Most studies of patients on placebo, they believe they're taking drug, they feel better. This is the opposite of that. So it really talks to the tremendous effect that zuranolone has on these depressive states.

Got it. And what's your latest views on how one might approach pricing and obtaining access for an acute use of really paradigm-changing depression therapy like zuranolone?

Right. I mean it's too soon to talk about the specific price point. But what I can tell you is that we plan on launching zuranolone should the studies be positive and we receive a positive approval, but within the context of a value-based agreement. And based upon the data to date, our view is that if a physician believes that a patient -- an MDD patient or PPD patient can benefit by zuranolone, then we want to make sure that prescription is filled right away without step-throughs, without significant prior auth hurdle. So we'll work with payers in the context of a proactive value-based agreement to make sure that that's our ask. And in return, we'll offer certain protections to payers, things like numbers of doses, numbers of patients, all those details that will be worked out in discussions with payers. But early conversations we've had indicates tremendous enthusiasm to partner with us to benefit these patients in need.

That's great. Good. And is there a certain magnitude of benefit that you think would be sort of the minimum that would resonate for both the KOLs you speak with, the prescribers as well as the payers that would facilitate this sort of access and uptake?

So there aren't specific numbers. And I emphasized this earlier, and I'll say it again. What's important here is that we have a statistically significant readout at day 15 relative to placebo. That, in and of itself, given the benefit-risk of zuranolone, is enough to convince the regulators, I believe, payers and health care providers that this is an important medicine. And if you step back, think about someone who's suffering from major depressive disorder or PPD. To give them an antidepressant that takes 6 to 8 weeks and works in about 1/3 of patients is very different than a drug that has proven to be effective on day 3 in a vast majority of patients. So that fast onset of action, returning people to their normative state, is a different benefit with a very clean safety profile that's just not been seen in the depressive landscape before.

Got it. Maybe one more question on zuranolone before we shift gears to the rest of the pipeline. What kind of a label or a label claim would you expect from the totality of the Phase III program for the drug, assuming, of course, that the studies read out positively? And I guess specific to the major depressive disorder indication, is your view based on your regulatory discussions that 2 weeks of dosing would be sufficient to have a label claim for the treatment of major depressive disorder? Or would you expect sort of a modified or a label specific to acute use and short duration episodic treatment?

So it's too early to put my words in the regulator's mouth, but we're certainly hoping for treatment claims for both MDD and PPD. But I'm confident we'll work with regulators to come up with an indication statement that allows medical policies in the payer side to be sufficient so if the physician writes for zuranolone either for 2 weeks or another 2 weeks or even another 2 weeks, which is a minority of patients, those scripts will be filled to benefit those patients.

Got it. Maybe shifting gears to 324. Some of the early work in essential tremor that you guys have done had shown a correlation between drug levels and tremor reductions. And the mechanism is also known for exposure correlating with adverse events. So how do you guys think about managing the therapeutic window in upcoming studies? Are there ways you can optimize this through dosing, frequency, titration, formulation, et cetera? What are you learning? And what's sort of an acceptable AE profile in this population?

Yes. So Brian, if we think about 324, it's a [ GABAergic ] PAM that's specifically designed for chronic treatment for movement disorders, starting with essential tremor. So we designed this drug specifically for chronic treatment with essential tremor. What we saw pre-kinetic was that with a single dose of SAGE-324, we saw a dramatic reduction in essential tremor amplitude that was sustained. What we did with KINETIC, and here's what we were looking for, we were looking for a 30% to 50% reduction in tremor amplitude that was sustained over the entire 28-day treatment window -- treatment period, without any surprise on adverse events. We got that out of KINETIC and more. We saw a statistically significant reduction in tremor amplitude at every single time point. And when we look at the more severe patients, those over 12, we saw a 41% reduction in tremor amplitude sustained over the period. And importantly, and this should not be overlooked, the change in tremor amplitude was statistically significantly correlated with activities of daily living. That's not necessarily something that is always seen with these kinds of medicines. So we were thrilled to see that we saw a statistically significant correlation between ADL and tremor amplitude. Why is that important? That's important because as we work with regulators on the Phase III, should they want functional endpoints like activities of daily living in the primary endpoints, we now know that we have that statistically significant effect as well. So we can design a trial to do so. Now what we're looking at right now in the next Phase II with our partners, Biogen, is a dose-ranging study. So what's the dose and frequency that confers benefit but allows patients to stay on the drug. So we're looking for an adverse event profile that has a very good adherence rate. And that's what we're studying right now. I'm very confident, based upon all the data out of KINETIC, that we're going to find a dose and frequency that allows chronic treatment with a very high adherence rate.

Great. And then you mentioned the 28-day endpoint for KINETIC. What does the 324 data tell you about the durability of effect for this mechanism overall to the lack of tachyphylaxis? What does that mean for the potential even beyond 324 for some of your other programs, zuranolone or other GABA modulators in your pipeline, to potentially be dosed more chronically? Is that something you might explore?

Absolutely. So clearly, starting with SAGE-324, this will be a chronic treatment. So now that we see effect size out to 28 days without tachyphylaxis, that's a very good read-through for chronic dose and chronic dosing for long periods of time. And that now gives us confidence with other programs that focus on GABAA that we can, in fact, work on diseases of chronic treatment. So it's a very good read-through for not only 324, but the rest of the pipeline.

Got it. And then just in the last 5 minutes, I would love to touch upon 718. You've shown some intriguing data on executive function improvements without amphetamine-like attention effects from several studies. I guess how interpretable are the open-label studies and the -- I guess how validated are the scales that you're looking at in Parkinson's versus Huntington's? And how confident are you that you can mitigate theoretical mechanism-based toxicity concerns?

Yes. So look, I'm incredibly excited by SAGE-718 and have been since I took over as CEO in mid-December, and made those views pretty clear. And the reason is -- I mean I like what's going on with zuranolone and 324 as well. What we have with 718, though, is we have a really important biologic insight that the Sage team figured out. And that was that in patients with neurogenic diseases, specifically Huntington's, there's a loss of 24-Hydroxycholesterol. And that specific receptor loss was identified by the Sage team. And then innovatively, we have SAGE-718 that kind of upregulates the pathway, the NMDA pathway. So biologically, that was the hypothesis. What's exciting, Brian, is every single study we've done has replicated or recapitulated that biologic hypothesis. So all the preclinical data, the early human data, the challenge studies we did, the Huntington's studies and now recently, the Parkinson's study, showed basically effects in executive function, learning and memory, exactly the biologic hypothesis that SAGE-718 was founded on. So I'm excited that we've declared a path for Huntington's disease. We have a really good regulatory path. I'm confident in the endpoints we're studying. And now with PARADIGM also being positive in Parkinson's, again, working exactly where we thought it would work and not working where we suspected it wouldn't, without any deleterious effects, we continue to confirm the biologic hypothesis. So SAGE-718 is moving down the track for Huntington's disease for sure, and we'll be clear about what other next steps relative to Parkinson's and Alzheimer's are in our future.

Got it. And I guess kind of on that, just anything you could expand upon with respect to what might be the path forward there for 718?

It's too early to say what other paths forward right now. But we -- right now, we have a profile of a drug with extreme benefit and a very benign safety profile. So we're excited. And we have a dose to take forward Huntington's. I believe it's a dose that we'll also take forward in Parkinson's and potentially, Alzheimer's. Of course, we're waiting for those data later this year.

Got it. And just in the last minute or 2, anything else from the earlier-stage pipeline that you'd like to highlight or are most excited about?

Well, I think, Brian, what I'm excited about with Sage is how deep -- the deep understanding the team has in both GABA and NMDA. And the methodology that Sage has used, it led to a successful approval of ZULRESSO, positive data for zuranolone, positive data for 324 and now 718. So what I'm really confident of is that we really have a product engine. And as you know, we've guided that we're looking to produce 2 new INDs per year starting in 2023. It's based upon that deep understanding of biology and the conversion using neuroactive steroids in oxysterol chemistry that gives me the confidence that we truly have a high probability of success pipeline behind it.

Great. Well, Barry, this has been great. Really appreciate you coming on and being a part of the conference in what I know is a very busy time. So we look forward to the upcoming data readouts, best of luck, and thanks for joining us again.

Thanks, Brian. Appreciate it. Do well.

You too.