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

All right. Thanks, everyone, for joining us for our last session of day 1 of the SVB Leerink CNS Forum. We are very thankful and lucky to have Sage Therapeutics' management team here. The topic is addressing key areas of debate regarding all data. And we have Barry Greene, CEO; and Steve Kanes, the CMO. And so the goal here is to kind of do a little bit of a fireside chat, Q&A and hit some of the key issues. And so please if we're not hitting the issues that you want to hit, please send us an email, let me know, and I'll try to get the questions asked.

So I'll just kick it right off and just jump right into it, guys, and just say, look, number one, a key criticism of the data is that you over the powered the study such that the study with stat-sig, but not clinically meaningful. And what are your thoughts about that criticism? And obviously, you've heard, this is not the first time.

Yes. And Marc, thanks for having us, and I appreciate Leerink for hosting us. Hopefully, we can keep this entertaining for the end of the day for everybody. So obviously, we've heard the criticism out there, and I'll make a couple of comments. And as someone who's treated patients like this, we'll ask Steve to dive in. But what we're seeing consistently with zuranolone is the drug performance -- performs the same over and over and over again. We've tested it in over 3,500 patients. And what we see each and every time is a rapid response, statistically significant responses when we've studied it at day 3. What we've seen is consistency of response. And in WATERFALL, in particular, there really was no difference statistically between our day 15 response and our day 42 response. In fact, it was 86% numerically, but those numbers are not any different. And then we're seeing a very tolerated profile. And when you put that in the context of a drug, it gets patients better faster in just 2 weeks. I'd say it's a pretty remarkable profile overall. Now what's important isn't necessarily a difference from placebo because we proved that statistically, significantly. What's important here is how quickly we get better -- patients better and how long they stay better. So in each and every study we've done with zuranolone, these are the Phase III studies, we've seen a drop between 12 and 18 points using round numbers. And so the drop of 14.1 points is absolutely clinically meaningful because patients are getting better. And that's really what, at the end of the day, what we're trying to do here. Steve, do you want to comment further?

Yes, sure. Many of you guys know, I'm a psychiatrist. I can tell you, I just used to see patients like this all the time, a rapid reduction in depressive symptoms is absolutely what patients are looking for. And it's what psychiatrists and physicians more generally need in order to ensure their patients feel well. And I would say despite in this particular trial, high placebo response rates, we continue to see a statistically significant reduction in HAM-D scores, day 3, day 8, day 12, day 15. And perhaps most importantly, 85% of that response is maintained out through day 42. So what does that mean? Like if you're a patient, you take medication, you get better and then you stay well. And we've looked much, much longer than that. We've been following patients for a year, and we've seen very prolonged benefit in patients. This is the data from SHORELINE where patients remain well out through -- 70% of the patients only require 1 or 2 treatments over the course of a year. So that's a very differentiated profile. I think that's something that's going to be a really important tool for physicians and patients when we're approved.

And Marc, I shared this as early as January when we sort of kicked off the year and gave our goals and objectives, we're helping the healthcare community reimagine the treatment of depression. Other people saw -- call it as a paradigm shift. So when you reach out to people who've been thinking about depression for 30 years in a certain way, and we're presenting a very different way to treat depression. It's not surprising that the KOLs that you might interview might get some of it but won't get all of it until we have an opportunity with the totality of the data to educate. And I guarantee you, if you start interviewing patients and ask if they want to get better in 2 or 3 days and take a 2-week course of treatment rather than something they take chronically, which gives them sexual dysfunction, weight gain and other issues, which we don't see, you know which drug they'd pick.

Yes. Well, it's interesting because part of, I guess, the question really is, hey, think about all the drugs that are on the market today. Don't they have failed studies. I mean, forget about clinically meaningful stat-sig, all that discussion. There's so many failed studies that you would say to yourself, wow, I mean, does that drug even work? But of course, they work. They've been on the market for years. They're massively used, they're -- I mean, obviously, in volumes. So I guess the real question is, does anybody even evaluate the data to this level come when the drug is on the market, like, I guess, is the question.

But that's exactly right. We believe the totality of the data we have, we know the drug works, we believe the totality of the drug. Today, the data we have today suggests a drug that should get approved. That's our belief set. And as you said, what a healthcare provider will care about is your patient gets better and stays better.

So there's some confusion that this study, plus the Phase II 201 study is enough to file. So I guess, maybe you could just make a comment on that.

Yes. So what we described, again, going back to as early as January is that we, in conjunction with the agency, sat down and based upon the data we had to date, to find 3 additional unique opportunities, any one of which if positive, on top of the positive data we already had, presented a fileable package. So it was WATERFALL, CORAL and SKYLARK, 2 in MDD and 1 in PPD, to get an indication in MDD and PPD. So our belief set today, based upon the discussions we've had with the agency is that we have a fileable package. The question that remains is the ongoing pharmacological studies and then we've got CORAL out there and SKYLARK out there. So as we've said, our intention, and we have breakthrough designation, our intention is sit down at the agency, talk about the totality of the data and get alignment on filing strategy so that we can be clear with everybody what we're going to file on and how we're going to file on. But to be clear, from an efficacy perspective, the data we have today, we believe is a fileable package. And we could...

You don't need anything else as far as you're concerned. Like you're saying you're working on thinking about the full package. But at this point, you've got 2 positive studies. You've checked the boxes as far as you're concerned, you can file.

We actually have 3 out of 4 positive studies. So yes.

Right. And you've said consistently that you don't need CORAL data to file. On the conference call a couple of weeks ago, it sounded like maybe the message changed a little bit. You kept bringing the study back up on the call if you expected the FDA to need the data, maybe that was just our confusion, other people's confusion, I'm not sure, but maybe you could just comment on what you meant there.

Yes. And hopefully, Marc, the message didn't change, but if you were confused, then we need to be more articulate about it. So what we've said, out of these 3 different paths, any one of which were positive, we have a fileable package. However, if we start filing and enter a clinical package and CORAL starts to read out and the agency believes that CORAL might be a major review, then what we want to do is gain alignment on the agency of what needs to be in the package, so we can go through the review once. What we don't want to do is file, have a PDUFA date, have a major review out there and have a 3-month extension to that filing date, having another clinical review. In our experience, it'd be more efficient and effective to file a clinical package altogether based upon a discussion with the agency about what needs to be in. So CORAL, from a safety perspective, might be helpful, but we don't need another positive study in our opinion, to file.

And even from a safety perspective, I know like there are many companies, I remember, over the years that have told me, we don't even want to start another study because the FDA might actually ask for that study. So we want to get the approval first, then we'll follow-up with these -- you're different. You've got all these studies going on. So is it -- what you're saying is there could be a case where they say, look, I mean, why not, right? CORAL's reading out so soon, the FDA might actually ask for some more data, whether it's safety, efficacy or both. That's kind of what you're saying, they might change their mind, so to speak.

I wouldn't label it an agency mine change per se, but we identified paths. It's simply a -- let's back up. What you wouldn't want, if you were -- anybody is -- you're in the middle of review, a whole new data set comes out and then you're suggested that you didn't review a whole new data set, which could have been helpful. Unfortunately, as you know, it's highly regulated. There's specific interactions. So that's the conversation that we have to have in the right formats with them so that we are completely aligned on what the filing package looks like.

Right. Right, right, right. Okay. And one of the other questions that I -- even the KOLs that we've spoken to have mentioned is they want to make sure that they see the granular data. They want to make sure that zuranolone actually is treating the core symptoms of depression. The drug is not just making people go to sleep, so they feel better and that kind of thing. And so maybe what you can do -- I mean, I know you're not going to give us all the data right now, I wouldn't expect that. But can you give us a sense of don't worry, those KOLS -- we'll see the individual data, and they'll understand that this actually is helping depression. What can you give us today that would help us feel better?

Well, the first thing I'll say, Marc, is based upon only the data we presented to date, SHORELINE data, the previous studies and WATERFALL. On only those data, as we've been able to do webinars and scientific exchanges to educate healthcare providers on the totality of the data, we're seeing mind shift dramatically, from not understanding the 2-week course, not understanding the impact, how rapidly it works to actually understanding. And I know in the KOL interviews you did, the only criticism I could see was misconception about 12 versus 14.1. And I found it interesting that on another KOL call, one of the KOLs said, it would have been better to have 10 versus 6. Well, 14 is better than 10 from a [indiscernible] perspective. So it's not about the difference, it's about the magnitude of effect. So even on no new data, when we communicated the totality of the data we have, the stuff that was in our press release and our conference calls, what physicians are understanding is I've got a depressed patient by day three, their mood, their anxiety levels and their dysfunction of sleep are all improving already. They've never seen anything like that with a drug that then -- that lasts for an extended period of time with a safety profile that zuranolone has. So even with the data today as we're educating, we're starting to help people understand the data. And of course, in scientific meetings, medical meetings coming up, the totality of the data will be clear. And like we've done before, the idea would be to publish all of the data.

And when that totality of the data comes out, we'll see the individual components of MADRS and HAM-D. And everyone will say, aha, okay, it's working where it needs to be. The key components are working.

Well, Marc, what I'm saying is those ahas are starting now, even with just the data we presented. And yes, the individual pieces of data will be out there. But even today, those ahas are occurring.

Okay. Help us with the indication that we should expect on the label? Given what we've got right now. I mean, if you went to file with what we had today, what's the label?

Well, it's too early to speculate on the exact data in the label and the exact indication statement, but maybe Steve can provide some color on how we're thinking about it.

Yes. And this is a program that's designed to be treatment for MDD and for postpartum depression, full stop. So -- and we intend this for the widest of patients and uses that we can imagine. So we've been looking at patients who are naive to antidepression treatment. We've been looking at patients that are added on to their underlying treatment, and in the CORAL study, the coinitiation without antidepressants. Based on the overall benefit and risk profile, we think this is a medicine that should be available to all patients, and we're looking for the broadest label possible. That's why we have breakthrough in the first place. We're not looking at TRD patients. We're looking at general patients with major depression. And we think everybody should be able to get better quickly. And we're hoping for a very broad label. So as Barry said, too early to speculate on the exact wording, but we're looking for a general label for the treatment of major depressive disorder and PPD.

And I guess, how do you expect this -- given what we know today, right, given what we have. I mean, let's just say that the other studies are not positive. They don't help you. You've got what you've got, like, how do you think this drug is used in the real world with respect to first line, second line -- that kind of thing, what do you think it replaces, it moves in front of? Give us a sense of what you think?

Yes. Steve, do you want to start? And then I'll fill it out.

Yes. I mean, we view this is -- this should be the treatment of choice for people with major depression. I mean, think of it this way, who doesn't want to get better in day 3? There's no other field of medicine where that's something that's even in question. And the drug was designed to be that way. So we're looking to both ensure that we have the information necessary to give people clarity on how to do that. And we're also working very carefully with payers to demonstrate value and make sure that all patients for whom this is appropriate have access. And Barry, I'll turn it over to you to think -- to say a little bit more about how we're thinking about that strategically as well.

Yes. So Marc, we will have data for healthcare providers and patients to have choices. If someone wants to use it in conjunction with an established antidepressant, they can, if they want to add an antidepressant, they can, if they want to remove, well, we should have data that lets physicians understand all of that. And then we're going to make sure in a proactive way to make sure that we're working with payers so that if a prescription comes in, the patient who needs that drug gets that drug without having to fail therapy. And as Steve said, it's -- there's a paradigm shift here. The idea that you must fail therapies to get better is strange in this area, and that's something that we're going to have to change over time, and we will.

Well, I guess the question is, as you're negotiating with the payers, I mean, don't you expect them to say, okay, fine, you can be a priority drug, preferred drug, whatever you want to call, but the patient has to fail one SSRI or SNRI first or 2? I mean, isn't that kind of just expected as you're kind of going into this?

No, that's not what we expect. We expect to partner with payers in a proactive value-based agreement framework so that if a physician believes that patient needs the drug, even a naive patient, they can get this drug. And again, this concept, there's a paradigm shift here. This concept that it's a satisfied market is fine for someone to say from an academic institution, if you turn to what the patients want, they're not satisfied. They don't like the drugs they're on. They don't like to wait 6 to 8 weeks to work to see if the drug works, if ever. And there's a number of patients that are in crisis mode pretty quickly, moms with new babies. You want people to get better faster and stay better.

Yes. I would just think what if that study with the rapid response ends up working, right, with respect to throwing it on top of SSRI adjunctive therapy for the 14 days, I mean that would change everything, right? I mean if that study ends up being a positive study. Because then I can see what you're saying, the payers would have a hard time kind of pushing back, right? I mean it's -- but now it might be hard. I mean the -- I don't know, I guess I would see them pushing back some.

Well, about 1/3 of the patients we've studied to date, and this is true also with WATERFALL, we're already on an antidepressive. So they were already on an underlying antidepressive and those patients performed well, just like the more naive patients or the patients that weren't on antidepressant formula. So we have data to date. It says whether you're naive, whether you failed or whether you're on antidepressant, the drug works. It works the same way. You're better by day 3, you stay better out to day 42. And if you look at the SHORELINE data, and Steve referenced this already, again, about 1/3 of the patients, only 70% -- 70% of the patients only need 1 or 2 two-week doses in the course of the year. If you step back, would you rather take a drug every day that reminds you of your depression, that gives you weight gain, incontinence, potential cardiac issues or something like zuranolone that you take for 2 weeks. I think the choice is pretty clear.

I've got more questions, but I've got a couple of questions that have come in from the audience. I want to make sure I ask these and not get lost. So the first one is, why do you think the placebo response was so strong. The drug response was definitely good, but it seems odd that the placebo response is stronger than what you normally see in an MDD study.

Yes. Steve, do you want to do some placebo...

Yes. I mean, placebo is always a challenge in MDD trials. I mean, you referenced it earlier, Marc. Just useful for the folks on the call to hear a few of the facts. For Prozac, and we all know Prozac, we literally needed 11 trials to get 7 that were positive. Of these, maybe 2 that are positive to file. And so the idea of placebo varying dramatically in the course of programs is well recognized. We know the longer you go, especially for drugs like ours that are known to -- believed to be working, there's an expectation that patients are going to get better. And so that certainly plays into it. And recall, this is a study that was completed during a global health pandemic. People were home, isolated. The rates of depression were going way up. And for those who are participating in trials, they're coming back and forth to offices and having social interactions that many weren't having. Regardless of placebo response, and we talked about this earlier, the drug continues to perform, and that's what we look for. The lived experience for somebody who's taking new medicine is I'll get better quickly, I'll remain well, I will not have -- we've said it before, but it's important to recognize, patients did not have immediate bounce back of symptoms on the day after they stopped taking the medication. They have withdrawal symptoms, which you would see even with SSRIs after months of treatment. And we didn't see any additional ongoing sort of safety or health concerns after the treatments are stopped. So that gives us the information that we need to think about absolutely changing the game for patients. And that will be very appealing to folks. And it may not -- people who have treatment-resistant depression or people who have symptoms that come back right away. But for most people with depression, they'll have one in the context of a life-changing event like a pandemic or the birth of a child or loss of a job or a loved one. And they may not necessarily, we all know people, we don't necessarily have another major depressive episode until another major life stressor comes along. So again that we can sign everybody to take antidepressants chronically, which is the prior -- I used to teach psychopharmacology and teach the medical students and residents. The dogma is 3 for 6 months, maybe stop at least for a year. And if person has another episode get in their lifetime, treat them chronically, period. That's the dogma, and we're looking to change the game for that, and the science is pointing us directly in that way. So you say, when the folks who really dig into the data, KOLs, who really need to understand, they start to see this, and we talk to them, you really start to get it. And it's like the light bulbs go on one by one, and we're looking to just get that story out further and further. But we're really proud of the profile, we placebo 3 out of 4 trials, and the drug performs exactly the same. That's what we're looking for in this study.

Yes, Marc, that despite the kind of best placebo response any of us have seen in a depression trial to date, it's at least a point better than the last. Despite that, the drug outperformed placebo and performed consistently across 3,500 patients. So we clearly know the drug works, we clearly know that what we're seeing is not random events. It's not a placebo effect. It's a drug effect.

Steve, what is -- Steve, what's normal for a placebo response in a depression study?

It is -- there is no normal. It balances around. So one of the reasons why this sort of delta from placebo question, which I think comes from different traditions, oncology and so forth, it has so much sailings. We see anywhere from -- oh, I don't know, as we saw in this trial, this was higher about 12 points. I see, 7, 10 points depending upon the study. You look back to our paper in the New England Journal, which is our first MDD trial, you look at the data that we had from the ROBIN study, which was the prior PPD trial. And it tracks in terms of pattern with the drug effect that likely due to expectation bias. But the kinds of things that we're seeing are not outside of the realm of what's seen in trials. But what we look for routinely, and what we plan for in our studies is, make sure you beat the placebo, make sure that you have a meaningful change from baseline. Patients recognize that they've gotten better and that there isn't a bounce back, and that's what we showed.

Yes. Yes. Okay. Let me ask another question that was sent in. Can you outline the series of events that will play out over the course of the next few months as it relates to FDA feedback on the file-ability on WATERFALL and Phase II data versus requiring [ poll ]?

So I guess, at this point, what we can say is -- and Steve mentioned this, we have breakthrough designation, which gives us better access to the agency for a series of discussions. So what we need to do is sit down with them, go through the totality data and align on the filing strategy. And when we've done that, and we, in Biogen, could confidently say we've aligned with the agency, here it is, we'll come out and share it.

Yes. Another question. Let's talk about the day 42 effect for a second. I know going into this, there was a lot of discussion about trying to set up expectations for people to realize that, look, investment community, we may not hit stat-sig, we're not even expecting to hit stat-sig. We don't need to hit stat-sig. We'd like to just show some type of duration of effect that plays into our paradigm-shifting medication selling yet. So in the spirit, I guess, of that, because I do think that you guys did the best you could to set up expectations. I mean, we heard you many times over at a lot of different program and conferences. What is your view of the maintenance of effect out there at day 42? How do you explain the fact that you're losing a little bit of effect? Do you feel like you're in the range of what you expected? Just give us that sense of how you're thinking about that right now?

Yes. Just -- so what we've said very clearly, again, goes back to the beginning of the year is, we'd hit stat-sig at the primary endpoint in any 1 of these 3 studies to have the final Phase III to get the drug approved. That's where we stand today. You're right, we tried to set expectations. This is a drug with a completely different benefit risk profile than anything on the market or anything in development. So based upon the differential benefit risk, having now the third positive Phase III study and all the other data are, again, a pack that we believe is fileable. In terms of the day 42 effect on this study, as I mentioned, numerically, it was 86% of the day 15. It was exactly the same as the day 3. And if you look at the 2 numbers, they actually weren't statistically significantly different in terms of the day. So those are the numbers. The more important part is all the other data, Marc, and we mentioned this a couple of times already, when you look at the SHORELINE data, the fact that 70% of patients only in 1 or 2 two-week treatment in the course of the year. And the patients know they're not on drug. So it's basically a reverse placebo effect. They know they're not on drug and they stay better not asking for drug again. The totality of data suggests that we have a drug that for most patients is highly durable well after they're off the drug.

Yes. I think one of the confusing things to some investors were that the patients kept taking the drug. Or do they stop taking the drug, right, from day 15 out to day 42? But the reason that it's so interesting is they stop taking the drug. They only took for 2 weeks. So that was kind of something that wasn't well understood, would you say? I mean, I'm trying to gauge like the conversations you've had over the past couple of weeks, and what you've been surprised about, what you haven't been surprised that, I suppose.

Well, look, we -- the study was designed for 2-week course of treatment as all zuranolone has been taken across those over 3,000 patient subjects we've talked about. So what I keep pointing people to is the consistency of how the drug works time and time and time again. The drug -- it's a drug effect. It's not a placebo, drug works better. It works the same. People are better, fast, and they stay well longer. On the SHORELINE data, and we'll have more data later this year on the full 50-milligram, continue to reinforce that people get better fast and stay better longer after only 2-week course of treatment.

So how does this -- let's just say that we file, we've got the SHORELINE. We put that in there. Let's say, all the studies are positive, just for argument's sake. Like what -- everything is positive. So now you've kind of gotten to the end of the year, you've got everything, you're sitting there and you got the best scenario possible for your program. How do you envision what the label would say, what would be on it for the SHORELINE data, and what the indication would actually be for? Help us understand that.

Right. So I think Steve talked about this earlier on this call already. Based upon the data to date in hand, we believe that we have an opportunity for both MDD and PPD. Of course, I can't speculate on the exact label. That's part of the whole FDA negotiation. But based on data at hand, we see positive effects in MDD and PPD and believe that's where we should go. Now how all the data get put -- data are compiled in the clinical section. And we'll work with the agency to make sure that it's clear.

But the SHORELINE would be in the clinical section, right? I mean, teaching people about this episodic type of usage. That's where it's going to be, right? There's not going to be an indication.

If we get it right, we've got a drug to treat MDD and PPD without lots of constraints. So that's sort of what we're looking at and going for.

If the other PPD study fails, can you still file a PPD?

So what we agreed with the agency last year is that we had -- we would design 3 different studies, 2 MDD and 1 PPD, any one of which, if positive gave us a fileable package. I think Steve already said this, if you use the analog of Prozac, at 7 failed Phase IIIs out of 11 Phase III trials.

Right. But I've gotten this question, so there's no confusion. You could file for PPD and MDD right now, and your expectation is unless the FDA changes their mind, that's what you're going to, get an indication for both.

Correct. That's correct.

Right. So then that kind of makes the PPD study that we're waiting on, not as important, clearly, at this point, right? And so I guess it would be from the standpoint of just like you were talking about CORAL, maybe from extra safety data that the FDA might want or something like that, is that how you view it now?

Yes, we're really talk about strategically a separate strategy from tax. Strategically, we have the data, the clinical study data in hand for a fileable package, strategically. We believe that. There's other little studies that go on. The question now, the agency is, what else they want in review because we do have other studies up and running, and we've got to sit down and make sure we are aligned with them -- we don't want to speak for the agency until we sit down and get aligned with the agency.

Let me kind of throw this back at you and just say, what else are you hearing out there that you think is a little bit misunderstood or not well appreciated, and you kind of want to address it here?

Well, Marc, you've asked some of it. So what's misunderstood is what's important is the magnitude of effect, not whether it's 1.7 or 2.0 difference from placebo. And again, I told you, I was listening to a call where a KOL, she said, 10 versus 7 would have been better, that's wrong. 14 is better than 10. So that's just wrong. So the magnitude effect is what accounts. We saw a stat-sig at date at each of the time points, including day 15, the day 42 powering to see statistical significance would have been thousands and thousands of patients and think about it, patients stayed better after being off drug for 4 weeks. So we are seeing a benefit-risk profile that's unlike others. And the confusion, and Steve said this, it's a shift in how everybody has been taught to treat patients. And when you're shifting everyone's been taught, it's going to take data and education some time. But I'm confident we're going to get there because patients deserve better.

In the discussions with FDA, has there been any view by the FDA, they might actually give it some type of indication for rapid onset, where there is that type of commentary in the indication?

It's too early to talk about that. I'm not sure of anything but treating MDD and PPD from an indication perspective is important. Clearly, all the data will get into the label, and then we all work with what we can promote, what has to be scientific exchange. But that -- we got some time to figure that out.

Yes. I just wasn't sure I know at one time, there was a consideration that maybe there's a rapid onset type of indication or whatever. That's why I was wondering.

Yes, Marc, I would put it in the perspective this way. There's essentially no evidence that standard antidepressants even have an effect at 2 weeks. And zuranolone, we're talking about the complete treatment course with these large effect sizes being encompassed entirely with 14 days of therapy. So as -- I think Barry said it, we're talking about a label for the treatment of MDD. But the entire approach is really one that's unique, really important. It's one that I know is going to have a huge impact for patients, a very welcome tool for physicians.

Can we shift the conversation to an area that you might not want to talk about a little bit, which is pricing? But maybe you could talk about the commentary with this value-based deals, or how -- just maybe some optionality of what you might be talking about with the payers, what might -- I mean, I know you're not going to give us the exact cost right now. I mean, I'm not asking for that. But the thought process behind how we should be thinking that you're thinking about pricing and these deals? Because, obviously, treat -- a value-based deal for a cholesterol drug is very simple, but a value-based deal for an Alzheimer's drug or depression drug is obviously a lot different. So maybe you can talk about how we should be thinking about this. How you're thinking about this?

Yes. So again, we've said this, it's too early to talk about specificity of price. We'll price it based upon the value we believe that we're accruing to the patients of the healthcare system. The strategic concept between a proactive value-based agreement, proactive is key here, is that we signaled to payers that we want to partner, that we expect that if a physician writes for this drug, the patient gets the drug. And for that, we want to understand what concerns they have and offer certain guarantees around those concerns. So it's too early to say much more than that, but we will stand by how the drug works and support the drug works. And there are some examples of proactive value-based agreements, mostly in the orphan space, as you know, but those same concepts can apply to large markets.

And so there will be value-based type of deals. I mean, I guess I'm trying to think about this whole -- if you price the drug for people who are going to use it for, let's say, a month out of 12 months out of the year, right? I mean you've got to demonstrate that that's all they need in order to get a pretty high annualized price, right, for one month, I guess. I'm just trying to think through how these discussions are going, or how you're thinking about it.

Yes. Well, we will work -- we will present the totality of the data. And we -- again, we have SHORELINE that suggest how many 2-week courses the group needs, broken down by whom. And we'll have those to tell. We'll also have a significant amount of health economic and pharmacoeconomic data to sit down and talk to payers about. So there'll be a robust discussion to educate.

But the way you're thinking about it is that you can price it off of this 4-week treatment. And maybe you don't have to pay above the 4-week treatment if patients end up being treated just like in SHORELINE, right, where that's kind of where I'm getting it from.

So Marc, there's some concept that we'll price for value, and that value may include how many different 2-week courses the patient needs. We'll have data to support that. And there could be concepts that if the population in totality falls above that range because some patients who only need 2 two-week course. There'll be clinical rebates based on clinical experience. But that's obviously a better way to work with payers than generic rebates because they need rebates. And my experience with payers is that they want patients to get better. They just don't want to pay exorbitant cost when they're not guaranteed that if it doesn't work, they get some level of clinical rebates. So we'll approach it that way.

Yes. I mean, in the past -- and this is the last question on the pricing. But in the past, I know there's been a discussion of, okay, we're going to work backwards here, right? So the annualized cost of this therapy needs to be something that's not too expensive relative to what can be taken up there. So call it, $5,000 per year for a depression drug. And so if the patients take the drug for 14 days, divide $5,000 by the 14, we get cost per day or, if it's 28 days or it's 42 day, whatever it is. Is that the mentality of how you're thinking about the cost per pill? Is that fair?

I mean the logic you're applying is logic we'll apply, but there's a lot more that will go into the calculus.

Yes. And on the $5,000, that's not a number that you've ever referred to. And so I shouldn't be playing off of that $5,000 at all or any number?

Too early to talk about price.

I don't know why I'm thinking there was a $5,000 commentary in the past. And so...

Not that I'm aware of.

Yes, okay. So I mean, we don't have that much time left, but I just want to like -- what else? What else should we cover here? What else do you want to make sure? I know I asked you once already, but I want to give you another chance.

Right. I think -- so we're sort of in the short term, let's examine -- pseudo examine a Phase III readout. And to be clear, zuranolone is a very important program for both us and Biogen. And as Steve and I have commented, we're confident that we're going to have a drug in the market that has an opportunity to help millions and millions of patients suffering from depression. But the Sage story is a story of a leader in brain health and a story of becoming a top-tier biopharmaceutical company over the next 5 to 10 years because after zuranolone, we have 324 for essential tremor where we've demonstrated the drug works, and now we're getting ready to do a Phase II to demonstrate dosing frequency. We have SAGE-718, where Huntington's and Parkinson's, we've already demonstrated improvements in executive function and learning and memory. And we're going to have Alzheimer's data coming up soon. And we've declared that we have a path for Huntington's already. So I'm pretty excited by the profile. And on top of that, we've got mid- and early-stage programs and a product engine capable of 2 INDs a year, starting in 2023. So we've got to get through some of the confusion around zuranolone, but I'm very confident that we're building a top-tier biopharma company that's going to impact brain health disorders across a spectrum of diseases.

I think this was helpful. And it's kind of interesting. I think I said this at the beginning, but I want to repeat it, and that is, you don't hear the doctors talking about all the failed studies of Prozac. They just write a prescription for Prozac, right? So I just wonder if there's even a discussion about this whole, oh, was this efficacious? Or is that efficacious? I mean if the bottom line is you're going to be entering the market with a new mechanism and a rapid onset, and those 2 things alone are extremely positive and exciting, and the efficacy of the drug is clearly stat-sig. So you know what I mean, like I don't think the discussion will be there, I guess, when you hit the market. That's what I'm saying, like, we've got different KOLs saying different things right now, but I'm not even sure that's a discussion, come a year from now, 2 years from now, whatever, when you hit the market.

I agree. When people see what zuranolone does, and they recognize the differential benefit risk, we're pretty confident this is exactly what patients are going to want. And patients are going to want this and their healthcare providers can see their patients get better by day 3, it's the drug they'll reach for their depression patients.

Yes, yes. Okay. Well, I know you need to go. I really appreciate the 2 of you joining us.

Thanks, Marc.

And continue, good luck with everything, and we look forward to, obviously, plenty more data.

Awesome. Thanks for having us.

All right.