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

Good morning. Welcome to Sage Therapeutics conference call to discuss the results of the KINETIC study. [Operator Instructions] This call is being webcast live on the Investors and Media section of Sage's website at sagerx.com. This call is property of Sage Therapeutics, and recordings, reproduction and transmission of this call without expressed written consent of Sage Therapeutics is strictly prohibited. Please note, this call is being recorded. I would now like to introduce Jeff Boyle, Vice President, Investor Relations at Sage.

Good morning, and thank you for joining Sage Therapeutics conference call to discuss results from the KINETIC study with SAGE-324 in essential tremor. Before I begin, I encourage everyone to go to the Investors and Media section of our website at sagerx.com, where you can find the press release related to today's call. I'd like to point out that we will be making forward-looking statements, which are based on our current expectations and beliefs. These statements are subject to certain risks and uncertainties, and our actual results may differ materially. Please consult the risk factors discussed in today's press release and in our SEC filings for additional details. Joining me on the call today are our Chief Executive Officer, Barry Greene, who will provide a high-level overview of the results and discuss the path forward to SAGE-324; Steve Kanes, our Chief Medical Officer, who will review the data we're presenting today; and Jim Doherty, our Chief Research Officer, who will join for the Q&A portion of the call and can provide additional color on SAGE-324. So with that, I'll now turn the call over to Barry.

Thanks, Jeff, and thanks, everyone, for joining us this morning. I'm pleased to be with you today to review the positive and very exciting results from our Phase II KINETIC study in essential tremor. At the start, let me set some context for these results. This study was designed to understand use of SAGE-324 in reducing tremors over time versus the promising reduction we've previously seen with a single dose at a single time point. To optimize the likelihood of single-dose effects, we dosed at 60 milligrams, the high end of the dose range established in Phase I studies. We also wanted to understand the adverse event profile at the 60-milligram dose, specifically, when we see anything surprising when giving the daily morning dose over 28 days. And as you saw from the press release, we're very pleased with the answers we got from the study. Now more specifically, as we've been saying for some time, we were looking for a big effect, in the range of 30% to 50% sustained reduction in tremor amplitude from baseline and an adverse event profile consistent with previous SAGE-324 data in GABAA PAMs in general. In other words, we are looking for the high end of the dose range have the meaningful effect, with tolerability profile that provides support and insights to guide next steps in the development of SAGE-324. This was the high bar, and we believe we exceeded it. Steve will walk us through the data in greater detail, but let me share a few highlights. First of all, the study achieved its primary endpoint with SAGE-324 demonstrating a statistically significant reduction from baseline in the TETRAS Item 4 upper limb tremor score at day 29 in the total study population compared to placebo or the ITT analysis, which corresponds to a 36% reduction from baseline in upper limb tremor amplitude in patients receiving SAGE-324 versus the 21% reduction in patients receiving placebo. And the safety profile was generally consistent with previously reported data from SAGE-324. Other highlights from the study to point out. Patients with a more severe tremor baseline, those representing the moderate-to-severe patient population, demonstrated a statistically significant reduction from baseline in the TETRAS Item 4 upper limb tremor score at day 28, which corresponded to a 41% reduction in upper limb tremor amplitude compared to an 18% reduction for placebo. We believe patients with more severe tumor, that is TETRAS score of greater than 12, represent the majority of ET patients getting diagnosed and seeking treatment today. Importantly, the activities of daily living, or ADL scores, showed a statistically significant correlation with upper limb tremor scores at all time points. Now while not powered to fully examine TETRAS ADL, SAGE-324 was numerically superior to placebo at all time points during the study, demonstrating the clinically meaningful nature of these data and the importance to ET patients. So hitting statistical significance in the primary endpoint, achieving clinically meaningful reductions in tremor amplitude, seeing the ADL tremor correlation all with an AE profile that was in line with our expectations for the 60-milligram dose is encouraging and exciting outcome for this Phase II trial. These data reinforce our belief that the pharmacologic characteristics of SAGE-324 are well suited for development opportunities in essential tremor and possibly other indications. With these new data, we'll work with our collaboration partners at Biogen to optimize next steps for the continued development of SAGE-324 in essential tremor. Again, we need to align with Biogen on the specifics of the trial design, patient numbers, timing, et cetera. But as we've been saying for some time, the logical next step would be a Phase IIb study to explore dose frequency and possibly formulations. And as previously planned, we believe we'll be able to initiate the next step in development later this year. Now before I move on, I'd like to thank the participants of the KINETIC study, their families, trial investigators, everyone at the sites and certainly our Sage employees and collaborators who've made invaluable contributions to this program. With that, I'll turn the call over to Steve to discuss the results in more detail. Steve?

Thanks, Barry, and good morning, everyone. Let me take you through an overview of the KINETIC study design and then provide a little more color about the results from the study. KINETIC is a Phase II multicenter, randomized, double-blind, placebo-controlled study that evaluated the efficacy, safety and tolerability of SAGE-324 60 milligrams in 69 patients with essential tremor aged 18 to 80 years old. Patients were randomized 1:1 to receive either 60 milligrams of SAGE-324 or matched placebo once daily in the morning for 28 days, with a follow-up period of an additional 2 weeks. As Barry highlighted, the trial evaluated treatment of SAGE-324 at the high end of the dose range. And the daily dose could be titrated down to 45 or 30 milligrams if 60 was deemed to be not well tolerated. Study patients were not taking other medications for essential tremor during the treatment period. The primary endpoint in the study was changed from baseline compared to placebo on day 29 in upper limb tremor score as measured by Item 4 of the TETRAS performance subscale. TETRAS is a physician-administered scale designed to provide an accurate, comprehensive assessment of essential tremor motor symptoms and has been shown to correlate with TETRAS activities of daily living. We're very pleased that the study met its primary endpoint, a significant reduction in TETRAS Item 4 upper limb tremor score from baseline at day 29 in the prespecified full analysis set compared to placebo with a p-value of 0.49, which corresponds to a 36% reduction from baseline and upper limb tremor amplitude in patients receiving 324 versus a 21% reduction in patients receiving placebo. Turning to patients with more severe tremor at baseline, that is patients at or above the median TETRAS performance subscale upper limb tremor Item 4 score of 12 or higher, SAGE-324 demonstrated a significant reduction from baseline in the Item 4 upper limb score at day 29 compared to placebo with a p-value of 0.007. And that corresponds to a 41% reduction from baseline and tremor amplitude at day 29 compared to an 18% reduction for placebo. Of note, these patients compromise the majority of patients in the study and represent the most significant unmet need. Additionally, in the study, activities of daily living scores showed a statistically significant correlation with upper limb tremor scores at all time points. From a clinical point of view, this is an important consideration, and we're encouraged by these data. 62% of the patients who received SAGE-324 down titrated in dose as allowed by the study protocol, and discontinuations were noted in 38% of patients receiving SAGE-324. And again, we dosed at 60 milligrams, the high end of the dose range, to better understand the AE profile. And I'm pleased to report that SAGE-324 performed as we anticipated. Adverse events were generally consistent with the safety profile of SAGE-324 seen to date and with other GABAA PAMs. Treatment-emergent adverse events that occurred in 10% or more of patients in the 324 treatment group and at a rate at least twice as high as that of patients in the placebo group were somnolence, dizziness, balance disorder, diplopia, dysarthria and gait disturbances. These findings are in line with our expectations for the 60-milligram dose. And as Barry said, we believe our next steps will focus on optimizing dose and frequency for the ongoing development of SAGE-324 in essential tremor. I'm highly encouraged by the results we're sharing today. While still early, these data are a meaningful step towards our development of SAGE-324 for essential tremor, a brain health disorder that impacts millions of patients, represents an area of high unmet need, as we highlighted during our 2020 future cast, and importantly, has not had a new pharmacologic treatment approved by the FDA in more than 50 years. Lastly, I'd also like to thank everyone involved in the study, particularly the patients, their families, the trial investigators and Sage employees. As Barry said, your contributions are invaluable. I'll now turn the call back over to Barry for closing remarks.

Thanks, Steven, and thanks again, everyone. As a reminder, essential tremor affects an estimated 6.4 million people in U.S. and more than 50% of patients experienced suboptimal response with current standard of care treatments. And while essential tremor is often associated with aging populations, it can begin much earlier in life with a progressive disease course. And as tremor severity increases, the likely of anxiety and depressive symptoms also increase. We believe these data provide clear support insights for further development of SAGE-324 in an area of significant unmet medical need. People with brain health disorders have been conditioned to accept the status quo due to limited innovation or the lack of truly transformative medicines in recent year. And that's certainly been the case with essential tremor. However, at Sage, we believe that people suffering from brain health disorders deserve better, and we aim to help achieve that. Now I'll turn the call back to Jeff to handle Q&A with the operator. Jeff?

Thanks, Barry. Before I turn it back over to the operator, I'm going to ask that you limit yourself to 1 question. If you have an additional question, feel free to return to the queue. But we want to make sure we give everyone time to ask a question. So with that, I'm going to turn it over to the operator. Operator?

[Operator Instructions] Our first question comes from Ritu Baral with Cowen.

I want a little more color just on the side effect rates. Can you give any color on any Grade 2 or 3 somnolence and sedation. And when, during the course of the day, those were experienced? And when over the course of the treatment did those result in down titration or discontinuation?

Thanks, Ritu. We're not going to be able to provide a lot of color, but let me set some context here and then turn it to Steve. And I said it on the call, and I'll remind you, Ritu and others. What we are looking for is a 30% to 50% reduction in tremor amplitude over time. We saw that. And again, from an adverse event profile, we're looking for no surprises. And that's exactly what we got here. Steve, do you want to provide a little bit more context and maybe reference some of the rates we actually saw in the Phase I study?

Sure. Some of the things that you're asking about, Ritu, are ones that we'll be looking into as we dig into the PK and PD relationships. But as Barry said, we were expecting to see these kinds of adverse events in this trial. For example, in the Phase I study at 60 milligrams, all patients experienced somnolence. So we knew that we were going to see this. The particular details and so forth are ones we'll dig into. I think the important point here is that now we know that we have a drug which shows the effects, the effects didn't wear out. It didn't wear off or tachyphylaxis over the course -- the time of treatment, and it truly gives us direction on where to go next as we think about accessory development.

The next question comes from Yasmeen Rahimi with Piper Sandler.

Congrats on the data. Can you maybe comment on when -- what was the earliest time point where you saw a separation between treatment and placebo in the TETRAS upper limits performance score in both populations, the full analysis populations as well as the patients with the severe tumor? I know that the assessments were taken in day 1, 8, 15 and 22. So if you could just comment on the time to see separation would be helpful.

Yes. Let me turn that over to Steve. I will say -- I mean, we saw statistical significance at all time points. But Steve, maybe you could provide some more color.

Sure. I would just remind everybody of the goals of treatment ultimately for essential tremor. This is very important to recognize that these tremors are chronic for patients, and so the treatment itself ought to be chronic as well. So we focus, of course, on what is the benefit at the end of treatment. Of course, we saw separation, as Barry said, throughout the dosing period. We're reporting here on the prespecified primary, and we'll get into the particulars of all of the -- when you start to see it and so forth as we start to work through all of the data and present this at conferences. Suffice it to say that these are truly clinically meaningful results. And the idea that they were correlated even over a relatively brief dosing period, which was a month with changes in the ADL scales, is quite encouraging.

Our next question comes from Laura Chico with Wedbush.

I guess, congrats, it sounds like you're not seeing anything unexpected on the adverse event side. So I was wondering if I could ask maybe a more regulatory focused question. I'm wondering if you could discuss how the agency might be weighing the significance of improvements on functional endpoints versus straight tremor reductions and thinking about that in terms of what might be valid endpoints in the subsequent studies?

Yes. Thanks, Laura. Thanks for the interest, and I appreciate the congratulatory note as well. I'll make some comments and then maybe Steve and Jim might have some more. So obviously, it's too early to talk about specific Phase III endpoints. The next step as we highlighted is a dose-ranging Phase II study. We'll explore dose frequency possible formulations where we'll be looking for a profile that is more -- a profile to go into Phase III and kind of a commercial profile. That's not what we're looking for here. So I will say though that the fact that the changes to tremor amplitude were statistically significantly correlated with the change activities of daily living. And at every time point, activities of daily living were superior for drug versus placebo gives us tremendous flexibility as we negotiate with the agency. Because we have both now seen changes in tremor amplitude, and importantly, as Steve highlighted, changes in activities of daily living. So that gives us great optionality in those regulatory discussions. Steve, Jim, anything to add?

Yes. I'll just say at a high level, Laura, this is -- essential tremor is something we've been working in for quite some time. I mean we started early on with SAGE-547, which is now ZULRESSO and really start to look into what endpoints are most meaningful, which ones are associated with changes in activity of daily living, working with experts in the field, working with patient groups and so forth. And that's why we're focusing on upper limb tremor score. We think that's really important. And as we think through how to systematically work with a new drug and a new mechanism, it's important to keep in mind which questions we asked here and where -- and which ones we'll need to answer down the road. And maybe, Jim, you can talk about the approach that we've taken for development and how we're thinking about sorting through some of the questions related to endpoints and regulatory.

Yes. Happy to, Steve. Good morning, Laura. I think the point that Barry and Steve both have raised, I'd just like to reinforce. When you think about the fact that we're talking about a patient population where there hasn't been any innovation a long time, of course, we are looking at the profile of 324. And as you're hearing today, I think the key message is that we were able to see a continuation of the kinds of responses that we had seen originally with some other GABAA PAMs and then in the Phase I results with SAGE-324 that we've talked about previously. So the focus for this Phase IIa study was really in showing that effect in a placebo-controlled study, and that's maintained over a 28-day time period. You've heard those results. But that doesn't mean that we're also not thinking about how the program continues to build. And so there are a number of other endpoints that are included in the study with an eye for thinking about what ultimately the regulatory studies might look like. But it's really just too kind of early to speculate on how we will put those together. But I do think, as Barry has mentioned, it's really very important that we see good correlation between the primary endpoint and effects on activities of daily living. So it really gives us confidence that we're seeing meaningful effects for essential tremor patients. And moving forward, we'll work with the agency to figure out what the pivotal studies need to look like.

Our next question comes from Salveen Richter with Goldman Sachs.

Could you help us understand how you're thinking about the design of the upcoming Phase IIb study? Would you be looking to enroll only the severe patients -- the more severe patients? Maybe help us understand what formulation work you believe would be required here. And whether you're comfortable with the dose-finding work to date?

Yes, Salveen, let me do some highlights. And then again, we're not able to say that much because we're still working with Biogen. But let me provide some context and maybe Steve or Jim can provide a little bit more color. So what I've said kind of in our prepared remarks and we've been saying this since January is what we're looking for in the study, I won't repeat that. But if we're seeing that 30% to 50% reduction over time with no surprises on adverse events, then the next step would be exploring dose and frequency and possibly formulation. Based upon these results, it's not clear that formulation will be required at all. It's something that we've been saying from the beginning. So it'd be exploring doses in the morning, exploring doses in the evening. Before dose escalation, things like that, to get to that Phase III-worthy commercial profile. Steve or Jim, anything to add?

Yes. The only thing I'd say is this is the first step into understanding how best to articulate the role of SAGE-324 in the treatment of essential tremor. And answering these fundamental questions for ourselves was absolutely essential before we started to go down the road of how best to dial in the profile of the drug. And so as Barry said, this was designed -- this study was designed to give us any particular information about the effect and to ensure that they didn't tachyphylaxis. The patients continue to demonstrate benefit over the course of therapy. And the mechanism really held up in a placebo-controlled trial. Incidentally, of course, this is the first time TETRAS and the Item 4 has been used in a development program. So even understanding some fundamentals as the placebo response and which patient groups are the best ones to use, gives us a lot of thought for really working out the details. So your questions are the right ones. They're the ones that we're going to be sorting through as we think with Biogen about what the true next steps might be.

Our next question comes from Cory Kasimov with JPMorgan.

This is Turner on for Cory. So just one quick one from us. I'm just curious whether or not the severe population of greater than 12 on the TETRAS score was a prespecified subgroup for analysis?

I'll turn it to Jim on that question, but let me just add some context there, Turner, and thanks for the question. We do believe that the TETRAS greater than 12 kind of patients will be moderate-to-severe patients would be the majority of patients seeking treatment. So logically, having a bigger number to see a more dynamic range that corresponding to moderate-to-severe patients, all fits together logically. Jim, do you want to talk about the specific -- any prespecifications?

Yes, sure, Barry. And I think the answer there is that in a Phase II study, we prespecified a number of analyses based on demographics and entry criteria. And so yes, the intent was always to look at the effect of the various baseline scores. So again, as we think about moving forward, all of these factors go into design for pivotal studies where you're talking about true prespecified analyses. But yes, we do always specify looking at how the intensity of baseline scores affect outcomes.

Our next question comes from Paul Matteis with Stifel.

This is Alex on for Paul. Do you mind walking us through how you define the full analysis set here for the primary and how dropouts were accounted for? Just kind of trying to understand how -- whether or not this was really a completers analysis or not? And then a quick follow-up, do you mind letting us know whether or not there are any falls that happened in the trial and how you're thinking about that moving forward?

Yes. Let me kick that to -- well, let me say that this -- what you're seeing for the p-value was the intent-to-treat or full analysis set, not a responder analysis. And we saw no falls in the trial. But Steve, do you want to comment more?

Sure. And again, it goes down to what -- how you move forward into an early Phase II study. So many of the parameters that we are needing to establish are ones that we need to bootstrap ourselves. So yes, for the straightforward question whether there were falls, no, there weren't falls. For the full analysis set, this is the intent to treat, everybody who was exposed to drug gets included. And the statistical approach we take, which is standard, allows for understanding the -- how to handle missing data. The strength of this, quite frankly, is the fact that even with a dropout rate, what is at 30-some-odd percent, we still have a significant p-value here. And recall that we had never -- we weren't necessarily even looking for a statistically significant p-value in the intent to treat. That's terrific to see so early on in the program, and it allows us to look at some of the other subgroups with more precision. So this is exactly the kind of performance that we've expected from a study that's positive at this phase.

Our next question comes from Tazeen Ahmad with Bank of America.

Just wanted some points of clarification. Can you just give us a little bit of color on whether or not more patients, when they down titrated, did they go more to the 30 mg or to the 45 mg? And related to that, how did you calculate the discontinuation rate? So included in that, could there have been some patients are first down titrated before discontinuing?

Steve, do you want to provide some color and maybe Jim could provide us more of the -- as much as we can share based on top line?

Sure. I would just say that the discontinuation rate includes both of those groups. And Jim, I don't know if there's other details that you can share on how those were calculated.

Yes, not really at this time. I think in the strictest sense, discontinuation means, as I said, that someone exited the trial at a certain time. I think given that people have the ability to down titrate as we move forward, we'll kind of be able to walk everybody through what the numbers look like across the different possibilities. But I think that level of detail doesn't change the key findings of the study.

Okay. And can you give color on whether it was the 30 or 45 that most people went?

So people were able to down titrate to 45 and then to 30, -- some -- it depends on the subjects.

And the breakout of numbers is something that will come later with publication in the presentation.

Our next question comes from Brian Abrahams with RBC Capital Markets.

This is Leo on for Brian. I just wanted to follow-up a bit on those 45- and 30-milligram doses. I'm curious if you can directionally speak to the tremor reductions you saw at some of the lower doses? And potentially on the subject of down titrations, if you had most down titrations on placebo and what those may have looked like compared to the 60-milligram dose?

Yes. I guess, Leo, what -- and I'll ask Steve or Jim for other color, but we have presented the top line data kind of in its totality. So the -- that 30% to 50% reduction, the 36 to 41 represents all patients in the down titration. We're not breaking out specifics. And I think Steve said this earlier, the fact that we had down titration, the fact that we had a discontinuation rate knowing that 60 was at the high end of the dose, gives us tremendous confidence in what we saw. In fact, if we think about sort of dose and frequency, there's certainly an opportunity, particularly with those patients that are moderate to severe over 12, to see even greater tremor amplitude production than we saw in this data set if we develop a dose frequency that people are on longer. So I'm very encouraged by what we're seeing. Steve or Jim, anything to add?

Yes. Yes. The only thing I'd add to that is, obviously, there's a lot of information here that we're going to really be able to use to identify everything from the patient population to the appropriate dosing scheme. And some of it are really the kinds of questions that you're asking. So PK/PD relationships, lower exposures and so forth. And that's all to come. The news of the day is, in the most clean analysis, which is the prespecified primary, specifically significant reductions and performance as we expected. And that's the starting point for really optimizing the overall profile of the drug as we move forward with Biogen as our co-development partner.

Yes. Just to reiterate, I mean, we're really pleased with what we found here in addition to that results that we're describing this morning. Of course, there's a lot more information. And part of the purpose of the Phase II study is to generate that information. So as you know, we take a very close look at things like PK/PD relationships, and we'll continue to do that with the data that have come in here. But for me, at least, I think what's really exciting is that when you think back to the profile of SAGE-324, the properties of the molecule allow us to dose with a long half-life and for a chronic therapy that, coupled with, as Barry was describing earlier, that at 60 milligrams, we're really kind of at the top end of the range that we're interested in looking at. This is -- really gives us a lot of opportunity to think about how to do this flexibility moving forward. And those are the kind of things that we'll be looking at next, and that will deal with things like dose and frequency and all of that. So more to come, more work to do, but it's certainly very exciting at this point.

Our next question comes from Danielle Brill with Raymond James.

I was hoping to get a little more color on the dose responses as well. Very curious if there was a correlation between the patients who experienced somnolence and TETRAS score improvement?

Danielle, great question. I'm not sure how much more we can share, but I'll look to Steve, if there's any other color.

Again, I think that's what we'll be looking at. I mean, what we could say is when we talk about there being 68%, there's obviously patients that are, in terms of somnolence, there's actually patients who are in that group who both responded and didn't experience somnolence. And we'll be picking through all those details as we move forward. So these are exactly the kinds of things that are part of the deeper analysis and what we hope to be looking to do internally and then sharing as we work through it.

Our next question comes from Andrew Tsai with Jefferies.

So my understanding is Inderal, which is a standard of care, doesn't necessarily work in all patients, maybe 50% of them. So just curious if in this study, most patients for 324 did show some kind of response.

Andrew, great question. Steve, do you want to take that?

Actually, I would actually refer to Jim so he could talk to you about who the patients were and whether and how we thought through their use of other treatments.

Yes. So Andrew, absolutely, you're correct that for Inderal, not all patients responded, back off the 50% of patients don't respond. So earlier with this study, we've asked subjects to wash out of other medications. And as you can see, we have also looked at things like baseline severity. I think the -- when you see a 36% reduction and 41% reduction, that shows you that we're seeing response across a substantial number of patients. I think that we'll just have to see moving forward what that fraction is. I think it's unlikely that any therapy is going to work for every single subject. But I do think that what we're seeing, when you look at the significant effect in the full analysis set, is that we are seeing good response across a substantial population.

The next question comes from Doug Tsao with H.C. Wainwright.

Congrats on the data. Just at this point, do you see a correlation between patients who saw the biggest benefits in terms of reduction in tremor versus improvements in functional measures?

Yes, Doug, thank you for congratulatory note. I guess what I will comment, and I'll leave it to Jim maybe for other color, but there's a limit to what we can provide. So I'll go back. What we've said from the beginning, both Steve and I over and over is we're expecting a 30% to 50% change in essential tremor from baseline. We were not -- a pretty vocal about this. We were not necessarily expecting statistical significance versus placebo, so thrilled we got that. We're looking for an adverse event profile that didn't provide any surprises. So we definitely got that. And then I was really pleased that we had a statistically significant correlation between tremor amplitude or TETRAS upper limb 4 score and activities daily. That's not necessarily something we fully expected. So that was an added benefit that came out of the kind of opening the envelope, if you will, that was not in a range at the bases. Like I commented earlier, the fact that they're correlated the way they work, statistically significant, gives us really important regulatory flexibility moving forward. Jim, anything else to offer?

Yes. Just -- I agree that it's really very, very interesting and very encouraging that we're seeing a good correlation across response to the primary endpoint, the TETRAS scale as well as response to the ADL. Recall that activities of daily living, there are a number of elements in that scale. And what we'll have to look at as we look at the next set of trials is you would expect to see in a larger populations whether or not you could talk about statistical significance there. But it really is encouraging that the 2 were correlated.

Our next question comes from Vamil Divan with Mizuho Securities.

So maybe just on the safety side again. Can you just talk a little bit -- I don't know if you can give any more color in terms of what the actual rates were on the placebo arm, discontinued or in terms of somnolence and dizziness? Just to get a little better sense of the comparison there. And just what rates of somnolence and dizziness especially would you think are acceptable? I'm sure, obviously, at the end of the day, how much efficacy the drug is giving. But is there sort of a bar you think you need to come under for those sort of key measures, connect this an attractive product for patients?

Yes. Let me provide some context and then ask Steve to provide some more color. So as we think about the next step with SAGE-324, now that we have the answer to the question we asked here, which are very positive. Now the question is, what's the right dose and frequency that both provides the benefit we saw, this 30% to 50% reduction, maybe even greater. And is the adverse event profile or tolerability profile, where patients can take the drug chronically. So exploring things like evening doses where sleepiness is nice, it's actually a good thing for patients, not a bad thing. Maybe starting at lower doses and increasing doses rather than titrating down doses. So those are the kind of things we're going to be exploring. And clearly, what we want to come out of the next step with the profile of the drug that we believe people will take chronically for extended periods of time because that's where this drug needs to be used. And pretty confident, based upon these data, that we'll be able to do that in the next step of the trial. Steve, anymore on the profile here?

Yes, absolutely. I think what we said is actually the way we look at it, which is these are the most common adverse events that occurred more in the drug group and placebo and greater than 2x. So it gives you a sense of how that dials in. The other piece of it and I think it's a really important one is we don't just sit here and try to scale this out ourselves about what would be an appropriate profile. We talk to patients. We talk to experts in the field. We talk to advocacy groups. And now that the study is done, we'll actually be following up with patients who participated in the trials to understand what their perception of the profile is and what would make it sort of desirable. So that's the kind of thing that you do. And it comes down to benefit and risk. If we think about more severe patients, they may have more tolerability to adverse events. So as we dial in what our metrics might be, that's going to be part and parcel of the work that we do now that we have something substantial to talk with them about. So a lot more to come in that area. As we said, this gives us a baseline in which to really dial in an ideal profile. And that work starts first with real data that we have today.

Our next question comes from Tim Lugo with William Blair.

This is Lachlan on for Tim. And congrats on a target here. I was wondering if you could provide any more color on sort of when the discontinuation in this down titration occurred over the 28-day course of the trial.

All right. Yes. I guess that's -- those are data that will come later with later presentations. I think we provided as much as we can provide at this point. Thank you.

Our next question comes from Sumant Kulkarni with Canaccord.

We know these are data with early promise, and your primary endpoint was based on upper limb tremor amplitude. So do you have any details on how 324 impacts a specific variable such as grip stunt, for example, and is it fair to assume 324's mechanism helps with other locations of essential tremor in the trunk, head and voice as well?

Yes. Steve, do you want to take that?

Sure. So there's a few things. I mean, obviously, these are the data that we have today, and we're focusing on our prespecified primary endpoint. And I would say we focus on upper limb tremor because that's what's that is what's correlated. It was thought to be correlated with disability. Of course, that's what we're showing. So yes, of course, when you start to look more deeply into the data across the spectrum of all symptoms of patients might exhibit, we'll understand more about the effects on these other aspects of tremor. There's no reason to think we're not having such effects. But the ones that are going to be important, and this is going to be important for payers, it's going to be important for patients are the ones that affect their lives. So right now, we're focusing on what we think is really associated with what patients are telling us they're most important to them. The physicians are using as the goal of their treatment. And the ones that ultimately affect how patients function. And that's what -- that's something that we've learned over time. And we've been working now for the better part of 7 years in essential tremor and really trying to make sure that the endpoints that we start to include, both from a scientific perspective as well as in our initial discussions with regulators, really speak to what's going to be most important for patients and what's going to move the needle for them from a functional perspective. So a lot more to come in terms of picking apart the data and which of the measures actually do move. But this was -- this study was specifically designed and powered with the goal of looking at upper limb tremor, and that's what we're sharing today.

Our next question comes from Jay Olson with Oppenheimer.

Congrats on the results. Can you comment on the magnitude of the placebo response? Was that in line with what you expected? And what kind of steps can you take in your next study to try to minimize the placebo response?

Yes. Thanks, Jay. Jim, do you want to talk about placebo and the next steps?

Yes, absolutely. Yes. So Jay, yes, the placebo response is more or less in line with what we are hearing from KOLs and others working in the space. I think to your question, you're always thinking about what can be done to minimize placebo effect. We know placebo expect is going to be an issue in any clinical trial that's run. Yes, it's back to that concept, though, there's a lot of information in the trial. And so we're going to work to get deeper into that information. And importantly, to -- as Steve was saying earlier, work with KOL. So that would be part of the conversation is what specifically can be done to minimize placebo effect as much as possible.

Next question comes from Yatin Suneja with Guggenheim Securities.

This is Eddie on for Yatin. Congrats. Just another follow-up on the down titration that you've seen given that you had over 60% of the patients. I was wondering if you could give us sort of some qualitative assessment of like a dose response when it came to tolerability, were those somnolence events driven by the 60-milligram dose? And do you have any sort of extra color there? And then I know you mentioned that you didn't see any falls, but can you confirm that you didn't see any loss of consciousness?

Yes. Again, I'm not sure how much more we can add. But Steve, do you want to take that?

Yes. Obviously, if there is -- if that -- if we have seen loss of consciousness, we'd be talking about it. We've been talking about what the most common adverse events are. And as I said, they're well within our expectations. With regard to some of the other pieces, we'll be getting to those data. It actually speaks more to understanding the PK/PD relationships and how best to use this profile to identify the appropriate dosing regimen. So a lot more to come in that area as well.

Our next question comes from Gary Nachman with BMO Capital Markets.

This is Irwin Fung on for Gary Nachman. Can you just give us a sense of how much improvement you saw in the TETRAS ADL score? And how exactly that's correlated with the TETRAS performance score? And in a pivotal trial, would you need to hit both of these endpoints as well?

Yes, Irwin, thanks for the question. Let me start with the last comment, and then I'll kick it to Steve for any other color on correlation. So as I mentioned before, with these data and the next step data. Importantly, we'll sit down with regulators and kind of map out what the right primary endpoint looks like, whether it's activity daily living or TETRAS upper limb score or full TETRAS score. The fact that we saw statistically significant correlation, though, again, gives us tremendous optionality as we work with regulators on what's the most important and on what's most important to patients. Steve, do you want to comment further?

Sure. Yes. I would just go to remind everybody that we're talking about our first Phase IIa trial in the first of its kind medicine with this drug class dose for over -- for just over 4 weeks. And so the fact that we're seeing improvements in ADLs and they're correlated with tremor, just -- it simply confirms our hypothesis and really give support to the idea of focusing on upper limb tremor scores is really the right kind of endpoint that allows us to dial in what's most important to patients. But as Barry said, that's not the same as -- which is going to be the primary endpoint that we've used for regulatory filing, how best to demonstrate these effects to payers and so forth, that's all to come. For us, this is, first and foremost, a scientific confirmation that dosing over the course of a month with this mechanism has true differences. Those differences are maintained over time that the mechanism that we've seen repeatedly with multiple drugs, related drugs in this class does holds up into the most, I would say, rigorous scrutiny in a randomized, placebo-controlled trial. And that gives us great confidence to move forward. But a lot of the questions around how much do we need to see and so forth, those are all to come. There are a lot -- I think Jim pointed this out. There are a lot of other endpoints that we've included, patient reported outcomes and so forth. And we'll begin the work with patients experts in the field, regulators to dial in, how best to use the really important information that we found today to dial in both the next studies as well as our Phase III program.

Our next question comes from Neena Bitritto-Garg with Citi.

So I was just wondering about -- on the primary endpoint, I believe you used on-site radars rather than a central radar to look at the TETRAS upper limb score. I guess can you just talk a little bit about potentially the effect on the results there and whether there were any particular sites that participated in the study that may have performed a little bit differently just based on use of on-site rating?

Yes. I can't comment that the training was quite rigorous. But Jim, do you want to provide more?

Yes. Just to say, you are correct that there is often a decision to be taken in these kinds of trials on whether to use a central radar or onsite rating. And we made in discussion with a number of KOLs, we made the decision to go with on-site rating. And I think, yes, much of that has to do with their challenges to things like central rating. Often that's done on video, of course. And then we felt that on-site rating was the best way to get a clear and precise assessment of tremor activity.

And I'm not showing any further questions at this time.

Well, then I'll wrap. So thanks, operator. Thanks, everybody, for being with us on the call today. If you've heard from all of us, we're very excited by the results we got from the KINETIC study. And we'll work with Biogen to map out next steps and report more when we do. So thank you for your attention. Appreciate it.

Ladies and gentlemen, this does conclude today's presentation. You may now disconnect, and have a wonderful day.