AtaiBeckley Inc. (ATAI) Earnings Call Transcript & Summary

All right. Hi, everybody. Thank you for joining today's virtual company panel day. If you don't know me, my name is Neena Bitritto-Garg. I'm one of the biotech analysts here at Citi. And for our next panel, we'll be discussing mood disorders and mental health. And I'm really pleased to be joined by executives from 4 companies working in the space, including at Atai Life Sciences, Bionomics, COMPASS Pathways and Praxis Precision Medicines. And before we do get into Q&A, I just wanted to flag that if you do have any questions, you can feel free to send me an e-mail directly. My email address is listed on the webcast. You can find me on Bloomberg and ping me that way, whatever works for you. So now I just wanted to turn it over to our panelists. Maybe if we go alphabetically by company names. So maybe Srini, if you want to start, introduce yourself and give some opening remarks on Atai and then we can go around, yes, alphabetically by company name.

Hi. That sounds good. Thank you so much, Neena, for the invite and also thanks for this opportunity. Yes, briefly, my name is Srini Rao. I'm the Chief Scientific Officer and Co-Founder of Atai. Atai Life Sciences is a biopharmaceutical company essentially that's focused on developing really differentiated, highly effective therapies for mental health disorders, kind of spanning in a number of areas of significant unmet medical need, including treatment-resistant depression, forms of anxiety, opioid use disorder, et cetera. Our platform encompasses both psychedelic and nonpsychedelic drug, but we also have digital therapeutics as well, so really using a very broad-based approach to treating mental health. And the end goal here is beyond just getting the MADRS down. It's really affecting long-term behavioral change. And again, this combination of digital therapeutics and drugs potentially with enrichment strategies using biomarkers is our game plan to get there. So by the end of this year, we'll have 10 assets in the clinic. So it's going to be a very active year. Many more things are going to be going into Phase II next year. So we have multiple Phase Is. We'll get into that a little bit later, I suspect. We have $335 million in the bank at the moment. So obviously, covering all of the development that -- through proof-of-concept, certainly for all of our assets. So with that, I'll hand it off to you, or the next person.

Okay. Maybe, Errol, if you would like...

Sure. Thank you, Neena. And I'll echo Srini's comments. It's really a pleasure to participate on this panel with both colleagues and friends. I'm Executive Chairman of Bionomics, which is based in Adelaide, was listed on the ASX for a couple of decades, and in December 2021, we did our IPO and listed on NASDAQ. We were pleased to get the IPO done at a really tough time. Bionomics is targeting ion channels through allosteric modulation. And we are developing drugs for the treatment of both psychiatric and neurological disorders. Our business strategy has a balanced approach between proprietary development and external partnerships. So within this scope of proprietary development, we have a lead asset called BNC210, which is a negative allosteric modulator of the alpha-7 nicotinic receptor. It's in the clinic currently, in 2 clinical trials. One is for the acute treatment of social anxiety disorder. This is 150-patient trial, which started at the beginning of the year. We hope to complete this by the end of this year. And we received Fast Track Designation from the FDA in November of 2021 for this indication. In addition, we're also in a chronic treatment trial for posttraumatic stress disorder. This is a trial that started in July of 2021. And with the chronic treatment [ nature ], we'll read out in the middle of next year. This is a 200-patient trial. Again, both of them are U.S. trials, the PTSD trials across 25 centers. Now balance with the proprietary development, we have a long-standing relationship with Merck in the U.S. for the development of drugs to the treatment of cognitive deficits in Alzheimer's disease, cognitive impairment in schizophrenia and attention deficit disorder. And in collaboration with Merck, we have 2 compounds in clinical development, which have completed safety studies, and there's several biomarker studies that are ongoing. And then just to finish up, the money that we've raised in 2021 through private placements as well as the IPO will take us to readouts in both the clinical trials that I've talked about. The social anxiety disorder trial, the PTSD trial will allow us to make some investments to get Phase III ready to establish some operations in the U.S. and effectively takes us to the end of 2023 in terms of the runway that we would have. So with that, let me stop and turn it back to you, Neena.

Okay. Guy?

Okay. Yes. Hi, Neena. Thank you very much for the invitation to be on the panel. Just as a starter, I'll be making forward-looking statements, like the others, I guess, and I just refer the audience to risk filings with the SEC. So my name is Guy Goodwin. I'm Chief Medical Officer of COMPASS Pathways. COMPASS is a company focused on transforming mental health care. So it's ambitious, and we want to treat disorders that are difficult disorders where other treatments have failed patients. Our lead candidate, COMP360 is psilocybin with assisted [indiscernible] supportive psychotherapy. And we're just about to enter Phase III trials later this year to really see whether we have confirmation of the Phase II studies for treatment-resistant depression. In the treatment-resistant depression start -- the IIb trials, we showed highly statistically significant difference between different doses of psilocybin, and we were extremely content also to see reasonable durability and the effects. So we felt that we've got a good evidence of the drug effect, which is obviously critical to approval by the FDA. We were very confident also that we -- in about 1/4 of the patients we treated, we were seeing significant durability. And together with that, we were seeing quite important supportive measures from the patients themselves. We were obviously relying on the MADRS rates remotely to judge the outcome formally. But we also informally, from the patients themselves, got excellent estimates and the responders of quality of life and the subjective reductions of distress to levels that are really comparable with the patient population. So we've felt that, that has been an excellent start, and we look forward to progressing forward later this year. Together with that, we just announced that we are starting a small Phase II study in PTSD. That's another indication that we think is one of considerable unmet need for the clinical population, and we look forward to making those results available as we go forward this year and next. So thanks very much.

Okay. Great. And Marcio?

Thank you so much. Thanks, Neena. Thanks for the invitation and just joining by all my colleagues here and friends as [indiscernible]. I'm very excited, right, to be talking. We're all like developing therapies like different but complementary, I would say, for patients with CNS conditions. At Praxis, we're focusing in 3 areas, psychiatry being like a large one of them, and I'm sure we're going to be discussing much of this today. We have 3 readouts in psychiatry alone before the end of the year. So it's kind of a very interesting and intense time for us here in major depressive disorder coming up, 2 studies in PTSD, which I'm sure is going to be part of our conversation. And then another area of focus for the company is movement disorder. We just had positive results last week. We're very excited with the central framework and much more to come between now and the end of the year and beyond. And then the last area that serves as, I would say, a platform for discovery and for proof-of-concept for us is epilepsies. Each one of them, as you all are very aware very, have large unmet needs, very large markets as well, I would say, fortunately and importantly since there's all these patients suffering that we can help, but like very interesting in terms of developments. And the interest on the community and the understanding of those networks, of those markers of the disease progress has increased very substantially over the last few years. And I think that's what ourselves, at Praxis, and many of our colleagues here in the panel have been taken advantage of and why we are really revolutionizing the way CNS is now tackled, in general. So very excited. Six readouts between now and the end of the year as a company, so that's very exciting as well, and nerve racking a little bit. And I'm sure we're going to discuss some of that. So thanks again for the invites.

Of course. Awesome. Well, I just want to ask one general question. Whoever wants to kind of answer this, feel free to step in. But one of the questions or one of the, I guess, concerns that often comes up in my conversation with investors is this just psychiatric conditions and mood disorders specifically are very difficult to run clinical trials and detect the treatment effect as obviously you all know largely due to placebo effects. So I guess how do you guys think about designing clinical studies to really minimize the risk of an outsized placebo effect and maximize your ability to actually detect the treatment effect? Whoever wants to step into...

Maybe I'll take a crack at it on a common disorder, which is PTSD that others can comment on. So we're really approaching this issue on a number of fronts, including providing comprehensive training to the investigators, the clinical staff in our ATTUNE trial, educating study participants, I think, during the screening process and on study in our protocol designs. All the trial staff involved in the rating, the participants received comprehensive training on what a placebo response is and what can contribute to the placebo response and the best practices they can put in place when managing the trial participants. Now at the participant level, each participant has read a short script at the visit prior to doing the rating questionnaires that reminds them that they may be either on placebo, active drug and that there is no expectation for them to improve, stay the same or worsen, and they are asked to provide an honest response. Now we're also aware that in psychiatric clinical studies, the placebo response can be enhanced by the number of times the participants are required to go through the efficacy questionnaires. So in our Phase IIb PTSD ATTUNE study, which includes a 12-week treatment phase, the protocol has been designed such that the participants complete their efficacy measurements at only 3-time points after starting BNC210 or placebo. For us, it's 4 weeks, 8 weeks and 12 weeks. Additionally, in the IIb study, the participants are randomized equally in number to either 210 or placebo, again, making sure that there's an equal expectation as to whether they're receiving active drug treatment or placebo. So hopefully, this will get the discussion going, but it is a tough one for all of us, I think, it's going to be dealing with the placebo response.

And Errol, I'm going to jump in if that's okay, as well. Like Errol covered like a big part of that, right? Placebo reminder scripts, right, it's something that he just mentioned, we are using as well in all our trials. I would say maybe I'll take a step back. There is an availability bias on expectation of inflation of placebo. That is not actually supported by the literature, right? If you look into the actual like literature in the last 15, 20 years, there's no inflation on placebo. There are some trials that actually didn't have very good controls, didn't put things in place, like Errol just mentioned, that created this impression that placebo was not under control. Upon proper controls, the ones that were discussed, and I would say, in our case, a few other things like using a secondary or external validation of the severity on the diagnosed, we use safer, as I'm sure it's available like for others being used successfully, and reducing the number of visits, reducing the number of assessments, as Errol just mentioned. And on top of that, I would say, specifically for depression, which is a little bit different behavior for PTSD in terms of the time-dependent placebo effects. If PTSD tends to be more stable over time, depression tends to be a little bit more amplified because depression is not a chronic condition and PTSD tends to be in general. So we also separated the primary endpoint from the last visits to really create this, in our case, right, stacking the cards to some extent in terms of the expectation of the effects versus the durability of the effect. So for us, it's day 15, for PRAX-114 is the primary. Patients are treated all the way to the day 28, and they have an assessment at day 29. So we are assessing 2 different things. And that is being shown as well. Together with having just a minimum number of arms, as Errol just mentioned, that those things together have been shown to help. I'll say -- the last thing I'll say on this topic is it starts and end with the right patients in the trial. If you get the wrong patients in this trial, you're going to have deep amplification of placebo. So we spend a significant amount of time making sure the right patients are getting in, right diagnosed, right severity, proper site and so on and so forth. So with that, should not be much of our history.

I'll jump in here.

Sorry. Go ahead, Guy.

Yes. I mean I was just going to say that this is an issue, obviously, when drugs unblind you. I mean placebo is fine if you can't detect the active treatment. But I mean the problem for a number of our drugs and particularly the ones we work with, they do unblind the patient potentially. So one of the solutions is obviously not to have placebo, but to use different doses because obviously proof-of-concept comes from comparing a low dose with a high dose, which is approach we took in our Phase II. The only problem with that is, of course, that it may very well be that the FDA would request to see placebo because of safety, they want to see a safety baseline against your active treatment. But I think in terms of improving efficacy, avoiding placebo is not about idea for proof-of-concept and how do you address that in Phase III, obviously, is a moot point because that is not usually what people do, but we'll see.

Yes. I mean I will say, I guess, one comment to that just generally is that I agree that the magnitude of unblinding could be much greater with the psychedelic. But of course, this is something that many different compounds, many different class of compounds have grappled with over the years, if you think about [indiscernible]. Even think about gabapentin or pregabalin, they are very sedating for most people and most people can sort of pick that up. So again, it's just something to think about. This has been a long discussion across many different areas. Going back to the placebo effect. I mean I guess there's a couple of additional things. We've talked about the number of arms minimizing that. But in sort of other general things that are -- that drive placebo effect include duration of trial. And part of that is with respect to adherence or drug that's daily. And of course, a multicenter draw, right? I mean just generally speaking, the more trial sites you have, the more issues you have, your effect size goes down. I think speaking also for Guy and certainly with the psychedelic compounds, the expectation is that the effects size is quite large, right? So we do have a number of such compounds in development. That has kind of a virtuous circle type of fact. We obviously can run smaller trials because the signal detection is much greater, so thus, there's fewer centers, et cetera. Rapidity of onset is really key, right? So I certainly ask a senior, if you think about an SSRI, it takes multiple weeks. There's sort of fatigue and people start to drop off. I think many of the -- I think all the compounds that are being developed with this particular -- by the folks of this particular panel, all of those are relatively rapid acting and certainly, that's true psychedelics. And that may also help mitigate things depending on when your primary endpoint is. And we've talked about adherence, so I just mentioned adherence. Again, some of these are intermittent therapies, clearly not an issue around adherence. So as you got -- you've got absolute control over that. So these are additional factors that we can sort of hold in, particularly with some of the psychedelic compounds to mitigate the placebo response.

Absolutely. Great. So I just want to talk a little bit about PTSD now just because I know each of you is working on a PTSD -- at least one PTSD program currently. Maybe if we could just go around quickly and each of you can kind of give us the highlights of your PTSD program, what you're working on. And ultimately, what you think would be a win in PTSD with your program? And then just kind of a side question for Errol and Srini. Your 2 companies do have a collaboration as well in PTSD. Maybe you could also weave that in, that would be great.

Yes. I can do that. Yes. I mean I can jump in. So we have a structure where we have a tie itself and then we have a number of subsidiary companies. It's kind of a decentralized model. One of the company that's actually developing something for PTSD is a company called EmpathBio. And we're essentially developing a derivative of MDMA. So there's another company, a not-for-profit, called MAPS, that has been through multiple trials at this point, Phase III trials, in fact, for MDMA in PTSD. There was a publication, I guess, maybe a year ago or 1.5 years ago. It shared some really nice results with that compound. So essentially, it was 3 separate visits, some weeks apart, 2 therapists each time, giving you -- basically, the subject is given MDMA and then it's essentially a reflection of talk-type therapy that occurs during the entactogenic effect. So that's the basic model there. Our derivative has some of aspects pharmacology that are different, which I think could be beneficial. And that, of course, is something -- that's a hypothesis that we need to confirm. But MDMA is super complex. It definitely releases serotonin, but it also does all sorts of stuff like impacting the dopamine system, the norepinephrine system. The latter is driving -- peripheral release of norepinephrine drives in the cardiovascular effects. There were some pretty marked blood pressure effects that were noted with MDMA. PTSD folks do -- can be -- can have higher BMIs because of potentially as a coping strategy for the PTSD. So it's something that's of some concern. We anticipate our derivative having much less of an effect. And the question, well, how much is central versus how much is peripheral something that is obviously also TBD. I also speculate that by mitigating this norepinephrine effect you might actually get better efficacy. But of course, that's very speculative, right, because putting -- the whole idea with MDMA is giving -- allowing you to sort of tap into experiences that you don't otherwise want to touch, right? So giving you a better therapeutic alliance with yourself is how what I normally quip about this. Norepinephrine, of course, can make you feel anxious. And the question is, is that impacting efficacy? We don't know, but we have that potential here -- potentially to improve on that.

Well, let me just follow up because what I'll do is I'll just talk about BNC210 as monotherapy for PTSD and then bridge what Srini built upon as to what we're thinking about in our collaboration together with EmpathBio and with MDMA. So BNC210 has shown in animal models, both an anxiolytic-like effects, antidepressant-like activity and has demonstrated sort of enhanced fear of extinction. All of these are attributes that would show sort of validity in terms of its utility for the treatment of PTSD. But let's focus now on the clinical data of the compound that would bring in, I think, more validity. So in clinical studies, we've demonstrated clear-cut reductions in anxiety, reductions in anxiety related to threat avoidance. In a panic situation, the patients, or in this case, it was healthy volunteers, recover from the panic attack much faster. And we had actually completed a full PTSD study with albeit with a different formulation before where we had actually shown an earlier time points, at a 4-week time point clear-cut, significant antidepressant effects and good trends towards lowering anxiety, but just missed the CAPS-5 endpoint. So you may ask, "Well, where's the hypothesis when you miss the primary endpoint?" That's when you look at it from a dose efficacy standpoint. But thankfully, we've taken blood levels as samples and looked at blood levels of the drug. And when you do a pharmacometric analysis and look at the relationship between exposure and efficacy, we saw a very significant relationship. But it also told us the blood levels that we need to be able to see a clinically meaningful reduction in past slide, which for the most part, I think, the KOLs would say would be about a 6-point reduction that you would need. And so we've developed a new formulation, which time does not permit me to go into, but clearly exceeds the levels that were predicted. And so we're back in the clinic now with the treatment paradigm that I talked about before. So BNC210 has great utility, we think, as a monotherapy. Now let me bridge to the collaboration that we have with EmpathBio and why it would make sense for us to think about a combination therapy. And it's really a broader discussion around way psychedelics and other treatments are going to go with combination therapy. So based on what Srini described, the MAPS trial, what it showed was some really nice intriguing data where MDMA by itself was not a treatment for PTSD. Let me make that clear. It's an adjunct to psychotherapy, but it requires intensive, 3 intensive 8-hour in-clinic sessions plus additional psychotherapy sessions, okay? So that's a heavy burden both on the patient, a patient population that really doesn't want to come for psychotherapy, along with the payer. So what we're thinking about is once -- depending on where the FDA is with MDMA itself and our collaboration with EmpathBio, imagine a treatment paradigm where the patient comes in for the first session with MDMA, go through that, and then they get BNC210 in the outpatient setting, which may reduce the need for the second and third session or other psychotherapy sessions, or take it from an 8-hour session down to 1-hour session, which has huge benefit for patients, for prescribers and for payers. So that's the treatment paradigm. This isn't that different than what people are looking at with the ketamine and SSRIs. And so for me, the psychedelic paradigm, which is really opening up I think new treatment paradigms, eventually, we're going to have to start thinking about these combination therapies. And from a Bionomic standpoint, we want to be ahead of the game in terms of thinking about them rather than playing catch up once, for example, MDMA gets approved. So hopefully, that addresses both BNC210 and our collaboration with EmpathBio.

Just one other follow-up point. I think you hit upon a really key issue around MDMA proper and that, of course, is the scalability. And one approach, I think that -- well, these are complementary approaches. One approach is to actually indeed have these follow-on therapies like BNC210. The other element that we're really focused on, of course, is digital therapeutics. So I mentioned reflection therapy. There are ways of potentially simplifying and/or automating aspects of that. And we have -- for the outpatient work, both preparatory as well as the follow-on elements, we are using more standard digital therapeutics in a sense, app-based things that actually provide activation commitment therapy, behavioral -- [ DBT ]. So we have all these different modalities that we can deliver electronically on a more frequent basis as opposed to coming in every 2 weeks or what have you. So it's something that's continuous and beneficial, presumably beneficial for the patient. And there's certainly some data outside of the MDMA space certainly for these digital apps having the same.

Let me jump and talk a little bit about 114 and why I think it's also complementary to everything you're saying. You're right, maybe take a step back again and talk about the paradigm in PTSD, which is that it's pretty much nothing that works for these patients right now. It's a very large part of the population. I think that is a [ highly used ] version of PTSD, unfortunately. But there is a reality of PTSD with so many ways and pathways patients get there, different etiologies, I would call, of the disease. And so many people not really wanting to talk about it's, really, the effect and so on. So the kind of some of the characteristics of the disease, as Errol mentioned, is like fear avoidance, right, or the reconstruction of the event or the sleep disorders, anxiety in general, like kind of the depressive states, kind of all combined to some extent. When you look into PRAX-114 preclinically and clinically, we're able to drive down the anxiety pretty, I would say, nicely in animal models and in health volunteers, patients with depression, not with PTSD ye. And the fear part of, at least in the animal model, is very pronounced, the reduction on the fear avoidance on those animals. Like it's supported by some experimental data in patients with PTSD. They have lower levels of neuroactive steroids as well the endogenous neuroactive steroids. So the scientific rationale is very strong to use something like PRAX-114 into this population. And the key question has always been can we give this like safe, right? We want these patients to feel better, not to feel worse, like they feel with some of the mitigations right now. So at the dose we are planning to use or we are using right now in the clinic, 40 milligrams and potentially 60 milligrams after 15 days. We have very low levels of somnolence or sedation, if any, in this population. So it allows for the structure of the sleep to be recovered, for the anxiety to be reduced, to fear avoidance and so on without really the all the side effects because that's one of the main reasons why they discontinued therapies right now is they're not helping and they're making them feel worse. So that's the scientific rationale. We're running our 80-patient trial right now, which should result by the end of the year.

So maybe I'll chime in on the back of that and say we're also interested in this area because of these effects on depression and anxiety. We've obviously documented with a single dose, which lasts for up to 12 weeks. And a significant number of patients, if that were to produce an effect in this PTSD group, it would be extremely useful without the need for formal psychotherapy addition. So we're doing a very exploratory study. It's just going to be 20 patients. It's open label. There's not going to be a control group. But we're going to find out about feasibility, acceptability, adverse effects and, hopefully, some indication whether there's a significant important reduction in symptoms. And clearly, we're looking to see and to see whether this is an area where we should also be putting more resources. But we'll be very excited to share the results in due course. It should be interesting. And there's a mechanistic component of imaging, which will help to objectify some of the changes that we do see them. So that's where we are.

Great. So I just want to talk a little bit about depression, too. I know each of you is working either on a kind of depression or anxiety. Maybe specifically to Marcio. I know you are having your MDD readout for PRAX-114 in June, so coming up soon. So I was just wondering if you could talk a little bit about what really you're looking for from an efficacy and safety and tolerability perspective. I know you mentioned what you've seen so far on somnolence. Yes. I guess we'll start there. If you could talk a little bit about what you want to see on efficacy, whether you think you can see kind of a marginal benefit versus some of the other GABAA PAMs that have been in development for MDD? And then what you think from an adverse event perspective would be kind of differentiating from other programs as well?

Absolutely. We're very excited, right, in the next few weeks. So June is just around corner here. We're going to see the results for Aria. It's our study. We enrolled 216 patients with moderate-to-severe depression. I want to start there because the first thing before we even ask about the fact is, are those patients truly moderates-to-severe depression, right? We talked a little bit about that. And we went to great length to make sure that proper patients are on those trials. It's a little bit of a sad, I would say, realization that we have running a trial here in the U.S. on how many people that were not really having a stable moderate-to-severe depression, had to screen out or fail on the trial, a very significant number in order to guess really the population was ideal for clinical trials. And I think that influences other experiments, especially during COVID on having the right patients in. So starting from that base, right, that you have those patients with MDS higher than 23 at baseline, probably going to land anywhere between 25, 26 on average, just like guessing at this point. What we want to see here, what we expect to see, right, is a very significant change from like the baseline where they are with very like significant depression to day 15, which goes through day 4 and day 8 that we have 2 visits. They are virtual visits and a presential visits for those patients. That is around 0.4 in the effects side. And what that means, right, in terms of like modeling is 3 points or more separation from day 15 [indiscernible] 17. Now that is -- basically what means is it's much faster than anything that is out there. It's a much larger effects size when we standardize, like I mentioned, 0.4 or so. But that is not all the [indiscernible] all these patients needs, right? If you then discontinue the drug and they don't like back or they become erratic, that we're not really serving these patients, right? We have to look into the patients first. So treating them into day 28 and [indiscernible] day 29, the secondary endpoints here, we expect to see basically a straight line from day 15 to day 29 for the treatments and whatever happens with placebo, they are may be giving 1 point or so. But it's still a separation there that would -- because our hypothesis for 114, that we need to continue treating until the episode is resolved or until the patient is really in remission for depression. So maybe just to recap the expectation, right, it's very like significant, clear at day 15, like numerical separation, at least expected, hopefully status quo as well day 29, although it's not required, with very low side effect in terms of like especially the ones related to -- like more related to this mechanism, like sedation, for example. So we expect to see less than 10% or so in that regard, which would be adequate for a chronic treatment.

Great. And then I guess just a follow-up on that, just what -- do you think, I guess, what about the molecule and about the study specifically do you think could drive maybe a larger effect size than what we've seen with other GABAA PAMs in MDD recently?

Yes. So the hypothesis for GABAA PAMs specifically the ones that are highly extrasynaptic referring, like 114 is, is this ability to drive like the overall like anxiety or depression, insomnia, like 2 levels that are like in either remission, or much better, without the drawbacks that are known for the mechanism, right? So like full sedation tolerance and so on and so forth. So from a more chemical standpoint and biochemical standpoint, 114 is the most extrasynaptic referring GABAA PAMs, like tenfold to the extrasynaptic versus synaptic. But then from a distribution standpoint and from like a more PK-related properties or drug properties, we are pretty certain that the drug is in the system. And maybe that is something that was not quite understood before, right? You need given concentrations in the brain to be there present always for the effects to be driven the way we just described. And to our knowledge, 114 is the only drug in late-stage developments that has this ability be taken with or without foods, to be given the exposure that are necessary or giving this before bedtime, so it's fairly complicated to control something else other than the drug itself like if you need like a significant amount of fat, for example, that others might require, it's not desirable. As we've heard before, it's not only PTSD patients that have problems with appetite disturbance, depression patients have issues there as well. So they are forcing something on them. So it's a combination between the drug itself, the way it's delivered, the concentrations we get into the brain and then the design of the trial. As I mentioned before, from the get go, we separated the primary endpoint from the last visits to really show the [ rapid ] effects, but to maximize the [ provisional ] success and to show the durability within the trial. So that's what we're going to be showing, and fingers crossed, in the next few weeks.

Great. And then Srini, I did want to ask you a little bit about GRX-917, so the deuterated etifoxine program just because there is a -- some might say, it's a similar mechanism to kind of a GABAA-PAM. So I guess maybe this is a question both for you and for Marcio. I know you're focusing more on anxiety, and there's some investor questions just around whether targeting GABAA and kind of the neurosteroid pathway may be a better target for anxiety or MDD patients with elevated anxiety. So I guess maybe thoughts from both of you or anybody else who would like to comment as well on that.

Well, maybe I can start with GRX-917 and some of the rationale there and then obviously, Marcio jump in. So as you mentioned, GRX-917 is a deuterated form of etifoxine. And etifoxine is a compound that's been around for quite some time. So this was developed and marketed -- well, route to market in 1979 in France. And it's been available there and in several other territories, but really not been in any of the major territories, like Germany, in Europe or the U.K. and certainly not in the United States. People really didn't know what it did pharmacologically for the longest time. When it came on the market, everybody thought it was a benzo. But it really pretty obvious it wasn't. When you -- I don't know if you've ever had a benzo like for a procedure or something, you end up feeling essentially drunk, right? I mean you are -- there's cognitive impairment. There's heavy sedation. There's ataxia. There's a lot of different things that occur acutely. And then, of course, chronically, there's dependence that -- and abuse liability that can be a problem. So this compound had none of that. And a lot of -- and most of this data was available pretty early on. So people then thought maybe it's activating some other part of the GABAA receptor, which is a fair assumption. But it was not particularly potent in doing that. So that didn't quite makes sense either. So over the course of the last decade or so, it's been realized that this compound is an agonist of TSPO, which is a mitochondrial protein that's actually responsible for producing endogenous neurosteroids. So correct me if I'm wrong, but 114 is a neurosteroid essentially. And this is essentially increasing the production of things like GABAA-PAM. And that's the mechanism by which we think it's doing something very -- it's obviously very similar in some ways to what 114 is doing or zuranolone is doing. And so that's obviously the reason for our excitement around this. In terms of indication, I totally agree regarding depression. Of course, brexanolone is approved for postpartum depression. And we've heard about where 114 is headed at the moment. I think the part that is swaying us at least at this point more towards anxiety is all the data that's available for this compound, for etifoxine, for the parent, right? So this was really approved, if you will, for anxiety and anxiety disorders, specifically adjustment disorder with anxiety. And it's been broadly used in that context for a very long time. So it behooves us to kind of start there. But I think there's many opportunities for label expansion beyond that, including anxious depression.

Yes. I think that's a good segue, right? Neuroactive steroids, I think it's pretty undisputable that they acts on one you can get like a safe way to activate that pathway in anxiety. I think what is a little bit more confusing maybe lately is anxiety and depression and the overlap. Patients with moderate-to-severe depression, the vast majority of them have high levels of anxiety periods. So that is like I think sometimes maybe it's oximoronic to try to separate those things, right? Those patients are coping with a myriad of symptoms and anxiety as part of that. But it's very hard to find patients with like high HAM-D or MADRS that don't have anxiety. I think that's a little bit where you're going as well, right?

Yes, agree.

It's a constellation that you're trying to treat. It would make as much sense to go more importantly on anxiety [indiscernible] going as MDD or in other conditions like that were related and discussed today, like avoidance in general or adjustment disorders in general. We have included and throughout the development of 114 patients with high levels of anxiety not because we are trying to find them, but because that's who those patients are. And you're seeing very significant reduction levels, but only. And I think that, that's the important part about this mechanism. If all we're doing is reduce the anxiety, I think there are like capable drugs right now that do that very fast. But what you're trying to do here is to maintain the levels of anxiety under the control, insomnia under control and the core depression, things that are not related to anxiety, for example. So we're really looking for and you're going to be reporting in a few weeks both sides of the equation here, right? You need to be able to drive the core depression down, so these patients would feel better. They would relate better with society and with themselves, and hopefully, we'll be able to help them.

Great. So Guy, I wanted to bring you back into the conversation. I know you recently had your end of Phase II meeting with the FDA for COMP360 and treatment-resistant depression. I know you'll share updates from that after you have the minutes from the meeting. But I guess in thinking about the Phase III study and the read through kind of from the Phase IIb, I mean, how are you thinking about defining the patient population? Some of these other aspects that we talked about earlier in terms of study design, the number of dosing sessions, the appropriate comparator arm, all those sorts of things just based on your continued analysis of the Phase IIb and conversations with the FDA?

I mean I think the patient population is obviously the critical one. And I'd echo all the issues about choice of subjects that were made earlier. This is -- the key failure is to recruit into studies people who are essentially going to give you a placebo response, not matter what happens. Or in our case, perhaps overenthusiastic responses depending on what they read in the New York Times. That's the sort of challenge is to get patients in who represent or representative of the treatment pathways that ultimately we want these treatments to fit into. So we're going to be taking as much care of that as we did in Phase II. We think there are a number of ways in which technically we've screened people that we may change. We just think we could do better. And we're going to be very careful about site selection because ultimately that's also very critical to the quality of the trial you deliver and not having too many. So these points are all generic, I think, and we'll be following them. As you know, Neena, we're not going to comment much on the Phase III design until mid-summer are also when we finally make agreement on what it should look like. And at that point, you will know everything. So look forward to that.

Sounds good. Totally understood. So Errol, I did want to ask you a question as well around your -- on your social anxiety disorder study. I know that's a pretty unique study design using a public speaking challenge and I think the design that people are probably less familiar with. So maybe if you could tell us a little bit about how that actually works, the regulatory precedent there, and what you're going to see from that study?

Great. Thanks. Maybe I can just start off by setting up social anxiety disorders since the other panelists really are not looking at this particular indication. So the attractiveness, it's a social anxiety disorder, the only approved treatment are the SSRIs. And as we know, they act in a protracted manner, requires several weeks of treatment, and in spite of that, only work in about 30% to 40% of the patients. So there is no acute on-demand treatment that's approved by the FDA. So what the docs use totally off-label are either the benzodiazepines, which we've discussed, or the beta blockers. Now the beta blockers will help with the cardiovascular manifestations of it. The benzos are great from the standpoint of producing the anxiety, but have the liabilities of addiction, motor impairment, sedation to deal with. Now BNC210 has shown efficacy in 2 separate clinical trials. One, as I mentioned, in reducing panic symptoms, and in a very nice study in generalized anxiety disorder patients that was carried out at King's College in London. And there we went actually head-to-head with lorazepam, the standard of care, and showed we were at least as good as lorazepam if not better. So why do we go after social anxiety disorder? It was actually driven by the regulatory endpoint and where a company, like VistaGen, that has a short-acting intranasal interferon, got the FDA to agree to a Phase III endpoint, which is essentially an experimental endpoint. And you'll get this when I describe a particular study where you bring in patients, challenge them with public speaking, in their case, and that's the endpoint for registration. I mean what could be more attractive than that? So with that as a backdrop, now let me describe to you our study. So in our case, just like VistaGen, study participants command the screen for social anxiety disorder with marked-to-severe symptoms, that's a score of 70 using the Liebowitz Social Anxiety Scale. Then they return to the clinical site for the public speaking challenge that is performed 1 hour after receiving just a single dose of 210. In our study, we've got 3 arms, placebo, a dose of 225 milligrams that in previous studies we've shown reduces efficacy at least as good as the benzo, or 675. So we're really exploring the dose range. Now the participants are then asked to select the speech topic from a specified list, and they have 2 minutes to prepare their speech. This is called the anticipation phase. Frankly, this is the most anxiety-provoking phase. They are then required to speak for 5 minutes in front of a small audience. This is the performance phase of the task. Now during the anticipation phase, the performance phase and at the end of the speech, the participants complete self-assessments, including the subjective units of the stress scale. Now this is a new scale. It's actually the primary endpoint that the FDA has agreed to. But we'll also look at the Spielberger State-Trait Anxiety Inventory and self-statements during public speaking scale. So this is -- it's an experimental endpoint, which at the end of the day, is our Phase II, but it will also be the Phase III endpoint. So that's it. It's a very straightforward study. One of the things I may add to that. This is a scale that there isn't a lot of historical precedent, other than what VistaGen has shown in the Phase II trial. And there, they've shown a reduction of about 13 points in the subscale. So that's about what we are looking at in terms of -- when you think about powering the study, it wasn't the easiest thing because you don't have a database here to really power. So that's approximately what we're looking at in terms of sort of the expectation of a significant effect.

Great. And Srini, I did want to come back to you just to ask about the data that we're expecting towards the end of this year in treatment-resistant depression with your R-ketamine program, PCN-101. I guess maybe if you could talk to us a little bit about what would be kind of a win there from a safety and efficacy perspective? And what you're looking for specifically in terms of the therapeutic window just considering that you are developing that with the intended use in an at-home setting versus clinical?

Yes, absolutely. So yes, I think it's -- maybe just to get to level set everybody. So we're developing R-ketamine for treatment-resistant depression. Of course, S-ketamine, the other enantiomer is approved for treatment-resistant depression as SPRAVATO. One of the challenges with SPRAVATO is that the efficacious dose is also the dose that is associated with disassociation, right? It has disassociations part of the side effect package. That may be one of the reasons that it gets -- that at-home use may be very challenging for that compound. So both preclinical data and very limited third-party clinical data suggest that R-ketamine may give you efficacy at a dose of substantially less than that dosage associated with just now that has disassociation. So that's really the premise here. That's the hypothesis for why this drug to be developed for at-home use. So we completed a Phase I trial. I guess it was last year and looked at a range of doses. And perhaps not surprisingly, at very high doses. We know that this an [ NMDA ] antagonist, low potency, but at high doses it looks rather similar to S-ketamine or racemic ketamine. However, we picked some doses that were well below that. And it's one that was would devoid of any kind of activity. It was somewhat consistent with what the previous open-label study had done by the third-party again. And then also, we picked another one that was just sub-associative. And that's what we've taken forward into our Phase IIa trial. In some ways, looks rather similar to what COMPASS did with their Phase II. Their Phase IIb, it's little smaller, obviously. But essentially, it's placebo versus 2 doses of R-ketamine dosed intravenously in this context and looking at MADRS at 24 hours. And so it's 31x 3 essentially. So that's the general -- it's a single administration. We, of course, anticipate multiple doses with this or repeated dosing, not unlike any other ketamine. So that's kind of -- that's the background here. In terms of what we are looking for and what would constitute a win, obviously, we're looking for statistical significance. But MADRS change, minimum clinically important difference on MADRS is 2. So we obviously want to see something greater than that. And one of the positive trials with S-ketamine, that number was 4. And with COMPASS' Phase IIb, obviously, it was over 6. So it's -- that obviously is quite effective. But substantially greater than 2 would be a win at 24 hours. But also persistence is sort of comparable to what you see with S-ketamine. So if it works at 24 hours, but then starts to tail off more rapidly, that could be something we'd have to take under consideration. But some preclinical data suggests longer or more durable efficacy. So if we're looking at a week and as well as 14 days now, so we'll have that data. The key, of course, is this efficacy in combination with minimal or no dissociation. So as I mentioned, there's 2 doses, one that really has no impact on disassociation. The other one that has minimal impact on disassociation. But at least in healthy volunteers, it's essentially within the normal range. This is only CADSS score. We don't need to get into the details of that, but certainly within the normal range. So that's what we're looking for here. I think that would make for a very compelling argument to the FDA about for at-home use.

Great. So in the last 5 minutes here, I did want to ask kind of a wrap-up general question if each of you wants to kind of chime in with some thoughts. So each of you, as we discussed, you're all developing therapy for mood disorders and other mental health conditions that I think do represent a paradigm shift versus the way that a lot of these conditions are currently treated. Many of them are considered -- even depression today, right, is often treated as a chronic disease even though, as Marcio pointed out earlier, it really isn't for a lot of people. So I guess how do you think of that from -- how do you think about that concept from a commercial perspective? And how you really kind of change the treatment paradigm with some of these therapies? And whoever wants to start, and then we'll wrap up after that.

I'll jump in first. The first maybe, say, May is Mental Health Awareness Month, right? So I think what those patients needs us to do is to develop effective safe therapies for them and for their families and for the society. So on that regards, however, we get there, we're going to figure out a way to get the economical model to work and keep that in focus. And what does it mean for us here with 114 specifically, with other molecules we're considering in like 40, that is one of our preclinical programs is getting them better, keeping them better so they don't relapse, so they don't come back. There are different ways. There are likely going to be more combinations in the future than we have right now. We talked about like digital treatments today, talking about psychedelics. And I believe that GABAA PAMs especially the safe ones, the durable ones are going to be a very clear like part of their armamentarium in the future. We are considering a very straightforward, maybe the most straightforward way to market that -- this drug if you were to get to the market as we expect to, that is as a continuous treatment, right? We can call that chronic. We can call that episodic. But at the end of the day, you take it. That is not a lot of utilization of the health care system, which is important. If you keep like starting and stopping a drug and switching drugs, that's not only create anxiety on the patients, it create anxiety on the health care providers as well and utilize the system, but it's something that it is relatively simple to introduce. So on that regards, the paradigm shift for Praxis is better, faster drugs, that are safe. It's not necessarily the way these patients are taking or interacting with the drug right now. And we think that that's probably enough to do our part to help. So that's how we're looking into. And we think that for an oral GABAA PAM, that's the one that address most of the markets.

Just kind of building on that, we're splitting the difference a little bit, right? So we've got psychedelic compound as well as non-psychedelic compounds that we've talked about. We have the etifoxine derivative, but we also have R-ketamine, both of which could be -- we're anticipating would be at-home use, obviously. I think an important element for us is just the broadly recognized bit that therapy of some kind is really important to affect long-term efficacy, right? So we're really interested in helping these folks, getting them intermission and keeping them there. And as someone had mentioned, all of -- particularly the psychedelics, the medication-assisted therapy and the therapy piece is key. Of course, the issue is you don't have enough therapists, right? And that's an important element. That's where we think the digital therapeutics could play a really critical role. That's where we think we can really make a big difference. I have mentioned it on previous calls, but I grew up in a small town in Southeast Kentucky that is always what I think about in this context because no therapist there, lots of OUD, lots of treatments in question. What can we do about it? It isn't necessarily going to be easy for someone to go and repeatedly have to go to a clinic to get their medications and therapy, but we think we can potentially mitigate that with some of our digital approaches.

Yes. Just to follow up on Srini. I think there's such a huge unmet medical need for what we're looking at. And the theme is really faster and fewer side effects. I think what we've heard from all the participants today and Bionomics with BNC210, that is the theme fast-tracking with BNC210 and reducing anxiety, less side effects versus the standard of care in terms of getting there. And the feedback, I'll tell you, we've gotten on the commercial side in terms of a market analysis that we've just completed with payers, community psychiatrists as well as KOLs, is they're desperate for a profile such as presented where we've got, I think, a unique opportunity really to do right by the patients and shareholders. And what could be better in the fields that we're in, in terms of making a difference.

I think you're right, Neena, to emphasize that this is sort of paradigm shifting and this whole approach that we're -- all of us, to some extent, involved in. I think, for us, we just had visits with 50 providers in the U.S. Our market access team went around and visited all of them in person. And we get a strong sense that there is a growth of these intermediate treatment facilities based on S-ketamine increasingly reimbursed, based also on neurostimulation, and we feel that we fit right into that. It's certainly true that I think there's a great place for keeping patients well with digital therapeutics, and we completely agree about that. I think that the emphasis on therapy in association with the drug is a little misleading for many people. It's sort of based on the Max model that we're dealing with assisted psychotherapy. In fact, the drug does the work in the case of psilocybin. And the therapist, if you want to call them a therapist, really is there to prepare the patient. We have 90% of our patients who've not had a psychedelic before. So this was a fair test of the capacity and the safety to do this in our Phase II trial. That preparation needs to be focused. And we can -- there was scope for improving it and getting quality control over it using digital technology again. But we think that's the key part. It's the preparation sport on the day. And then it may well be that in the long run, it helps to have kind of simple behavioral activation, [ CVT-type ] apps, which will help to keep the people well. And as with all these fast-acting treatments, we're going to have to keep track of the patients in order to detect relapse and to replace or supplement or whatever we have to do to keep people well, and that's going to be a challenge on its own. But it's a great time to be in this particular field, I think.

Absolutely. Fantastic. Apologies for running a couple of minutes over. I want to thank each of you for taking the time to join us today. And anybody who is listening on the line, there is a break for about 30 minutes, and then we'll have our next session at 12:30 Eastern. But thank you, again, everybody. Great.

Thank you, everyone.

Thank you. Everyone. Be well.