Should I start? Or, okay. All right, we're going to get started with our next session. I'm Andrew Tsai, Senior Biotech Analyst at Jefferies. Thanks for joining us. It's my pleasure to have Atai, Srinivas, CSO, or soon-to-be CEO, joining me today. Welcome.
Thank you, Andrew.
Do you want to spend maybe a couple of minutes talking about the Atai story, what programs you have, milestones? We can go from there.
All right, sounds good. So yes, Atai was founded in sort of late 2018 with a mandate really around mental health, so treating mental health disorders with large unmet medical need and using innovative approaches to do it. So the initial focus was actually an investment in Compass. Compass Pathways is a company that's developing psilocybin for the treatment of treatment-resistant depression, but then kind of branched out from there. I joined shortly after the company was formed. My background is really around drug development, M.D., Ph.D., and neuropharmacology with a Ph.D. in neuropharmacology, and been heavily involved in drug development for over 20 years at this point. So my mandate was really to flesh this company out in terms of a pipeline, making it an operating company. So that's what we've done. We had a very unique kind of hub-and-spoke model, so we had lots of subsidiaries.
We had incubated companies, so subsidiaries were partially owned, a sub-majority, somewhat majority owned, and then wholly owned assets as well. We've kind of simplified that a little bit. So over time, we've really kind of come down to a core pipeline of four companies or four assets, and then we've got some additional things with Beckley that are strategic investments, so Beckley and Compass as well. At this point, we have a heavy concentration in psychedelics, and certainly our focus in terms of the long run is really around interventional psychiatry. But we do have a very exciting program in cognitive impairment and schizophrenia that we'll be reading out for the phase II-B trial next year.
I see. And so if someone were to ask you, "Why should I invest in psychedelics today?" what would you say? How has the regulatory environment changed or evolved? I'm sure we'll get more to that in a bit. How have industry data sets played out and so forth?
Yeah, so I mean, we'll talk about the outcome, but big picture, I think there's a lot of - there remain a lot of tailwinds, frankly. So let's start with the data sets. We obviously saw the Compass data set a little while ago, recently saw MindMed's, a very nice trial conducted by MindMed in GAD, again showing good efficacy with this class of compounds. In terms of commercial, the progenitor, if you will, in this space is Spravato, and Spravato is expected to hit $1 billion this year. They've been critical to actually building out infrastructure. We'll talk about that a little bit later. I mean, that's where our - using their infrastructure is where our focus is as well. And again, on the regulatory side, things continue to be good, additional breakthrough designations, several more successful end-of-phase II meetings. So yeah, I think things are continuing to progress quite well.
And then in the very near term, actually this week, we obviously had the AdCom for MDMA, which unfortunately had a negative vote. So what were your takeaways, good and bad, from the AdCom?
Yeah, so I think a lot of people were surprised. I think some of us were a little less surprised. So just kind of stepping back, if you—so Rick Doblin is the founder of MAPS, founded in the 1980s. He is a pioneer in the space, and I would say that he is, if not single-handedly, he has played an outsized role in the psychedelic renaissance that's occurred over the last several years. MAPS started with MDMA when it became a Schedule I compound. So it was a unique model that was basically only donations, and that's why it took them as long. So the IND for MDMA was put in place in 2021. So just look at that timeline that's gone on. The upshot of the donation-based thing, in addition to this long time, is also they were always a little bit on the underfunded side.
So the trials are very small. It was, to me, always an unusual data package because you had two almost phase II-sized trials as opposed to robust phase III trials. And then some of the other things is there was - and some of the allegations as well - is that there was kind of a bias there, really want to get MDMA out to everybody, preaching to the choir, I mean, really kind of focusing on true believers within the company. But also the allegations were around patients and patient bias and who the people, the patients that came in, 40% of folks had taken MDMA before. So I mean, there were a lot of different elements there that there were a lot of issues with MAPS going into it.
Now, from the actual AdCom itself, there were a number of things that are very specific to MAPS, and then there's a couple of things that are a bit broader. I mean, functional unblinding is one that came up multiple times, and we can talk about that, so.
Right. And so you are, I mean, I'm assuming you would say you're better funded, you can run studies with more rigor, and so forth. And so what were the issues that MAPS had that are idiosyncratic in your view?
Yeah, idiosyncratic. We talked about the small trials. We talked about some of the selection bias of the patients, picking the true believers, having a lot of folks that have experience with the drug. Lots of peculiar things, like they didn't do safety labs, so they only had a very small database for safety labs. Normally, in a clinical study, you do labs before and after, even in a phase III trial, so you have a good data set. There were concerns about cardiovascular. They didn't do ECGs as follow-up when someone had persistent tachycardia, for example. So yeah, I mean, there's quite a few, again, relatively idiosyncratic things for these trials.
And then on functional unblinding, it is more maybe prevalent?
It's prevalent.
Yeah, in a sense. So how are you tackling that with your studies?
Yeah, I mean, the big picture is that functional unblinding isn't new or unique to psychedelic compounds, right? So this is a phenomenon that's been described for many years, and people have always been like, "How do we deal with this?" Right? So functional unblinding just means that the patients can tell whether or not they got the drug, right? So that can be - in the case of psychedelics, it's a bit more obvious, I guess. You have a psychedelic experience. But with other drugs, even if you get an SSRI, like if I gave you fluoxetine, you would be pretty wired and pretty nauseous. If I gave you paroxetine, you'd be pretty sleepy and nauseous. And then, of course, there's gamma-hydroxybutyrate, which is approved - that's Xyrem, which is approved for narcolepsy. That drug can put a decent number of people into a functional coma, right?
Because you're taking a lot of it, you're in a sleep state. It looks architecturally like sleep, but you're not arousable. So there are lots of ways that a patient can tell. There's many drugs a patient can tell when they got it. So how do you deal with that? There's a couple of different elements to that. One is to show a dose response. MindMed did a really nice job with that, by the way, recently with their trial. But if you show a dose response, then it becomes a little bit—you're demonstrating the pharmacology of the drug is also driving things. In the case of Compass, they tested 1 mg, 10mg, and 25 mg in their phase II trial. They showed that despite the fact that both 10 mg and 25 mg were associated with robust psychedelic effects, 10 mg was less effective than 25 mg.
So that speaks to an underlying pharmacological effect that's above and beyond the functional unblinding. So that was lacking in the MAPS studies.
Right. One thing I think that the AdCom brought up is manualized therapy in their trial. Are you guys doing that at all?
Yeah. So that's a really interesting point. So MAPS had a very unique approach to this, right? They were calling it MDMA-assisted therapy. So they were insisting that therapy was primary and MDMA was sort of secondary to facilitate the therapy. So that's a little bit of a strange positioning when you go to the FDA because the FDA doesn't control therapy. So another element here was, and this came up in the AdCom, it's normal to stick a drug on top of standard of care, right? So if you had one form of therapy for PTSD and then you stuck the drug on top of it, that's a completely accepted approach to running a clinical trial. In this case, they came up almost with an ad hoc therapy that was not standardized.
So different therapy - they had very broad guidelines for what they could do, and different therapists were treating different patients differently. So that's a problem. And then it was almost like you had two actives, and that came up. That's why they kept the AdCom there talking about factorial designs. They wanted to see what that sort of therapy was doing by itself and then see the interaction with the drug. The therapy was actually even further complicated. We just talked about functional unblinding. I was really referring to functional unblinding in the context of the patient. Here, there was also unblinding of the therapist. That's a whole level of complexity that normally you don't see in drugs. All those other drugs I was talking about, you don't see the site or other people really having the functional unblinding. It's really all about the patient.
Here, you had this whole other level, and the therapist was before, during, and after. So again, introduced another set of variables.
I see. For your studies, will you be using therapists?
No. So all of the other companies, ourselves included, are—so let's step back for a second. MAPS, again, was doing some form of therapy that was, in some cases, more cognitive behavioral, etc. That's different than psychological support. So psychedelics do require a bit more prep, certainly, than "normal drugs." And they also do require some kind of follow-up, right? So things can come up during the psychedelic event that requires some work. So you check in on the patient the next day, typically. And then there's normally one other sort of scheduled visit. And of course, if something is problematic, you then make sure the patient's adequately taken care of. So that's really a more standard of care type thing.
I see.
So that's what we're doing. That's very different than actual therapy.
Right. Right. And then I guess another topic was abuse potential, safety model, or whatever. How are you tackling that?
Yeah. So it wasn't, I mean, the drug, MDMA, does have some abuse potential. There's no doubt about that. That's not really the problem. This drug is going to be given in a supervised setting. In this context, however, what happened was they didn't actually capture the adverse events of special interest. That's something you always do, right? It's kind of like you always do labs. You also always capture adverse events, and particularly those of special interest. And those are things like hallucinations, dissociation, euphoria, etc. These are all things that are associated with abuse. So you have to capture them. They didn't do that. So they couldn't characterize adequately the abuse potential of the drug. And of course, they also didn't do HAP studies and all these human abuse potential studies, etc., which you'd normally do.
Right. Okay. So there's that. FDA will make their decision August 11th, and we'll see.
Yeah, we'll see.
Okay. And speaking of MDMA, you actually do have a program in MDMA. So if the FDA voted negatively, does that affect your strategy for your own program?
I don't think so. There's a couple of reasons for that. I mean, we talked about a number of issues with the program itself, and I think those are all addressable, right? So that's the important bit. We're not focusing on therapy as an example. But what has intrigued us about our MDMA is that the initial hypothesis was that it might be a better, more tolerated version of MDMA. So one thing that did come up yesterday or on Tuesday was cardiovascular risk. So we had some thoughts that this could be an improvement on a cardiovascular safety profile. What we found instead is that the profile was very different overall. So it had some MDMA-like features, so entactogenic features, but it also had some fairly pronounced psychedelic features. And so MDMA, one of the things that characterizes it is that it's almost an external-facing kind of experience.
The people want to interact. The patients want to interact with someone, right? And that's why it's used in raves and things like that. The experience in the phase I was all internal, more like a psychedelic. So people were processing things, and in phase I, these folks were coming out saying how much this was meaningful to them and that we have to develop this drug. But it was much more of an internal experience, which is, again, a very different profile. So that's what makes this really exciting for us.
That's interesting. And so technically, it is the right isomer of MDMA, your program, or how is it different?
Yeah. Yeah.
Right isomer. And it's patented.
That's right. Yep. Well, yeah, we have a number of, I mean, obviously, we don't have composition there, but we do have a number of other patents in process.
Got it. And then my last question on that one is, what's the next steps for this program that you have?
Yeah. We'll hopefully have some clarity on that. We're looking at some phase II-type trials, exploratory phase II trials to further—well, to replicate the results that we found. I mean, these are all phase I results. And then obviously take a look at efficacy as well.
I see. Okay. And so shifting gears to another program you have with data this year is BPL-003 for TRD. It's a placebo-controlled study in actual patients. And you've shared some initial phase II-A data. So maybe talk us through what you saw that makes you excited about this compound.
Yeah. So a couple of things. So the phase II-A, just to kind of get that out of the way, smaller study, 11 subjects evaluable, and looking at basically outcomes at multiple time points all the way up to 12 weeks. What we found intriguing there was that we had good remission and response rates. Their response rates were around 50%, actually remission pretty close to that as well, that were stable over time. So one of the other things around functional unblinding is that there's an expectation that if it's all functional unblinding, that something around expectation, your rates will drop off. Your efficacy rates drop off as a function of time. Here we found that to be very stable. So that was exciting. Now, in a broader picture around this compound, why we actually invested in this in the first place is that it's a different moiety.
I mean, we're developing internally, we have a program around DMT that's a different moiety with different pharmacology, 5-Methoxy-DMT. It's using a transmucosal formulation. It's an intranasal transmucosal. It's got a very nice PK profile, so not too aggressive. And something that fits into the sort of two-hour window. The psychological effects, psychedelic effects do wrap up within two hours. So again, the patient should be ready for discharge then, again, kind of taking this provider paradigm. So that's what we've been focused on, and that's why we were excited about this in the first place.
I see. And so now the phase II-B, again, we'll read out second half. I think it's evaluating 8 and 12 milligram doses versus placebo in 225 patients. That's a lot. Could this be a supporting pivotal study?
I mean, it's definitely going to be supportive, right? So the way this works typically is you would never go into this with a single phase III trial. And of course, look at Compass as a model. But they had robust phase II-B results. It was a well-designed, well-controlled study. If one of those phase IIIs is 0.07, certainly this is very supportive of that. So that's the way to kind of think about it. Is it a standalone in the context of only one phase III? I guess that's something that you discuss with the agency. It's a possibility, but normally one wouldn't conduct the program that way.
Right. And the primary endpoint is four weeks. So what kind of placebo-adjusted change on MADRS would we want to see? And I guess when I think about TRD, the Spravato, how would that compare to what Spravato showed?
Yeah. I mean, we haven't really guided on that, but Spravato is not a bad proxy. I mean, what you see in their 4-week study, they had a kind of an odd mishmash of trials, to be honest with you. But they did have a four-week trial that showed a four-point separation. However, that four-point separation over four weeks took eight visits, right? So it was two per week. Each one was two hours in the clinic from administration to discharge. You're not driving home, so you got to secure some kind of transportation. You got to get transportation there. There's some additional time in the doctor's office. You can do the math. It's a lot of hours, right? I mean, it's at least on the order of six hours, if not more per week.
Times four weeks is 24 hours versus this one, which is a single administration, two hours. Obviously, in a clinical trial context, it's longer. But one expects in a commercial context, it'll be two hours of monitoring plus the additional time. So it's just a very different paradigm. Similar effects to that with a much better paradigm. That's why.
Right. And it is an eight-week-long study, I believe. And so what are you expecting on durability out to eight weeks?
I mean, that's obviously something that we're very intrigued by. The phase II-A gave us higher pretest probability that you'll get durability that persists out there. Of course, we anticipate that it's going to decay over time. The idea of this trial is to give us the data we need to then design the phase III program. Like, when do you redose? How do you set those criteria up for that?
Right. And do you think you would file for breakthrough designation? As I think about other programs, they recently had breakthrough designations after placebo-controlled data. Is that your intention?
Yeah. In general, I would say that that's an unreasonable expectation to at least file for.
Right. And there is another program working on 5-MeO. Maybe compare, contrast your program to theirs?
Yeah. So this is a program by GH Research. It's an inhaled product. And then the administration protocol is up to three doses. So there's a couple of reasons that I – when we started with DMT, we obviously had inhaled as an option, right? Inhaled has a lot of intrinsic problems. I worked with a company called Alexza a long time ago. We had a program with them. They really struggled. They had a very intermittent use, presumably once in your lifetime, compound. It was inhaled loxapine. The FDA made them jump through a lot of hoops for lung toxicity. So the FDA just doesn't like stuff going into the lungs. That's a general statement. So that's the first thing. And then the pharmacokinetics in that case are kind of brutal, right? It's inhaled. It's almost entirely absorbed. Tmax is in tens of seconds.
And so you get this huge spike, and that kind of just puts you over the edge, right? That is a very different profile than what we were looking for with DMT and what we ultimately found with BPL-003, where Tmax is on the order of 10 minutes. So it's a much more gentle thing and then kind of wraps up over the course of, well, within two hours.
I see.
The other interesting thing is that they're doing this three-dose paradigm, up to three doses, which means you have to block the day anyway. It's not different than psilocybin, right? So we're focusing on this two-hour window. But that one, you have to schedule for longer. And I think it was around two-thirds of patients did get two-thirds, two doses, and I think another third got three doses. But you still have to schedule for the entire thing. So going back to the AdCom, they want to see dose response. You don't really have that here, right? You can't do that. Or at least it's more challenging. I mean, I guess there are different ways you could do different trial design that incorporated one, two, or three doses or something like that.
The other thing that they're doing was a seven-day endpoint in their phase II-B and then going straight to open label afterwards. That means you can assess durability. And so that's another important element to assessing functional unblinding. So there's a couple of challenges there.
Right. Okay. Ultimately, you think you can be first to market with your compound?
Yeah. I mean, certainly the current phase II-B trial is being conducted internationally, but in particular in the United States. IND was allowed back in, I think it was February or March of last year. GH currently is under clinical hold, partially, yeah, from what we understand, I think publicly disclosed around lung tox and additional respiratory assessments in animals.
Okay. In the last two-three minutes, is there one program you would like me to dig deeper on?
Yeah. I mean, there's a couple, well, there's two or three. No. Well, VLS, we've got some more data coming out. So we'll be talking about that. We are conducting a follow-up phase I trial with our new formulation. Long and short, we're very encouraged by the preliminary results. That's why we guided recently for starting a phase II trial later this year. EMP-01, I talked about a little bit. That's our R-MDMA program. We'll talk about that a little bit more and kind of flesh out some of the interesting results that I discussed. And then I did mention the RL-007 program, cognitive impairment associated with schizophrenia, again, currently on track for middle of next year, large phase II-B trial.
I see. Maybe one question is, you've been the CSO. You're transitioning to this CEO role. It's been a pleasure, Florian, if you're listening, to work with you, wishing you the best. So what was the impetus for the CEO?
Yeah. I mean, this is something that's been discussed for a while. And it's also what I alluded to earlier. I mean, before there was a more complex structure for the company where there was much more like an investment-type focus. Now we've simplified down pretty significantly. So we've got these two assets that are strategic investments, Beckley and Compass, and essentially everything else. The things that we've talked about are 100% owned or functionally, like in the case of Recognify with RL-007, is functionally in many ways an internal program as well. So it's just a different paradigm. And of course, these are all kicking off into phase II programs. So it's really about that clinical execution piece that's internal. And that's obviously something that I've been focused on for my career.
That's great. Okay. Cash?
We have a runway out to 2026.
Okay. So plenty of runway.
That's pretty much plus securities.
Right. So enough runway to see multiple phase IIs.
See these through, yeah.
Okay. Very good. I think we'll wrap it up there.
All right. Very good.
Thank you, Srini, for talking. Thanks, everyone, for listening.
Appreciate it.