Excellent. Thanks, everybody, for continuing to listen to our great panels today. It's my pleasure to be introducing the Compass Pathways team. We're going to talk a lot about the upcoming Phase III readouts and some of their preparation on the regulatory and commercial side. Before that, maybe we can start. Kabir, thank you so much for joining. You can just give us a quick snapshot of Compass and the update on the 005 and 006 studies and timelines, and then we'll dig in. Thank you again.
Thanks very much, Paul. Thanks for the invitation. Just as a reminder, we'll be making some forward-looking statements, so I'd refer you to our risk statements and our various filings with the SEC. Yes, we are very much in the middle of Phase III with COMP360, our proprietary synthetic formulation of psilocybin for treatment-resistant depression. The first of those studies, 005, single dose placebo-controlled, we will see six-week primary endpoint data towards the end of next quarter. The second of those studies, 006, which is the three doses with a second repeat fixed dose of three weeks. For that study, we are expecting 26-week data in the second half of 2026. I'm happy to confirm that we are on track for both of those timings at this point.
Excellent. Maybe let's start just with a broader question on psychedelics. Since you joined Compass, how has the landscape changed? Where do you think we are currently in just the world's acceptance of psychedelics as a class when you think about all your stakeholders, right? Patients, physicians, and bigger, more traditional pharmaceutical companies.
Yeah. I think the good news is what you've seen over the last few years is not only us, but a number of other companies are now really generating robust clinical evidence. I mean, the whole thrust in this field has been to move beyond what was kind of anecdote and experience to the type of robust evidence that can be delivered to regulatory authorities with the potential ultimately for approval and reimbursement. Not only us, but other companies are moving forward on that. I think you're seeing a clear separation of those companies that are doing that. I think, and I'll actually hand to Guy and Lori also to talk about what they're seeing from a physician and commercial perspective. I think you're clearly seeing a lot of interest from physicians and patients, but also a desire to see data.
There is an acknowledgment that we are still relatively early in understanding the appropriate use of whether it is of psilocybin or these other things as well. On the commercial side, clearly, the development of SPRAVATO and the fact that that is now a blockbuster is a good leading indicator of where psychedelics could go. Let me just ask Guy and Lori to add their comments on that and how they see the field overall. Guy?
Yeah, thanks, Kabir. Yeah, I think one is very conscious of the interest and excitement in the academic sector. If one goes to conferences, indeed conferences where there are not just academics, but also clinicians, the sessions about psychedelics are always packed out. There is that interest, but frankly, there is not really the knowledge yet to see translation into real world. I think that is where I can segue to my colleague who can talk a little bit about that. The enthusiasm, no problem. Actually, the knowledge of how to use these drugs, that is the task. Lori's the best person to tell us how to address it.
Yeah. Hey, thanks, Guy. I agree with you, Paul. I think the landscape has definitely shifted, and there's probably a good reason. There's been quite a significant lack of innovation in the mental health space, especially in these difficult-to-treat patient populations for quite some time now. If you think about what the psychedelics are actually studying and generating data in, in some cases, like selfishly, TRD, for example, there are only two pharmacologic products approved right now to treat TRD. If you think about PTSD, again, only two that are approved. It's also been 30 years since the last one was approved or 30-plus years since the last one's been approved. Really, I think the psychedelic space and the data that these molecules are starting to generate is what's causing some of the excitement.
On top of that, let's not forget, I think the FDA also is starting to shift and become pretty excited about psychedelics as well. By my count, there are, what, five molecules being developed that have breakthrough therapy designation, which is fairly impressive in this space. As a reminder, breakthrough therapy designation helps advance the development and the speed to market on products that show promising data above and beyond what's available right now.
Yeah. Yep. Okay. Great. Maybe let's go through a couple of questions on the upcoming 005 readout. Maybe just set the stage on the design of this study and what are the couple of key questions you're trying to answer with this trial specifically?
This trial is a single administration of 25 mg of COMP360 compared with placebo. That is not a design we would have favored from the point of view of validity because it's quite likely to be unblinding. Nevertheless, it's one that we were obliged to do to get a true safety baseline from the placebo arm. Its primary outcome is at six weeks, and the primary outcome measure is the MADRS, which is assessed remotely so that the assessor is blind to the status of the patient. The trial continues out to 26 weeks as a blinded activity. During that time, patients are eligible for retreatment. We're going to get some sense of how optional retreatments can operate. Patients who elect not to have retreatment with the original dose, which is effectively the design, can elect to go on to antidepressants at that point.
They continue to the 26-week time at which all patients become eligible for an open-label 25 mg dose of COMP360. What we will get in that extended run follow-through is evidence of the tolerability of 25 mg open-label in patients who are taking antidepressants. We'll get some sense of open-label versus blinded, which is actually something that's almost never been done in this field. Of course, we'll get the full 52-week safety data that we think is also important for understanding and contextualizing the benefits of treatment with this drug. That's 005.
Yeah, makes sense. I know it's a different trial, right? But does the complete lack of placebo response in GH's phase 2b make you more confident in the odds of success in this study and further inform your view that the 005 study might be the easier of the two given, again, this point around functional blinding?
I think there are some unique features to the GH approach, first being that the primary outcome is defined at eight days. That seems a little short. A placebo response to those circumstances, particularly when patients know that they will get an active treatment almost immediately after it, is very much like a waiting list control rather than a true placebo comparison. The difficulty with, and indeed what we saw effectively, was a nocebo response in the placebo-treated patients. Because, again, of the design of multiple administrations on the day, patients will be, I think, will be 100% certain that they have received an inert placebo. The difficulty is that the active arm is difficult to interpret because it consists, we always think of active arms of consisting of the drug effect and a placebo effect.
Of course, we're not getting any control for the placebo component of the active arm. At some point, one is going to have to see a more valid placebo to evaluate the true strength of that effect observed in the existing study.
How are you handling antidepressant initiation in this trial? Understanding, again, that if me or someone I knew came into this study, they're expecting a psychedelic experience, they don't get one, they still feel awful, right? I mean, there's an ethical issue there. How does that play into this?
We are encouraging patients to stay with it through the six weeks, which is tough for some patients. It certainly will be. Obviously, if clinical need supervenes, they can have any treatment that is necessary. Patients give informed consent. They know that they are doing what is effectively an experiment, and we thank them very much for their altruism in that respect. When they qualify for having the option to go back onto antidepressants, we expect quite a number of them will, and we assume that will be particularly noticeable in the placebo arm. We have specified a list which gives a certain amount of freedom of choice. Of course, they can then continue on that drug, which may very well give some sort of partial response. It remains to be seen. That is the way we have handled it ethically.
We're trying to offer patients as early as it's practical the option of going back onto antidepressants. At the same time, we want them not to because as late as possible because we want a valid comparison at six weeks.
Right. Yeah. There is a way to handle that statistically if they do go on one. I know you've shown multiple analyses in the last trial. Can you just review that for people? This is still, I think, a point of confusion out there.
Yeah, I know. Our primary analysis last time did interpolate for patients going back onto antidepressants. In other words, if you went back onto an antidepressant, it was assumed you might improve due to it. So the scores were worsened systematically for patients who did go back onto antidepressants. At the end of the day, this did not actually change the primary comparison very much from if you did not do that. We have elected in both the new Phase III trials to simply treat patients in an intention-to-treat basis without interpolation. It will be easier for people to understand, but we accept that for those patients going back onto antidepressants, they may be weakening the effect that we would have seen in an ideal world. We think that is probably more sensible just because it is more understandable.
Yep. Yep. Okay. The other thing too, and I know you guys have done a good job of being very consistent about this, but it still never ceases to surprise me in how people end up being surprised about these things. You've been very specific about the top-line disclosure, what you're comfortable disclosing, and what you don't want to disclose, right? As I understand it, right, the holdback, right, is really related to you preserving the integrity of some of the other data we're going to get in the 06 trial and just overall maximizing the regulatory probability of success here. Can you expand upon that? Is my interpretation correct? Where does this kind of thought process stem from?
Yeah, I'll start, and Guy, please build on it. It is. I mean, it's preserving the integrity not only of the second trial, but indeed within 005 itself. As Guy said, the trials do run fully blinded all the way through to 26 weeks. Therefore, in making the determination to disclose the primary endpoint data only of 005, again, to your point, it will literally just be that primary endpoint. It will be the difference in the change from baseline between the arms with the associated p-value and confidence interval. The reason we're doing that is exactly to preserve the integrity of that blind on 005, but also to minimize any impact on 06.
As a reminder, for 006, the second study, which owing to its design with the three different doses, is one that we actually do believe addresses many of the agency's concerns around functional unblinding. For that, we will only be looking at data at 26 weeks, not at six weeks at the primary endpoint. That is exactly to preserve the integrity of that all the way through to 26 weeks because those claims in the second part of 06 are going to be material to our profile and our label.
Yeah. Yep. Okay. Fair enough. The other disclosure, right, that we're all looking forward to relates to safety in the top line and the whole question around suicidal ideation. I guess two questions there. One is, I think you've said that the DSMB is going to opine on their perspective as to whether or not there is a signal. I'd be curious kind of how you think about where the line is for the DSMB, given that there's small numbers, right? Like if there's two events on drug and one on placebo, is that a signal? Is that noise? Then just second, what's the team's confidence or where does the team's confidence come from that there's not a real signal here? This was more just kind of an artifact of noise and small numbers in the last trial.
Essentially, I think whenever one looks at an adverse event in which suicidality is implicated, and bear in mind, these are often quite complicated events. It might be an admission to hospital, and that admission to hospital may very well be due to worsening, and that worsening may include suicidality, but it may not be uniquely suicidality. It is essential that the DSMB look at the events in that way, in clinical judgment, taking each case in its own context and make a judgment about whether they think it seems to represent some specific risk to either the design of our study or to the drug that we are studying. Simple numbers will not do it. That is essentially the reason why we are requiring clinical judgment to evaluate.
The second part of your question, which is how will we get comfortable about the quality of risk, as it were, you have to go to something more sensitive than serious adverse events because they're too infrequent, and they're much more likely to be driven by simply chance. For that reason, we are systematically collecting data on a scale, the Columbia Suicidality Severity Rating Scale. We will use that to look at what are effectively sub-threshold changes in suicidality, which we can then relate to changes also in depressive symptoms. It will be equal scores, if you like, on those sub-threshold measures in all of the patient populations that will give us some confidence about the true underlying risk relating to dose or relating to the arm of the study.
Right. Right. Okay. Okay.
The only thing I would add is obviously the DSMB is conducting regular reviews of both studies. The most recent was February. If there had been any sign of an imbalance that they actually had clinical concern around, they would have told us by now, whereas hitherto it's just been proceed as planned.
Yeah. Great. Good clarification. I guess we'll look forward to getting more color there really soon. Maybe just broadly, in psychiatry, I feel like there's often a fixation in the investor community on the bar. Sometimes it frustrates me when it feels like for the market, there's a big emotional difference between a three-point MADRS difference and a four-point difference and a two-point difference when most of the prescribers of Prozac probably didn't realize the drug failed in more trials than it worked in. Nonetheless, these headlines matter and how they're reported and how they're described. When you think about a therapy like COMP360, right, which is certainly more complicated to use than Zoloft, do you feel like the efficacy bar for this to be an exciting commercial product is higher?
Like it has to have an elevated effect size that is materially better than average for an antidepressant?
I'll start and then hand to Guy and Lori. The only place I'll start is a reminder that we're talking about treatment-resistant depression, not MDD, yes? We're talking about people who have typically been failed by multiple drugs at this point. Therefore, I would argue that the bar in TRD is different from that in MDD. Guy?
Yeah, I think that's right. I mean, in truth, it's very tricky to extrapolate from Phase III trials to the real world and to patient satisfaction with the overall experience, etc., etc. It's hard to do. I don't think we should be setting the bars higher. I don't think that's reasonable. Just the nature of these trials, as we know, they're quite prone to failure. They're difficult to conduct. I think piling more pressure on to do better is unreasonable, frankly. I'm certainly not looking at our results in that way personally. I don't expect the field to, frankly.
Yeah. Okay. Fair enough. We'll get some more data soon. I think the other just kind of couple of development questions before I want to ask a few in the commercial to wrap things up, just as it relates to enrollment, right? You've given us color, right, later to Q4 for this first 005 readout. For 006, what can you say about just where you are and your comfort that that readout is not going to slip?
Yeah. No, it's a good question, Paul. What I would say is, while we haven't thus far disclosed numbers, we are well underway with recruitment and just clarification. It did start while 005 was underway. These are not sequential studies. 006 is already recruiting aggressively, not only in the U.S. and Canada, but in a whole host of European countries. As a reminder, 005 was U.S. only. 006, in terms of distribution, looks much more like 001, with around 50% U.S., 50% ex-U.S. Just by the nature of TRD, where the patients are, the medical systems in Europe, recruitment tends to be significantly quicker in Europe than it is in the U.S. We've also said as 005 wraps up imminently, the best-performing sites from 05 will move into 06 in the U.S.
They haven't had to stop screening in 005, so there's a bonus of patients there for 006. It's the same indication. For now, we are very confident on that second half 2026 guidance, recognizing it's a wide guidance, and we would hope over the course of the remainder of this year to narrow that down somewhat.
Yeah. Okay. Makes sense. And then the 06 study, how different is it from the Phase 2B?
Not very, except, of course, that it lasts for blinded up to 26 weeks and for full observational reasons up to 52. The design otherwise is very similar in that we're going for a comparison between 1, 10, and 25. The 10 is a psychoactive dose that we believe is unlikely to be efficacious, but therefore sits in and kind of keeps everybody uncertain about what the dosing is for any individual patient. There are two administrations of the drug, which is the key change. That means we're expecting a larger benefit. That was something that was predicted by the PIs, and many of the patients commented in 001 that they thought two treatments would help them. We're going to be very excited to see that. I think otherwise we made minimal changes to the entry criteria.
I suppose a little difference is that we're having a few more patients with previous experience with psychedelics and with psilocybin in particular. That has allowed us to stratify the randomization on that past history. We will get some sense of whether those patients are different in any way from the group who claim to be naive to the drug. Those are the very, very minor changes that we've made. The double treatment is the thing that's really interesting.
Yeah. Okay. Okay. Great. All right. Two commercial questions, or I guess one of them is sort of regulatory, sort of commercial, and then we can wrap up. Hopefully, it's okay with you if we go a minute or two over since we started a couple of minutes late. Just as it relates to the monitoring in the real world, can you talk about why the team is comfortable that you're not going to need a therapist in the real world for patient monitoring? When you think about the commercial scalability of this, how important is it in getting regulators on board globally that this is really not therapy-guided efficacy and that the therapy piece or the therapist piece is really just about caution, safety, monitoring, etc.?
Guy, do you want to speak about it from a clinical standpoint, and then I can certainly answer from a commercial standpoint?
Yeah. I think, I mean, I think obviously the need for a psychotherapist was driven by the FDA. In Europe, we are not necessarily using psychotherapists. We're using nurses in particular and other healthcare professionals. I think essentially we think that the key thing is the preparation for the experience. That's not something that we would want to diminish, but we don't see that as requiring a psychotherapist. That can be delivered in part with a really good psychoeducational package and also someone in person who is going to sit with the patient on the day. We think that relationship helps a good deal to make sure that patients feel safe and therefore are more able to relax and accept the experience. Those are the components that are important. How they're delivered, I know that Lori is thinking a lot about that.
Maybe I should hand over the thinking to her.
Yeah. Thanks, Guy. We are obviously learning a lot through what's happening in the clinical trials, and we're working with our collaboration partners as well to understand what will happen in a real-world setting. Given that, especially during the administration period, the therapist is really there just for safeguarding and monitoring. We feel pretty comfortable that we will only be limited to a healthcare provider during that time period. That can be obviously whatever the list is at the site's discretion. A good analog is SPRAVATO, right? SPRAVATO REMS just is limited to a healthcare provider being present during administration and available during the monitoring session. We feel like it will be a very comparable situation.
Yeah. Okay. I realize that a couple of minutes left does not do this question justice, Lori, so I'm sorry. I think the one most interesting commercial question that I regularly get just honestly relates to the economics. I mean, it feels like figuring out the economics and the reimbursement with SPRAVATO has been key to it taking off. I think that has been a big part of it. As it relates to COMP360, given the longer duration, how might the economic model here be similar or different? What are the implications for scalability?
Yeah. We agree. Provider economics is going to be important to the adoption of the product. Obviously, something we're spending quite a bit of time thinking about and working through. Again, we're kind of leaning on learnings from our collaboration partnerships that we have in place, as well as additional just commercial preparation work. If you think about SPRAVATO, best-case scenario for SPRAVATO is potentially three patients per day per room. If you think a two-hour administration session, turnover, administrative burden of scheduling, that is how their day kind of looks. Hours and some of the work that I was most impressed with when I came on board to Compass was the work that was done around getting the CPT III codes adopted and the adoption of those, obviously, after approval. Having those set up in place allows for the centers to bill for the administration and support time.
Effectively, if it's an eight-hour, six to eight-hour administration time period, the center can bill exactly the same time that they would take as cycling patients through. From a provider economics, we think it will be highly comparable. We also know SPRAVATO, especially the high-volume centers, are doing buy and bill. We don't see any reason to believe that they wouldn't do the same for COMP360.
Great. Makes sense. Thank you. I know we could easily talk for another half hour about more of your data and commercial and stuff, but I really appreciate it. We look forward to your results about two to three months away. Best of luck and thanks so much.
Thank you, Paul. Thanks very much.
All right. Good seeing you. Thank you. Thanks everyone for joining.