Hardman: It would indeed be interesting if one of the MEQ subdomains was predictive of antidepressant effect. Does the fact that both the 8 mg and 12 mg group performed similarly complicate the dose-response story? In Compass Pathways’ Phase 2b, for example, they had a clear dose-response relationship where the highest dose led to the largest antidepressant effects. Presumably, in your Phase 3 program, one of your studies is going to have three doses again, so what might the middle dose look like?

Feilding-Mellen: I’ll let Rob talk about the phase three design, but I think we were in a very different position to Compass when we designed this. For Compass, there was an enormous amount of published data on psilocybin; the 25 milligram dose had been identified as an effective, well-tolerated dose. So, there was a hypothesis that the middle arm was really used to show a dose response.

Because there’s so much less that’s known about 5-MeO-DMT, we really did want to give ourselves two chances of finding an effective dose. So that’s why we chose 8 mg. The Phase 2a was 10 mg, so we kind of split the difference with the Phase 2b and went with 8 and 12 mg.

Ultimately, we’ve shown that 8 is a really good fit. I think what’s really exciting is that with both doses they were well tolerated and showed efficacy, and statistically significant, clinically meaningful reductions from the day after dosing all the way out to eight weeks of blinded follow-up.

That was, I think, a major question in this whole field that we’ve gone a long way to answering: Can you replicate the durability of effect that the longer psychedelics have? If anything, we’ve shown we have at least as good, if not better, durability than psilocybin. So that’s really, really exciting.

Conley: You were asking about the Phase 3 design, too.

Obviously, we’re going to work with the FDA, and less directly with the EMA, on that front. But from the standpoint of dosing, as Cosmo pointed out, Compass was doing a very different thing. 10 versus 25 mg was just a much larger difference than 8 versus 12 mg.

But other than that, the FDA really looks at a couple of things. One is blinding. The 0.3 mg dose was actually there to help with blinding, you’re going to still get some physical and physiologic effects. It’s not a psychedelic dose, but it’s not a dose without effect.

So I think we sort of answered the question about blinding. Now, of course, that’s my opinion. I think it depends on what the agency and others think about our blinding, but I think there’s a pretty good chance they’ll be more interested in us accruing a lot of placebo data, because placebo data is always how you get a better side effect profile.

I’ve had other programs during my work at [Eli] Lilly where, if the drug has a pretty low side effect profile, and by my judgement this does, the FDA might say: ‘Okay, you don’t need a lower dose, because in the clinic you’re not going to ever want to give a lower dose.’

The other reason they want it is for blinding, because then dose ranging helps suggest what is due to the drug.

So there’s a couple of balls up in the air about that one. We’ve planned for a scenario where we need a lower dose, but obviously we’d like the most efficient studies possible. If we all come to an agreement that we can just go with one dose into the clinic, or one dose with a couple of administrations; the fewer doses the better in Phase 3.

So that’s kind of where we’re at. We’re going to work with the agency to figure that out.

Feilding-Mellen: One of the things we were really pleased about was that the 0.3 mg, kind of active comparator arm, showed a reduction on MADRS. That is what you want to see, it shows a properly controlled study. It very much is comparable to what Compass showed in the Phase 2b with their 1 mg dose. More importantly than that, it’s very similar to what Spravato showed in their studies with the placebo.

So, all that’s suggesting we really do have a properly controlled, blinded study as well.

Hardman: Did you administer any blinding integrity questionnaires in this study?

Conley: We did, yeah. We’re analyzing that now, so I don’t have the final word on that, but I can tell you, we did fool some people.

I think we’ll see something that’s pretty similar to what you see with antidepressants, with a daily oral medicine, when they’re asking those blinding questionnaires, typically, about a quarter or third of the people on placebo guess they were on drug. So, 70% think they’re on placebo, but some are fooled.

That’s different than, for example, GH [Research]. They did three doses in one day, and the people on placebo were asked every time: ‘Do you need another dose?’ 100% of them said yes. So that’s kind of evidence they did guess that they weren’t on the drug.