There is a moment in a psilocybin trial that never makes the press release. It happens roughly four hours after dosing, when the participant — eye shades on, weighted blanket pulled up, two therapists in chairs on either side — comes back to the room. Most of them don't speak right away. They sit up slowly. Someone hands them water. The clinical team writes notes. Then, three or six weeks later, a researcher in a different room opens a spreadsheet, and a Montgomery-Asberg Depression Rating Scale score either drops or it doesn't.
That number is what 2026 has spent the year arguing about.
Two trials, one hard question
In February, Compass Pathways reported topline results from COMP005, the first of two pivotal Phase 3 trials of its synthetic psilocybin formulation, COMP360, in treatment-resistant depression. A single 25 mg dose, paired with psychological support, produced a statistically significant reduction in symptom severity against placebo — a 3.6-point MADRS difference at week six, p<0.001. The second trial, COMP006, read out months later: two 25 mg doses three weeks apart, a 3.8-point difference against a 1 mg active comparator, and 39 percent of patients meeting a clinically meaningful response threshold of at least a 25 percent reduction in symptoms. Both trials reported a tolerable safety profile, with treatment-emergent adverse events mild or moderate and most resolving within 24 hours.
It was the largest body of Phase 3 data the field has ever produced.
One month after the COMP005 readout, the EPISODE trial landed in JAMA Psychiatry. A 2-center, triple-blind, active-placebo-controlled Phase 2b study out of Germany of 144 participants with treatment-resistant depression, EPISODE found that 25 mg of psilocybin with adjunct psychotherapy was associated with a clinically meaningful symptom reduction — but did not meet its primary outcome at the prespecified endpoint. The investigators also flagged a higher incidence of suicidal ideation on dosing days in the high-dose group compared to controls, the kind of safety signal that, in clinical research, is reported flatly and stripped of any reassuring varnish.
"Psilocybin, 25 mg, with adjunct psychotherapy, was associated with a clinically meaningful reduction in depressive symptoms in individuals with treatment-resistant depression, although findings did not show a significant effect on the primary outcome." — Authors, EPISODE trial, JAMA Psychiatry, March 2026
Two trials, one substance, one diagnosis, two different verdicts that aren't actually contradictory. That is the part most coverage missed.
What the numbers can and can't say
Treatment-resistant depression is, by the diagnostic definition the trials used, depression that has failed at least two adequate courses of standard antidepressants. The people enrolling in these studies are not curious wellness consumers. Many have been sick for years. The bar a new therapy has to clear isn't whether it works at all — it's whether it does something the existing options can't.
The COMP360 trials read as a meaningful "yes." The EPISODE trial reads as "something is happening, but our specific way of measuring it didn't quite catch it." Both are findings. Neither is the end of the story. Phase 3 programs typically require multiple successful trials, and Compass has said it expects 26-week data from COMP006 to read out in the second half of 2026, with a regulatory filing dependent on that follow-up holding up.
There is also the problem the field calls the blinding problem. When the active drug produces a several-hour altered state of consciousness, neither the participant nor the clinician can pretend not to know who got the real thing. Researchers have tried active-placebo designs, microdose comparators, and niacin controls. None of them fully solve it. JAMA Psychiatry published a companion paper alongside EPISODE specifically interrogating how unblinding shapes outcomes, which is a thing serious clinical science does and content marketing does not.
What the numbers also can't say is whether the drug is doing the work, or whether the drug is creating conditions in which the therapy does the work, or whether the eight hours of preparation and the two therapists in chairs and the weighted blanket and the integration sessions afterward are doing the work. The protocols braid all of these together. Unbraiding them is the next decade of research.
The view from the cheap seats
None of this is the version of the story you'll get from a stock alert or a culture piece. The stock alert wants a number that goes up. The culture piece wants a clean redemption arc — broken person takes mushroom, person is fixed. The actual finding, after fifteen years of methodical work by groups at Johns Hopkins, NYU, Imperial College London, MAPS, Usona, Compass and others, is more interesting than either version: a class of compounds long dismissed as cultural curiosities appears to do something real for a population modern psychiatry has been failing for decades, and the rigor of what we now know about that is hard-won and incomplete.
Real means measurable. Incomplete means careful. Both can be true in the same sentence.
The people pushing this research forward — the trial coordinators, the participants who sat with their hardest material for eight hours, the statisticians arguing about blinding, the regulators reading 200-page protocols — are not in it for vibes. They are doing the slow, unglamorous work of turning a thing people have believed about psychedelic compounds for sixty years into a thing that can or cannot be defended in a clinical document. That work is its own kind of rebellion. It is rebellion that shows up on time, takes notes, and cites sources.
If you know, you NOE.
Sources:
- JAMA Psychiatry — Efficacy and Safety of Psilocybin in Treatment-Resistant Major Depression: The EPISODE Randomized Clinical Trial
- STAT News — Compass says its psilocybin drug helped patients with severe depression in two trials
- Psychiatric Times — COMP360 Psilocybin for Treatment-Resistant Depression: Positive Phase 3 Efficacy Data
- Compass Pathways — Successfully Achieves Primary Endpoint in Second Phase 3 Trial Evaluating COMP360
- HCPLive — COMP360 Psilocybin Meets Primary Endpoint in Second Phase 3 Trial for TRD
- Science Media Centre — Expert reaction to JAMA Psychiatry papers on psilocybin and blinding in psychedelic therapy
This is editorial reporting and historical context. Nothing here is medical advice. Psychedelic compounds are controlled substances in most jurisdictions. If you're struggling, talk to a licensed professional.