Psilocybin has had an unusually good press for a drug still in trials. A study published in JAMA psychiatry in 2026 is a useful corrective, not because it shows the treatment fails, but because it is the kind of result that rarely makes it into the coverage: a well-designed trial that did not work out, reported plainly by researchers who call it inconclusive themselves. [pubmed-depression-jun23-2026-source]
What the researchers did
The EPISODE trial recruited 144 adults aged 25 to 65 with treatment-resistant depression from two outpatient centres in Germany. Participants were withdrawn from their antidepressant medication before starting.
Everyone received two dosing sessions six weeks apart, embedded in psychotherapeutic sessions, and was randomly assigned to receive 25mg of psilocybin, 5mg, or nicotinamide, an inactive vitamin used as the placebo. Investigators, participants and raters were all blinded.
The trial’s stated main test was the proportion of people whose depression score halved by week six, before the second dose.
What they found
Response rates were 17.0% on 25mg of psilocybin, 12.5% on 5mg, and 10.6% on the placebo.
Response rates as reported in the EPISODE trial (Mertens and colleagues, 2026), JAMA psychiatry. The difference between the 25mg group and the placebo group did not reach statistical significance.
The ordering is what you would hope for: more drug, more response. But the gap between the full dose and the placebo was not statistically significant, meaning a difference that size could plausibly have arisen by chance in a trial of this size.
Because the researchers had committed in advance to a fixed order of tests, and the first one failed, formal testing stopped there. Everything reported afterwards is exploratory, including the secondary measures that did suggest a clinically meaningful effect on symptom scores.
That distinction is not a technicality. It is the difference between a trial that demonstrated something and a trial that produced a hint worth chasing.
The safety signals
Adverse events clustered on dosing days, as expected for a drug with acute effects. Two are worth naming.
Suicidal ideation was reported on around 4% of psilocybin dosing days, against 1 to 2% in the comparison conditions. In a treatment-resistant depression population these numbers are all low and all difficult to attribute cleanly, but the direction is the wrong one and it should not be edited out.
Two serious adverse reactions followed the 25mg dose, one of which was hallucinogen persisting perception disorder: visual disturbances that continue long after the drug has cleared. It is rare and it is not trivial.
Most participants tolerated the treatment without difficulty. Both things are true.
What is genuinely striking
Look again at the response rates. Fewer than one in five people responded on the active drug, and roughly one in ten on placebo.
That is what treatment-resistant depression looks like from the inside of a trial. These were people for whom the standard treatments had already failed, taken off their medication for the study, given an intensive psychotherapeutic protocol alongside the drug, and most of them did not improve by the trial’s definition.
It is a sobering number and it is the context in which any new treatment for this group has to be judged. The bar is low because the problem is hard, and a treatment that reliably moved these numbers would matter enormously.
How to read it alongside the rest of the field
Earlier psilocybin trials have reported more encouraging results, which raises the obvious question of why this one did not.
Several features make it a harder test than some of its predecessors: genuinely treatment-resistant participants, withdrawal from existing antidepressants, a triple-blind design, and a comparison that included a low active dose rather than only an inert pill. Harder tests produce smaller effects, and that is usually a sign the harder test is closer to the truth.
The reasonable position is that psilocybin remains a plausible candidate whose evidence base is more mixed than its reputation, and that the next few trials matter more than the coverage of any single one.
If you are struggling now
Treatment-resistant depression is exhausting in a specific way: the sense that the options have been used up. They usually have not. Combinations, augmentation strategies, established brain stimulation treatments and structured psychotherapy all remain available, and a second opinion from a specialist service is a reasonable thing to ask for.
If you are having thoughts of suicide, contact your local emergency services or a crisis helpline now.
The source
These findings are drawn from “Efficacy and Safety of Psilocybin in Treatment-Resistant Major Depression: The EPISODE Randomized Clinical Trial” (Mertens LJ, Koslowski M, Betzler F, et al., 2026), published in JAMA psychiatry. Read the full study on PubMed.