A rigorously designed trial delivers a negative result
STUDY: Mertens LJ et al, JAMA Psychiatry 2026;83(5):448–460
STUDY TYPE: Randomized controlled trial
FUNDING: Federal Ministry of Education and Research (Germany)
Background
Psilocybin is a serotonin 2A receptor agonist that, paired with psychotherapy, has shown antidepressant effects in earlier trials. Most of those trials were small and methodologically weak. This trial set out to do it right — triple-blind, active placebo-controlled, conducted at two German university hospitals.
The Study
- 144 adults with treatment-resistant depression, off antidepressants, average illness duration 14 years
- Randomized to psilocybin 25 mg, psilocybin 5 mg, or nicotinamide (active placebo), with 14 hours of structured psychotherapy
- Primary outcome: 50% reduction on the Depression scale (HAMD17) at 6 weeks
At 6 weeks, response rates were 17% with psilocybin 25 mg, 13% with psilocybin 5 mg, and 11% with nicotinamide — not a significant difference. The primary endpoint was negative.
Secondary outcomes showed positive signs. The HAMD17 score dropped 4.6 points more with psilocybin 25 mg than nicotinamide at week 6, a clinically meaningful difference. At week 1, response was 34% with the high dose versus 6% with placebo, suggesting a rapid but fading effect.
Side Effects
Most adverse events were limited to the dosing days. Suicidal ideation (4% on psilocybin vs 1-2% on comparison treatment). One patient developed hallucinogen persisting perception disorder.
Limitations
The trial was underpowered for the primary outcome because researchers overestimated the treatment effect. Most participants (86%) correctly guessed they had received the high-dose psilocybin, making true blinding impossible. The sample was almost entirely White and socioeconomically homogeneous, with no follow-up beyond 12 weeks.
Practice Implications
- This is the most rigorous psilocybin trial in treatment-resistant depression to date, and it came up short on the primary outcome.
- Most trials generate positive secondary outcomes, usually due to random scatter in the data, so these are not reliable.
- Hallucinogen persisting perception disorder is a known risk, and occurs in approximately 4.2% of people who these compounds, particularly with LSD.
— Chris Aiken, MD
Director, Psych Partners
Editor in Chief, Carlat Psychiatry Report
