Single-dose psilocybin-assisted therapy in major depressive disorder: A placebo-controlled, double-blind, randomised clinical trial

Interest in the therapeutic potential of psilocybin for major depressive disorder (MDD) has risen alongside recognition of limitations of current antidepressant treatments and reports of rapid, durable antidepressant responses following psilocybin administration. The authors note that many earlier studies suffered from small samples, open or waitlist comparators, likely functional unblinding of raters, and incomplete adverse event (AE) assessment, and that several larger trials have used short primary end points that leave questions about longer-term clinical utility in a chronic condition such as MDD. Raison and colleagues set out to address these gaps with a randomized, double-blind, multi-site Phase II trial comparing a single oral 25-mg dose of psilocybin to an active placebo (100-mg niacin) administered with standardised psychological support. The study aimed to characterise the timing, magnitude, durability (to six weeks), and safety of antidepressant effects while reducing rater unblinding through centralised remote assessments.

This was a randomized, two-arm, Phase II clinical trial conducted at 11 US sites between December 2019 and June 2022. Eligible participants were medically stable adults aged 21–65 meeting DSM-5 criteria for current MDD of at least 60 days’ duration, with a central rater-assessed Montgomery–Åsberg Depression Rating Scale (MADRS) score ≥28 at screening and limited improvement during a 7–35 day period to permit medication tapering where needed. Key exclusions included personal or first-degree family history of psychosis or mania, moderate/severe substance use disorder, recent psychedelic use (past 5 years or >10 lifetime uses), active suicidal intent, or recent suicidal behaviour. There were no exclusions for number of prior episodes or most prior treatments, although a history of deep brain stimulation or vagus nerve stimulation was exclusionary. Participants were randomised 1:1 using permuted blocks stratified by site to receive a single 25-mg oral dose of synthetic psilocybin or 100-mg niacin in identical capsules. Blinding included participants, site personnel, sponsor, central outcome assessors, and statisticians. To standardise ‘‘set and setting,’’ both groups received a protocolised psychological support package: 6–8 hours of preparatory sessions, a 7–10 hour supervised dosing session with facilitators (encouraged use of eyeshades and a playlist), and 4 hours of postdose integration. Lead facilitators were doctoral-level clinicians and co-facilitators had minimum bachelor-level mental health training; all completed study-specific training. The primary efficacy outcome was the between-group difference in mean change in central rater-assessed MADRS from baseline to day 43. The key secondary outcome was change in MADRS from baseline to day 8. Centralised remote raters conducted MADRS via telephone to reduce functional unblinding. Additional secondary outcomes included change in Sheehan Disability Scale (SDS) from baseline to day 43 and proportions with sustained response (≥50% MADRS reduction at days 8, 15, 29, 43) and sustained remission (MADRS ≤10 at those same visits). Safety assessments captured adverse events from enrollment through day 43; solicited events included suicidal ideation (C-SSRS), elevated vitals requiring medication, headache, nausea, and visual perceptual effects. The trial was powered for the day 43 primary end point (sample size target 100) after an FDA-recommended change from an initial day 8 primary end point. The intent-to-treat (ITT) population included all randomised participants analysed by randomised group; a per-protocol (PP) set excluded major protocol violators. Primary analyses used mixed-effects models for repeated measures adjusted for baseline score, site, sex, and treatment-resistant depression (TRD) status, with time categorical and a treatment × time interaction, and without imputation; multiple imputation sensitivity analyses were also performed. Sustained response/remission were analysed with logistic regression adjusted for sex and TRD. A sequential testing procedure controlled the family-wise alpha for primary and prespecified secondary end points.

Among 1,529 prescreened respondents, 347 consented, and 104 participants were randomised and received study drug (51 randomised to psilocybin, 53 to niacin); due to one dosing error, 50 participants received psilocybin and 54 received niacin. Median time from enrollment to randomization was similar across groups. Baseline characteristics in the ITT population showed mean ages of about 40–42 years, roughly half of participants were male, and the sample was predominantly White (89%). Prior lifetime psychedelic use was reported by 19.6% in the psilocybin group and 24.5% in the niacin group. Median current episode length was long in both groups (median weeks 53 vs 81), and about 12.5% of the ITT sample met pre-specified criteria for TRD. By day 43, attrition differed by group: 1 participant in the psilocybin group and 9 in the niacin group had withdrawn or been lost to follow-up; three participants in each group initiated antidepressant medication before day 43. No participant withdrew because of an adverse event. On the primary efficacy outcome, psilocybin produced significantly larger reductions in depressive symptoms than niacin: mean difference in MADRS change from baseline to day 43 was −12.3 points (95% CI −17.5 to −7.2; P <.001). The key secondary end point (baseline to day 8) showed a similar effect (mean difference −12.0; 95% CI −16.6 to −7.4; P <.001). These effects were consistent at intermediate visits (days 15 and 29) and in multiple-imputation sensitivity analyses. More participants receiving psilocybin achieved sustained depressive symptom response across the postdose visits; the extracted text does not clearly report the ITT counts for sustained response, but in the PP analysis sustained remission was significantly more common with psilocybin (12/44, 27%) than niacin (3/36, 8%); adjusted absolute difference 18.9% (95% CI 3.0%–34.9%), OR 4.0 (95% CI 1.0–15.7; P = .04). Exploratory outcomes favoured psilocybin on global disease severity, self-reported depressive and anxiety symptoms, and quality of life, while no effect was seen on measures of emotional blunting. Safety analyses (by treatment received) showed more adverse events in the psilocybin group. Through day 43, 44/50 participants (88%) receiving psilocybin and 33/54 (61%) receiving niacin reported at least one AE. Three serious AEs occurred prior to randomization (nephrolithiasis, obstructive incisional hernia, and appendicitis). From day 1 to day 9, 41/50 (82%) psilocybin participants experienced at least one drug-related treatment-emergent AE (TEAE) versus 24/54 (44%) niacin participants (difference 38%; relative incidence 1.8). Severe related TEAEs through day 9 were reported by 4/50 (8%) in the psilocybin group (including migraine, headache, illusion, panic attack/paranoia) and 0 in the niacin group. Solicited events were mostly mild; the most common were headache (66% psilocybin vs 24% niacin; difference 42%; RI 2.7) and nausea (48% vs 6%; difference 42%; RI 8.6). Visual perceptual effects were reported by 22/50 (44%) on the day of dosing and resolved by study end. No suicidal or self-injurious behaviour was observed; instances of suicidal ideation were passive and increases in C-SSRS score from baseline to end of trial occurred in 1 psilocybin and 5 niacin participants. No clinically significant vital sign or laboratory abnormalities were observed.

Raison and colleagues interpret their findings as showing that a single 25-mg dose of psilocybin administered with standardised psychological support produced statistically and clinically meaningful reductions in depressive symptoms compared with an active niacin placebo, with benefits evident by day 8 and maintained through the 6-week primary end point. The between-group difference of 12.3 MADRS points exceeds commonly cited thresholds for substantial clinical improvement and is larger than active-placebo differences reported in some prior depression trials. Improvements were also seen in functional disability and a range of exploratory measures of disease severity, anxiety, and quality of life, and psilocybin did not produce the emotional blunting sometimes reported with conventional antidepressants. In terms of safety, the investigators report that psilocybin was generally well tolerated but associated with more overall, solicited, and severe TEAEs than niacin, most occurring during or shortly after the dosing session; no serious TEAEs or clinically significant laboratory or vital-sign changes were observed. The authors note that psychedelics may produce adverse effects not fully captured by standard scales and that new psychiatric syndromes could conceivably be induced even as target symptoms improve. The paper acknowledges several limitations. Allocation blinding was not formally assessed and functional unblinding from acute psychoactive effects likely occurred despite use of centralised remote raters; this may have influenced outcomes. Use of niacin as an active placebo could have increased comparator believability and influenced placebo response, although placebo responses here were smaller than typical daily-pill trials. The six-week primary end point, while longer than many prior trials, does not address longer-term durability; the relatively low proportion of participants meeting TRD criteria in this sample may also affect generalisability. Fidelity to the psychological support protocol was not formally assessed, leaving open the possibility that facilitator variability contributed to outcomes. Finally, the sample lacked ethnic and racial diversity, limiting external validity to underrepresented groups. The authors emphasise the need for larger, longer-term, and more diverse trials and for studies that better characterise the contributions of psychological support and placebo-related factors.

In this randomised, double-blind Phase II trial, a single 25-mg dose of psilocybin given with protocolised psychological support led to greater reductions in depressive symptoms and improvements in functional disability than 100 mg niacin placebo at six weeks, without any serious treatment-emergent adverse events. Psilocybin was associated with increased rates of mostly acute, expected adverse events relative to niacin. The investigators conclude that, when administered within a structured psychosocial programme, psilocybin may hold promise as a novel treatment for MDD, while underscoring the need for further research to define longer-term efficacy, safety, and applicability across diverse populations.