Single-dose psilocybin for a treatment-resistant episode of major depression: Impact on patient-reported depression severity, anxiety, function, and quality of life

Treatment-resistant depression is associated with greater illness severity, longer duration, higher disability and suicide risk, and larger economic costs than treatment-responsive depression. Prior research, including pilot studies in patients with life-threatening cancer and small trials in major depressive disorder, has suggested that psilocybin — a tryptamine alkaloid found in psilocybe mushrooms — may have antidepressant effects. However, earlier work has been limited by small samples, single-site recruitment, crossover designs, and intensive concurrent psychotherapy that complicates attribution of effects to the drug itself. Goodwin and colleagues designed the present Phase II, dose-finding trial to identify an efficacious and acceptable dose of a proprietary, synthetic psilocybin formulation (COMP360) administered with time-limited psychological support, and to assess safety in adults with a treatment-resistant major depressive episode. The trial aimed to evaluate single doses of 25 mg and 10 mg against a 1 mg control, with the primary outcome being change in depressive symptoms at 3 weeks as measured by the MADRS (Montgomery–Åsberg Depression Rating Scale).

This was a Phase II, double-blind, parallel-group, randomised, dose-finding clinical trial conducted at 22 sites in 10 countries (Europe and North America) between March 2019 and September 2021. The sponsor, COMPASS Pathfinder, provided the COMP360 psilocybin formulation; trial operations and rating assessments were run by contract research organisations, and MADRS assessments were performed by remote raters blinded to trial details. Ethical approvals and informed consent were obtained at each site. Adults aged 18 years or older meeting DSM-5 criteria for a single or recurrent major depressive episode without psychotic features were eligible if they met an operational definition of treatment resistance (failure of two to four adequate antidepressant trials, each ≥8 weeks, per the MGH ATRQ). Recruitment was via clinical referrals and advertisements. A run-in period of 3–6 weeks included tapering and discontinuation of prohibited CNS-active medications at least 2 weeks before baseline and at least three preparatory therapy meetings. Participants were excluded if judged to be at clinically significant suicide risk (the extracted text notes such persons were excluded but does not provide precise criteria). Randomisation was 1:1:1 to single oral doses of 25 mg, 10 mg, or 1 mg psilocybin, stratified by country and prior psilocybin experience. Administration sessions lasted 6–8 hours in purposefully nonclinical rooms with two trained therapists present; participants used eyeshades and listened to a standardised playlist. Integration sessions occurred on day 2 and at week 1. Participants were requested to remain off antidepressants for the first 3 weeks but could restart medication if clinically necessary. Therapists completed a standardised training programme and did not conduct outcome ratings. The primary efficacy end point was change from baseline (day −1) to week 3 in the MADRS total score, administered by blinded remote raters at multiple time points up to week 12. Key secondary end points were response (≥50% reduction in MADRS at week 3), remission (MADRS ≤10 at week 3), and sustained response (week 3 response maintained through week 12). Safety assessments included adverse-event reporting coded by MedDRA, suicidality using the Columbia Suicide Severity Rating Scale, vital signs, laboratory tests, and 12-lead ECGs. The planned sample size (216; 72 per group) was calculated to provide 90% power to detect a 6-point difference in MADRS change at week 3 assuming SD 11. Efficacy analyses used a modified intention-to-treat set (participants who received treatment and had ≥1 postbaseline assessment). A hypothetical-strategy estimand imputed worsening MADRS scores for participants who initiated new antidepressant treatment or withdrew for lack of efficacy or adverse events; other missing data were handled as missing at random. The primary analysis used a mixed model for repeated measures (MMRM) with treatment, visit, pooled site, treatment-by-visit interaction and baseline MADRS; Rubin's rules combined estimates across imputations. Response and remission were analysed with generalized linear mixed models and sustained response with logistic regression. A hierarchical testing procedure controlled type I error across primary and the three key secondary end points (25 mg then 10 mg comparisons sequentially). Safety analyses were descriptive on the safety analysis set (all randomised participants who received study drug).

Of 428 individuals screened, 233 participants were randomised and received a single dose of psilocybin (safety and modified intention-to-treat sets): 79 in the 25-mg group, 75 in the 10-mg group, and 79 in the 1-mg group. By week 12, withdrawals were 5 (6%) in the 25-mg group, 9 (12%) in the 10-mg group, and 10 (13%) in the 1-mg group. Baseline characteristics were similar across groups: mean age 39.8 years, 52% female, 92% White, 95% with prior depressive episodes (mean 6.9 lifetime episodes), and 86% with current episode duration >1 year. Two thirds were taking antidepressants at screening, and mean baseline MADRS scores were approximately 32 across groups. Prior exposure to psilocybin was reported by 6% of participants. Primary efficacy: the least-squares mean change in MADRS from baseline to week 3 was −12.0 points (25 mg), −7.9 points (10 mg), and −5.4 points (1 mg). The adjusted difference for 25 mg versus 1 mg was −6.6 (95% CI −10.2 to −2.9; P<0.001). The 10 mg versus 1 mg comparison yielded a difference of −2.5 (95% CI −6.2 to 1.2; P=0.18). Because the 10 mg comparison was not significant, the prespecified hierarchical testing halted, and the trial did not claim statistical significance for subsequent secondary end points. Secondary and other outcomes: the incidence of response at week 3 was 37% in the 25-mg group, 19% in the 10-mg group, and 18% in the 1-mg group (odds ratio 25 mg vs 1 mg 2.9; 95% CI 1.2 to 6.6). Remission at week 3 occurred in 29% of participants in the 25-mg group versus 9% and 8% in the 10-mg and 1-mg groups respectively (odds ratio 25 mg vs 1 mg 4.8; 95% CI 1.8 to 12.8). Sustained response through week 12 was observed in 20% (25 mg), 5% (10 mg), and 10% (1 mg); the odds ratio for 25 mg versus 1 mg was 2.2 (95% CI 0.9 to 5.4), with the confidence interval including 1. Sensitivity, per-protocol, and post hoc analyses adjusting for sex or number of lifetime episodes produced results consistent with the primary analysis. Safety: adverse events were reported in 84% of participants in the 25-mg group, 75% in the 10-mg group, and 72% in the 1-mg group. On the day of dosing, the most frequent events in the 25-mg group were headache (24%), nausea (22%), dizziness (6%), and fatigue (6%). Severe adverse events on day 1 were uncommon (4% in 25 mg). From day 2 to week 3, severe adverse events occurred in 9% (25 mg), 7% (10 mg), and 1% (1 mg). Serious adverse events included suicidal ideation and intentional self-injury in the higher-dose groups; no serious adverse events were reported in the 1-mg group during day 2 to week 3. After week 3 up to week 12, serious events included suicidal behaviour in three participants in the 25-mg group; these three had prior histories of suicidal behaviour or nonsuicidal self-injury and did not show a week-3 treatment response. At baseline, passive or active suicidal ideation was present in 27% (25 mg), 36% (10 mg), and 24% (1 mg). Worsening of suicidal state from baseline to week 3 occurred in 14% (25 mg), 17% (10 mg), and 9% (1 mg). No clinically significant or consistent changes were observed in vital signs, clinical laboratory tests, or 12-lead ECGs according to the extracted text.

Goodwin and colleagues report that single-dose psilocybin given with a manualised, time-limited psychological support package was feasible in a treatment-resistant major depressive episode population followed for 12 weeks. A 25-mg dose produced a statistically significant greater reduction in MADRS score at the prespecified primary time point of week 3 compared with a 1-mg control; the 10-mg dose did not differ significantly from 1 mg. The authors note that secondary end-point results (response and remission at week 3) were generally directionally consistent with the primary outcome but that hierarchical testing rules prevent definitive claims for these endpoints, and sustained response at week 12 did not show a clear statistically significant advantage for 25 mg. In discussing safety, the investigators emphasise that adverse events such as headache, nausea, dizziness and fatigue occurred more commonly with higher doses, and that suicidal ideation, self-injury and suicidal behaviour were observed in all dose groups but were numerically more frequent in the higher-dose groups. The authors underline that some participants worsened in suicidal state, and therefore suicidality requires careful clinical vigilance in future trials. The trial was designed to address limitations of earlier studies by using a parallel-group design, multicentre recruitment, remote blinded raters, and limiting psychological support to a structured preparatory and integration protocol rather than ongoing psychotherapy. The authors observe that the week-3 response rate for 25 mg (37%) is numerically lower than response rates reported in some first-line antidepressant trials but higher than rates seen in later-line treatments in the STAR*D sequence. Pharmacokinetic and receptor-occupancy data providing dose-dependent subjective effects are offered as a possible explanation for the dose–response pattern seen here. Limitations acknowledged by the investigators include the lack of an active comparator, limited ethnic diversity of participants, exclusion of people judged at clinically significant suicide risk, and inherent difficulties in maintaining blinding because of the distinctive acute subjective effects of psilocybin (the trial did not assess participants' ability to guess dose assignment). They also note that it is unknown whether other psilocybin preparations would yield similar results. The authors conclude that larger and longer trials, including comparisons with existing treatments, are required to determine the efficacy and safety of psilocybin for treatment-resistant depression.