This open-label trial (n=15) evaluates the efficacy and safety of psilocybin (25mg) in veterans with severe treatment-resistant depression (TRD). It finds that 60% of participants met response criteria and 53% met remission criteria at 3 weeks post-treatment, with 47% maintaining response and 40% maintaining remission at 12 weeks.
Papers cited by this study that are also in Blossom
Aaronson, S. T., van der Vaart, A., Miller, T. et al. · JAMA Psychiatry (2024)
Background
The enduring and severe depression often suffered by Veterans causes immense suffering and is associated with high rates of suicide and disability. This is the first study to evaluate the efficacy and safety of psilocybin in Veterans with severe treatment-resistant depression (TRD).Methods 15 Veterans with severe TRD (major depressive episode failing to respond to ≥5 treatments, or lasting >2 years) received 25 mg of psilocybin. Primary outcome was change in Montgomery-Åsberg Depression Rating scale (MADRS) at 3 weeks posttreatment. Response was defined s ≥ 50 % reduction in MADRS, and remission as ≤10 MADRS score. Psychedelic experience was assessed using the Five-Dimensional Altered States of Consciousness scale (5D-ASC). Safety measures included assessment of suicidality and adverse events. Participants on antidepressants were tapered to avoid drug interactions.
Results
Of 15 participants, 60 % met response and 53 % met remission criteria at Week 3. At 12 weeks, 47 % maintained response, and 40 % remission. Co-morbid PTSD did not significantly influence study outcomes. The psychedelic experience reported in 5D-ASC did not correlate with response. Participants judged to need antidepressants were restarted and considered non-responders from that timepoint (n = 4). No unexpected adverse events occurred.
Limitations
Limitations include the small sample size, and the uncontrolled and unblinded nature of the study.
Conclusions
In this first study on psilocybin for Veterans with severe TRD, a surprising response and remission was seen. Many Veterans had PTSD though no moderating impact of response was observed. The degree of psychedelic experience did not correlate with depression changes. Further study is warranted.
Major Depressive Disorder (MDD) is a leading cause of disability worldwide and is especially prevalent and complex among US military Veterans, who have higher rates of depression, PTSD and multiple medical comorbidities compared with the general population. A substantial proportion of patients with MDD do not remit after standard treatments; treatment-resistant depression (TRD) is commonly defined as failure to respond to at least two pharmacologic trials, and among Veterans TRD is frequent, chronic, and associated with higher healthcare costs and worse outcomes. Psilocybin, a serotonergic 5-HT2A agonist, has shown preliminary antidepressant effects in civilian samples and in earlier open-label and some randomised trials, but Veterans—who often have greater chronicity, comorbidity and prior treatment exposure—have not been specifically studied. Ellis and colleagues therefore designed an initial, first-in-kind open-label pilot to evaluate the safety and preliminary efficacy of a single 25 mg dose of pharmaceutical-grade psilocybin (COMP360) with psychological support in US military Veterans with severe TRD, operationalised as failure of ≥5 treatments in the current episode or episode duration >2 years. The study aimed primarily to measure change in clinician-rated depression severity at three weeks, and secondarily to assess self-reported symptoms, functioning, safety and the relation of acute psychedelic experience to clinical response.
This was an open-label pilot enrolling 15 US military Veterans with severe TRD. Participants, investigators, raters and statisticians were not blinded and all received the active intervention (single 25 mg oral dose of synthetic psilocybin COMP360) together with manualised psychological support. The study was conducted at VA Palo Alto Healthcare System and Stanford University between November 2021 and February 2023; participants were recruited via VA providers, advocacy groups and ClinicalTrials.gov. The Institutional Review Board at Stanford and the VA research office approved the protocol and all participants provided informed consent. Inclusion criteria specified age 18–65, current VA care, a diagnosis of MDD meeting DSM-5-TR criteria, and severe TRD defined as either ≥5 failed treatments in the current episode (pharmacotherapy and somatic interventions counted; psychotherapy did not count) or episode duration >2 years, plus a Hamilton Rating Scale for Depression (HRSD-17) score ≥18. Key exclusions included current or past psychotic or bipolar disorder, current substance use disorder, recent suicidal behaviour, unstable medical conditions, recent psilocybin exposure, and pregnancy. Participants on contraindicated serotonergic medications were required to discontinue these under clinician supervision prior to dosing. Screening comprised structured diagnostic interview (MINI), HRSD, clinician-rated C-SSRS for suicidality, self-report measures (QIDS), physical exam, labs, urine drug screen and ECG. Preparatory psychotherapy included two 60-minute sessions and a 90-minute joint session the day before dosing with trained, certified therapists. Dosing occurred in a supportive, controlled environment: participants received 25 mg psilocybin in the morning, were monitored for 6–8 hours by two licensed therapists, and completed the Five-Dimensional Altered States of Consciousness (5D-ASC) scale that evening. Integration sessions were provided the next day (90 minutes) and at least two further integration visits occurred during follow-up. An optional neurocognitive/fMRI substudy was offered but its results are to be reported separately. The primary outcome was change in Montgomery–Åsberg Depression Rating Scale (MADRS) score from baseline to Week 3; response was defined as ≥50% reduction in MADRS and remission as MADRS ≤10 at Week 3, with sustained response/remission assessed at Week 12. Secondary outcomes included self-rated QIDS and the Sheehan Disability Scale (SDS). Safety outcomes comprised adverse events and suicidality (C-SSRS and clinical review). The investigators used mixed-effects models for repeated measures (MMRM) to analyse longitudinal change (chosen for small-N designs and handling missing data), including fixed effects for time, age, race, gender and comorbid PTSD, and a random subject effect. Observed-case analyses were reported, with participants who withdrew or restarted antidepressants counted as nonresponders from that timepoint; a sensitivity intent-to-treat (ITT) analysis including all observations was also performed. Correlations between 5D-ASC scores and MADRS change were examined using Pearson's r after confirming normality.
Bogenschutz, M. P., Forcehimes, A. A., Pommy, J. A. et al. · Journal of Psychopharmacology (2015)
Carhart-Harris, R. L., Giribaldi, B., Watts, R. et al. · New England Journal of Medicine (2021)
Carhart-Harris, R. L., Bolstridge, M., Rucker, J. et al. · Lancet Psychiatry (2016)
Davis, A. K., Barrett, F. S., May, D. G. et al. · JAMA Psychiatry (2021)
Garcia-Romeu, A., Griffiths, R. R., Johnson, M. W. · Current Drug Abuse Reviews (2015)
Goldberg, S. B., Pace, B. T., Nicholas, C. R. et al. · Psychiatry Research (2020)
Goodwin, G. M., Aaronson, S. T., Alvarez, O. et al. · New England Journal of Medicine (2022)
Griffiths, R. R., Johnson, M. W. · Journal of Psychopharmacology (2016)
Moreno, F. A., Wiegand, C. B., Taitano, E. K. et al. · Journal of Clinical Psychiatry (2006)
Pereira, L. · European Neuropsychopharmacology (2023)
Raison, C. L., Sanacora, G., Woolley, J. D. et al. · JAMA (2023)
Roseman, L., Nutt, D. J., Carhart-Harris, R. L. · Frontiers in Pharmacology (2018)
Sloshower, J. A., Skosnik, P. D., Safi-Aghdam, H. et al. · Journal of Psychopharmacology (2023)
Nielson, E. M., Tai, S. J., Lennard-Jones, M. et al. · Frontiers in Psychiatry (2021)
Schindowski, E. M., Jungwirth, J., Schuldt, A. et al. · EClinicalMedicine (2023)
Papers in Blossom that reference this study
Ellis, S., Bostian, C., Donnelly, A. et al. · Journal of Affective Disorders (2025)
Calnan, M., Blest-Hopley, G., Busch, C. et al. · Brain and Behavior (2025)
More than 200 people expressed interest; 15 Veterans were enrolled between March 2022 and February 2023 and completed follow-up through May 2023. One participant withdrew at Week 2; the authors report that sensitivity ITT analyses did not change the main findings. Primary outcome: baseline MADRS mean was 35.3 (±7.3). MMRM indicated an average MADRS decrease of −23.00 (SD 11.34) from baseline to Week 3 (t = −7.85; p < 0.001), with a large repeated-measures Cohen's d = 2.23. At Week 3, 60% (9/15) met response criteria and 53% (8/15) met remission. Effects persisted for many participants: at Week 12, 47% (7/15) remained responders and 40% (6/15) remained in remission. Overall MADRS scores were significantly lower than baseline across study timepoints (F(3,40) = 32.73; p < 0.001). The model found no meaningful influence of race, comorbid PTSD, gender or age on change in MADRS, and post-hoc comparisons suggested scores did not change significantly between the early post-dosing timepoints, consistent with an early sustained effect. Secondary outcomes: self-rated QIDS scores fell by an average of −12.10 (SD 5.28) from baseline to Week 3 (t = −8.87; p < 0.001) and were significantly reduced at each follow-up (F(3,30) = 38.94; p < 0.001), with no significant further change between Week 1, Week 3 and Week 12. Functional impairment as measured by SDS also decreased across time (F(3,26) = 13.65; p < 0.0001). The random-effects terms for age and race showed statistical associations with SDS, but the authors note wide individual variability and that zero fell within confidence intervals for the race effect, limiting interpretability. Exploratory analyses: the acute psychedelic experience (5D-ASC total score) was not significantly correlated with change in MADRS at Week 3 (r = 0.41, p = 0.15) or Week 12 (r = −0.13, p = 0.66), and no 5D-ASC subscales showed significant correlations with symptom change. Safety: there were no serious adverse events attributable to psilocybin or study participation. Common transient adverse events included mild-to-moderate headache in 7/15 participants (47%), nausea in 3 participants, dry heaving in 2, and mild back pain in 2; two participants experienced transient increased psychological distress during or immediately after dosing, without increased suicidal ideation or behaviour. One participant reported mild worsening of tinnitus that persisted after dosing. One participant had blood pressure outside the acceptable range prior to dosing but, after clinical consultation, was dosed and experienced no cardiac events. There were no increases in suicidal ideation across the follow-up period as measured by C-SSRS, although one participant had transient increased suicidal ideation during the antidepressant taper prior to dosing. Medication taper/restart: seven participants (47%) underwent medication taper; four of these subsequently restarted at least one antidepressant during follow-up (two before Week 1, one at Week 6, one at Week 9). The authors treated participants who restarted antidepressants as nonresponders from that timepoint in the primary analyses.
Ellis and colleagues interpret these data as preliminary evidence that a single 25 mg dose of psilocybin with psychological support can produce clinically notable reductions in depressive symptoms and improvements in functioning in a small sample of US Veterans with severe, chronic TRD. At the pre-specified primary endpoint of three weeks, 60% met response and 53% met remission; appreciable proportions (47% response, 40% remission) retained improvement at Week 12. The magnitude and durability of effects in this severely treatment-exposed Veteran sample were highlighted as notable, particularly because many participants had long-standing illness (mean reported duration ~13.5 years) and multiple prior failed treatments. The authors situate their findings relative to other psilocybin trials in civilian and mixed samples, noting similar response/remission rates in several open-label and smaller controlled studies, while recognising that their Veteran cohort had greater chronicity and prior treatment exposure. Contrary to some previous reports, they did not find a relationship between the intensity of the acute psychedelic experience (5D-ASC) and clinical outcome. Functional improvements (SDS) accompanied symptom change, suggesting potential benefits beyond symptom reduction, though the small sample size limits inference. Key limitations acknowledged include the small sample, absence of a control or placebo arm, and open-label, unblinded design, which may introduce expectancy and other biases. The investigators also note that classifying participants who restarted antidepressants as nonresponders may be overly conservative and that longer-term follow-up and larger, controlled trials are necessary to determine generalisability, durability and safety in this population. The authors emphasise the need to manage participant expectations carefully in future studies given the variable responses and some participants' lack of benefit. They report that 12-month follow-up and related neuroimaging findings will be presented in subsequent papers.