Psilocybin-assisted cognitive behavioral therapy for major depressive disorder: A pilot trial

Psilocybin-assisted therapy has shown antidepressant effects in major depressive disorder, but earlier studies have largely focused on the drug component and have provided limited detail about the psychotherapy used alongside psilocybin. The authors argue that this leaves uncertainty about what the adjunctive psychological treatment is doing, how standardised it is, and whether a manualised, evidence-based psychotherapy might improve acceptability, durability, and dissemination of treatment. They also note that prior psilocybin studies often used supportive, non-directive psychotherapy and rarely reported manuals or clear therapeutic orientation. Weintraub and colleagues therefore set out to test psilocybin-assisted cognitive behavioural therapy (PA-CBT) in adults with active major depressive disorder symptoms. Their aims were to assess feasibility and acceptability, and to obtain preliminary data on effects on depressive symptoms, psychosocial functioning, emotion regulation, and cognitive schemas over 7 months. They hypothesised low attrition, favourable ratings from participants and clinicians, and improvements in the clinical and psychological outcomes measured.

This was an open, single-group pilot trial conducted at the University of California, Los Angeles Semel Institute between June 2023 and August 2025. Participants were recruited from the greater Los Angeles area and had to be aged 21-60, meet lifetime criteria for major depressive disorder, and have current active depressive symptoms above the stated Hamilton Depression Rating Scale threshold. The extracted text indicates several exclusions, including active suicidality, hypertension or tachycardia, clinically significant ECG abnormality, serotonergic medication use, Cluster B personality disorder, current CBT, recent substance use disorder, personal or family history of bipolar or psychotic disorder, significant cardiac or neurological history, and psychedelic use within the prior 12 months. Eligibility was assessed in stages: phone screening, medical screening with physical examination, ECG, blood draw, urine drug testing and pregnancy testing where relevant, followed by psychiatric diagnostic interviews using the SCID-5 and a family history screen, and finally physician confirmation of eligibility. The study team consisted mainly of licensed psychologists or psychiatrists, with one master's-level co-therapist/monitor. One experienced CBT clinician served as the primary therapist for each participant. PA-CBT was delivered over 4 months as 12 individual 1-hour psychotherapy sessions, arranged as 9 weekly sessions followed by 3 biweekly sessions. Psilocybin was administered twice, at 10 mg after session 3 and 25 mg after session 6. The psychotherapy was adapted from previous psilocybin-assisted therapy protocols and the Unified Protocol of CBT for emotional disorders, and it was manualised. Early sessions focused on preparation, psychoeducation about psilocybin, mindset and intentions, and guided imagery; mid-treatment sessions introduced CBT theory and behavioural skills such as behavioural activation and pleasant events scheduling; later sessions focused on cognitive skills, cognitive reappraisal, and relapse prevention planning. An additional booster session could be offered if more time or support was needed. The psilocybin sessions lasted 6-8 hours and followed standard supportive procedures: participants lay on a couch with eyeshades, listened to a preset music playlist through headphones, and were continuously monitored by two clinicians, one of whom also delivered the psychotherapy. Vital signs were checked hourly, and a study physician was on-call throughout. Treatment adherence was checked with a CBT component checklist; 32 sessions were rated, inter-rater reliability was high (κ = 0.88), and adherence was reported as 96.9%. Primary assessments occurred at baseline, post-treatment at 4 months, and 3-month follow-up at 7 months. Feasibility was based on screening, consent, enrolment, and retention figures. Acceptability was measured with the Client Satisfaction Questionnaire-8 and an internally developed participant/clinician questionnaire. Safety was assessed through adverse events coded using CTCAE categories. Depressive symptoms were rated with the HAM-D and PHQ-9, mood states with the Profile of Mood States, and functioning with the Global Assessment of Functioning and Clinical Global Impressions scales. Exploratory outcomes included the Difficulties with Emotion Regulation Scale and the Brief Core Schema Scale. The authors used repeated-measures ANOVA for symptom, mood, and functioning outcomes, and mixed models with random participant intercepts to examine associations between changes in emotion regulation, schemas, and depressive symptoms.

Eighty-four people completed phone screening, 42 consented to further screening, and 16 met full eligibility criteria and were enrolled. The sample included 7 women and 9 men, with a mean age of 43.3 years. All 16 participants were retained through the 7-month study. One participant (6.3%) did not complete the second drug session because of a medical event that the authors state was unrelated to the study. All participants completed psychotherapy, with a mean of 12.4 sessions. Treatment satisfaction was very high: the mean CSQ-8 score was 30.8 out of 32. Participants and clinicians generally rated the programme components positively, although some participants felt there were too few CBT sessions. Three recurring suggestions for refinement were reported: starting CBT content earlier in the preparation phase, including more acceptance-based cognitive skill practice, and delaying goal-setting until after the first psilocybin session. Adverse events were mostly mild. The most common were headache during or immediately after 54.8% of psilocybin sessions, nausea in 35.4%, and gastrointestinal discomfort in 22.5%. There were no serious adverse events, and reported events resolved within a few days. Depressive symptoms improved significantly. On clinician-rated HAM-D, the repeated-measures analysis showed F(2,30) = 29.7, p < 0.001, with Hedges' g = 2.7. On participant-rated PHQ-9, F(2,28) = 14.3, p < 0.001, with Hedges' g = 1.9. Improvements were significant from baseline to post-treatment and were maintained at 3-month follow-up. At post-treatment, 10 participants (62.5%) had a large response on HAM-D, 9 (56.3%) met remission criteria, 3 (18.8%) had a moderate response, and 3 showed minimal or no improvement. Global functioning also improved significantly and remained improved at follow-up: GAF F(2,30) = 22.7, p < 0.001, g = 2.2; CGI F(2,30) = 18.6, p < 0.001, g = 2.1. Mood ratings across five time points showed a significant within-subjects effect for depressive mood, anxiety, and self-esteem. The authors describe a pattern in which mood and anxiety improved sharply after each psilocybin session, then partially increased by post-treatment but remained better than baseline, with stability through follow-up. Self-esteem showed the opposite trend. Exploratory measures also improved: emotion regulation improved on the DERS, and both positive and negative self-schemas improved on the BCSS. Positive schema change was greater than negative schema change. In mixed models, better emotion regulation and more positive or less negative self-schemas were significantly associated with lower HAM-D scores over time.

The authors interpret the trial as showing that CBT can be a feasible, acceptable, and potentially beneficial adjunct to psilocybin for major depressive disorder. They emphasise the high retention, high satisfaction ratings, and generally positive feedback from both participants and clinicians, who mostly felt the number of sessions was adequate and that the drug and psychotherapy components worked well together. In their view, the findings suggest that PA-CBT was associated with meaningful improvements in depressive symptoms and functioning across treatment and 3-month follow-up, with minimal attrition. Weintraub and colleagues position their results as broadly consistent with earlier psilocybin-assisted therapy studies reporting large antidepressant effects, while arguing that their effect sizes were at least as large as those seen in prior pilot work. They also suggest that the longer follow-up window in this study may indicate more durable benefit than in many previous trials. The authors propose that CBT may help prolong early antidepressant gains from psilocybin, drawing an analogy with earlier work suggesting CBT can extend ketamine response. They further interpret the associations between improved emotion regulation, altered cognitive schemas, and symptom reduction as preliminary support for their proposed treatment targets. The discussion highlights participant feedback suggesting possible refinements, including introducing CBT earlier, emphasising acceptance-based skills, and reconsidering when treatment goals are set relative to psilocybin sessions. The authors suggest that a greater focus on positive emotion and positive self-schemas may be especially relevant in psilocybin-assisted psychotherapy. The main limitations they acknowledge are the absence of a comparator condition, the small sample size, the short follow-up for a pilot study, and the possibility that the large effect sizes partly reflect nonspecific trial factors such as intensive care and attention. They also note the risk of assessor bias in an open-label design, especially because clinician-rated effects were larger than self-rated effects, and they say that mood ratings just before each psilocybin dose would have improved attribution of drug-specific effects. Generalisability was limited by exclusions related to physical health, serotonergic medications, and bipolar-spectrum depression.

The authors conclude that the study examined the acceptability, feasibility, and clinical benefits of pairing psilocybin with CBT for major depressive disorder, and they suggest that other psychotherapies such as ACT or internal family systems approaches might also be useful. They argue that the psychotherapy component of psychedelic-assisted therapy should be refined and optimised to improve reach and effectiveness. They call for larger randomised controlled trials comparing different psychotherapy intensities and modalities over longer follow-up periods, and for studies examining whether changes in schemas, emotion regulation, or other cognitive and behavioural variables mediate response and help identify subgroups most likely to benefit.