Psilocybin-assisted group therapy in patients with cancer diagnosed with a major depressive disorder

Depression is common among patients with cancer and is associated with poorer treatment adherence, reduced quality of life, and higher mortality. Earlier pharmacological and psychotherapeutic approaches have shown limited and inconsistent benefits in this population, with some meta-analyses finding no clear superiority of antidepressants over placebo; side effects, delayed onset, and drug interactions further complicate their use in oncology. Psilocybin, a serotonergic 5-HT2A agonist, has emerged as a candidate intervention because it can modulate neural networks implicated in rigid negative cognition and may produce mystical or spiritual experiences that correlate with improved wellbeing. Previous clinical trials have demonstrated safety and antidepressant effects for psilocybin in individual therapy settings and in cancer-associated psychological distress, but conventional delivery models have relied on dyadic care (two therapists per patient), isolated administration settings, and samples mixing anxiety and depressive disorders. Agrawal and colleagues set out to evaluate a novel cohort-based delivery model: psilocybin-assisted therapy administered simultaneously to small groups (cohorts of three to four participants) within a community outpatient cancer centre, combining one-to-one therapist support in individual rooms with group preparation and integration sessions. The study aimed to examine safety, feasibility, tolerability, and preliminary efficacy in patients with curable and noncurable cancer who met criteria for major depressive disorder (MDD). This Phase II, open-label trial tested a single 25-mg psilocybin dose with structured preparatory and integration psychotherapy adapted for the group format.

This was a single-site, Phase II, open-label trial conducted in a community oncology practice. Participants were enrolled in eight cohorts of three to four people between November 2020 and May 2021. Key eligibility screening included the Mini International Neuropsychiatric Interview and the 17-item Hamilton Depression Rating Scale; participants with significant suicide risk (per the Columbia Suicide Severity Rating Scale within the past year or at screening/baseline) were excluded. Prior recreational psychedelic use was allowed if it had occurred more than 12 months before screening. Participants tapered psychiatric medications under medical supervision before the intervention when applicable. The extracted text does not clearly report additional inclusion or exclusion criteria in full. Treatment comprised a single 25-mg oral dose of psilocybin administered simultaneously to cohort members. Therapeutic support included individual sessions (a reported 4.25 hours in a 1:1 therapist–patient ratio) and group sessions (a reported 3.75 hours), with a total of approximately 8 hours of preparation and integration therapy across two preparatory visits, the dosing day, and two postdosing integration sessions. Preparatory work included a 2-hour individual session focused on psychoeducation, coping strategies, and establishing a therapeutic alliance; group preparation and group integration sessions were also provided. During dosing, participants were in individual rooms using eyeshades and a pharmacokinetically timed music programme; one-to-one therapists provided in-room support while two lead clinicians monitored sessions via live video. Participants remained at the facility for 8 hours on dosing day. Safety assessments captured adverse events (AEs) up to 8 weeks postdosing, vital signs, 12-lead ECGs, laboratory tests, and suicidality via C-SSRS. Efficacy was primarily assessed with the clinician-rated Montgomery–Åsberg Depression Rating Scale (MADRS); response was predefined as ≥50% reduction from baseline and remission as MADRS <10. Additional clinician- and self-rated instruments measured depressive symptoms (QIDS-SR16), anxiety (HAM-A, STAI), pain (pain VAS and EuroQoL pain subscale), demoralization (Demoralization Scale II), disability (Sheehan Disability Scale), quality of life (EQ-5D-5L), psychosocial-spiritual healing (NIH HEALS), and altered states of consciousness (5D-ASC). For analysis, descriptive statistics summarised recruitment, retention, baseline characteristics, and AEs. The Full Analysis Set included all treated participants. Efficacy and exploratory endpoints were analysed with mixed models for repeated measures adjusted for baseline, age, visit, gender, and interactions (baseline-by-visit, age-by-visit, gender-by-visit), with visit as the repeated effect; comparisons with baseline used a two-sided .05 testing level. Effect sizes were reported as Cohen's d. The extracted text indicates no statistical testing was performed for recruitment and retention metrics.

Thirty participants were enrolled and completed the protocol across eight cohorts, with an attrition rate of 0%. Mean age was 56 years (SD 12); 70% identified as female and 30% as male. The sample was predominantly Caucasian (80%), with 46.7% having curable cancer and 53.3% noncurable metastatic disease. Breast cancer was the most common diagnosis (33.3%), followed by leukaemia/lymphoma (26.7%). Half of participants reported prior antidepressant use and 70% had received at least one line of cancer therapy. Safety: No psilocybin-related serious adverse events were reported. Treatment-related AEs such as nausea and headache were generally mild and expected; no laboratory or ECG abnormalities were observed and rescue medications were not required. There were no reports of suicidality on the C-SSRS during follow-up. Primary efficacy: By week 8 posttreatment, MADRS scores decreased by a mean of 19.1 points from baseline (95% CI, -22.3 to -16.0; p < .001), corresponding to a large effect (Cohen d = 2.55). The study defined response as ≥50% MADRS reduction and remission as MADRS <10, but the extracted text does not provide frequencies for response/remission in the main Results section. Exploratory outcomes: Measures of pain, demoralization, disability, and psychosocial-spiritual wellbeing improved from baseline to week 8. Pain VAS showed a mean change of -1.2 (95% CI, -2.0 to -0.4; Cohen d = 0.56), a 37% reduction; the EuroQoL pain subscale decreased by -2.3 (95% CI, -2.9 to -1.6; Cohen d = 0.66), a 26% reduction. Demoralization fell by -6.8 points (95% CI, -9.1 to -4.4; Cohen d = 1.02), a 44% reduction, with larger effects in those with higher baseline demoralization. Functional impairment on the Sheehan Disability Scale declined by -7.5 points (95% CI, -9.6 to -5.3; Cohen d = 1.19), a 65% decrease. Trait anxiety attenuation at 8 weeks was noted in 73% of participants, per the Discussion text.

Agrawal and colleagues interpret the findings as demonstrating that cohort-based, psilocybin-assisted therapy delivered in a community cancer centre is feasible, tolerable, and preliminarily efficacious for patients with cancer and MDD. The authors highlight that simultaneous dosing with one-to-one therapist support in individual rooms, combined with group preparation and integration, allowed efficient delivery while maintaining safety; no psilocybin-related serious adverse events occurred. They suggest the group-oriented model may improve scalability, reduce the clinician resource burden compared with the conventional two-therapist-per-patient approach, and increase accessibility by situating treatment in a community oncology setting rather than an academic or inpatient psychiatric centre. The authors note that participants with both curable and noncurable cancer experienced similar baseline depression severity and comparable improvements by week 8, indicating potential benefit across prognostic categories. They also discuss observed reductions in trait anxiety and pain-related measures, proposing that psilocybin may foster psychological flexibility and resilience that persistently alter traits and reduce pain-related impairment; however, mechanisms remain speculative. Agrawal and colleagues place their results in the context of prior psilocybin trials that reported rapid antidepressant effects and reductions in cancer-related distress, and they regard the current cohort-based approach as a novel delivery variant deserving further study. Limitations acknowledged by the authors include the open-label design, modest sample size, single-site setting, and absence of a control arm, which constrain causal inference and generalisability. The extracted text does not provide a detailed list of other study limitations such as potential selection bias or unblinded outcome assessment, nor does it report longer-term follow-up beyond 8 weeks. The authors recommend larger, controlled trials to confirm efficacy, to compare delivery models, and to refine preparatory, dosing and integration procedures that address whole-person care in cancer treatment centres.

The study concludes that psilocybin-assisted therapy delivered in a group-cohort model within a community cancer centre was safe, feasible, and associated with clinically meaningful reductions in depressive symptoms among patients with curable and noncurable cancer. The authors argue these outcomes support further investigation in larger randomised trials, including control arms, and exploration of group-format variations and comprehensive psychosocial care strategies integrated into cancer treatment settings.