HOPE: A Pilot Study of Psilocybin Enhanced Group Psychotherapy in Patients with Cancer

Depressive and anxiety symptoms that accompany a cancer diagnosis have been the subject of multiple psilocybin-assisted therapy studies, which have typically used an individual-format model and reported rapid, often durable improvements in mood and existential distress. However, the individual-centred protocol used in most trials is resource intensive—often requiring many clinician hours per participant—and therefore poses barriers to scalability and wider clinical implementation. No published modern study had tested whether a full group-format intervention, including a group psilocybin dosing session, is safe, feasible, and retains therapeutic efficacy. Lewis and colleagues set out to pilot a full-group psilocybin-assisted psychotherapy model in patients with cancer and a DSM-5 depressive disorder. The study aimed to assess safety and feasibility, and to collect preliminary efficacy data on depressive symptoms and related psychosocial outcomes, using a single high-dose (25 mg) group psilocybin session embedded within group preparatory and integration sessions for cohorts of four participants.

Design and overall procedures: The HOPE trial was a single-arm, open-label pilot study conducted under Institutional Review Board, FDA, and DEA approvals. Each cohort comprised four participants and the intervention spanned three weeks: three 120-minute group preparatory sessions (with 30 minutes of 1:1 time per participant within each session), a single group psilocybin administration session (25 mg, day 0), and three 120-minute group integration sessions. Preparatory and integration groups used a supportive-expressive process and psychoeducation; therapists emphasised nondirective, supportive engagement during dosing. Sessions were delivered with a 1:1 therapist-to-participant ratio plus an additional lead therapist; a multidisciplinary 10-person therapist team had completed standard training in psychedelic-assisted therapy. COVID-19 precautions (masking) were observed throughout. Participants and screening: Recruitment was open to patients with any cancer type or stage who met DSM-5 criteria for a depressive disorder; referrals came from interdisciplinary care teams. Screening included chart review, telephone pre-screening, and an in-person psychiatric and medical assessment with PHQ-9 and Columbia Suicide Severity Rating Scale (C-SSRS). Laboratory screening included a comprehensive metabolic panel, urine drug screen and pregnancy test when applicable. The protocol specified a 42-day window for screening/enrolment after the in-person visit. Three cohorts (total n = 12) were enrolled between October 2021 and May 2022; participants were paired with an individual therapist prior to starting preparatory sessions. Intervention specifics and safety monitoring: The group dosing took place in a partitioned clinical space with reclining chairs and a communal music playlist; a private break-out room was available but not used. Vital signs were monitored frequently during the dosing day (every 30 minutes for two hours, then hourly). Adverse events were assessed using Common Terminology Criteria for Adverse Events v5.0 and the C-SSRS throughout preparatory, dosing and integration sessions. Participants were followed for 27 days for emergent suicidal ideation or behaviour and had contact information for their therapist and study team. Outcomes and timing: Feasibility benchmarks were recruitment and retention, defined as completing ≥75% of preparatory and integration sessions plus the dosing session. The pre-specified primary clinical outcome was change in the 17-item Hamilton Depression Rating Scale (HAM-D) score at two weeks post-intervention (day 27), with baseline HAM-D taken on the day of the first preparatory session. Secondary outcomes included the Functional Assessment of Chronic Illness Therapy—Spiritual Well-Being Scale (FACIT-Sp), the Death Transcendence Scale, and the Mystical Experience Questionnaire (MEQ-30) administered at the end of the dosing day. Assessments occurred at seven time points from baseline through 26-week follow-up. Statistical analysis: The investigators performed power calculations based on prior within-group effect sizes, targeting a sample size of 10 evaluable patients to detect d = 1.20 with 84% power and enrolling 12 to allow for dropout. Hypothesis testing used mixed effects models with time as the primary fixed effect and a random intercept to account for repeated measures; a secondary analysis model included all seven time points with an AR(1) covariance structure. Effect sizes are reported as Cohen's d.

Enrollment and sample characteristics: Of 42 potential patients prescreened, 14 proceeded to formal screening and 12 were enrolled into three cohorts of four participants (October 2021–May 2022). The final six-month follow-up contacts occurred in December 2022. The sample was majority female (8 female, 4 male), mean age 48.2 years (range 30–71), and predominantly Caucasian. Cancer types, stage and prognosis were heterogeneous; most participants had terminal diagnoses while three had likely curable cancers. All participants met DSM-5 criteria for a depressive disorder. The extracted text does not provide a full breakdown of cancer stages or all baseline comorbidities. Retention, attendance and safety: All 12 participants completed the intervention and primary endpoint assessments at two weeks and six months; overall attendance was 98% (two participants each missed one integration session). No serious adverse events were attributed to psilocybin. During dosing, no participant required the private break-out room or emergent psychiatric medications. Four participants received short-term medications for general medical complaints during the dosing day (two received promethazine 25 mg for nausea, one received acetaminophen 325 mg for headache, one received propranolol 10 mg for sustained hypertension). One participant experienced nausea followed by hypotension, vomiting and diarrhoea the next day; these prolonged symptoms were later attributed to a viral gastroenteritis affecting the participant’s family. No emergent suicidal ideation or suicidal behaviour was detected during monitoring. Primary depression outcomes: The pre-specified primary outcome, change in HAM-D score from baseline to two weeks, showed a clinically substantial reduction: mean HAM-D decreased from 21.5 at baseline to 10.09 at two weeks (F1,9.00 = 33.81, P < 0.001, d = 1.71). At 26 weeks the mean HAM-D was 14.83 (baseline 21.5 to 14.83; F1,6.49 = 15.58, P = 0.006, d = 1.28). Six of 12 participants (50%) met remission at two weeks, defined as HAM-D < 7. Regarding categorical change magnitudes, three of 12 showed a clinically significant change of 4–6 points and eight of 12 showed a clinically substantial change of 7–12 points. Across all time points the main effect of time was significant (F6,36.98 = 14.02, P < 0.001). Secondary and exploratory outcomes: Several FACIT subscales showed significant improvements from baseline to the two-week follow-up: FACIT-Emotional, FACIT-Functional, FACIT-Spiritual Meaning, FACIT-Spiritual Peace, and FACIT-Spiritual Faith reached significance (other FACIT subscales did not). On the Death Transcendence Scale, the Mysticism subscale showed a significant time effect (F1,8.62 = 9.54, P = 0.014, d = 1.08). On the MEQ-30, six of 12 participants (50%) experienced a complete mystical experience (defined as >60% on all MEQ subscales). MEQ-Total scores correlated with HAM-D change between baseline and two weeks (r = -0.71, P = 0.015); MEQ-Mystical (r = -0.63, P = 0.038) and MEQ-Ineffability (r = -0.68, P = 0.021) subscales were also significantly associated with HAM-D change. Participants who achieved a complete mystical experience had greater reductions in HAM-D than those who did not (t = 2.35, P = 0.04, Cohen's d = 1.42).

Lewis and colleagues interpret these pilot data as supporting the safety, feasibility, and potential efficacy of a full group-format psilocybin-assisted psychotherapy protocol for cancer patients with depressive disorders. In three cohorts of four participants there were no psilocybin-attributed serious adverse events, 12 transient adverse reactions across six participants, 100% completion of the intervention, and 98% session attendance—outcomes the authors contrast with lower completion rates typically reported for other group interventions in cancer populations. The observed reductions in HAM-D scores at two and 26 weeks align with results from prior individual-format psilocybin trials in cancer populations, and improvements were also seen in several domains of quality of life, spiritual well-being and death transcendence. The investigators highlight an association between mystical-type experiences (MEQ-30) and reductions in depressive symptoms, noting that half the sample achieved a complete mystical experience and that MEQ scores correlated with HAM-D change. They also quantify a plausible efficiency advantage for the group model: reporting that participants collectively received 240 therapy hours distributed among five therapists (300 individual therapist hours to run the study) versus an estimated 480 therapist hours if delivered with a standard 2:1 individual-session model, representing a 37.5% reduction in therapist hours. Several limitations are acknowledged. The open-label, single-arm pilot design lacks a control group and is therefore vulnerable to expectancy and selection biases. The sample was small and predominantly Caucasian, limiting generalisability; five of 12 participants reported prior classic psychedelic use and five had prior THC exposure, suggesting possible selection bias relative to population prevalences. The intervention used a non-manualised supportive-expressive group approach, which may hinder replicability and makes it difficult to disentangle effects of psilocybin from group-therapy processes. Heterogeneity in cancer types and prognoses further limits inferences about differential effects by disease status. The authors also caution about the potential for amplified social suggestibility in group settings—which may have therapeutic value but also raises ethical concerns and underscores the need for rigorous oversight. In closing, the study team suggests that the full group model is a promising, more resource-efficient format that appears to preserve benefits observed in individual psilocybin trials, and they recommend further controlled research (including direct comparisons of individual versus group formats and investigations of mechanisms) before broad implementation is pursued.