Psilocybin produces substantial and sustained decreases in depression and anxiety in patients with life-threatening cancer

Cancer patients commonly develop chronic psychosocial distress characterised by depressed mood, anxiety, and reduced quality of life, with up to 40% meeting criteria for a mood disorder. Existing pharmacological treatments (antidepressants, benzodiazepines) have limited and mixed evidence in this population, and psychological interventions have produced small-to-moderate effects. Earlier uncontrolled and small controlled studies from the 1960s–70s, and more recent pilot work, have suggested that classic hallucinogens such as psilocybin may reduce psychological distress in patients with life‑threatening illness, but rigorous modern trials comparing active versus inactive conditions have been limited. Griffiths and colleagues designed a randomised, double‑blind, counterbalanced cross‑over trial to provide a more rigorous test of psilocybin's effects on depression, anxiety, and related measures in psychologically distressed cancer patients. The study compared a very low (1 or 3 mg/70 kg) versus a high (22 or 30 mg/70 kg) oral dose of psilocybin in 51 participants, with sessions separated by about 5 weeks and a 6‑month follow‑up. Procedures sought to minimise expectancy effects and to assess both clinician‑rated and self‑reported outcomes, observer ratings, and putative mediators such as mystical‑type experience.

Participants were adults with a potentially life‑threatening cancer diagnosis and a DSM‑IV psychiatric diagnosis including chronic adjustment disorder with anxiety or depressed mood, dysthymia, generalized anxiety disorder, major depressive disorder, or combinations thereof. Of 566 people screened by telephone, 56 were randomised and 51 completed at least one session; 65% had recurrent or metastatic disease. Detailed inclusion/exclusion criteria were provided in supplementary material and the Johns Hopkins IRB approved the protocol; all participants gave written informed consent. The study used a two‑session, double‑blind cross‑over design with participants randomised to receive either the low dose in Session 1 and the high dose in Session 2 (Low‑Dose‑1st Group) or the reverse (High‑Dose‑1st Group). Sessions were typically separated by a mean of 38 days and the total participation duration averaged 275 days. A preparatory period included multiple meetings with two session monitors before the first session, in‑session support by two monitors in a living‑room‑like environment with music and encouragement to focus inward, and follow‑up meetings after sessions; many contacts were by telephone for participants living at a distance. Psilocybin was administered in opaque capsules with lactose filler; low doses were 1 or 3 mg/70 kg and high doses were 22 or 30 mg/70 kg. The high dose was reduced from 30 to 22 mg/70 kg early in the trial because of discontinuations and prior dose‑effect data, and the low dose was reduced from 3 to 1 mg/70 kg to increase the likelihood of pharmacological inactivity. Blinding and expectancy minimisation were pursued by telling participants and monitors that psilocybin would be given in both sessions and that dose levels could range widely; they were encouraged to seek maximal therapeutic benefit from each session. Primary outcomes were clinician‑rated depression (GRID‑HAM‑D‑17) and anxiety (HAM‑A) assessed at baseline, about 5 weeks after each session, and at about 6 months after Session 2. A clinically significant response was prespecified as a ≥50% decrease from baseline; remission required both ≥50% decrease and a score ≤7 on the respective clinician scale. Numerous secondary self‑report measures assessed depressive and anxiety symptoms (BDI, HADS, STAI, POMS, BSI), quality of life and meaning (MQOL, Purpose in Life, LAP‑R coherence), death acceptance and transcendence, optimism (LOT‑R), spirituality measures, community observer ratings, and participant ratings of persisting effects. Acute session measures included cardiovascular indices and monitor ratings; mystical‑type experience was assessed with the MEQ30 and related scales. Statistical analyses used repeated measures regressions (SAS PROC MIXED) with appropriate covariance structures, planned comparison t‑tests, and Friedman's Test for dichotomous outcomes. Effect sizes were reported as Cohen's d. Exploratory correlations examined relationships between MEQ30 scores and enduring outcomes, with partial correlations controlling for perceived intensity; mediation of dose effects by MEQ30 was tested using bootstrap analyses (PROCESS macro in SPSS) to estimate indirect effects and 95% confidence intervals. Analyses tested combining low‑dose levels and inclusion/exclusion of one 30 mg participant and of six participants who initiated psychotropic medication between Post‑session 2 and 6 months; these checks did not alter main conclusions.

Acute session effects showed orderly, dose‑related increases in blood pressure, heart rate, and all 16 monitor‑rated dimensions of mood and behaviour during sessions, with high‑dose peak effects typically at 90–180 minutes and resolution toward baseline over the remainder of the 6‑hour monitoring window. End‑of‑session subjective measures (e.g. HRS, 5D‑ASC, Mysticism Scale, MEQ30) were significantly greater after the high dose than the low dose. On the 17 therapeutically relevant measures assessed longitudinally, 16 showed significant effects overall and 11 met conservative criteria for demonstrating an effect of psilocybin dose (i.e. a between‑group difference at Post‑session 1 and a within‑group difference between Post‑session 1 and Post‑session 2 in the Low‑Dose‑1st Group). For these 11 measures the mean between‑group effect size at Post‑session 1 was 0.82, the mean within‑group effect size for Post‑session 1 versus Post‑session 2 in the Low‑Dose‑1st Group was 0.66, and the combined Baseline-to‑6‑month effect size was 1.55. Six additional measures did not meet the most conservative criterion yet still showed significant improvement from baseline to 6 months, with a mean Baseline‑to‑6‑month effect size of 1.35. Clinician‑rated response and remission rates showed large, sustained effects. Five weeks after Session 1, 92% of participants in the High‑Dose‑1st Group met the prespecified clinical response criterion on the GRID‑HAM‑D‑17 versus 32% in the Low‑Dose‑1st Group; 79% of the High‑Dose‑1st Group maintained response at 6 months. For HAM‑A the comparable rates were 76% versus 24% at 5 weeks and 83% response for the High‑Dose‑1st Group at 6 months. Collapsing across groups, overall 6‑month clinician‑rated response rates were 78% for depression and 83% for anxiety, with remission rates of 65% and 57%, respectively. Analyses excluding six participants who started antidepressant or anxiolytic medication between Post‑session 2 and 6 months produced similar or slightly higher rates. Participant self‑reports, community observer ratings, and spirituality measures all showed significant positive changes that were generally sustained at 6 months. The Persisting Effects Questionnaire indicated that the high dose produced significantly greater positive long‑term changes in attitudes about life and self, mood, social and behavioural domains, and spirituality; negative ratings were low and not different between conditions. Community observers independently reported improvements in participants' psychosocial functioning. Mystical‑type experience (MEQ30) assessed at the end of Session 1 correlated significantly with 18 of 20 measures assessed 5 weeks later, including meaningfulness (r = 0.77), spiritual significance (r = 0.75), increased life satisfaction (r = 0.53), GRID‑HAM‑D (r = -0.41), and HAM‑A (r = -0.59). Partial correlations controlling for participant‑rated intensity retained significant associations for 11 measures, indicating an effect of mystical‑type experience beyond overall intensity. Bootstrap mediation analyses indicated that MEQ30 score significantly mediated the effect of psilocybin dose on seven outcomes, with reported indirect effect point estimates and bias‑corrected 95% confidence intervals for several outcomes: meaningfulness 1.43 [0.72–2.44], spiritual significance 1.19 [0.59–2.10], life satisfaction 0.60 [0.218–1.19], HADS Anxiety -1.50 [-3.50 to -0.33], HADS Depression -1.11 [-2.79 to -0.02], and HADS Total -2.62 [-5.74 to -0.72]. The extracted text truncates the mediation detail for HAM‑A and so the full estimate/CI for that outcome is not clearly reported here. Safety and tolerability: no serious adverse events were attributed to psilocybin. Transient cardiovascular increases occurred more often at the high dose (systolic >160 mm Hg in 34% high vs 17% low; diastolic >100 mm Hg in 13% high vs 2% low) but did not require medical intervention. Nausea or vomiting occurred in 15% of high‑dose sessions and none of the low‑dose sessions. Physical discomfort (21% vs 8%), psychological discomfort (32% vs 12%), anxiety episodes (26% vs 15%), and one transient episode of paranoid ideation (2% of high‑dose sessions) were reported; there were no cases of hallucinogen persisting perception disorder or prolonged psychosis. Blinding integrity analyses suggested partial success: mean monitor ratings of perceived dose differed (7.0 vs 1.7, p<0.001) but distributions overlapped and some monitors made incorrect inferences about study design, indicating that expectancy was not fully determinative of outcomes.

Griffiths and colleagues interpret the findings as demonstrating that a single high dose of psilocybin, given under psychologically supportive, double‑blind conditions, produced large, rapid and sustained reductions in clinician‑rated and self‑reported depression and anxiety in patients with life‑threatening cancer, along with increases in quality of life, meaning, death acceptance and optimism. The authors highlight that 11 of 17 therapeutically relevant measures met conservative criteria for dose‑dependent efficacy, and that clinician‑rated response and remission rates were substantial and largely sustained at 6 months (overall 6‑month response 78% for depression and 83% for anxiety). The study team positions these results as an extension of earlier pilot findings with lower doses and smaller samples, noting that the time‑course and qualitative features of the high‑dose session resemble prior psilocybin studies in healthy volunteers. A notable mechanistic observation is the strong relationship between mystical‑type experience on session day and enduring therapeutic outcomes; partial correlations controlling for intensity and bootstrap mediation analyses support a mediating role for mystical‑type experience for several outcomes. The authors acknowledge several limitations. The cross‑over design, with participants receiving the alternative dose after 5 weeks, meant that long‑term double‑blind between‑group comparisons were not possible. The very low dose intended as a placebo may have retained some pharmacological activity, and full blinding is difficult when studying drugs with highly discriminable acute effects. Other limitations include a relatively small sample size (n = 51), a predominantly White and highly educated sample limiting generalisability, reliance on some measures that lack independent validation (Persisting Effects Questionnaire, Community Observer Questionnaire), and the absence of blinded clinician raters. The approach also requires substantial preparatory and integration support and is constrained by exclusion criteria, which may limit scalability and patient willingness to participate. Finally, the neurobiological mechanisms underlying enduring therapeutic effects remain speculative; as a 5‑HT2A agonist, psilocin affects cortical and subcortical networks and may alter network dynamics and neuroplasticity, but precise pathways linking acute subjective experiences to long‑term change remain to be established. The authors suggest that their results justify further research, ideally multisite trials in larger and more diverse populations to confirm generalisability and safety.

Under psychologically supportive, double‑blind conditions, a single high dose of psilocybin produced substantial and sustained reductions in depressed mood and anxiety and improvements in quality of life and death anxiety in patients with life‑threatening cancer, with effects reported by patients, clinicians, and community observers at least through 6 months. Clinician‑rated 6‑month response rates were 78% for depression and 83% for anxiety. The authors conclude that multisite studies in larger, more diverse patient samples are warranted to establish the generality and safety of psilocybin treatment for cancer‑related psychological distress.