Psilocybin Research: A Scientific Overview

Psilocybin Research: A Scientific Overview

This page provides a comprehensive review of the clinical evidence base for psilocybin-assisted therapy — organized by indication, with an honest assessment of what the data shows and what questions remain open.

How to Read This Overview

Level of evidence terminology:

• Phase 2 RCT: Randomized controlled trial, moderate sample size (typically 50–200 participants). Establishes efficacy signal, not definitive approval evidence.
• Phase 3 RCT: Large-scale randomized controlled trial (200+ participants, often multi-site). The standard for FDA approval.
• Phase 2 open-label / pilot: No control group; provides safety and dose-ranging data.
• Effect size: Cohen's d or percentage change relative to baseline or control. Larger effect size = more clinically meaningful.

The psychedelic therapy field is primarily in Phase 2 as of 2026. COMPASS Pathways is the only program with Phase 3 data for psilocybin specifically.

Major Depression (MDD and TRD)

COMPASS Pathways — COMP360 (Synthetic Psilocybin)

Phase 3 — Treatment-Resistant Depression (TRD)

The largest controlled psilocybin trial conducted to date.

• n=233, randomized to 1mg (control), 10mg, or 25mg COMP360
• Primary endpoint: MADRS score at week 3
• Results: 25mg condition showed significant reduction in depression score vs. control (p<0.001); response rate ~35–40% vs. ~15% control
• Durability: 60-65% of week-3 responders maintained response at week 12
• Adverse events: Headache, nausea, fatigue in first 24 hours; psychological distress during session manageable with support

NDA submission expected: Late 2026; FDA review 12–18 months; earliest approval Q4 2027–Q1 2028

Johns Hopkins — Psilocybin for MDD

Phase 2 RCT

• n=24, psilocybin + supportive therapy vs. delayed treatment
• 71% showed significant clinical response at 4 weeks; 54% in remission
• 12-month follow-up: Majority maintained improvement
• Effect size: Large (Cohen's d > 2.5 for some measures)

Imperial College London — Psilocybin vs. Escitalopram

Phase 2 RCT (landmark)

• n=59, psilocybin (25mg × 2) vs. escitalopram (antidepressant)
• Primary endpoint (QIDS): Not significant between groups
• Secondary endpoints (most depression scales): Psilocybin showed significantly greater improvement
• Remission rate: 57% psilocybin vs. 28% escitalopram
• Interpretation: Controversial; the primary endpoint was negative but most secondary measures favored psilocybin

Tobacco Addiction

Johns Hopkins — Matthew Johnson

Phase 2, open-label

• n=15, long-term smokers who failed multiple cessation attempts
• 2–3 psilocybin sessions with CBT support
• 80% abstinence at 6 months (compare: ~35% for varenicline, best pharmacotherapy)
• 67% abstinent at 12 months
• Effect size: Very large

Follow-up RCT: Currently ongoing at Johns Hopkins (psilocybin vs. nicotine patch). Results expected 2026–2027.

Alcohol Use Disorder

NYU Langone — Bogenschutz et al.

Phase 2 RCT

• n=93, psilocybin (25mg × 2) vs. diphenhydramine (active placebo)
• Both groups received 12 weeks of Motivational Enhancement Therapy
• Percent heavy drinking days: 83% reduction psilocybin vs. 51% placebo
• Abstinence: Significantly higher in psilocybin group
• Published: JAMA Psychiatry, 2022

Effect size for heavy drinking days: Cohen's d ~0.6 — moderate to large, substantial clinical significance in this population.

PTSD

UCSF / UC Berkeley Joint Trial

Phase 2, mixed population

• n=60, veterans, firefighters, law enforcement with PTSD
• Psilocybin (25mg × 2) with manualized therapy
• 60% showed clinically significant PCL-5 reduction at 3 months
• 47% no longer met full PTSD diagnostic criteria
• No serious adverse events

This is Phase 2 data — promising and important but not yet Phase 3 scale. MAPS's MDMA data (different molecule) is stronger specifically for veteran PTSD.

Cancer Anxiety and End-of-Life

Johns Hopkins and NYU (Cancer Anxiety Studies)

Phase 2, both sites

Multiple trials — jointly: n>80 across sites

• Psilocybin (single session) for existential anxiety in terminal cancer diagnosis
• Hopkins: 92% of participants showed significant anxiety reduction; 83% clinically significant at 6 months
• NYU: 83% showed sustained reduction at 6 months
• These numbers represent some of the largest effect sizes in any psychedelic trial

The mechanism differs somewhat from depression: the therapeutic work centers on confronting existential distress directly rather than disrupting depressive cognitive patterns.

OCD

Moreno et al. — University of Arizona

Phase 2, open-label (n=9)

• 9 OCD patients, three dose levels each in crossover
• All 9 participants showed significant Y-BOCS reduction at all dose levels
• Including low doses without full psychedelic effects — unusual finding

Yale RCT (ongoing): First controlled trial. Early data suggests significant Y-BOCS reduction in psilocybin arm. Full results expected 2026–2027.

Anorexia Nervosa

Imperial College London Pilot

Phase 2, open-label (n=10)

• First clinical trial of any psychedelic for eating disorders
• Significant EDE-Q reductions at 1-month follow-up
• Several participants described qualitatively different body perception during and after

UCSD RCT (enrolling): First randomized controlled trial. Results expected 2027–2028.

The Evidence Gaps

What we don't yet know:

• Long-term durability: Most trial follow-up is 6–12 months. Does improvement persist at 2–5 years? Data is emerging but limited.
• Opioid use disorder: Mechanistic rationale is strong; controlled trials are barely starting.
• Optimal dosing and number of sessions: Is one session enough? Two? Does more always mean more?
• Who responds vs. who doesn't: The mystical experience correlation is consistent but doesn't fully predict outcomes.
• Long-term safety at scale: The current trial populations are carefully screened. Real-world outcomes with less screening are unknown.

What we know confidently:

• Psilocybin is physiologically safe at clinical doses in screened, medically stable individuals
• Effect sizes in depression, addiction, and anxiety are among the largest seen in psychiatric clinical trials
• Therapeutic context (preparation, facilitation, integration) is essential — this is not a pure pharmacological effect
• Mystical-type experience intensity correlates with treatment outcomes across indications

The Mechanism

The most compelling mechanistic framework is Carhart-Harris's REBUS model:

Psilocybin reduces default mode network (DMN) activity, temporarily flattening the brain's predictive hierarchy. In depression, OCD, and addiction, this hierarchy is stuck — maintaining pathological self-referential patterns with excessive confidence. Flattening it creates a window in which new patterns can form.

Simultaneously, post-psilocybin neuroplasticity (BDNF elevation, dendritic growth) is elevated for approximately 2–4 weeks, making this window therapeutically actionable.

Therapy in the post-session window leverages both mechanisms: the opened hierarchy allows new cognitive models to be experienced, and the elevated neuroplasticity allows them to consolidate.

Resources

• ClinicalTrials.gov: For current enrolling trials
• Psychedelic Alpha: psychedelicalpha.com — regulatory and data tracking
• MAPS: maps.org — MDMA research and clinical infrastructure
• Heffter Research Institute: heffter.org — foundational psilocybin research funding