Exploratory Controlled Study of the Migraine-Suppressing Effects of Psilocybin

Migraine is a highly prevalent and disabling neurological disorder for which existing treatments often have limited efficacy or undesirable side effects. Earlier anecdotal and some clinical reports have suggested that agonists at the 5-hydroxytryptamine 2A (5-HT2A) receptor, including psilocybin and LSD, can produce long-lasting reductions in headache burden after a single or a few oral doses, a pattern distinct from most conventional migraine preventives. However, controlled investigations of psilocybin in headache disorders have been lacking. Schindler and colleagues undertook an exploratory, proof-of-concept study to test whether a single low oral dose of psilocybin would suppress migraine frequency over the subsequent two-week period and to assess safety in a controlled setting. The study used a double-blind, placebo-controlled, cross-over framework, with the aim of informing future, larger trials and mechanistic work if signals of efficacy were observed.

This was an exploratory double-blind, placebo-controlled, cross-over experimental study conducted under an Investigational New Drug application and approved by relevant institutional review boards. Eligible adults were aged 21–65, met International Classification of Headache Disorders criteria for migraine, and had at least two migraine attacks per week. Exclusion criteria included serious cardiovascular, cerebrovascular, central nervous system or psychiatric disorders (notably psychotic or manic disorders in the participant or a first-degree relative), recent substance abuse, and recent use of serotonergic antidepressants or certain other medications; triptans were allowed within specified limits. Recruitment occurred from the community and specialist centres, and screening included medical, laboratory, and structured psychiatric assessments. Synthetic psilocybin (0.143 mg/kg) was prepared and chemically validated (100% purity by HPLC) and encapsulated to match placebo (microcrystalline cellulose). The study design involved two experimental sessions separated by at least 14 days. Under the approved blinding procedure, all subjects received placebo in the first session and psilocybin in the second, so each participant served as their own control and potential long-lasting effects of psilocybin would not contaminate a later placebo session. Experimental sessions took place in an outpatient neurobiological research unit with physiological monitoring, an intravenous line for rescue medication if needed, and psychological monitoring; subjects remained on site for at least 6 hours after dosing. Subjects kept daily headache diaries starting 14 days before the first session (baseline) and continuing until 14 days after the second session. The primary outcome was change in weekly migraine days in the 2 weeks after dosing compared with baseline. Other primary outcomes included weekly migraine attacks and ratings of associated symptoms and functional impairment; secondary outcomes included abortive medication use, time to next migraine attacks, acute vital sign changes, psychotropic ratings, and adverse events. Acute subjective effects were measured throughout sessions with visual analogue scales and post-session with the validated 5-Dimensional Altered States of Consciousness (5D-ASC) scale. Statistical analyses were primarily within-subject: paired t tests compared changes from baseline between placebo and psilocybin, linear mixed models assessed repeated measures (vitals and subjective effects), Spearman correlations tested associations between acute effects and migraine outcomes, and Fisher's exact test compared adverse event counts. The study targeted 12 participants for exploratory power considerations but completed analyses on 10 subjects.

Between November 2017 and December 2019, 69 candidates were prescreened, 14 underwent secondary screening, 12 entered study procedures, and 10 subjects were included in the final analysis. Two participants were excluded: one for scheduling inability to attend the second session and one because their baseline period lacked sufficient migraine attacks. The analysed sample comprised seven females and three males with a mean age of 40.5 years (SEM 4.4). Two subjects had prior exposure to psilocybin (not specifically for migraine). Primary outcome: The reduction in weekly migraine days from baseline over the 2-week period after dosing was significantly greater following psilocybin (mean change -1.65 days/week, 95% CI -2.53 to -0.77) than after placebo (-0.15 days/week, 95% CI -1.13 to 0.83), with p = 0.003 (t(9) = 4.11). Responder rates after psilocybin were 80% with at least 25% reduction, 50% with at least 50% reduction, and 30% with at least 75% reduction; corresponding figures after placebo were 20%, 20%, and 0% respectively. The difference was statistically significant at the ≥25% threshold (p = 0.023, Fisher exact). Secondary and other migraine outcomes: Psilocybin produced significantly larger reductions than placebo in weekly migraine attacks, pain severity, attack-related functional impairment, and weekly days using abortive medications. There were no significant between-treatment differences in attack duration or in ratings of associated symptoms (photophobia, phonophobia, nausea/vomiting). Time to next attacks: Time to the first post-dose migraine was statistically equivocal between conditions, whereas time to the second migraine attack was significantly longer after psilocybin (mean 10.30 days, SEM 1.61) than after placebo (mean 5.00 days, SEM 1.13), p = 0.012 (t(9) = 3.14). The authors note ceiling effects for this measure because some participants had only one or no attacks in the 2 weeks after psilocybin. Acute drug effects and psychotropic ratings: Mixed-model analyses showed a significant drug × time interaction for the VAS “overall drug effect,” driven by psilocybin but not placebo; a similar interaction was observed for the feeling of “peace/harmony.” End-of-session 5D-ASC total scores (expressed as percent of the total possible) were higher after psilocybin (19.35% (SEM 7.55)) than after placebo (3.08% (SEM 1.80)); p = 0.026 (t(9) = 2.65). Importantly, neither peak “overall drug effect” nor the 5D-ASC total score correlated significantly with the percent change in weekly migraine days (Spearman r = 0.469, p = 0.17 and r = 0.418, p = 0.23 respectively). Safety and adverse events: No unexpected or serious adverse events occurred and no participants withdrew because of adverse events. Transient symptoms reported during sessions included lightheadedness and muscle tension/soreness; within 24 hours both placebo and psilocybin sessions were followed by reports of tension/sore muscles, general headache, and migraine attack. Physiologically, mean arterial pressure (MAP) showed a significant drug × time interaction across the test day, with post hoc analyses indicating a MAP increase after psilocybin starting at 45 minutes and lasting until about 4 hours; the maximum acute MAP increase over placebo was 12.2 mmHg (95% CI 4.61 to 19.73) at 1.5 hours. All adverse events were transient and self-limiting, and at approximately 3 months follow-up no subjects reported persistent physical, psychological, or cognitive changes.

Schindler and colleagues interpret their findings as demonstrating that a single low oral dose of psilocybin produced statistically and clinically meaningful reductions in several migraine measures over the two-week period following administration. They note this is, to their knowledge, the first controlled study of psilocybin in a headache disorder and that the observed persistent clinical effect after a single dose is novel compared with most conventional migraine preventives that require ongoing dosing or long-lasting biologic presence. The investigators highlight an apparent dissociation between acute psychotropic effects and the sustained anti-migraine effect: higher psychotropic scores did not predict larger reductions in migraine burden, and some prior reports and congeners with reduced psychedelic effects have also suggested therapeutic benefit. From this they suggest the possibility that therapeutic mechanisms of select 5-HT2A receptor compounds in headache might be separable from their psychedelic effects and that sub-hallucinogenic or chemically modified agents could be explored. The authors discuss potential biological mechanisms including immunomodulatory and anti-inflammatory actions involving serotonin, sigma-1, and toll-like receptors, as well as neuroendocrine and sleep-related pathways, all of which have relevance to migraine pathophysiology. They emphasise that mechanistic studies alongside clinical trials would be necessary to identify the relevant targets. Limitations acknowledged by the study team include the small sample size and lack of demographic diversity (all subjects were Caucasian), the short diary window that introduced ceiling effects for some time-to-event measures, and the absence of validated migraine severity instruments. They also discuss the potential for unblinding because psilocybin produced noticeable acute effects, although some overlapping effects were reported after placebo and subjective acute-effect measures did not correlate with efficacy; the authors recommend using active placebos (for example niacin) and formal measures of blinding and expectation in future trials. Finally, the investigators call for replication in larger, fully randomised studies, dose-ranging work, trials of repeated administration schedules, and concurrent mechanistic research to characterise efficacy, dose dependence, durability, and safety before clinical application.