People ask about LSD for migraine as if there were a stack of studies somewhere. There isn’t. ClinicalTrials.gov lists no trial of LSD for migraine at all. Its chemical cousin psilocybin has one small Yale study that looked promising and a second psilocybin migraine study that couldn’t clearly beat an antihistamine.
Cluster headache, a rarer and far more brutal disorder, has a little more: decades of patient surveys, a handful of tiny trials and two LSD trials that haven’t reported yet.
My view is that the evidence is thin mostly because these drugs are hard to study, not because anyone looked carefully and found nothing. Meanwhile the standard treatments leave a large minority of patients without relief, and medicine and its regulators have taken their time.
Psychedelics and cluster headache: what patients report
The story starts with cluster headache: attacks of one-sided pain, usually around the eye, last 15 minutes to three hours, come up to eight times a day and may be among the worst pain a person can feel (Schindler and Burish, BMJ, 2022). About one person in 1,000 gets it (Fischera et al., 2008). In a survey of 1,134 American patients, 55% reported suicidal thoughts (Rozen and Fishman, 2012).
In 1998 patients online began swapping stories about LSD and psilocybin mushrooms, and in 2002 Bob Wold turned that community into the nonprofit Clusterbusters. Harvard researchers at McLean Hospital interviewed 53 self-treating patients for a 2006 paper in Neurology. Of 26 psilocybin users, 22 said it aborted attacks, and 7 of 8 LSD users said it ended a cluster period (Sewell, Halpern and Pope).
In the Clusterbusters medication survey of 496 people, psilocybin and LSD were rated as good as or better than most conventional drugs, even at doses too low to cause hallucinations (Schindler et al., 2015). A 2022 review of nine surveys with 5,419 respondents found the same pattern (Rusanen et al.). The survey’s free-text answers were later published under one patient’s line: “You will eat shoe polish if you think it would help” (Schindler et al., 2021).
Migraine has no movement on that scale, but in an online pain survey, people with migraine said psychedelics relieved pain better than their usual medication (Cavarra et al., 2023). In a British cohort born in 1958, people who had ever used a classic psychedelic had 25% lower odds of frequent bad headaches (Bjurenfalk et al., 2025).
None of this is proof. Online surveys recruit believers, and anyone who has handled survey data knows how easily a self-selected sample tells you what you hoped to hear. Episodic cluster periods also end on their own, so a patient who took mushrooms in week six of a cycle may be thanking the wrong thing. What the reports do is make a hypothesis too strong to ignore.
LSD’s family has treated migraine since 1925
The idea came from neurology’s own medicine cabinet, not from hippies. Arthur Stoll at Sandoz isolated ergotamine from ergot in 1918, and a doctor named Maier first used it against migraine in Switzerland in 1925 (Tfelt-Hansen and Koehler, 2008). LSD came out of the same Sandoz ergot chemistry, as I describe in my history of LSD.
The closer relative is methysergide, made from lysergic acid. In 1959 the Florentine headache researcher Federigo Sicuteri introduced it as the first effective migraine preventive. Within about five years it was linked to retroperitoneal fibrosis, scarring at the back of the abdomen, after long-term use (Koehler and Tfelt-Hansen, 2008). US production stopped in 2002 (Wikipedia), and since 2014 Europe has allowed it only for severe migraine and cluster headache when standard drugs have failed (MHRA).
Then the twist. Studying methysergide, the Rotterdam pharmacologist Pramod Saxena found an “atypical” serotonin receptor in 1984, and that finding helped prompt the development of sumatriptan, the first triptan (same source). Today’s most widely used migraine drugs descend partly from research on a close chemical relative of LSD. Sicuteri also wrote about LSD and its derivatives as migraine preventives in 1963 (Kempner and Schindler, 2025). Then LSD was banned, and the idea sat largely untouched for decades.
What do the psilocybin migraine studies actually show?
Emmanuelle Schindler, a Yale neurologist, has run the only randomized trials of psilocybin in migraine I could find. In the first, ten adults each received one dose of psilocybin and one of placebo, two weeks apart. In the two weeks after psilocybin, migraine days fell by 1.65 a week, against 0.15 after placebo, a statistically significant difference (Schindler et al., Neurotherapeutics, 2020). The benefit did not track how intense the trip was.
Enthusiasts quote the follow-up less often. Eighteen adults got zero, one or two doses of psilocybin a week apart, and diphenhydramine, a sedating antihistamine meant to make guessing harder, whenever they didn’t get psilocybin. All three groups ended up with roughly 50% fewer migraine days over eight weeks, and the differences were not significant (Schindler et al., Headache, December 2025). In the first two weeks, 80% in each psilocybin group halved their migraine days, against 17% on the antihistamine, a gap that came close to significance.
So the controlled evidence for migraine is one positive pilot, one inconclusive trial and 28 analyzed patients. Yale is now running a 50-person study of how psilocybin works in migraine (NCT06464367). For LSD and migraine there isn’t even a registered trial.
Psilocybin, BOL-148 and the LSD cluster headache trials
Yale gave 16 cluster headache patients psilocybin or placebo in three doses about five days apart, a schedule copied from what patients had worked out themselves. Attacks fell by 3.2 a week on psilocybin and essentially not at all on placebo, but with 14 people analyzed the difference was not significant (Schindler et al., 2022). When ten returned at least six months later for another round, with no placebo group this time, their attacks fell from 18.4 to 9.8 a week (Schindler et al., 2024).
In an open-label Copenhagen study of ten chronic patients, three doses of psilocybin cut attacks by 31% on average, and one patient had 21 weeks of complete remission (Madsen et al., 2024). COVID-19 stopped recruitment at ten (NCT04280055).
The oddest result came from Hannover Medical School. Five patients whom verapamil, the standard preventive, had failed took BOL-148 (2-bromo-LSD, a version of LSD that causes no hallucinations) three times in ten days. One went into a remission still going six months later, two chronic patients had over a month without attacks, and a fourth had far fewer (Karst et al., Cephalalgia, 2010). With five patients, no placebo and two authors holding a patent on the drug, it’s a lead rather than a finding. But if a non-hallucinogenic analogue works, the trip may not be the medicine.
In August 2026, sixteen years later, its developer Ceruvia Lifesciences raised $8 million for a phase 1/2 trial planned in Germany, with first results, from healthy volunteers, due in mid-2027 (BioSpace).
As for LSD itself, University Hospital Basel started a placebo-controlled crossover trial in January 2019, planning 30 patients who each get three full doses and three placebos. The registry lists it as active but no longer recruiting, with completion expected in October 2027 (NCT03781128). In the Netherlands, Radboud University Medical Center and two partner hospitals are testing a much smaller “minidose” against placebo in 65 chronic patients (NCT05477459). Australia’s George Institute is planning a 40-patient psilocybin trial (George Institute).
What standard treatment offers, and who it leaves behind
Headache medicine isn’t empty-handed. For cluster headache, pure oxygen through a mask left 78% of attacks pain-free within 15 minutes in a London trial, against 20% with air (Cohen et al., JAMA, 2009). Injected sumatriptan brought relief within 15 minutes in 74% of attacks, against 26% on placebo (NEJM, 1991).
The gaps are wide, though. Migraine is the world’s second-biggest cause of disability (Steiner et al., 2020). In a German registry of 2,284 patients in specialist care, 42.5% had failed at least one triptan and 13.1% at least two (Ruscheweyh et al., 2023). In STRIVE, a trial behind the 2018 approval of erenumab, the first drug to block the CGRP pathway (AJMC), half the patients on the higher dose cut their migraine days by at least half, against 26.6% on placebo (Goadsby et al., NEJM, 2017). That is real progress, and it leaves the other half.
Cluster headache prevention is weaker. A 2023 consensus of European specialists called it sparsely studied and limited by relatively low efficacy and side effects (Lund et al.). Galcanezumab, a CGRP antibody, was approved in the US in 2019 for episodic cluster headache (AJMC), after 71% of patients halved their attacks by week three, against 53% on placebo (Goadsby et al., 2019). In the chronic form it failed (Dodick et al., 2020), and so did erenumab, with 31.7% responders against 45% on placebo (Mecklenburg et al., 2025). In a 2026 Australian survey, 35% of patients called their treatment ineffective and 27% only partly effective (Haghdoost et al.).
Why is it so hard to study LSD for migraine and cluster headache?
First, the law. LSD and psilocybin are Schedule I drugs in the US, so researchers need a DEA registration on top of FDA approval. In 2013 David Nutt and colleagues called research under such controls difficult and often close to impossible (Nature Reviews Neuroscience). Congress eased the paperwork only in July 2025, in the HALT Fentanyl Act (Public Law 119-26).
Second, money. The molecules are long out of patent, so no company has a reason to fund big headache trials, and headache research is starved anyway: an analysis of NIH funding from 2011 to 2023 found headache disorders among the most underfunded of ten brain disorders relative to their burden (Berglas et al., 2025). The Yale trials list private foundations and a drug developer as collaborators. When psychedelic medicine did take off, it went to psychiatry: the FDA’s guidance on psychedelic trials, finalized in July 2026, is framed around psychiatric and substance use disorders (Applied Clinical Trials). The sociologist Joanna Kempner and Schindler call this neglect of pain a “pain gap.”
Third, blinding. A sugar pill fools nobody once the walls start breathing, and patients who know they got the real thing may report what they expect. Schindler’s antihistamine partly fixed that, which may be exactly why her second migraine trial looked weaker. Her group has since reviewed better “active placebos” (Aday et al., 2025).
Fourth, cluster headache trials are hard even with legal drugs. A 2022 review counted 27 placebo-controlled prevention trials; of the 15 no longer running, five had stopped early, and seven of the ten completed ones enrolled fewer patients than planned or didn’t say (Dodick et al.). Entry criteria are tight: the Dutch LSD trial wants at least eight attacks a week and excludes anyone with a family history of psychosis or bipolar disorder.
The case for caution
The skeptics have a point. The best-controlled migraine trial so far was inconclusive, and small positive studies often shrink when they grow up. A little critical thinking says a striking story from a self-selected sample is where research starts, not where it ends.
There are physical risks too. Even 15 micrograms of LSD raised blood pressure in healthy volunteers (Cavarra et al., 2025). LSD and psilocybin resemble methysergide and other drugs that damaged heart valves when taken regularly, and a 2024 review warned that months of microdosing could carry a similar risk nobody has measured (Rouaud et al.). Anything bought illegally has unknown contents. None of this argues for doing nothing. It argues for the large, careful trials that still don’t exist.
What remains when standard treatment fails?
- A headache specialist. In the Australian survey, a quarter of patients weren’t seeing anyone about their cluster headache. Specialists can try other triptans, nerve blocks, steroids and, for chronic cluster headache, neurostimulation (Lund et al.). The American Migraine Foundation keeps a directory of doctors.
- Newer drug classes. The European Headache Federation suggests that people who have failed two triptans be considered for gepants and ditans, the newest acute migraine drugs (Ruscheweyh et al.).
- Clinical trials. On ClinicalTrials.gov, search by condition and filter for “recruiting”; you can add psilocybin or LSD as the intervention. The EU’s Clinical Trials Information System does the same for Europe. Expect strict criteria.
- Patient groups. Clusterbusters, OUCH UK and the Migraine Trust connect patients with clinicians and research, and they lobby: Clusterbusters says it spent ten years pushing Medicare to pay for oxygen.
Some patients self-medicate, and pretending otherwise helps nobody. In the Clusterbusters survey, illicit substances came up in 35.5% of respondents’ comments, nearly three times as often as side effects, and about a quarter wrote that nothing had worked. That sample leans toward users, but the reason is plain: when nothing legal has helped and the pain is this bad, the legal and medical risks start to look small.
My take
I think the case for psychedelics in headache is stronger than the evidence, and the gap between them is mostly medicine’s fault, not the patients’. Sicuteri was writing about LSD derivatives for migraine in 1963. Sixty-three years later, psilocybin trials have analyzed 28 migraine patients, LSD trials none, and the Basel trial in cluster headache has been running for nearly eight years without a result.
The second Yale trial may be the first sign of a shrinking effect. If psilocybin turns out to be no better than an antihistamine for migraine, I want to know, and so do the patients spending money and taking legal risks on it. Either answer beats the current shrug. What I object to is a system in which treatments for one of the most painful conditions known wait for patient groups and private money to fund the trials.
Regulators have started to move: Congress has eased Schedule I research, the FDA has finalized its guidance for psychedelic trials, and an LSD drug for anxiety has produced positive phase 3 results (Psychiatric Times). Headache should be next in line rather than an afterthought. Until it is, the patients who got there first will keep doing what they have done since the 1990s: running the experiment on themselves.
Questions people ask
Does LSD help with migraine?
Nobody knows yet. As of November 2026 there is no registered clinical trial of LSD for migraine, only patient reports. The closest evidence comes from two small Yale trials of psilocybin: one found fewer migraine days after a single dose, the other could not separate psilocybin from an antihistamine placebo.
Does psilocybin work for cluster headache?
Patient surveys say yes, but the controlled evidence is thin. A Yale trial found a drop in attacks that was not statistically significant, and a Danish open-label study of ten patients found attacks fell by 31% on average. Larger trials of LSD are under way in Switzerland and the Netherlands, and a psilocybin trial is planned in Australia.
What is BOL-148?
BOL-148, or 2-bromo-LSD, is a chemical relative of LSD that does not cause hallucinations. In a 2010 case series of five cluster headache patients in Hannover, most had long remissions or far fewer attacks. Its developer, Ceruvia Lifesciences, expects first data from a new phase 1/2 trial in 2027.
How do I find a clinical trial for cluster headache or migraine?
Search ClinicalTrials.gov by condition and filter for studies that are recruiting. In Europe, the EU’s Clinical Trials Information System at euclinicaltrials.eu lists trials as well. Expect strict entry criteria, and ask a headache specialist whether you are likely to qualify.