Psilocybin for PTSD: where the evidence stands

Search for psychedelic therapy and PTSD and the headlines are mostly about MDMA, not psilocybin. MDMA-assisted therapy has been tested in two large randomized trials (Mitchell et al., 2021; Mitchell et al., 2023). In August 2024 the US Food and Drug Administration (FDA) declined to approve it and asked for another phase 3 trial (Lykos Therapeutics, 2024).

Psilocybin for PTSD is a much earlier story. We found two small, uncontrolled studies with results, a pilot of 12 veterans (Armstrong et al., 2026) and a trial of 22 adults (McGowan et al., 2026), plus larger trials that are registered but have not reported. This page describes what each study did and found, and where it falls short. It is not medical advice, it does not say what anyone should do, and it covers supervised research only.

Does psilocybin work for PTSD?

Nobody can say yet. Both completed psilocybin studies we found were open-label and had no comparison group, so everyone knew they were getting psilocybin and there was nothing to compare against. Both reported lower PTSD scores after treatment. That is a signal worth testing, not a result that shows the drug works.

The authors of one study put it plainly: psilocybin had “not yet been investigated in PTSD” before their trial (McGowan et al., 2026). A 2021 systematic review that searched the literature up to October 2019 found nine PTSD trials of psychoactive drugs: five of ketamine and four of MDMA, and none of psilocybin (Varker et al., 2021). Everything on psilocybin and PTSD described below arrived after that.

Why is the best-known PTSD result about MDMA, not psilocybin?

Because MDMA is the drug that went through large, randomized, placebo-controlled PTSD trials. MDMA is also a different kind of drug. One research team describes it as not a serotonergic psychedelic (Armstrong et al., 2026). Psilocybin is one. Results for one do not carry over to the other. For how psilocybin differs from other substances, see psilocybin, LSD and DMT compared.

The two MDMA phase 3 trials

The first (known as MAPP1) randomized 90 people with severe PTSD to MDMA or placebo, each with a course of manualized therapy. Scores on a clinician-rated PTSD scale (CAPS-5) fell by 24.4 points on average with MDMA and 13.9 with placebo among those who completed treatment, measured two months after the last drug session (Mitchell et al., 2021). The second (MAPP2) randomized 104 people with moderate to severe PTSD. Scores fell by 23.7 with MDMA and 14.8 with placebo. Severe adverse events were reported by 5 of 53 people (9.4%) given MDMA and 2 of 51 (3.9%) given placebo, with no deaths or serious adverse events (Mitchell et al., 2023).

Those are the results behind the headlines. Note that the placebo group also improved by a lot, and both groups received therapy. That matters for what comes next.

What did the FDA decide in August 2024?

On 9 August 2024 the sponsor, Lykos Therapeutics, announced that the FDA had issued a complete response letter for its MDMA application. The FDA requested an additional phase 3 trial. The company said the agency’s concerns were about the durability of the data, expectancy bias among participants who had used MDMA before, and the role of the psychotherapy (Lykos Therapeutics, 2024).

Two months earlier, on 4 June 2024, the FDA’s advisory committee had voted on the evidence. Asked whether the data show that the drug is effective in PTSD, 2 members said yes and 9 said no. Asked whether the benefits outweigh the risks with the FDA’s proposed safety programme, 1 said yes and 10 said no (FDA minutes, 2024). Committee members who voted no pointed to functional unblinding, meaning participants could tell which group they were in, and to the lack of management of expectation bias. Those who voted yes noted the same concerns but said they could not overlook how large the effects were (FDA minutes, 2024).

The FDA’s briefing document said about 90% of participants given MDMA and 75% of those given placebo guessed their assignment correctly in a post-study survey. It also noted that every treatment arm in every study included psychotherapy, so the contribution of therapy to the overall effect had not been characterized (FDA briefing document, 2024). These are the same two problems that every psychedelic trial, including psilocybin ones, has to solve.

Where things stand as of 4 October 2026: MAPS, the non-profit that initiated the original research, said in September 2025 that the FDA had made the letter public for the first time (MAPS, 2025). In August 2026 a report that the sponsor, now called Resilient Pharmaceuticals, had resubmitted its application without a new phase 3 trial drew a MAPS statement responding to it (MAPS, 2026; HCPLive, 2026). We did not find an FDA decision on the resubmission or an approval. Check the FDA’s own announcements for the current position.

What psilocybin studies in PTSD actually exist?

Only two prospective studies with reported results met our search, both open-label and uncontrolled. The table sets them beside the MDMA trials. Scores come from the same clinician-rated scale, but the designs differ so much that the numbers must not be compared across rows. We leave out dose amounts on purpose.

McGowan et al., 2026: 22 adults, one session

This phase 2 trial ran between June 2022 and February 2024 at King’s College London, Mount Sinai Hospital in New York and Sunstone Therapies in Maryland. It was funded and sponsored by COMPASS Pathfinder. 22 adults with PTSD (64% female, average age 39) each received one session of the company’s psilocybin formulation (COMP360) after preparation sessions, with psychological support. The main goal was safety and tolerability, and PTSD scores were secondary outcomes (McGowan et al., 2026; ClinicalTrials.gov NCT05312151).

On safety, there were 117 treatment-emergent adverse events in the 22 people. Headache (50%), nausea (36%), crying (27%) and fatigue (27%) were common, and none were serious or led to anyone leaving the trial. On scores, the average CAPS-5 total fell by 29.9 points at week 4 and 29.5 at week 12. By the authors’ definitions, 81.8% had responded and 63.6% were in remission at week 4, and 77.3% and 54.5% at week 12 (McGowan et al., 2026).

The authors describe it as a feasibility study with significant limitations: small, open-label, without a comparator arm, and with people who had complex PTSD or significant suicide risk excluded. They said larger controlled studies are needed. They also note that MDMA trials defined response and remission differently, so applying the MDMA definitions would give a higher response rate but a lower remission rate. That is one reason cross-trial comparisons mislead (McGowan et al., 2026).

Armstrong et al., 2026: 12 veterans, two sessions

This pilot at Ohio State University enrolled US military veterans aged 21 to 64 with severe PTSD that had not responded to treatment. Of 668 people who completed an online pre-screener, 13 enrolled and 12 completed the study. Each had about eight hours of therapy, then two psilocybin sessions two to three weeks apart, then about eight more hours of therapy. The main goals were safety and suicidal thinking (Armstrong et al., 2026).

No serious adverse events occurred. Non-serious events included headaches, anxiety and dizziness, and the authors report no increase in suicidal ideation or behaviour. Clinician-rated PTSD scores fell by 27.5 points on average from baseline to one month after the second session, and 75% (9 of 12) were counted as responders and in remission (Armstrong et al., 2026).

One detail shows why an open-label design is hard to read. Participants’ scores improved significantly during the preparation phase, before they received any psilocybin, and the size of that early improvement was linked to the outcome later. The authors list the limits: a small sample, no control group, open-label design, short follow-up, self-referral, a sample that was 92% white, and strict exclusions such as suicide history (Armstrong et al., 2026). The study is registered as NCT05554094.

What about other research?

• A trial of psilocybin for demoralization in long-term AIDS survivors reported lower scores on a PTSD checklist, but it was not designed to study PTSD, as Armstrong’s team notes (Armstrong et al., 2026).
• Uncontrolled studies of veterans who attended psychedelic retreats outside clinical trials reported lower PTSD checklist scores afterwards. One followed 58 veterans at psilocybin or ayahuasca retreats (Calnan et al., 2025); another followed 21 veterans with a history of traumatic brain injury at psilocybin retreats (Blest-Hopley et al., 2025). Both lacked a control group and relied on self-selected participants. They were not clinical trials and tell us little about safety or cause.
• Reviews describe the idea of psilocybin for PTSD as promising but largely untested. One 2025 review sets out how psilocybin might act and the challenges for future studies, without reporting new patient data (Miller and Zoladz, 2025).

Why do people confuse psilocybin and MDMA for PTSD?

There are real reasons, which is why it helps to separate them.

• Similar treatment format. Both are tested as a drug session wrapped in preparation and follow-up therapy. In the psilocybin trial above, there were preparation sessions, a 6 to 8 hour session with two staff, and follow-up. The MDMA trials had three drug sessions plus preparatory and integrative therapy (McGowan et al., 2026; Mitchell et al., 2021).
• Both are grouped as “psychedelic-assisted therapy”. A review notes that both have received an FDA breakthrough therapy designation, but for different conditions: psilocybin for resistant depression and MDMA for PTSD (Bird et al., 2021). That detail is easy to lose.
• Researchers compare them. Armstrong’s team set their 75% result beside the 71% reported for MDMA-assisted therapy (Armstrong et al., 2026). Yet an uncontrolled pilot of 12 cannot be set against a placebo-controlled trial of 90.
• They share the same methodological problem. Both produce strong effects that make blinding hard, which is why the FDA’s concerns about MDMA trials also apply to psilocybin trials. For depression, see our look at psilocybin depression trials.

Which larger psilocybin PTSD trials are registered?

A search of ClinicalTrials.gov on 4 October 2026 for psilocybin and PTSD returned 29 records, most of them small. These are examples of the larger or better-controlled ones. A registration shows a trial was planned, not that it worked. Dates and sizes are the registrants’ own and may change.

The largest, Redefine, has the kind of design the existing psilocybin studies lack: random assignment and masking. By the registry’s own dates, results are years away. COMPASS sponsored one of the completed psilocybin studies and is the sponsor of Redefine, which is worth knowing when reading their results.

What does this evidence not show?

• It does not show that psilocybin is an effective, safe or approved treatment for PTSD. The only completed studies we found had no comparison group.
• It does not show that results for MDMA carry over to psilocybin. They are different drugs tested in different designs.
• It does not separate the drug from other factors. Every study included preparation and therapy, people knew they were receiving the drug, and in Armstrong’s pilot scores improved before any dose.
• It does not show how long any change lasts. The follow-up in the studies above was one to three months.
• It does not tell us about people the studies left out. They excluded or screened out many groups, including people with significant suicide risk, bipolar or psychotic disorders, and in McGowan’s trial, complex PTSD.
• It does not test self-treatment or retreats. Trial participants were screened, prepared and supervised by clinicians. Nothing here applies to taking mushrooms outside such a setting.
• It does not reflect everyone with PTSD. Armstrong’s sample was 92% white and McGowan’s was a small group at three sites (Armstrong et al., 2026; McGowan et al., 2026).
• It does not show who might be harmed. Studies of 22 and 12 people cannot reveal uncommon risks (McGowan et al., 2026; Armstrong et al., 2026).

Is psilocybin an approved treatment for PTSD in Canada?

As of 24 October 2026, Health Canada’s page (last modified 20 October 2026) states that production, sale and possession of magic mushrooms are illegal in Canada, and that there are no approved therapeutic products containing psilocybin in Canada or elsewhere (Health Canada). Approval positions can change, so check that page for the current position.

Some of the registered trials above list Canadian sites, in Calgary, Kelowna and Toronto. Taking part in a trial is not the same as access to a treatment. The same Health Canada page describes the Special Access Program, under which authorized practitioners can request psilocybin for patients. We have not reviewed how it works in practice and this article does not advise on it (Health Canada).

What is still unknown?

• Whether psilocybin lowers PTSD symptoms more than a comparison condition. No randomized results are published yet, as far as we could find.
• How much of any change comes from the drug versus the therapy, the preparation, expectancy and the natural course of the condition.
• How long any change lasts, and what the risks are in larger and more varied groups.
• Whether results from veterans, from the general adult group in the 22-person trial, and from survivors of specific kinds of trauma will match.
• What happens to the MDMA application, and what the FDA will accept as evidence for psychedelic therapies in general.

For now, the accurate summary is that psilocybin for PTSD is a question being tested, with small early signals and no controlled answer. For background on the compound, see what psilocybin is and what magic mushrooms are. For another condition studied in trials, read our look at psilocybin and alcohol use disorder, or the regulatory story of COMP360 and the FDA.

Shroomverse sells mushroom products. This article is general information about clinical research. It is not medical, legal or purchasing advice, and it does not recommend any product.