The Psychedelic Setback: What MDMA Rejection Taught Drug Developers

In August 2024, the FDA declined to approve MDMA-assisted therapy for post-traumatic stress disorder. The decision followed a June advisory committee meeting in which panelists voted nine to two that the data had not shown MDMA was effective, and ten to one that the benefits had not been shown to outweigh the risks. For a psychedelic industry that had treated approval as nearly assured, the rejection was a sobering event.

The sponsor, Lykos Therapeutics, had run two Phase 3 trials of MDMA combined with a proprietary form of psychotherapy: MAPP1 and MAPP2. The CAPS-5 improvements in the active arm were large, and the effect size looked, on paper, like a blockbuster. Yet the agency and its advisors identified a constellation of problems that no favorable effect size could paper over.

The first was functional unblinding. MDMA produces powerful subjective effects that make it nearly impossible to keep participants uncertain about whether they received the drug, a problem well documented in the broader psychedelic trial literature. Because the primary endpoint depended on self-report and clinician interviews, expectancy became indistinguishable from pharmacology.
The second was selection bias. A substantial fraction of enrolled participants had used MDMA before and entered the trial convinced of its benefits, a population unlikely to generate regulatory-grade efficacy data.

The third category of problem was more concerning. FDA inspections uncovered unreported adverse events at multiple sites, and the CRL noted that Lykos had not adequately characterized the drug's safety, acute effects, duration of impairment, or abuse potential. Serious allegations of therapist sexual misconduct at a Phase 2 site in Canada led to a journal article retraction. The therapy manual was non-standard and difficult to separate from the pharmacology, raising the question of whether the trial was testing a drug or a drug-plus-therapy combination that the FDA has no clear pathway to regulate.

The psychedelic development community absorbed these lessons rapidly. Sponsors now design trials with central remote raters who never see the participants in their altered states. Psychological support is standardized, non-directive, and explicitly decoupled from trauma-focused psychotherapy, so that any effect observed is attributable to the drug. Dosing and therapy sessions are commonly video-recorded for independent adherence rating, a practice pioneered in the MAPS MDMA program and reinforced as a bias-mitigation strategy in the FDA's 2023 draft guidance on psychedelic drug development. Blinding integrity is measured and built into sensitivity analyses. Abuse potential is assessed directly through validated instruments. Cardiovascular effects are characterized continuously rather than sampled.

These design improvements matter because the underlying hypothesis, that certain compounds acting on the serotonin 5-HT2A receptor might produce therapeutic effects unachievable with conventional medications, has not been disproven.

Synthetic psilocybin, intravenous psilocin, 5-MeO-DMT, and other candidates remain in active development. An early Phase 2 open-label study of COMP360 psilocybin in PTSD reported signals of durable symptom improvement out to 12 weeks after a single dose, and larger controlled studies are underway.

The MDMA rejection was not the end of the psychedelic story in PTSD. It was the end of an era in which enthusiasm could substitute for methodological discipline. The next era requires both, and the patients waiting for better treatments deserve no less.

Contact us today to learn how we can help.