The FDA's request to study psilocybin for depression

The FDA's request to study psilocybin for depression was a pivotal redirection of psychedelic research: regulators who had been approached about trials with cancer patients urged researchers to investigate depression in the broader population. The request signaled that the initial clinical findings might matter beyond existential distress at the end of life, but it was an invitation to test a possibility, not approval of a treatment.

Background: A cautious return to clinical research

By December 2016, results from cancer studies at New York University and Johns Hopkins University had brought psychedelic therapy back into public and professional view. Published together in a special issue of the Journal of Psychopharmacology, the studies reported that about 80 percent of participating cancer patients had clinically significant reductions in anxiety and depression, with benefits lasting at least six months. Across both trials, the intensity of participants' mystical experiences closely correlated with the degree of symptom relief. The findings attracted prominent mental-health commentators, including two former presidents of the American Psychiatric Association.

The evidence nevertheless came from just 80 subjects in all. Larger studies would be necessary before regulators could consider rescheduling psilocybin or approving a treatment. The results were promising enough to prompt calls for further research, and medical schools and funders began expressing interest in future trials. The moment was therefore a return to respectability, not a settled verdict on efficacy.

The request and its surprise

In early 2017, Roland Griffiths and Stephen Ross brought the cancer-trial findings to the FDA, hoping to obtain approval for a larger phase 3 study of psilocybin in cancer patients. Instead, FDA staff surprised them by asking whether the researchers could broaden their proposal and test psilocybin for depression in the general population. In the regulators' view, the data contained a strong enough signal of possible antidepressant benefit that it would be a missed opportunity not to investigate, given the scale of need and the limitations of available therapies.

The request reversed the researchers' original calculation. They had chosen cancer patients partly because they believed it would be easier to secure permission to study a controlled substance among people who were already seriously ill or dying. Ross later described the meeting as surreal and said he remained stunned by its outcome. The FDA declined to confirm or deny his account, explaining that it did not comment on drugs in development or under regulatory review. The episode thus conveyed both an unexpected regulatory opening and a limit on what could be independently confirmed about the conversation.

A parallel signal from depression research

The FDA's proposal resonated with a separate development in Europe. In 2016, researchers approached the European Medicines Agency about using psilocybin for anxiety and depression in people facing life-changing diagnoses. The agency noted that existential distress was not an official diagnosis covered by national health services, but suggested that a possible depression benefit warranted a larger, multisite trial.

That response drew not only on the cancer-trial findings but also on a feasibility study led by Robin Carhart-Harris in David Nutt's laboratory at Imperial College London. The initial study gave psilocybin to 12 people with treatment-resistant depression—six men and six women, all of whom had already tried at least two treatments without success. There was no control group, and participants knew they were receiving psilocybin. After one week, all showed improvement and two-thirds were depression-free; seven still had substantial benefit at three months. The study later expanded to 20 volunteers. At six months, six remained in remission while others had relapsed to varying degrees, suggesting that the treatment might need to be repeated. It was small and uncontrolled, but reported rapid benefits, good tolerability, and no adverse events in this group.

Promise, limitations, and the question of treatment

The request did not resolve the central evidentiary problem. Psychedelic studies are difficult to blind; they combine a drug session with therapy; and psilocybin remained illegal. Early trials also tend to involve small, selected groups and unusually attentive, experienced therapists. Researchers and clinicians may be invested in success, while the novelty of a treatment can strengthen placebo effects. For these reasons, early findings across several conditions were better understood as promising signals than as definitive proof of cure.

That caution had historical force. In Saskatchewan, LSD therapy had come to be regarded as a promising treatment for alcohol addiction, and provincial policy made it a standard option. But when Toronto's Addiction Research Foundation tried to replicate the work under more controlled conditions, clinicians placed volunteers in neutral rooms, refrained from engaging with them apart from administering questionnaires, and sometimes restrained or blindfolded them. The results failed to match the Saskatchewan trials, and some participants had terrifying experiences. The disagreement exposed how strongly therapeutic outcomes could depend on the setting and human support surrounding a drug session. A later reassessment of Timothy Leary's Concord Prison Experiment similarly found that his striking claim of reduced recidivism was exaggerated and that the groups showed no statistically significant difference. These precedents made enthusiasm for new findings inseparable from scrutiny of methods.

The FDA request nonetheless widened the field's horizon: research initially framed around the needs of people with cancer might also test a treatment for depression on a much larger scale. Whether that promise would survive larger trials remained open. Moving from a regulatory invitation to phase 3 research would require studies at multiple sites with hundreds of volunteers and substantial funding; the request changed the question researchers were being asked to pursue, not the standard of evidence they would have to meet.