Compass Pathways released twenty-six-week data on Monday from COMP006, its second Phase 3 trial of COMP360 psilocybin in treatment-resistant depression. In nearly six hundred participants whose current depressive episodes had lasted, on average, more than three years, thirty-nine percent of those receiving the 25-milligram dose achieved a clinically meaningful response by week six and maintained that response through week twenty-six. The first trial, COMP005, had shown twenty-five percent responding to a single dose. The second trial gave two fixed doses. The difference between twenty-five percent and thirty-nine percent is the difference between a promising signal and a filing-quality dataset.
The filing is underway. Compass is in the middle of a rolling New Drug Application with the FDA, with final submission on track for the fourth quarter of this year. In April, FDA Commissioner Marty Makary awarded Compass one of three National Priority Vouchers under a pilot program that compresses the standard ten-to-twelve-month review to one to two months. The other two vouchers went to the Usona Institute for psilocybin in major depressive disorder and to Otsuka Pharmaceutical, which is acquiring Transcend Therapeutics, for methylone in post-traumatic stress disorder. Three psychedelic therapies are now on an accelerated regulatory clock. Compass is furthest ahead.
This would be the first classic psychedelic ever approved by the FDA. The agency rejected Lykos Therapeutics' application for MDMA-assisted therapy for PTSD in August 2024, citing unblinding, safety concerns, and potential clinical trial misconduct. The advisory committee noted that nearly all participants in the MDMA trials could correctly identify whether they had received the drug, raising questions about placebo-controlled results when the drug's effects are unmistakable. Compass avoided that problem by running a conventional two-arm trial with a 1-milligram control dose rather than an inactive placebo.
The regulatory path after FDA approval is not a formality. Psilocybin is currently Schedule I, the same federal classification as heroin and LSD. The Drug Enforcement Administration must reschedule COMP360 before it can be prescribed. The precedent is Epidiolex, GW Pharma's cannabidiol formulation for epilepsy, which the FDA approved in June 2018. The DEA rescheduled it to Schedule V within approximately ninety days. But the DEA rescheduled only the specific approved product, not cannabidiol generally. If the same model applies, COMP360 would move to a lower schedule while psilocybin itself remains Schedule I for everyone else. The company gets a regulatory moat that no patent could replicate: the molecule is simple to synthesize, but the scheduling makes it illegal to prescribe in any formulation except theirs.
President Trump signed Executive Order 14401 in April 2026, directing the federal government to accelerate access to psychedelic therapies for serious mental illness and mandating expedited rescheduling review for Schedule I substances that have completed Phase 3 trials. The executive order does not change the law. It changes the speed at which the agencies move through it. Whether that speed holds through the months between FDA approval and DEA action is the first of several gates that have to open in sequence before a patient can sit in a chair and take this drug.
That chair is not optional. COMP360 will require a Risk Evaluation and Mitigation Strategy, meaning it must be administered in a certified healthcare setting under the supervision of a trained therapist. The session lasts six to eight hours. The patient stays in the facility for the duration. This is not a prescription a physician writes and a pharmacist fills. It is a procedure: scheduled, supervised, and site-dependent. Oregon and Colorado already permit supervised psilocybin sessions under state law, at costs ranging from $1,500 to $5,000 per session. The federal model will look different, but the fundamental constraint is the same. The drug cannot be separated from the setting.
The mechanism explains why. Psilocybin acts primarily through agonism of the 5-HT2A serotonin receptor on cortical pyramidal neurons. But unlike selective serotonin reuptake inhibitors, which modulate serotonin levels and require daily dosing to maintain their effect, psilocybin triggers a cascade of downstream changes that persist long after the molecule has cleared the body. It promotes dendritic spine growth, upregulates brain-derived neurotrophic factor, and induces neuroplasticity through glutamatergic signaling. The clinical term is psychoplastogen: a compound that rewires neural circuits rather than adjusting their chemistry. An SSRI turns a dial every morning. Psilocybin reroutes the wiring.
The six-month data is the evidence that the rewiring holds. Patients in COMP006 had failed at least two prior antidepressant treatments. Their average episode had persisted for more than three years, and they had experienced more than six lifetime episodes. These are the patients the existing pharmacopeia has already given up on. Among those who responded by week six, nearly thirty percent later achieved full remission after retreatment in Part B of the trial. Serious adverse events were low and comparable between the 25-milligram and 1-milligram arms, at 5.7 percent and 6.3 percent respectively. The vast majority of side effects were transient and resolved on the day of dosing.
This creates a commercial problem that has no precedent in pharmaceutical history. Approximately 16.6 percent of American adults currently take antidepressants, according to a 2025 survey. SSRIs account for roughly half of all antidepressant prescriptions and anchor an $18 billion annual market built on one assumption: that depression requires daily chemical management, indefinitely. The patient takes a pill every morning. The pharmacy refills it every month. The manufacturer collects recurring revenue from a recurring condition. The entire economics of the category depend on the treatment never ending.
COMP360 ends. Two doses, six months of effect, and a subset of patients who reach remission. A drug whose clinical superiority is measured by how infrequently the patient needs it is the inverse of the model that built the antidepressant market. The better it works, the fewer sessions each patient requires. The fewer sessions each patient requires, the lower the lifetime revenue per patient. A cure is a worse business than a treatment. This is not a new observation in pharmaceutical economics. It is the reason the industry has always preferred chronic conditions to acute ones. But it has never been tested at this scale, in this category, against this standard of care.
The market has priced Compass Pathways at approximately $1.74 billion. The company holds $466 million in cash, enough runway through 2028 and beyond a potential launch in the first half of 2027. For comparison, Vertex Pharmaceuticals announced yesterday that it will pay $10 billion for Crinetics Pharmaceuticals, a company whose lead asset treats acromegaly, a rare hormonal disorder. Vertex projects $5 billion in combined peak revenue from the Crinetics portfolio. The treatment-resistant depression market is valued at $2 billion to $3.8 billion and growing, but those estimates are built on existing therapies. A drug that works where existing therapies have failed could redefine the addressable population.
The stock fell in premarket trading on Monday despite the data. Markets process psychedelic medicine the way they process any asset whose value depends on a sequence of regulatory gates, each of which is binary. The FDA can approve or reject. The DEA can reschedule or delay. Insurers can cover or exclude. Certified treatment centers can scale or bottleneck. Each gate is independent. Each must open. The clinical evidence is no longer the question. The questions are institutional.
Six hundred patients sat in supervised sessions, took a synthetic compound derived from a mushroom that has been used for thousands of years, and thirty-nine percent of them got better for six months from a condition that had resisted everything else. The molecule is simple. The data is clean. The regulatory pathway exists. The executive order is signed. What remains is the part that has nothing to do with science: whether the institutions that control access to medicine can accommodate a drug that works by being taken as rarely as possible.