Compass Pathways puts caution at the center of its psilocybin launch plan
Compass Pathways is moving toward what could become the first FDA-approved psychedelic therapy in the United States, but chief executive Kabir Nath made one thing plain in Boston: the company intends to “go slow before we go fast.” The message came as Compass prepares to finish a rolling FDA submission this quarter for COMP360, its synthetic psilocybin treatment for treatment-resistant depression, with a launch targeted for early next year if regulators clear the product.
The caution is not just corporate theater. The broader psychedelics sector is trying to move from scientific promise to clinical deployment under intense public scrutiny, and early missteps could shape how clinicians, payers, regulators, and patients view the class for years. A bad first experience for patients or prescribers could reverberate well beyond one product.
Nath framed the opportunity in blunt terms. “I think this is a really big deal for patients, for providers, and incidentally for Compass Pathways as well,” he said during the STAT Summit in Boston. The company is preparing for a highly regulated launch environment, including an FDA REMS, or risk evaluation and mitigation strategy, if the therapy wins approval. Compass is also investing in provider training for the supervised administration model it expects to accompany the drug.
A supervised six-hour session, not a standard antidepressant prescription
Psilocybin is being developed in a very different clinical format from conventional antidepressants. Rather than daily dosing with a slow onset of action, COMP360 is designed around a supervised treatment session lasting roughly six hours. Nath said that in Compass’s Phase 3 trials, about a quarter of patients had a clinically meaningful response after a single administration, and that figure rose to about 40% after a second dose three weeks later.
The difference is more than logistical. A treatment that can produce rapid and lasting improvement after one or two administrations changes the way depression care is organized, especially for patients who have not responded to existing options. If those benefits hold up after approval, the barrier will not be pharmacology alone. It will also be whether health systems can deliver long monitored sessions at scale without compromising safety or access.
Compass has also pointed to longer-term data suggesting that patients who respond can maintain benefit for as long as a year after two or three administrations. Even without inventing a broader claim than the data support, that kind of durability would be unusual in depression care, where many treatments depend on continuous adherence and repeated dose adjustments.
The supervised model also explains why reimbursement looms so large. The cost is not limited to the drug itself; it includes several hours of clinical monitoring. Nath said the company has spent years working on reimbursement mechanisms for that monitoring and is in discussions with commercial insurers and the Centers for Medicare and Medicaid Services. Coverage decisions will likely determine whether this therapy remains a specialty-center intervention or becomes more widely usable in psychiatric practice.
Blinding remains one of psychiatry’s hardest trial problems
Steven Hyman, a Harvard University professor and former director of the National Institute of Mental Health, used the same stage to press on the unresolved scientific questions around psychedelic treatment. The first is familiar to anyone following this field: blinding is hard when a drug produces unmistakable subjective effects.
“The unblinding remains a challenging problem,” Hyman said.
In a field where placebo response and expectation effects can be unusually strong, the issue is not academic. If participants can infer which arm they are in because of the psychedelic experience, trial interpretation becomes more difficult. That concern was central in the FDA’s 2024 rejection of an MDMA-based therapy from Lykos Therapeutics. For psilocybin programs, the methodological challenge is to separate a true drug effect from the psychological force of knowing one has taken a psychedelic.
Hyman also raised a deeper question: whether the psychedelic experience itself is required for clinical benefit. Several companies are now trying to develop compounds that might deliver antidepressant effects without hallucinations or other intense subjective effects. If that approach works, the practical implications would be considerable. Patients might be able to take such drugs at home, avoiding the need for long supervised sessions and potentially lowering costs.
“It would be nontrivial if you could separate these, because then somebody could write a prescription and send somebody home,” Hyman said. “But we don’t know.”
The psychotherapy question is still unsettled
Psychedelic treatment has long been linked with psychotherapy, but Compass has deliberately separated psychotherapy from its clinical trials. Nath argued that the company’s task is to develop a drug, not to dictate how physicians deliver psychological care. He said recorded treatment sessions were analyzed and found to be silent roughly 80% of the time.
That detail matters because it cuts to the heart of how much of the clinical effect belongs to the molecule and how much belongs to the surrounding therapeutic environment. Hyman said psychotherapy after a psychedelic experience could still help patients process intense emotions, memories, and other psychological effects. Yet he emphasized that researchers have not determined how much benefit comes from the pharmacology of psilocybin itself and how much comes from the experience layered around it.
The unresolved split between drug effect and therapeutic setting is central to how psychiatrists may eventually use these medicines. A product approved as a drug can be regulated, reimbursed, and distributed in ways psychotherapy cannot. But if the psychological context is part of what drives benefit, stripping away too much of it could weaken real-world outcomes.
Credibility, safety, and the risk of moving too quickly
The discussion also reflected a broader anxiety in the field: acceleration without discipline. Officials in the Trump administration have expressed support for faster development of psychedelic treatments, but Hyman warned that the field’s credibility will depend on rigorous research, responsible marketing, and long-term safety monitoring.
“I don’t mean this as a pun,” Hyman said. “But being really sober about the way this research is presented — the only way to gain credibility is to be absolutely rigorous.”
He compared the moment to cannabis, where early assumptions about safety did not fully capture the risks that emerged as products became more potent. The parallel is not perfect, but the warning is direct. Once a therapy enters routine use, enthusiasm can outrun the evidence unless long-term monitoring is built in from the start.
Even if COMP360 wins approval, the commercial path will remain difficult. Nath declined to set a pricing benchmark against Johnson & Johnson’s Spravato, an esketamine treatment for depression, saying only that it provides a reference point. He also did not offer a specific price. Access will hinge not only on what insurers cover, but on whether clinics can support the staffing and monitoring needed for each session.
Hyman closed with the kind of caution that reflects the state of psychiatry more broadly. The field is still searching for objective biomarkers that could help identify which patients are most likely to benefit from a given treatment. “We’re awash in subjectivity still,” he said, pointing to how much remains unknown about the biological underpinnings of depression and why some patients respond while others do not. The approval path may be close, but the science behind patient selection, mechanism, and long-term safety remains unfinished.