From Signal to System: FDA Approval Does Not Mean Treatment Can Begin the Next Morning

Authored by

REMS for psychedelic therapies

FDA approval would be a major milestone for psychedelic therapies, but it would not make treatment immediately available to patients. For sponsors, clinicians, and health systems, the next challenge is turning an approved product into a workable care model. That means navigating controlled-substance scheduling, risk-management requirements, site readiness, and the delivery infrastructure needed to bring treatment to patients safely and consistently.

That post-approval challenge is the focus of this third blog in our psychedelic drug development series. In blog two, we explored the evidence questions shaping psilocybin development in treatment-resistant depression. Here, we look beyond the approval decision to examine what must be in place for these therapies to reach patients safely, reliably, and at scale.

Approval and Scheduling Run on Parallel Tracks

Psilocybin, MDMA, and lysergide, commonly known as LSD, remain Schedule I controlled substances under federal law. This affects the leading programs across depression, anxiety, and PTSD, including Compass Pathways’ and Usona’s psilocybin programs, Resilient’s MDMA-based development for PTSD, and Definium Therapeutics’ lysergide programs in MDD and GAD.

FDA approval and controlled-substance scheduling address related but different questions. FDA determines whether a specific drug product is safe and effective for its proposed use and whether it can be approved with appropriate labeling and risk controls. DEA determines the level of control required because of the drug’s abuse potential, dependence liability, public-health risk, and patterns of misuse.

The agencies also sit within different parts of the federal government. FDA is an agency within the Department of Health and Human Services (HHS), whose primary orientation is public health, medical evidence, product safety, and therapeutic benefit. DEA is an agency within the Department of Justice and is responsible for administering and enforcing the Controlled Substances Act. Its remit includes diversion, unlawful manufacture and distribution, prescribing controls, registration, security, recordkeeping, and enforcement. The same molecule is therefore viewed through two different windows: one asks whether it can be used as a medicine; the other asks how tightly it must be controlled once it is.

The process does not simply begin after FDA reaches an approval decision. When a new drug application involves a stimulant, depressant, or hallucinogenic drug that appears to have abuse potential, HHS must transmit that information to the Attorney General. In practice, FDA’s Controlled Substance Staff evaluates abuse-related data during development and NDA review so that the scientific and medical scheduling recommendation can be prepared in coordination with the approval process.

The sponsor’s application must therefore support two related assessments. The NDA must establish efficacy, safety, manufacturing quality, labeling, and the overall benefit-risk profile. The abuse-potential package must characterize the pharmacology and chemistry of the drug, its subjective effects, reinforcing properties, physical or psychological dependence, similarity to known controlled substances, evidence of misuse or diversion, and the implications of the proposed formulation and route of administration.

HHS, relying heavily on FDA’s scientific review, then conducts the analysis required under the Controlled Substances Act. This is commonly called the eight-factor analysis.

The first factor is the drug’s actual or relative potential for abuse. This is not limited to whether participants report liking the drug. It considers whether the substance produces effects associated with misuse, whether people may take it for nonmedical purposes, and how its abuse potential compares with substances already controlled.

The second factor is the scientific evidence of its pharmacological effects. This includes receptor activity, central nervous system effects, dose-response relationships, psychoactive effects, and findings from human abuse-potential studies and nonclinical models.

The third factor is the current state of scientific knowledge regarding the drug. FDA and HHS consider the overall body of evidence, including chemistry, pharmacology, clinical experience, safety, formulation, route of administration, and any information relevant to how the product may be used or misused.

The fourth factor is the drug’s history and current pattern of abuse. Psilocybin, MDMA, and LSD are not entirely new chemical entities with no societal history. Each has a distinct pattern of nonmedical use, public perception, illicit availability, and prior enforcement experience that must be considered separately from the controlled pharmaceutical product.

The fifth factor is the scope, duration, and significance of abuse. This asks how extensively the substance has been misused, how long those patterns have existed, and how serious the associated consequences have been.

The sixth factor is risk to public health. This may include acute psychiatric or cardiovascular events, accidental injury, impaired judgment, unsafe use without supervision, interactions with other substances, diversion, and the potential consequences of broader availability.

The seventh factor is psychic or physiological dependence liability. The analysis considers whether the drug produces psychological reinforcement, tolerance, withdrawal, compulsive use, or other forms of dependence, and how that liability compares with drugs already assigned to a schedule.

The eighth factor is whether the substance is an immediate precursor of another controlled substance. This factor is generally less central for the finished psilocybin, MDMA, and lysergide products themselves, but the statute requires it to be considered. The full eight-factor framework is specified in Section 201 of the Controlled Substances Act.

After completing its scientific and medical evaluation, HHS sends DEA a written recommendation on whether the drug should be controlled and, if so, in which schedule. HHS’s scientific and medical findings are binding on DEA as to those scientific and medical questions. If HHS recommends that a substance should not be controlled, DEA cannot place it under control. HHS does not, however, issue the final scheduling rule.

DEA then applies the Controlled Substances Act from the enforcement and diversion-control perspective. It considers the HHS analysis, the statutory criteria for the proposed schedule, relevant law-enforcement information, patterns of illicit manufacture and distribution, and the controls necessary to protect the public. The scheduling authority formally belongs to the Attorney General and is delegated to DEA.

For a newly approved drug with abuse potential, the statute provides an expedited pathway. If HHS recommends placement in Schedule II, III, IV, or V, DEA must issue an interim final rule no later than 90 days after the later of two events: receipt of the HHS scientific and medical evaluation and scheduling recommendation, or notification that FDA has approved the drug application. The interim final rule becomes effective immediately, while still allowing interested parties to submit comments and request a hearing before DEA later issues a final rule.

This distinction is important for psychedelic sponsors. FDA approval may establish an accepted medical use for the specific approved product, but it does not by itself move the substance out of Schedule I. DEA must complete the scheduling action before the product can move into ordinary prescribing and controlled distribution.

Schedule III is widely viewed across these development programs as the most workable target because it would recognize accepted medical use while retaining meaningful controls for abuse and diversion. But Schedule III is not awarded simply because a sponsor requests it or because the product has been approved. HHS must determine that the scientific evidence supports the Schedule III criteria, including a lower abuse potential than substances in Schedules I and II and no more than moderate or low physical dependence or high psychological dependence. DEA must then implement that recommendation through the Controlled Substances Act process.

Even after federal rescheduling, the operational work is not finished. States may need to revise their own schedules. Manufacturers, distributors, pharmacies, prescribers, and treatment settings may require DEA registrations or modifications. Security, inventory, recordkeeping, ordering, storage, disposal, and diversion-control procedures must be operational. International treaty obligations may also need to be considered.

Approval opens one lock. Scheduling opens another.

And the two locks are installed by agencies with different mandates, different statutory authorities, and different definitions of a successful outcome. FDA asks whether the evidence supports use of the product in patients. DEA asks how that product can enter medical practice without losing control of the substance around it.

For Compass, MDMA-based sponsors, and Definium, scheduling is therefore a parallel development workstream that must be supported by the abuse-potential program, anticipated in the label and REMS strategy, and connected to manufacturing, distribution, site certification, and launch readiness. These steps will shape the operational pathway long before launch.

The result could be an approved medicine with uneven availability across the country.

Then there is also the practical infrastructure. The approved pharmaceutical product must also be clearly distinguished from unapproved or nonmedical forms of the same substance. That distinction may be obvious to regulators and drug developers, but it may not be obvious to patients or the public.

REMS Could Become the Operating System

Scheduling controls the substance. A REMS may control the conditions under which the approved product is used.

For psychedelic therapies, REMS could become the operating system linking the clinical trial model to routine delivery.

The questions are intensely practical. Must the prescriber be certified? Must the healthcare setting be certified? Who determines patient eligibility? What training is required? How long must the patient be observed? Can the patient leave alone? What emergency capability must be available? How is re-treatment authorized? What information returns to the sponsor or regulator?

Every decision affects both safety and access.

A complex REMS may create consistency, but it may also shrink the treatment network. A lighter model may be easier to scale but may not adequately control risks related to prolonged psychoactive effects, patient vulnerability, or inconsistent provider preparation.

The goal should be requirements that are risk based, not fear based.

We should not build six locks on the door simply because the treatment is unfamiliar. We should build the right locks for the actual risks. A REMS that protects patients is essential. However, a REMS so complex that only a few urban centers can operate it may turn access into a privilege.

That concern becomes even more significant if durability requires periodic re-treatment. Each additional administration brings the patient back through the same system, including scheduling, certification, observation, transportation, staffing, and reimbursement.

Access Is More Than an Approval Date

HRSA’s recent request for information shows that federal agencies are already asking how potential FDA-approved psychedelic therapies could be delivered in ambulatory clinics, rural health centers, and medically underserved communities. The request addresses training, pre-administration preparation, delivery models, follow-up, and the infrastructure needed for safe care.

This moves the discussion beyond the controlled world of Phase 3 trials.

A treatment session may require a dedicated room, trained clinical personnel, patient preparation, hours of observation, emergency procedures, transportation planning, controlled-substance storage, and follow-up after discharge.

Large academic centers may be ready with this infrastructure. A small psychiatric practice or rural clinic may not. Yet excluding those settings would leave many patients behind.

The people carrying the greatest burden of depression may have the least ability to travel several hours, take a full day away from work, arrange childcare, secure transportation, or navigate a complicated authorization process.

The product may act quickly. The patient’s journey to receive it may not.

Development programs should therefore collect operational evidence, not just clinical outcomes. How long does each stage of care take? Which personnel are truly essential? What causes cancellations? How does distance affect participation? Which follow-up can occur remotely? How does clinical-site experience influence safety and outcome?

The delivery system is not separate from the value of the treatment. It determines how many patients can actually receive it.

The VA Has an Opportunity to Learn, Not Simply Deploy

The HHS-VA partnership is important because the VA is not merely another potential treatment network.

It has national reach, longitudinal records, mental-health expertise, pharmacy infrastructure, and the ability to observe outcomes over time. The collaboration announced this week includes research, clinician preparation, evidence-based protocols, real-world evidence, cost information, and readiness for possible future deployment.

The VA also cares for patients whose clinical reality often extends beyond one diagnosis. Veterans may live with PTSD, MDD, TRD, chronic pain, traumatic brain injury, anxiety, substance use, sleep disorders, suicidality, or several of these at once.

The real patient does not read the inclusion criteria before arriving. That makes the VA especially valuable as a learning healthcare system.

It can help answer who benefits, who does not, how comorbidities affect safety, how frequently re-treatment is used, whether outcomes vary by geography or provider experience, and whether improvement translates into fewer hospitalizations, emergency visits, or other healthcare utilization.

Its greatest contribution may not be faster adoption but disciplined learning to advance the field. A deployment program that counts doses will tell us very little but a system that connects treatment, safety, durability, function, utilization, and cost can help shape the field for years.

Real-World Evidence Must Be Designed Before Launch

Registration trials are designed to answer specific questions in controlled populations.

Clinical practice immediately becomes less orderly.

Patients have more comorbidities. Medication histories are incomplete. Providers differ in experience. Supportive models drift. Follow-up becomes uneven. Re-treatment becomes individualized.

Some patients may improve in ways that a depression scale does not fully capture. Others may show early response and then decline before a formal relapse endpoint is reached.

Real-world evidence should therefore be designed as the next chapter of development, not the appendix written after launch.

We need to understand who receives treatment, what they have previously tried, what other illnesses they carry, how acute psychoactive effects are documented, how long benefit lasts, when relapse occurs, how frequently re-treatment is used, and whether the therapy changes hospitalizations, emergency care, employment, relationships, and day-to-day function.

This is where comparison with standard of care becomes especially important.

A psychedelic therapy should not be judged only by whether it produces a larger short-term score change than placebo. It should be evaluated against the burden and outcomes of existing treatment pathways.

Does it reduce years of daily medication? Does it postpone or avoid ECT? Does it reduce repeated esketamine visits? Does it produce longer remission than the therapy it replaces? Does it improve adherence because administration is intermittent? Or does it simply exchange one recurring treatment cycle for another, with more infrastructure around each visit?

Real-world evidence should not replace randomized trials. That would be like asking the rearview mirror to replace the windshield.

Randomized evidence and real-world evidence answer different questions. One helps establish whether a treatment can work under defined conditions. The other shows how it performs after those conditions loosen.

For psychedelic therapies, post-approval evidence must be part of the development architecture from the beginning.

Approval Is the Beginning

The FDA’s final guidance is an important step. It gives the field more clarity and confirms that psychedelic drug development is being treated as a serious area of medicine.

The broader federal activity is equally significant because it recognizes something drug developers already understand: the molecule is only one part of the challenge.

The next phase will test whether sponsors can define patient populations precisely, manage functional unblinding honestly, demonstrate benefit beyond the acute phase, characterize subjective safety events consistently, plan for re-treatment, navigate scheduling, design workable REMS programs, prepare clinical sites, and generate evidence after approval.

The field should avoid two extremes.

The first is the belief that these treatments appear transformative and should therefore be held to a lower standard.

The second is the belief that their complexity makes safe, scalable care impossible.

Neither serves patients.

Millions of people live with depression that has not responded adequately to available treatment. Many have cycled through medication after medication. Some have received repeated procedures. Others have experienced improvement only to watch the disease return.

The possibility of a new option deserves both urgency and careful access to patients.

The central question for psychedelics is whether patients remain well after the immediate effect has passed.

The next chapter will not be defined only by a statistically significant endpoint or an accelerated FDA review. It will be defined by whether the evidence remains credible, whether recovery extends beyond a rating scale, whether benefit lasts, and whether the treatment can move safely from specialized trial sites into the lives of patients who need it.

More than 21 million U.S. adults experienced a major depressive episode in a single year, including 14.5 million whose illness caused severe impairment. More than 49,000 Americans die by suicide annually. These are not abstract epidemiologic figures. They represent patients cycling through treatments, families watching relapse return, and clinicians running out of options.

The signal is clear. Now we have to build the road, and we cannot afford for it to end at approval.