RN Collins (Series 3) No.7 – Federal Regulatory Standards After the MDMA Rejection: What FDA’s Complete Response Letter May Mean for the Psychedelic Pipeline

RN Collins (Series 3) No.7

RN Collins has written a new series (3) of 18 articles for cannabis law report on 2026 Psychedelics & Legal Issues.

This is the 7th in the series

Contact RN Collins: https://www.linkedin.com/in/rn-collins/

July 2026

Federal Regulatory Standards After the MDMA Rejection: What FDA’s Complete Response Letter May Mean for the Psychedelic Pipeline

The CRL, the Compass pipeline, the April 2026 executive order, and the open legal and regulatory questions facing the psychedelic industry

RN Collins, M.S., JD Candidate

https://www.linkedin.com/in/rn-collins/

I. Introduction

MDMA-assisted therapy was, for several years, widely expected to be the first psychedelic compound to clear FDA approval. The clinical data from the pivotal Phase 3 trials, MAPP1 and MAPP2, showed PTSD remission rates of 67 to 71% in the MDMA-assisted therapy arm compared to 32 to 48% in placebo-assisted therapy, and FDA had granted Breakthrough Therapy Designation in 2017. From the outside, the path looked close to a formality. It was not.

On August 9, 2024, FDA issued a Complete Response Letter rejecting Lykos Therapeutics’ New Drug Application for MDMA-assisted therapy for PTSD. The rejection followed a 9-2 advisory committee vote against efficacy and a unanimous vote on safety concerns at the June 2024 Psychedelic Drugs Advisory Committee meeting. What the letter and the surrounding record actually identified as deficient, and what those findings may imply for every psychedelic sponsor that follows, is arguably more consequential than the headline rejection itself.

Three issues emerge from the record: apparent data integrity concerns tied to Lykos’s therapy training program, an unresolved methodological question about blinding in trials of a psychoactive substance, and an absence of a therapist-conduct risk mitigation framework. None of these findings is specific to MDMA as a molecule. Each touches the drug-plus-therapy structure that characterizes most psychedelic-assisted treatment protocols under development, including the psilocybin programs now advancing toward their own FDA submissions. Separately, Compass Pathways has reported two positive Phase 3 trials for its psilocybin candidate COMP360, has received FDA’s grant of rolling NDA submission, and had submitted portions of that NDA as of May 2026, with final submission anticipated in Q4 2026. The DEA’s 2026 aggregate production quotas, psilocybin at 50,000 grams and psilocin at 80,000 grams, both increased from 2025, may reflect growing recognition of clinical trial activity, though a quota increase is not itself a scheduling determination and should not be read as one. The April 2026 executive order directing the Attorney General to initiate scheduling reviews upon Phase 3 completion appears, on closer reading, to add little material acceleration, since DEA is already obligated by statute to consider rescheduling within 90 days of an FDA approval.

This article works through what the record actually says, and what it may, and may not, mean for the legal and regulatory posture of the psychedelic industry going forward. Section II analyzes the three CRL deficiencies and considers how broadly they might apply beyond MDMA specifically. Section III addresses the functional unblinding question in more detail. Section IV considers what a resubmission might need to include. Section V surveys the psilocybin pipeline and the scheduling backdrop. Section VI closes with a set of observations about what the CRL record could mean for how psychedelic drug development, and its surrounding legal and regulatory questions, are approached going forward, along with a short set of recommendations addressed to the agencies and institutions best positioned to resolve the open questions.

II. The CRL and Its Three Deficiencies

FDA’s Complete Response Letter to Lykos identified three primary areas of concern, based on the available record, principally the FDA Briefing Document prepared for the June 2024 advisory committee meeting and Lykos’s own public characterization of the letter, since the CRL itself has not been made public.

Deficiency 1: Data Integrity Concerns in the Therapy Training Program

FDA appears to have identified significant concerns about the fidelity of Lykos’s MDMA-assisted therapy training program. Based on the available record, FDA seems to have determined that therapist conduct issues, including boundary violations documented in multiple post-trial proceedings, may have compromised the agency’s ability to characterize the therapy component as a standardized, replicable treatment deliverable safely across a general population. This finding is not, on its face, a pharmacological one. It appears instead to be a finding about the drug-therapy combination that constitutes the proposed treatment. FDA had been asked to evaluate MDMA-assisted therapy as a combination product, a molecule paired with a structured psychotherapy protocol, and where the therapy protocol’s integrity is in question, the combination product as a whole may be difficult to approve regardless of the pharmacological data standing alone.

The implication for other psychedelic NDAs involving a drug-plus-therapy combination, which would include most psilocybin-assisted therapy protocols currently in development, may be fairly direct, though it should not be overstated: FDA appears likely to evaluate the therapy training program with a rigor comparable to that applied to the pharmacological data. It may follow that therapist training fidelity, therapist selection criteria, therapist conduct monitoring, and mechanisms for removing and replacing therapists whose conduct falls outside protocol bounds would need to be specified in some detail in a future NDA and demonstrated to be replicable at scale. Whether FDA would in fact require this level of specificity for every combination product, or only where a sponsor’s record raises the kind of concerns present in the Lykos trials, is not yet established with any certainty.

Deficiency 2: Inadequate Blinding Characterization

FDA appears to have determined that Lykos had not adequately addressed what is sometimes called the functional unblinding problem, the practical difficulty, perhaps near-impossibility, of maintaining subject blindness in a trial of a substance with pronounced subjective effects. The advisory committee’s 9-2 efficacy vote seems to reflect, at least in part, a concern that the observed treatment effect could not be reliably distinguished from an expectancy effect amplified by functional unblinding. FDA has not proposed or endorsed any particular solution to the functional unblinding problem, and it may not have one. The agency has, however, effectively signaled that future NDAs will likely need to document blinding methodology, attempt to measure blinding adequacy, and address what a blinding failure could mean for the estimated effect size.

The COMP005 trial data for Compass Pathways’ COMP360 showed a 3.6-point MADRS advantage at six weeks, a comparatively modest effect size relative to some antidepressant benchmarks. A more modest effect size may leave proportionally more room for functional unblinding to serve as an alternative explanation: if subjects who correctly guess their treatment assignment show greater improvement than those who do not, that pattern could account for a larger share of a modest effect than of a large one, though this remains an inference rather than a demonstrated fact in COMP360’s case specifically. This is the general area of scrutiny that a COMP360 NDA’s blinding characterization section would likely need to address, though how FDA will ultimately weigh it cannot be predicted with confidence.

Deficiency 3: Therapist Safety Risk

FDA’s third area of concern, as best can be determined from the record, was the absence of a risk evaluation and mitigation strategy (REMS) addressing the specific risk of therapist misconduct during MDMA-assisted therapy sessions. This concern appears grounded in documented instances of therapist boundary violations in the clinical trial context and in the inherent vulnerability of patients in an MDMA-altered state during multi-hour sessions, though it bears emphasis that documented instances of misconduct in a trial setting are not the same as a finding that fraud occurred, and this article does not take a position on whether any individual’s conduct met that legal standard. A REMS requirement for a combination product of this kind would likely mean that any FDA-approved MDMA-assisted therapy would operate under a restricted distribution system involving prescriber, dispenser, and patient certification requirements.

For Compass Pathways’ COMP360 psilocybin program and other psilocybin-plus-therapy combination NDAs, the REMS question may be fairly direct, though far from settled: a COMP360 NDA may need to address therapist misconduct risk with a REMS, or to explain why the pharmacological and structural characteristics of COMP360 therapy differ enough from MDMA-assisted therapy to avoid that requirement. An argument that single-session psilocybin therapy, given its shorter duration of action, narrows the relevant risk window may have some force, but it would need to be made explicitly and supported by session management data, and there is no indication yet of how FDA would receive such an argument.

III. The Advisory Committee Process and What a Resubmission May Require

FDA’s Psychedelic Drugs Advisory Committee voted 9-2 against efficacy and unanimously on safety concerns at the June 4-5, 2024 meeting. The 9-2 efficacy vote does not appear to have been a finding that MDMA-assisted therapy is ineffective. It seems better read as a finding that the committee could not, based on the data package before it, conclude with confidence that the apparent efficacy was attributable to the drug-therapy combination rather than to expectancy effects amplified by functional unblinding and to a treatment arm that received more therapist contact time than the comparator arm.

Resilient Pharmaceuticals, LLC, which acquired Lykos’s MDMA-assisted therapy assets following a reported $50 million Series B investment in late 2024, is reported to be developing a resubmission strategy addressing each of the three CRL deficiencies described above. A resubmission with a reasonable chance of success would likely need to include: (1) a restructured therapy training and certification program with demonstrated fidelity monitoring and conduct remediation mechanisms; (2) a pre-specified blinding adequacy analysis with sensitivity analyses addressing the functional unblinding effect; and (3) a REMS proposal addressing therapist conduct risk. Each of these elements is specific to MDMA-assisted therapy as a combination product, but each may carry implications worth considering for psilocybin NDA strategy as well, even though the degree of overlap is not yet established.

IV. The Psilocybin Pipeline: Compass Pathways and the Scheduling Backdrop

A. Compass Pathways: COMP360

Compass Pathways has reported two positive Phase 3 trials for COMP360 (synthetic psilocybin, Polymorph A). COMP005 (n=233) reported positive results in June 2025, with a statistically significant 3.6-point MADRS advantage at the six-week primary endpoint. COMP006 reported positive results in February 2026, with high statistical significance on its primary endpoint. FDA granted rolling NDA submission for COMP360 in Q1 2026; portions of the NDA had reportedly been submitted as of May 13, 2026, with final submission anticipated in Q4 2026.

A COMP360 rolling NDA would likely need to address the same three areas of concern identified in the MDMA CRL, though it is not certain FDA would weigh them identically for a different molecule and protocol. The therapy training and conduct monitoring question may be addressed, at least in part, by Compass’s structured COMP360 therapy protocol and its facilitator training program, which appears to have been developed with the CRL deficiencies in view. The functional unblinding question would likely require a COMP360 NDA to include a pre-specified blinding adequacy analysis addressing what the modest 3.6-point MADRS effect size could mean for functional unblinding as an alternative explanation. The REMS question would likely require Compass to propose a risk mitigation strategy for therapist conduct, or to demonstrate why COMP360’s protocol characteristics reduce that risk enough to avoid a REMS, an argument that may or may not persuade FDA, depending on data not yet public.

B. The Scheduling Backdrop

The DEA’s 2026 aggregate production quotas, psilocybin at 50,000 grams (up from 30,000 in 2025), psilocin at 80,000 grams (up from 36,000 in 2025), may signal growing recognition of clinical trial activity, but a quota increase does not constitute a scheduling determination and should not be read as a prediction of one. Both signals point in a similar directional sense toward eventual rescheduling discussion, without changing the operative legal standard for scheduling itself.

The legal standard for “currently accepted medical use” in scheduling determinations was established in Grinspoon v. DEA, 828 F.2d 881 (1st Cir. 1987), and the DEA’s operative five-part test, requiring known and reproducible chemistry, adequate safety studies, adequate and well-controlled efficacy studies, qualified expert acceptance, and wide scientific availability, was established in Alliance for Cannabis Therapeutics v. DEA, 15 F.3d 1131, 1135 (D.C. Cir. 1994), https://law.justia.com/cases/federal/appellate-courts/F3/15/1131/536307/. It could be argued that both standards are now largely satisfied for COMP360 following two positive Phase 3 trials, though this is an inference and not a settled legal conclusion, and reasonable observers might disagree about how far two positive trials go toward satisfying a five-part test built around FDA approval as much as trial results. The underlying question, whether FDA approval, as opposed to positive trial data alone, is a prerequisite for DEA scheduling action, remains open and is not resolved by anything discussed in this article.

The April 2026 executive order directing the Attorney General to initiate scheduling reviews upon Phase 3 completion appears, on the available reading, to add limited material acceleration: DEA is already required by statute to consider rescheduling within 90 days of an FDA approval, and the order does not appear to change FDA’s evidentiary standards for NDA approval, the Schedule I status of MDMA and psilocybin, or the requirement that sponsors complete clinical trials demonstrating safety and efficacy. The April 2026 DEA rescheduling action that moved certain cannabis products to Schedule III does not appear to have any direct application to psychedelics.

Advanced Integrative Medical Science Institute v. DEA (AIMS II), No. 22-1568, 2023 WL 7101927 (9th Cir. Oct. 27, 2023), directed DEA to reconsider its prior denial of a petition to move psilocybin to Schedule II for use in terminally ill patients. The case remains pending on remand; its eventual resolution may help clarify whether DEA can act on a rescheduling petition for a substance that lacks FDA approval for any indication, though that clarification has not yet occurred, and any prediction about how the remand will resolve would be premature.

V. What the CRL Record May Mean Going Forward

A. Implications for How Psychedelic Drug Development Is Assessed

The MDMA CRL appears to have shifted the landscape for evaluating psychedelic drug development in at least three respects, though each should be read as a directional signal rather than a settled rule. First, the record suggests FDA will evaluate psychedelic drug-therapy combinations as combination products, scrutinizing both the pharmacological and the therapeutic components together rather than treating the drug as severable from its delivery protocol. Any assessment of a psychedelic sponsor’s prospects that relies solely on Phase 2 efficacy data, without a parallel look at the therapy training program, therapist selection process, and conduct-monitoring infrastructure, risks resting on an incomplete picture.

Second, the record suggests functional unblinding has become a specific area of FDA attention that psychedelic Phase 3 trial designs may need to address from the protocol design stage forward, though how rigorously and in what form remains to be seen. Trial designs that do not pre-specify a blinding adequacy analysis could face the kind of advisory-committee skepticism that MAPP1 and MAPP2 encountered, though this is an inference drawn from a single advisory committee outcome rather than a guaranteed pattern. Third, the record suggests therapist conduct risk has become a safety question requiring active mitigation rather than passive monitoring. Sponsors whose therapist training programs lack explicit conduct monitoring and remediation mechanisms may be building Phase 3 programs that a reviewer, applying the CRL record as a reference point, could view skeptically, though this too is a matter of degree rather than certainty.

Compass Pathways’ two positive Phase 3 trials for COMP360, its rolling NDA submission underway as of May 2026, and its reported April 2026 CNPV voucher (said to target a one-to-two month priority review) together appear to represent the most concrete near-term FDA approval scenario currently in the psychedelic pipeline, though “most concrete” here is a relative and not an absolute characterization. If used for the COMP360 NDA, the CNPV voucher could accelerate FDA review from a standard 10-to-12 months to approximately 6 months, potentially producing an FDA decision by mid-2027, a timeline that remains contingent on the final NDA submission and on FDA’s review of the same three deficiency categories discussed above.

B. Open Questions and Recommendations for Regulators and Policymakers

Several of the questions raised by the CRL record are, at bottom, questions for FDA, DEA, and Congress to resolve rather than questions any individual sponsor or observer can settle through argument alone. A few observations follow, framed as considerations these institutions may wish to take up rather than as instructions to any particular market participant:

1. FDA could consider issuing generalized guidance on blinding adequacy analysis for psychoactive-drug trials, given that the functional unblinding problem identified in the MDMA CRL appears likely to recur across the psilocybin and broader psychedelic pipeline. Absent such guidance, each sponsor may be left to infer FDA’s expectations from a single advisory committee record, which may not generalize well across different molecules and effect sizes.
2. FDA might also consider clarifying, in guidance or through its handling of the COMP360 NDA, how it intends to evaluate REMS requirements for psychedelic combination products with materially shorter duration-of-action profiles than MDMA. The current record leaves open whether a shorter session length meaningfully reduces the therapist-conduct risk that prompted the MDMA REMS finding, and clearer agency guidance could reduce uncertainty for the broader pipeline.
3. DEA and the Department of Justice may wish to clarify, whether through rulemaking, guidance, or the pending AIMS II remand, whether a rescheduling petition can proceed for a substance lacking FDA approval for any indication. This question has been left open by the current record and its resolution would affect scheduling strategy across the psychedelic and adjacent cannabis policy space.
4. Congress may wish to address, through the ARPA-H matching fund mechanism discussed in connection with psychedelic research funding, the appropriations uncertainty created by the budget reconciliation process underway as of June 2026. The mechanism’s practical availability appears to depend on appropriations that remain subject to ongoing legislative uncertainty, and clearer, more durable funding commitments could reduce planning uncertainty for sponsors and research institutions relying on the mechanism.
5. Finally, the § 280E divergence discussed in the Oregon Playbook article remains relevant background for any entity operating both psychedelic and cannabis programs, since MDMA and psilocybin remain Schedule I substances while certain cannabis products have moved to Schedule III. Regulators addressing tax treatment of Schedule I versus Schedule III substances may wish to consider whether the current framework, developed primarily around cannabis, translates cleanly to the psychedelic therapeutic context, or whether a distinct approach is warranted.

None of the foregoing should be read as a prediction of how FDA, DEA, or Congress will in fact act. The CRL record raises a set of questions that remain, at this writing, unresolved, and this article does not purport to resolve them.

Endnotes

1. FDA, Complete Response Letter, NDA 215455, MDMA-Assisted Therapy for PTSD (Aug. 9, 2024) (Lykos Therapeutics) (the CRL itself is not publicly available; deficiency descriptions here rely on the FDA Briefing Document and Lykos’s own press release); FDA, FDA Briefing Document, NDA/BLA# 215455, Midomafetamine, Psychopharmacologic Drugs Advisory Committee Meeting (June 4, 2024), https://www.fda.gov/media/178984/download; Psychedelic Drugs Advisory Committee Meeting, June 4-5, 2024, FDA Docket No. FDA-2024-N-1352 (9-2 vote on efficacy; unanimous on safety); Lykos Therapeutics, Press Release (Aug. 9, 2024) (confirming CRL and three deficiency categories).
2. Compass Pathways, Q1 2026 Earnings Press Release (May 13, 2026), SEC Form 8-K (confirming COMP005 positive June 2025, primary endpoint 3.6-point MADRS advantage at 6 weeks; COMP006 positive February 2026; FDA rolling NDA submission granted Q1 2026; sections already submitted; final submission on track Q4 2026); Compass Pathways, Form 8-K (Feb. 2026) (COMP006 primary endpoint results).
3. DEA, Established Aggregate Production Quotas for Schedule I and II Controlled Substances for 2026, Docket No. DEA-1568E, 91 Fed. Reg. (Jan. 5, 2026), https://www.federalregister.gov/documents/2026/01/05/2025-24277/established-aggregate-production-quotas-for-schedule-i-and-ii-controlled-substances-and-assessment (psilocybin quota 50,000g; psilocin quota 80,000g); Grinspoon v. DEA, 828 F.2d 881 (1st Cir. 1987) (“currently accepted medical use” standard); Alliance for Cannabis Therapeutics v. DEA, 15 F.3d 1131, 1135 (D.C. Cir. 1994) (DEA five-part accepted medical use test), https://law.justia.com/cases/federal/appellate-courts/F3/15/1131/536307/; Advanced Integrative Medical Science Institute v. DEA (AIMS II), No. 22-1568, 2023 WL 7101927 (9th Cir. Oct. 27, 2023) (directing DEA to reconsider prior denial of psilocybin Schedule II petition).