In August 2024, the FDA rejected Lykos Therapeutics’s New Drug Application for MDMA-assisted therapy to treat posttraumatic stress disorder (PTSD). The decision surprised many as just seven years earlier, the agency had granted MDMA breakthrough therapy designation to treat PTSD based on positive preliminary reports. 

The rejection sparked strong criticism from psychedelics and mental health groups. Some characterized the decision as “the epitome of bureaucratic red tape,” arguing the agency “undermined its own regulatory integrity” by choosing politics over science. Others suggested a broader lesson, arguing that the FDA simply is “not equipped to handle the unique aspects of psychedelic therapies” and therefore needs “new regulatory standards.”

A broader narrative quickly took hold: because psychedelics are so unusual, they may be fundamentally incompatible with the FDA approval process. Yet the Lykos rejection tells a different story. It does not indicate that the FDA chose politics over science. Nor does anything in the rejection letter suggest that psychedelics are somehow incompatible with FDA approval. In fact, many of the flaws that ultimately doomed Lykos’s application were methodological ones the FDA had previously identified and offered solutions to.

It is true that psychedelics pose some complex and certainly unusual challenges for clinical trials, including functional unblinding, expectancy bias, and difficulties disentangling the effects of the drug itself from accompanying psychotherapy. But a closer examination of the Lykos saga reveals a more modest—and more important—lesson: psychedelic drugs are not incompatible with FDA approval; poorly designed clinical trials are.  

Using the Lykos saga as a case study, we argue that these and other challenges associated with assessing the safety and efficacy of psychedelic drugs are, at least in the current moment, overblown. In the end, the FDA rejected Lykos’ application not because it abandoned science or because psychedelics are incapable of satisfying the FDA’s safety and efficacy standards. It was that Lykos’s clinical trials failed to do so.

What are Psychedelics?

Psychedelics are a class of psychoactive substances that include classic 5-HT2 agonists like psilocybin (magic mushrooms) and lysergic acid diethylamide (LSD), enactogens or empathogens like methylenedioxymethamphetamine (MDMA), and alkaloids like ibogaine. These substances have attracted tremendous interest because of their potential to treat debilitating conditions including posttraumatic stress disorder, abuse, depression, and anxiety.

At the moment, however, many psychedelics are classified as Schedule I under federal law. Schedule I substances are considered to have a high potential for abuse and no currently accepted medical use. As a result, psychedelics generally cannot be prescribed, administered, or dispensed for medical use. For context, heroin is also classified as Schedule I.

If and when the FDA approves any of these psychedelic substances for a particular indication, that approval will likely prompt the DEA to reschedule to a lower schedule with fewer restrictions on prescribing and administering. Many patients and advocates are thus rooting for psychedelics to gain FDA approval.

What are the FDA’s Requirements for a New Drug?

The FDA, for its part, has maintained that psychedelics are subject to the same evidentiary standards as any other drug. Under the Food, Drug and Cosmetic Act, a drug’s effectiveness must be established by substantial evidence, defined as “evidence consisting of adequate and well-controlled investigations.”

Though there are many ways to satisfy this standard, the gold standard for an adequate and well-controlled investigation has long been a double-blinded study comparing effects of a drug on a control group and a placebo group.

But psychedelics complicate this model. Because they create visual and perceptual changes—commonly known as the “trip”—trial subjects can pretty easily figure out whether they have received the control or the placebo, nullifying any kind of blinding. Further complicating matters is the fact that psychedelics are typically administered in conjunction with psychotherapy, which makes it difficult to decouple the effects of the drug itself from the therapy.

Below we use the Lykos saga to illustrate many of these challenges. We note that the FDA had anticipated many of them and, in several instances, even proposed potential workarounds. Ultimately, Lykos’s application was likely rejected not because these challenges were insurmountable, but because Lykos apparently did not even try to address them.

The Lykos Saga

On December 11, 2023, Lykos Therapeutics submitted a New Drug Application seeking approval for MDMA-assisted psychotherapy to treat PTSD. The results initially appeared encouraging, as subjects who received the MDMA experienced a greater reduction in symptoms than those who received the placebo. But as discussed below, the trials suffered from several severe methodological flaws that ultimately led the FDA to question the reliability of these results.

A. Trial Design Issues

For starters, the FDA had concerns about functional unblinding. Approximately 90% of the clinical trial participants who were assigned the MDMA drug and approximately 75% of those who were assigned the placebo drug correctly guessed which drug they received.

The FDA, by the way, foresaw this problem well before Lykos even submitted its application. In its June 2023 draft guidance (which it finalized in July 2026), the agency warned applicants that certain features of psychedelic drugs such as intense perceptual disturbances—again, the trip—may make it difficult to use a traditional placebo as a control, and that applicants should consider other placebo alternatives such as microdoses of MDMA or other psychoactive drugs that mimic the trip.

The FDA was also bothered by the expectation bias problem—subjects who believed they received the MDMA might have fared better simply because they were expecting to. Meanwhile, those who believed they received the placebo might not have fared as well, because, well, they were expecting not to. The FDA anticipated this problem too and advised that one way to minimize such nocebo distortions is to implement a trial using a low, middle, and high dose without a placebo, and coupling that with a placebo-control trial.

Compounding these two problems was the fact that the trial’s in-session monitors figured out which drug the subjects received based on their behavior during the sessions. The FDA thus concluded that these problems likely impacted any of the observed treatment effects.

The FDA’s rejection letter also noted that about 40% of its participants reported prior MDMA use—way more than the general population with mental illness. This was problematic for several reasons. For starters, it limited the generalizability of the results because the enrolled participants simply did not reflect the general PTSD population. The study participants were also far more likely to recognize the effects of MDMA and therefore to anticipate experiencing clinical benefits. Put differently, the subject population itself amplified the functional unblinding and expectation bias problems.  

And just to be clear, these sorts of blinding and expectation bias challenges are not isolated to psychedelics. Similar ones likely also arose in the context of other medications with psychoactive effects, like opiates, benzodiazepines, gabapentin and—most recently—esketamine—all of which the FDA ultimately did approve.

The lesson, then, is not that such trial design issues should be ignored, but rather, they should be managed through thoughtful and careful trial design.

B. Reliability of the Data

The FDA also had serious concerns about the incompleteness of Lykos’s data.

While Lykos’s clinical trial protocol required reporting of adverse events, it limited adverse events to any undesirable, unfavorable, inappropriate, or untoward medical occurrence. It explicitly excluded “positive” or “favorable” effects such as euphoria and altered mood. The problem with that is MDMA acts on the central nervous system and therefore must be assessed for abuse potential. Even if such effects are perceived positively by participants, they must still be reported, as they would be highly relevant to any labeling or monitoring requirements.

The FDA even warned sponsors of this precise issue and even advised them to collect information regarding these positive events, as they are all associated with abuse. Lykos did not listen.

Furthermore, Lykos neglected to report several adverse events tied to at least two sites.

C. Failure to Decouple Effects of MDMA Itself From Effects of Psychotherapy

Lykos’s trials also failed to adequately decouple the effects of the MDMA itself from the effects of the psychotherapy (though the FDA did note this was not an approvability issue). This was important, as Lykos was seeking approval for MDMA in conjunction with therapy, and the FDA only regulates drugs, not psychotherapy.

Here again the FDA had already previously suggested some ways sponsors could address this, including by describing trial design elements intended to reduce potential bias or try to quantify the contribution of psychotherapy to the treatment effect of the drug itself or using a factorial design. But this, Lykos did not do.

Nor is this challenge unique to psychedelics. The FDA has previously approved other drugs designed to be used alongside psychotherapy or counseling, including buprenorphine for opioid use disorder and extended-release naltrexone for alcohol and opioid use disorder. These examples confirm that companies can, in fact, successfully navigate FDA approval for drugs that require therapeutic support.

D. Failure to Demonstrate Long-Term Results  

The FDA was also worried about the durability of any demonstrated effects. PTSD is a chronic condition. Neither of Lykos’s primary two eighteen-month studies (MAPP1 and MAPP2) followed subjects beyond eight weeks after the last MDMA dose.  

While Lykos did attempt to assess durability via one six-month follow-up study (MPLONG), that study—not surprisingly—also had serious methodological problems. For starters, only some of the participants from the two primary eighteen-month studies enrolled, which screams self-selection. Making matters worse, many of those who did self-select into the long-term study sought out other potentially additional therapeutic interventions before the six-month follow-up. How is the FDA supposed to determine whether any long-term effects then were due to MDMA, or other factors?

There is more. The FDA also recommended that Lykos include basic liver toxicity labs, which is crucial because MDMA is metabolized primarily by the liver. The agency further suggested Lykos analyze any interactions between MDMA and SSRIs, a concern which, again, the FDA voiced in its draft (and echoed in its final) guidance.

Looking Ahead

So, all in all, the emerging narrative that psychedelic drugs are incompatible with the FDA’s framework for new drugs appears to be a bit exaggerated.

Yes, challenges do arise with respect to clinical trial design—functional unblinding, expectancy bias, data reporting, and decoupling the contributions of psychotherapy from the MDMA itself. And yes, they should be taken seriously. But that Lykos’s application was rejected does not mean these challenges are insurmountable.

If anything, it’s the opposite. The FDA anticipated many of these challenges and even provided a roadmap for how to handle them. Had only one—or even just some—of these problems been present, the outcome might have been different. The lesson from the Lykos saga then is not that psychedelics require some sort of carveout from the FDA’s approval framework. It is that cutting corners simply won’t work.

This lesson is especially important in the current moment, when both states and the federal government are taking an unusual interest in shepherding psychedelics through FDA approval. To be clear though, we do not contend it is necessarily a bad thing for governments to nudge innovation in this space. But the Lykos saga is a good reminder that psychedelics, like many other drugs, must still undergo rigorous clinical trials that produce reliable, trustworthy evidence of safety and efficacy. In the end, the FDA’s rejection of Lykos’s application might not be evidence that the agency’s framework isn’t flexible enough. It might—just might—be evidence that the system is working.

Prachi Mehta is currently an Academic Fellow at the University of Texas School of Law. Her research focuses on how doctrinal and institutional design choices shape biopharmaceutical innovation, competition, and access.

Melissa Wasserman is the Associate Dean for Research and the Charles Tilford McCormick Professor of Law at the University of Texas School of Law. Her research focuses on innovation policy, particularly on patent law, administrative law, and empirical methods.

This post was written shortly after the May 2026 Conference on Psychedelics and the Law held at the University of Texas at Austin School of Law, with edits made in September 2026.