The State of the Art in Psychedelic Trial Design

“Going forward, for psychedelics to be approved, there will need to be a more complex design [in clinical trials],” said FDA Psychopharmacological Advisory Committee chair Rajesh Narendran during the June 4 meeting when Lykos Therapeutics’ application for MDMA-assisted therapy was rejected.

The advisory panel voted 9-2 that data gathered by Lykos in its clinical trials was insufficient to show that MDMA-AT is effective for PTSD. Lykos addressed these concerns including questions about whether the therapy is safe for patients and whether the research was biased. Scientists who structure trials involving psychedelics are challenged by the limits of the historical gold standards for clinical, experimental, and human subjects study designs. So, what would more complex clinical trial designs look like?

In May, a panel of four psychedelic study trial experts convened at the annual Horizons: Perspectives on Psychedelics conference at the New York Academy of Medicine to discuss the limitations of previous trials and their ideas for better designs.

One of the biggest challenges echoed by the FDA panel is the fact that studying psychedelic-assisted therapy is, essentially, studying the science of therapy, said Natalie Gukasyan, former Medical Director of the Johns Hopkins Center for Psychedelic and Consciousness Research  and current Assistant Professor of Psychiatry at Columbia University Irving Medical Center. 

Questions faced by psychedelic-assisted therapy study designers are the same unresolved questions that have lived at the frontier of therapy science for decades, she said.

For instance: when a therapist has a warm and friendly demeanor, offers some degree of safe and gentle touch, and aims to get to know someone better by asking deep questions with sincere attention, the participant is likely to feel better. These displays of humanity are not only part and parcel of a good therapy session, including a psychedelic-assisted one, but enhance placebo when used disproportionately in a study, Gukasyan said. 

The problem is that different elements of care are provided based on who receives the active drug. It’s simple chemistry: a psychedelic drug can profoundly alter a person’s behavior, which also changes the way the caregiver interacts with them. When people believe they have taken a placebo and not psilocybin, Gukasyan adds, she spends the first hour of the session managing their disappointment. This is a very different interaction than if the participant received psilocybin, which might elicit deeper questions that probe their well-being and build a rapport between the researcher and patient.

More Complex Designs

One of the more complex designs for psychedelic studies may be eschewing placebo conditions altogether. Julie Holland, an author and psychiatrist, argued that it may be unethical to conduct psychedelic-assisted therapy trials with the standard randomized controlled trial design. This structure means neither the clinician nor patient know whether they received placebo or the active drug. The disappointment effect Gukasyan points out has led to suicidality in some patients, said Holland. And for more sensitive populations, such as those with schizophrenia, Holland said it wouldn’t be ethical to subject them to a guessing game.

There is a more ethical study design that is likely to become popular in psychedelic clinical trials, says Holland, which is called a comparative efficacy study. In this trial design, instead of patients receiving a placebo, they receive treatment that is already established to be effective, and experts gauge whether patients who received the experimental treatment fared even better. For example, this design is employed in cancer research. When studying the effects of chemotherapy, one group of patients receives last year’s chemotherapy, and another group receives the new chemotherapy being tested. No one is left untreated. Holland argues that people who are experiencing mental and emotional anguish and are open to an experimental therapy would be better served by such a trial design.

One way that psychedelic-assisted therapy clinical trials, such as the Lykos MDMA-assisted therapy studies, have addressed this issue is to ensure that patients who initially receive placebo are eventually enrolled to participate in MDMA-assisted therapy with an active dose. But these months of waiting confer harms, says Gukasyan. Patients are often asked to go off their existing medications and remove other therapeutic interventions from their routine. As a result, they can be left with an untreated condition for months while waiting for the active treatment.

Better Blinded Studies 

The researchers on the Horizons panel noted that psychedelic trial design, par the course for all human subjects research, aims to correct for expectancy, such as placebo and nocebo effects; experiencing healing or harm based on what one expects to experience in a study. But a psychedelic experience, the main intervention in psychedelic clinical trials, is subject to set and setting, and its lasting therapeutic effect does not appear separable from the study’s design.

The FDA panel, which voted not to recommend the approval of MDMA-assisted therapy, expressed that they wanted to see better techniques for blinding in the studies. Harriet de Wit, an experimental psychologist at the University of Chicago who has studied LSD microdosing, offered several ideas for study designs at the Horizons conference.

Studying microdoses of a psychedelic could be useful to ascertain its effects, de Wit said. In theory, a researcher could find a dose that is small enough to be sub-perceptual to the consumer and caregiver. A problem de Wit encountered in her microdosing experiments, however, is that everyone’s threshold for detecting a microdose in their system is different. Likewise, the dose that shows an effect may not only be different for everyone, but in clinical populations, such as people with depression, people may react fundamentally differently to a psychedelic than non-clinical populations would. One of de Wit’s recent studies showed that people who scored higher on indices of depression experienced more pronounced positive mood shifts and greater reductions to their depression scores after receiving 26 micrograms of LSD, as opposed to  people who weren’t clinically depressed.

In other words, for a blinded design to study a drug’s effect, apart from any expectancy effect such as a subject responding positivity to a placebo, researchers would need to be careful to understand people’s individual sensitivities to the drug to examine its impact and make broader scientific conclusions.

Another idea de Wit proposed is to heighten the placebo effect of people in the control condition to match the same level of high expectations that surround expectations of the benefits of psychedelic compounds. For example, a researcher could tell a study participant that they will receive one of two very promising drugs that have been shown to have great benefits for mental health. This could equalize the expectancy effect, she said.

One way this could be done is to tell study subjects that they may get another drug that also has been shown to reduce depression, for example, but note that it has a different effect. By doing so, researchers could tease out the expectancy effect. But the ethics of this are questionable, says de Wit. More ethical would be a comparative efficacy study which gets more people treated. It would be “comparing apples to oranges, versus apples to placebo,” says de Wit. 

Finally, de Wit calls for more reporting on null studies where the data collected does not show the expected effect or support the researcher’s hypothesis. If scientists publish their null findings, and the media covers these studies, she said this would allow for greater understanding of the intervention and the drugs, and for whom they do and don’t work. Correcting for this publication bias could allow us to understand with greater precision these substances whose effects, like their experiences, elude the grasp of our full comprehension.

Changing Public Perception 

Tehseen Noorani, a scientist at the University of Auckland,  runs an international fellowship of psychedelic study design experts called “Reimagining Psychedelic Trials.” When it comes to FDA approval, Noorani believes that psychedelics are unlike anything the pharmaceutical industry has considered before. 

When studying substances “very sensitive to the mindset of the people using them,” whose effects may be pronounced or wholly influenced by expectancy effects such as placebo, he believes that  the scientifically-demonstrated effect of the substance itself will change as publication perception changes.

“Noorani calls this porous relationship between public perception and clinical studies into psychedelic substances “very sensitive to the mindset of the people using them.” In standard clinical trials, Noorani says it’s common to consider how an intervention works for a specific indication, but not consider the time and context also being integral to the proposed therapy. He says this means that the effects we’re seeing in the science of psychedelics could change, not merely by trial design, but as public perception of psychedelics changes. “We are not used to a drug where in ten years from now, the effects are different, because time and place changed” says Noorani. 

Gukasyan says there is a historical precedent in the first SRRIs which were widely accepted when no other drugs for mood disorders were available and the science showing big effect sizes, whereas now the effect sizes for SSRIs are “puny”.

With such implications, where does a psychedelic trial begin and end? Noorani and his colleagues are considering how to manage this porous relationship between public perception and clinical  studies. For instance, by highlighting elements of clinical trials with psychedelic compounds, even media reports of such investigations will influence the expectations of a study participant. This could change their experience, the results of the study, and the subsequent media stories that report the study. His group calls these recursive cycles “dark loops.”

These design ideas offered by the four Horizons expert panelists could provide useful ideas in the wake of the FDA decision on MDMA-AT expected this week.