Psilocybin-assisted therapy is emerging as one of the most closely watched approaches for treating depression, with early trials suggesting rapid and sometimes lasting relief after just one or two guided sessions. However, as studies multiply the growing volume of data is becoming harder to maintain a clear, contextual understanding of the overall evidence.

To keep pace, researchers at the University of Pennsylvania, Philadelphia, developed Synthesis of Psychedelic Research Studies (Sypres), a “living” systematic review and meta-analysis of psilocybin trials for depression.

The open-access database (sypres.io), described in a paper published online on April 6 in Nature Mental Health, tracks new studies as they are published, offering an ongoing synthesis of results.

Better Clinical Decision-Making

Developed by the Penn Lifespan Informatics and Neuroimaging Center (PennLINC), the platform launched with a review of 15 randomized controlled trials involving 801 participants.

Investigators focused the main analysis on a subset of 12 trials (n = 585) with similar designs, finding a large reduction in depression scores compared with control conditions (Hedges g = -0.90).

“While psychedelic-assisted therapy shows promise, we can’t rely on individual trials to guide clinical decisions,” said lead author S. Parker Singleton, PhD, a senior research and data scientist at PennLINC, where he leads research and development for the Sypres initiative.

He described Sypres as a “necessary counterweight to the hype,” offering a transparent resource for evaluating the latest synthesized evidence.

Sypres helps bridge the lag between publication of new findings and their clinical interpretation. The pace of the field, Singleton noted, has rendered traditional, one-off research papers nearly outdated by the time they are published.

“We’re trying to shorten the gap from when data first appear and how they’re understood,” said Singleton, who co-led the study with Brooke Sevchik, a data scientist at PennLINC.

“Even over the course of this paper, new trials were coming out. A static review just can’t keep up,” Sevchik said.

Unlike conventional meta-analyses that offer a static snapshot, Sypres uses an interactive dashboard to incorporate new findings as they become available. This allows users to conduct their own analyses, excluding specific trials or focusing on patient subgroups.

The database is maintained through a multi-institutional collaboration based at PennLINC and directed by Ted Satterthwaite, MD, a senior author on the paper. The team brings together researchers from Johns Hopkins University, Vanderbilt University, and the University of Rochester to centralize, manage, and scale the expanding field of psychedelic research.

‘Living’ Evidence

Sypres is built on the infrastructure of the Metapsy project, an international open-science initiative based at Vrije Universiteit (VU) Amsterdam, Amsterdam, Netherlands. Developed over more than a decade and launched online in 2022, Metapsy serves as a centralized hub for psychotherapy meta-analyses, with data standards and software tools to ensure findings are reproducible and open to scrutiny.

Led by Pim Cuijpers, PhD, professor emeritus of clinical psychology, and Eirini Karyotaki, PhD, associate professor of clinical psychology, the initiative spans 12 major mental health conditions — including depression, anxiety disorders, posttraumatic stress disorder, and eating disorders — and synthesizes evidence from more than 1000 randomized trials. This work has established a rigorous blueprint for “living” evidence models, enabling more transparent and comparable analyses.

Mathias Harrer, PhD, assistant professor at VU Amsterdam and a lead developer of the framework, said an evidence-based “living” model is essential in fast-moving and often polarized areas of research.

“Maintaining up-to-date information is especially valuable in psychedelic research, where the discourse between ‘true believers’ and skeptics can be quite heated,” Harrer said.

He noted that psychiatric meta-analyses often reach conflicting conclusions based on opaque analytic decisions, making it difficult to separate evidence from researcher assumptions.

Psychedelic trials introduce additional complexity, he added, because they combine pharmacologic and psychotherapeutic elements. This requires expert judgment when coding studies and evaluating the risk for bias.

“A living database, combined with tools that allow researchers to run their own analyses, fosters a more open and rigorous engagement with the evidence,” Harrer added.

He cautioned that early enthusiasm with regard to psychedelics frequently outpaces the data.

“In psychiatric research, highly enthusiastic early findings are often followed by more sobering results as studies expand to broader populations,” he noted.

Interest in psilocybin has grown rapidly, driven in part by the limitations of existing treatments. Depression remains a leading cause of disability worldwide, and many patients do not respond fully to antidepressants or psychotherapy.

Keeping Pace With the Data

This therapeutic gap has fueled a surge in research, with more than 130 trials currently underway. However, the evidence base is still evolving.

The largest trial to date, involving 233 patients with treatment-resistant depression, found that a 25-mg dose of psilocybin significantly reduced depression scores at 3 weeks, whereas lower doses showed no meaningful effect.

“Early studies made psilocybin look like a kind of miracle treatment,” Singleton said. “More recent data suggest a more measured outlook, indicating that psilocybin may work about as well as existing treatments like SSRIs [Selective serotonin reuptake inhibitors].”

Sypres’ meta-analysis not only showed a large overall effect size for symptom reduction but also identified several methodological challenges. Trials remain relatively small and heterogeneous, with substantial variation in design.

In addition, functional unblinding — where participants infer their treatment assignment because of psilocybin’s psychoactive effects — can undermine placebo-controlled designs and inflate perceived benefits through expectancy effects.

The risk for bias is another concern. Only a minority of studies included in the Sypres review were rated as low risk, while many had “some concerns” or were considered high risk. Participants were also highly selected, raising questions about how well findings translate to broader clinical populations.

A further complication is that psilocybin is almost always delivered alongside intensive psychological support. Because every trial in the Sypres review incorporated structured support before, during, and after dosing, the researchers could not isolate the drug’s independent effect. This makes it difficult to determine how much of the observed benefit is attributable to psilocybin itself vs the accompanying therapy.

With a growing number of studies in progress, Sypres is positioned to incorporate a broader range of findings across conditions. The platform currently focuses on psilocybin for depression, but the team is extending the model to other areas, including 3,4-methylenedioxymethamphetamine-assisted therapy for posttraumatic stress disorder.

For now, much of the published research reflects only a narrow slice of the broader landscape, with many additional trials expected to reshape understanding of these treatments. As psychedelic therapies move closer to wider clinical use, Singleton said expectations will need to remain grounded in data that continue to evolve.

“There’s a tendency for early results to set expectations,” he said. “What we’re trying to build is a resource that evolves with the data, so those expectations can change as the evidence does.”

Disclosure information for study authors is available in the original study publication.