In brief
- An analysis of an open-label pilot in people with Parkinson’s disease measured blood inflammation markers before and after two psilocybin doses; plasma could be analysed for 9 of the 12 participants who completed the protocol.
- An index combining five proinflammatory markers was lower than baseline the day after each dose and 30 days after the second.
- There was no control group and everyone knew they were taking psilocybin: this is a biological signal to investigate, not proof of an anti-inflammatory effect or of clinical benefit.
On October 21, 2026, JAMA Psychiatry published a research letter from a University of California, San Francisco (UCSF) team, by Bradley, Kelley and colleagues, with Aoife O’Donovan as senior author. The work explores whether psilocybin is associated with changes in markers of inflammation and mitochondrial stress in people with Parkinson’s disease, a question that until now rested mainly on cell and animal studies.
What was measured
The data come from a single-arm, open-label pilot trial (NCT04932434) in adults aged 40 to 75 with mild to moderate Parkinson’s disease (Hoehn and Yahr stages 1 to 3) and mood dysfunction. Twelve completed the protocol: 10 mg of oral synthetic psilocybin and, about two weeks later, 25 mg, with psychological support. Samples were taken at baseline, the day after each dose and 30 days after the 25 mg dose; plasma was analysed for 9 people.
The primary outcome was a proinflammatory index averaging five markers: IL-6, IL-8, TNF-α, interferon gamma and C-reactive protein (CRP). The team also measured IL-10, an anti-inflammatory molecule, and GDF-15 and FGF-21, two circulating indicators of mitochondrial stress. Each time point was compared with baseline using mixed-effects models, with correction for multiple comparisons.
Results
The index was lower than baseline at all three time points, with standardized mean differences of −0.72 the day after 10 mg, −0.60 the day after 25 mg and −0.62 at 30 days (adjusted P values of .003, .02 and .047; the last only just below the usual .05 threshold).
Individually, IL-6, IL-8 and CRP fell significantly only after the first dose; interferon gamma and TNF-α showed no significant changes after correction. IL-10 was higher the day after the 10 mg dose, but not at the other time points. GDF-15 fell after each dose, though not at 30 days, and FGF-21 rose slightly without reaching significance.
Limitations
The authors themselves point to the open-label design, the lack of a control group and the small sample. Without a comparator, a possible effect of psilocybin cannot be separated from placebo, regression to the mean, clinical heterogeneity or changes in stress and sleep. False positives may remain despite correction, and GDF-15 and FGF-21 are sensitive but not specific markers. This analysis also does not measure symptoms: a blood marker is not the same as an improvement in the disease.
The work was funded by Silo Pharma and an anonymous donor, which according to the article had no role in design, analysis or writing. Several authors report ties to pharmaceutical and psychedelic-industry companies, and O’Donovan reports grants from Silo Pharma for this work.
What earlier research says
Some of the same researchers published a systematic review of 61 studies in Brain, Behavior, and Immunity in September: 24 cell studies, 30 animal studies and 7 human studies covering 10 trials. Anti-inflammatory effects were reported by 19 of 24 cell studies and 25 of 30 animal studies, but by only 3 of 10 human trials. In the meta-analysis of 18 mouse studies, the drop in cytokines appeared in models with induced inflammation and not in healthy animals, and the authors conclude that the limited human evidence precludes conclusions about their potential in people. We read only the abstract of that review. The original pilot, published in 2025 in Neuropsychopharmacology, recorded no serious adverse events among its 12 participants; 10 had adverse events, mainly anxiety, nausea and increased blood pressure.
What it means
The finding justifies controlled trials with repeated sampling, follow-up beyond 30 days and more direct measures of immune and mitochondrial biology, as the authors propose. It does not support using psilocybin to treat Parkinson’s disease or inflammation. For people living with the condition, their usual medication and effects such as anxiety or raised blood pressure described in the pilot are reasons not to improvise outside a trial. On a related topic, we covered preclinical DMT findings in Parkinson’s disease, and broader context is in our psilocybin and mushrooms guide.
Sources
- Bradley ER, Kelley P, Szigeti B, et al. Psilocybin and Inflammatory and Mitochondrial Stress Markers in Parkinson Disease: A Nonrandomized Clinical Trial. JAMA Psychiatry, October 21, 2026. Full text consulted via the guest-access link; supplements not consulted.
- Europe PMC bibliographic record (PMID 42905733), with authors, journal and dates; the letter has no abstract.
- Trial registration NCT04932434 on ClinicalTrials.gov (UCSF).
- Lerche AS, Kelley DP, Benros ME, et al. Are psychedelics anti-inflammatory? A systematic review and meta-analysis of studies examining the immunomodulatory effects of serotonergic psychedelics. Brain, Behavior, and Immunity, September 28, 2026. Abstract only.
- Bradley ER, et al. Psilocybin therapy for mood dysfunction in Parkinson’s disease: an open-label pilot trial. Neuropsychopharmacology, 2025. Abstract only.
Illustrative image: “Blood testing equipment in Norway“. Photo: Marius Vassnes, via Wikimedia Commons, licensed under CC BY-SA 4.0. Blood-testing supplies at a Norwegian health centre, unrelated to the study.
Educational content produced from a harm-reduction perspective and with respect for individual freedom. It does not replace advice from a healthcare professional and is not intended to encourage or condemn any drug use.
