CSF galanin and noradrenaline downregulation by psilocybin therapy in major depressive disorder

Major depressive disorder remains a major cause of disability, and many patients do not achieve remission with first-line treatments, with a substantial subgroup meeting criteria for treatment-resistant depression. The introduction argues that this therapeutic gap reflects incomplete understanding of the biology of depression and the need for mechanistically informative biomarkers. Psilocybin has emerged as a promising rapid-acting antidepressant, but its biological mechanism in MDD is still unclear. Paslawski and colleagues set out to characterise the biological effects of psilocybin-assisted therapy in MDD using a randomised, placebo-controlled secondary analysis with a multi-modal biomarker strategy. The study focused on cerebrospinal fluid monoamines and metabolites, inflammatory cytokines, BDNF, plasma proteomics, and p11 in peripheral blood mononuclear cells, with the aim of identifying molecular signatures linked to treatment response and to psilocybin’s proposed neuroplastic and immunological effects.

This was a secondary analysis of a previously described randomised, placebo-controlled trial in 35 participants with MDD assigned to psilocybin or niacin. Samples were collected 1-2 days before dosing and again about 14 days later. Study medication was administered with 200 mL of water. Participants were monitored by psychologists, with a physician available if needed. Cerebrospinal fluid was obtained by lumbar puncture at the L3-L4 or L4-L5 level while participants were upright. Five millilitres of CSF was collected, immediately centrifuged, divided into 1 mL aliquots, and frozen until analysis. Seventeen participants in each group provided paired blood samples, and paired CSF samples were available from 14 psilocybin-treated participants and 11 placebo participants. Only individuals with paired samples were included in the statistical analyses. The biomarker programme included a targeted cytokine panel (IL-1β, IL-6, TNF-α, IFN-γ), monoamine neurotransmitters and metabolites in CSF, BDNF and proBDNF/BDNF, Olink Target 96 Metabolism proteomics, and p11 measurements in PBMCs and leukocyte subpopulations. The psilocybin and placebo groups were broadly similar in age, MADRS score, and sex ratio in the paired-sample subsets reported. The study had ethical approval from the Swedish Ethical Review Authority and the Swedish Medical Products Agency, and all participants gave written informed consent. For the follow-up analyses, changes in analyte levels from baseline to the post-treatment visit were compared between groups using baseline values as covariates. The extracted text does not give further detail on the specific statistical model beyond this adjustment.

In the CSF analyses, psilocybin was associated with lower galanin and lower noradrenaline at follow-up compared with placebo. In the proteomic panel, psilocybin also showed reduced SOST and increased CLUL1. The text notes additional trends in other proteins, but the main reported significant changes were galanin, noradrenaline, SOST, and CLUL1. No differences were seen in p11 levels in PBMCs or in leukocyte subpopulations. CSF proBDNF/BDNF showed substantial variability and no significant change. The targeted cytokine panel did not show significant treatment-related changes in IL-1β, IL-6, TNF-α, or IFN-γ. When baseline-to-follow-up changes in analytes were correlated with change in MADRS scores, the strongest association was for galanin. The next strongest correlations were for adrenaline, SOST, SDC4, and serotonin. Galanin and SOST were positively correlated with MADRS change, whereas adrenaline, serotonin, and SDC4 were negatively correlated with MADRS change. The extracted text does not provide the exact correlation coefficients or p-values for these associations.

The authors interpret the study as identifying a multimodal CSF biomarker signature of psilocybin-assisted therapy in MDD, with the most robust finding being reduced CSF galanin. They emphasise that this change occurred both within the psilocybin group and relative to placebo, and that galanin change was strongly correlated with change in MADRS scores. They frame galanin as a stress-inducible inhibitor of monoaminergic firing that is co-localised with noradrenaline in locus coeruleus neurons and with serotonin in dorsal raphe neurons, and they note that earlier post-mortem and genetic studies have linked galanin signalling to depression. The discussion argues that the simultaneous reduction in CSF noradrenaline is unlikely to be incidental. The authors propose that psilocybin’s 5-HT2A agonism may directly suppress locus coeruleus activity, lowering noradrenaline, or that both markers reflect a normalisation of locus coeruleus hyperactivity in depression. They also note that antidepressants that reduce locus coeruleus activity may decrease galanin expression, suggesting galanin may track locus coeruleus activity rather than simply oppose it. Beyond galanin and noradrenaline, the authors view the changes in SOST and CLUL1, and the trends in SEMA3F and NT-proBNP, as potentially reflecting a less stressed, less immune-activated, and less vascularly strained state, although none of these proteins had previously been linked to MDD, psilocybin, or niacin. They present this as broadly consistent with psilocybin’s proposed neuronal network resetting effects. The authors also discuss the absence of significant cytokine, BDNF, and p11 effects. They suggest that the timing of sampling may explain the cytokine null findings, because inflammatory changes may be strongest in the first week after treatment. For p11, they suggest mechanistic differences from SSRIs and possibly a different time course for rapid-acting agents, so the null result is presented as a constraint on future hypothesis testing rather than evidence against a role for p11. The main limitations identified are the modest sample size for biomarker analyses, the possibility that a single sampling time point missed time-sensitive molecular effects, and clinical heterogeneity in the cohort, including differences in severity, treatment history, and inflammatory subtype. The authors conclude that the study provides a neurochemical and proteomic characterisation of psilocybin’s effects in MDD and supports the galanin-noradrenaline axis as a key pharmacodynamic signature. They state that reduced CSF galanin is a novel biomarker candidate that needs validation in larger cohorts and further correlation with clinical outcomes.