Temporal dynamics in neuroimaging as correlates of therapeutic response to psilocybin in major depressive disorder: A systematic review and critical appraisal.
Level 1 - systematic review of randomized trials
Systematic review of neuroimaging studies in clinical depression trials
PubMed 40962065 · doi:10.1016/j.jad.2025.120335
What was done
This systematic review (PROSPERO CRD42024557973) synthesized neuroimaging studies evaluating psilocybin for major depressive disorder (MDD) and treatment-resistant depression (TRD). The authors examined 11 eligible studies utilizing modalities such as resting-state functional connectivity, task-based BOLD fMRI, amplitude of low-frequency fluctuations, dynamic functional connectivity, and magnetic resonance spectroscopy to assess temporal neuroimaging changes post-treatment (early: 0–4 weeks; long-term: 5+ weeks) and correlate them with validated clinical endpoints.
What was found
No numerical estimates, effect sizes, or test statistics were reported in the abstract. Early post-treatment (0–4 weeks) was characterized by decreased network modularity, increased global brain integration, and modulation of affective circuits including the default mode network, amygdala, and prefrontal areas; these changes significantly correlated with clinical reductions in BDI, QIDS, and SHAPS scores. Long-term changes (5+ weeks) demonstrated sustained large-scale network reorganization, particularly increased connectivity between the salience network and prefrontal and parietal cortices.
Why it matters
It organizes the temporal sequence of neural changes following psilocybin treatment, demonstrating that early affective modulation and global integration precede long-term large-scale network reorganization associated with antidepressant response.
Limits
The abstract reports no exact figures or effect sizes. The primary literature is heavily limited by a small total number of studies (11), extensive reuse of overlapping datasets, high exploratory flexibility in neuroimaging analyses, risk of bias, and limited generalizability.