Esketamine-induced dentate gyrus plasticity in treatment resistant depression: First-in-human evidence

Background: Treatment-resistant depression (TRD) involves impaired hippocampal plasticity. Preclinical work showed synaptic potentiation and dendritic spine growth in the dentate gyrus (DG) after esketamine (ESK), but human evidence is lacking. We aimed to detect early DG changes after ESK using advanced diffusion MRI, and to test whether baseline MRI metrics predict clinical response. Methods: Twelve adults with TRD and 24 matched controls were enrolled. TRD patients were assessed at baseline (V1), two weeks after ESK initiation (V2), and at treatment completion (V3). Depression severity was rated with the Montgomery-Åsberg Depression Rating Scale. MRI (3D-T1, multi-shell diffusion) was acquired at V1 and V2 in TRD and at V1 in controls. DG volumes were computed with FreeSurfer; diffusion tensor, Q-ball and Neurite Orientation Dispersion and Density Imaging metrics were extracted with the Ginkgo-Toolbox. Linear mixed-effects models tested time, age, and sex effects (FDR-corrected). Pearson correlations assessed MRI-symptom change associations. Results: DG volumes were stable from V1 to V2. Right-DG fractional anisotropy (FA) decreased (χ²=9.38, P _FDR =0.01) and left-DG orientation dispersion index (ODI) increased (χ²=10.65, P _FDR =0.003). Lower baseline left-DG FA correlated with greater improvement at V2 (r=–0.57, p=0.05), and FA reduction from V1 to V2 correlated with improvement (r=0.74, p=0.01). No significant correlations were observed at V3. Conclusion: ESK was associated with early DG microstructural changes - reduced FA and increased ODI - consistent with greater dendritic complexity. Baseline DG diffusion metrics predicted early improvement, supporting their potential as in-vivo markers of plasticity and candidate biomarkers for TRD treatment. Larger studies are needed.