Neurodevelopmental cortical morphometric alterations in medication-naïve adolescents with first-episode early-onset depression: a surface-based morphometry study

To investigate cortical and subcortical morphometric differences in adolescents with first-episode, medication-naïve early-onset depression (EOD). Structural MRI data from 161 adolescents with EOD and 161 healthy controls (HCs) were processed using FreeSurfer. Vertex-wise maps were smoothed with a 10-mm FWHM kernel and tested after adjustment for age, sex, IQ, and eTIV using 1000-permutation cluster-wise correction. Parcel-level group differences were assessed with general linear models. Benjamini–Hochberg false discovery rate (FDR) correction was applied within each morphometric domain. Exploratory partial correlations with HAMD-17 scores controlled for the same covariates, with FDR correction across 168 result-selected endpoints. Vertex-wise analyses showed widespread cortical thickening, broadly reduced cortical surface area, and mixed cortical-volume changes. Parcel-level analyses showed predominantly increased cortical thickness, reduced cortical surface area, and mixed cortical-volume differences. Subcortical differences included lower bilateral pallidal and right hippocampal volumes and higher bilateral accumbens, right thalamic, and right amygdala volumes. Within the EOD group, left entorhinal thickness was inversely associated with HAMD-17 scores under the exploratory HAMD-17 correction (partial r=−0.298, q=0.0255). This association did not survive the sensitivity correction combining both symptom scales. The sample showed group-level cortical and subcortical morphometric differences. The cross-sectional design does not establish altered developmental trajectories, proximity to illness onset, or individual diagnostic or treatment-selection utility.