Transcriptomic profiles link heterogeneous brain structural deviation to neurological subtypes in Wilson's disease.

Disease progression or physical decline varies considerably across patients with Wilson's disease (WD), limiting the understanding of pathological and biological processes underlying heterogeneity in the brain and clinical phenotypes of WD. Herein, using large-scale data (N = 1749), we developed a gray matter volume (GMV)-based 'brain age model' to calculate the deviation between predicted brain age and chronological age, known as the brain-predicted age difference (brainPAD) and brain structure deviation (BSD), across one hepatic subtype (HWD, N = 32) and two neurological subtypes, namely, parkinsonism (PWD, N = 78) and dystonia (DWD, N = 46). BrainPAD is associated with clinical severity and is further related to GMV, revealing heterogeneous patterns of BSD across the two neurological subtypes. On the basis of support vector machine (SVM) analysis, the GMV-based BSD pattern robustly discriminates PWD from DWD. The SVM-derived fingerprint is significantly related to brainPAD in both DWD and PWD patients, suggesting that heterogeneous BSD has a stronger effect on neurological WD. In addition, differences in cortical transcriptomic signatures between the two neurological subtypes suggest that heterogeneous cortical BSD patterns may be preferentially related to subtype-specific pathology, involving biological processes related to synaptic plasticity, dendritic development, and neuronal projection, as well as cell type-specific signatures associated with excitatory and inhibitory neurons, oligodendrocytes, and microglia. In contrast, BSD in subcortical regions exhibited largely shared transcriptomic signatures between PWD and DWD, suggesting that subcortical biological processes may predominantly reflect disease-level pathology in neurological WD. Moreover, BSD-related genes in both cortical and subcortical regions are enriched for genetic risks associated with multiple neurological disorders, suggesting a potential complex pathology underlying heterogeneous clinical manifestations in patients with neurological WD. Together, these findings highlight convergent subcortical and divergent cortical transcriptomic signatures underlying subtype-specific BSD patterns, potentially providing novel biological insights into heterogeneous neurodegenerative mechanisms across neurological WD subtypes.