Two neuroanatomical subtypes of major depressive disorder exhibit dissociable clinical profiles and distinct molecular signatures.

BACKGROUND

Major depressive disorder (MDD) is characterized by marked neurobiological and clinical heterogeneity; however, biologically informed neuroanatomical subtypes and their molecular correlates remain insufficiently defined.

METHODS

We applied the Subtype and Stage Inference (SuStaIn) model to large-scale multi-center structural MRI data from the REST-meta-MDD consortium, including 1276 patients with MDD and 1104 healthy controls, and externally validated the findings in an independent SRPBS cohort (254 patients and 701 controls). Imaging transcriptomic, bioinformatic, cell-type enrichment, and clinical analyses were performed to characterize subtype-specific features.

RESULTS

Two neuroanatomical subtypes were identified, characterized by distinct SuStaIn-inferred patterns of gray matter variation: one showing a cortical-to-cerebellar pattern and the other a cerebello-subcortical-to-cortical pattern. The subtypes showed differential clinical characteristics and brain-symptom associations. The cortical-to-cerebellar subtype was associated with greater guilt and hypochondriasis, with cerebellar structure related to affective and anxiety symptoms, whereas the second subtype showed an association between inferior frontal gyrus volume and insomnia-related symptoms. Cross-modal analyses identified distinct transcriptomic correlates associated with the two subtypes. The first subtype showed spatial enrichment of synaptic and mitochondrial pathways preferentially expressed in cortical projection neurons, whereas the second subtype showed spatial enrichment of pathways related to chromatin organization, epigenetic regulation, and calcium-MAPK signaling in cerebellar granule cells.

CONCLUSION

These findings identify two neuroanatomical subtypes of MDD with distinct clinical and molecular profiles, providing evidence for biologically informed stratification and offering a multiscale framework for understanding disease heterogeneity.