Brain-Behavior Depression Subtypes Distinguish Response to rTMS and Pharmacotherapy

Major depressive disorder (MDD) is a complex and heterogeneous condition. While various methods for subtyping MDD have been proposed, there is no consensus on whether these subtypes can guide treatment selection. In my talk, I will present a robust, data-driven model developed from a large, single-site dataset (n=328) that combines clinical assessment data with resting-state fMRI to identify MDD subtypes. Our analysis revealed four distinct MDD subtypes, each characterized by unique symptom profiles, atypical functional connectivity, and responses to transcranial magnetic stimulation (TMS). We further validated these subtypes by applying them to an independent dataset of individuals treated with escitalopram (n=130) and found that subtype membership predicted differential responses to either TMS or pharmacotherapy. These findings suggest that subtyping approaches to classifying neurobiological and symptom variability in MDD are stable, generalizable, and capable of predicting treatment outcomes across different interventions.