Excitation-Inhibition Balance and Fronto-Limbic Connectivity Distinguish TMS Treatment Outcomes in Refractory Depression.

Abstract

Background: Intermittent theta burst stimulation (iTBS) targeting the left dorsolateral prefrontal cortex (DLPFC) is effective for treatment-resistant depression, but outcomes remain variable. The neurophysiological mechanisms underlying response remain poorly understood, limiting precision neuromodulation.

Methods: We combined transcranial magnetic stimulation with electroencephalography (TMS-EEG) and connectome-based Jansen-Rit neural mass modeling to investigate the mechanistic basis of treatment efficacy in 90 patients with treatment-resistant depression. Individualized whole-brain models were fitted to pre- and post-iTBS TMS-evoked potentials, yielding subject-specific estimates of synaptic connectivity, excitation-inhibition (E/I) balance, and cortical dynamics. Brain states were defined by singular value decomposition of the model's population activity, and partial least squares analysis with nested cross-validation identified the brain-state pattern distinguishing responders from non-responders.

Results: We identified two neurophysiological signatures associated with treatment response. First, responders showed greater post-treatment reductions in 3-10 Hz oscillatory power, driven by shifts in E/I balance toward greater inhibitory control through reduced excitatory drive and increased inhibitory-to-pyramidal connectivity. Second, a pre-treatment brain state characterized by opposite-polarity dynamics between subgenual anterior cingulate cortex (sgACC) and left DLPFC distinguished responders from non-responders, consistent with prior fMRI findings on frontolimbic anticorrelation. Neural trajectory analysis further revealed greater dynamical reorganization and increased attractor stability in responders following treatment.

Conclusions: These findings provide a neurophysiologically-informed framework linking iTBS treatment efficacy to specific changes in cortical E/I balance and frontolimbic circuit dynamics, pointing toward candidate markers that, with subject-specific structural connectivity and independent validation, may inform personalized neuromodulation approaches in treatment-resistant depression.