Transcranial magnetic stimulation (TMS) is a promising treatment for Major Depressive Disorder (MDD), but its effectiveness varies significantly across individuals, necessitating biomarkers for pretreatment stratification. We developed an EEG-based digital biomarker, the dorsolateral prefrontal cortex (dlPFC) sensitivity index (dSI), to predict response to repetitive TMS. In Study 1, dSI was computed from EEG functional connectivity features using an L1-regularized logistic regression model. In Study 2, patients were prospectively stratified into sensitive, non-sensitive, and sham groups, all undergoing a personalized, neuronavigated bilateral dlPFC rTMS protocol for two weeks. Baseline EEG-derived dlPFC sensitivity index (dSI) significantly predicted clinical and cognitive outcomes following rTMS. Patients prospectively identified as ‘sensitive’ based on their pretreatment dSI exhibited greater improvements in emotional attention network task reaction times and more pronounced reductions in hostility and paranoia compared to ‘non-sensitive’ and sham groups. Post-TMS EEG analysis further confirmed network-specific modulation in the sensitive group. These findings support the predictive validity of this EEG-based task biomarker for personalizing TMS treatment in MDD.
Wan et al. (Thu,) studied this question.