Lower pre-treatment TMS-evoked cortical reactivity and alpha-band oscillatory dynamics predict efficacy of primary motor cortex neuromodulation for chronic pain

Repetitive transcranial magnetic stimulation (rTMS) targeting the primary motor cortex (M1) provides significant pain relief in approximately 45% of patients with chronic pain. Identifying markers that predict rTMS treatment responders to M1 before initiating treatment is crucial for informing decision-making and improving patient outcomes in clinical practice. In this secondary analysis of a clinical trial assessing the effects of rTMS on different clinical targets (NCT06395649), we employed a combination of transcranial magnetic stimulation and electroencephalography (TMS-EEG) to investigate cortical reactivity and oscillatory dynamics in 43 patients with chronic pain prior to 12 sessions of therapeutic 10 Hz rTMS applied to M1 over an 8-week period. Responders were defined as those reporting a ≥30% reduction in pain intensity on a visual analogue scale at the end of treatment. TMS-evoked cortical reactivity was assessed using global mean field power (GMFP), calculated across all electrodes, and local mean field power (LMFP), computed in 4 electrodes at the TMS stimulation site. TMS-evoked oscillatory dynamics were evaluated with event-related spectral perturbation (ERSP) and intertrial coherence (ITC) across three frequency bands: alpha (8–12 Hz), beta-1 (13–20 Hz), and beta-2 (21–30 Hz) at the stimulation site. 20 patients (47%) were responders at the end of the rTMS treatment. Compared to non-responders, responders exhibited lower GMFP and lower LMFP at the stimulation region (both P<0.05). They also showed lower alpha-band ERSP and ITC at the stimulation region (both P<0.05). These lower neurophysiological features were associated with greater reductions in pain intensity (all P<0.05). Exploratory supervised machine learning using three baseline TMS-EEG features (GMFP, alpha-band ERSP, and ITC) predicted responder status with acceptable accuracy (ROC-AUC = 0.70, PR-AUC = 0.76). Together, these results suggest that lower pre-treatment TMS-evoked cortical reactivity and alpha-band oscillatory dynamics activity predict better clinical response to rTMS in patients with chronic pain (≈ 3 in 4 Responders correctly identified). Future prospectively designed clinical studies should implement TMS-EEG assessment of the target to be used for rTMS before treatment is started, in order to prospectively test the utility of these personalized neuromodulatory interventions.