A Pilot Study of Stereoelectroencephalography Electrodes in a Patient with Refractory Chronic Migraine: Personalized Targets for Precise Deep Brain Stimulation

Background and ObjectivesThe complexity of neural circuits and the heterogeneity of brain networks are barriers for further improving the efficacy of DBS. This study aimed to establish a clinical paradigm to personalize the design of DBS in patients with refractory headache, which would constitute a milestone in this field. MethodsWe implanted 14 stereoelectroencephalography electrodes in a patient with refractory migraine for clinical monitoring and electrophysiological recording. During monitoring, we collected the VAS score in 5-min increments, and recorded electrophysiological data in real-time. Data were classified into two types of symptoms (high and low symptoms) for determining the spectral power features of specific brain regions reflecting pain fluctuations, which we called Biomarker, using statistical analyses and cross-validated machine-learning models. During stimulation, we tested the clinical effect through a systematic bipolar stimulation survey and blinded sham-controlled stimulation studies, and collected real-time electrophysiological data. Based on the identification of brain areas with clinical improvement, the optimal target for stimulate was determined by validating the clinical response against the biomarker, and phase-amplitude coupling finally. ResultsFor biomarker, RNAc-HFO was the most considerably correlated with VAS score (rho = 0.5292, p < 0.0001), and differed significantly between mild and severe pain levels (p = 0.0003), also with the greatest weighting in the characteristic ranking. The machine-learning model showed an accuracy and AUC remaining at 75.51% and 0.77, respectively, for RAC-HFO. For target, LdACC was identified as the most effective stimulation target, based on the VAS score reported over the stimulation period. VAS score (p = 0.006), RNAc-HFO (p = 0.0029) were significantly improved after stimulation compared to pre-stimulation in LdACC. The significant modulatory effect of RNAc-HFO by the low-frequency phase of LdACC also confirmed the modulatory effect of LdACC and RNAc during headache fluctuation. DiscussionAs a pilot study for exploring precise and personalized DBS in refractory migraine, we identified the biomarker and optimal target via the integration of clinical and electrophysiological data. The concept of the proposed data-driven approach to optimizing personalized treatment strategies for DBS may create a new frontier in the field of refractory headache and pain.