Can we predict drug response by functional connectivity in patients with juvenile myoclonic epilepsy?

Objectives: We investigated functional connectivity based on EEG using graph theoretical analysis in patients with newly diagnosed juvenile myoclonic epilepsy (JME), and whether it could play a role as a biomarker predicting antiepileptic drug (AED) response. Methods: We consecutively enrolled 38 patients with JME and 40 normal controls. The initial EEG was undertaken at the time of diagnosis of JME in a drug-naive state. The second EEG was done after at least 12 months from the time of the initial EEG. We classified the patients with JME into two groups according to AED response at the time of taking the second EEG. We investigated functional connectivity based on graph theoretical analysis using connectivity measures of the coherence and phase locking value. Results: In the analysis of functional connectivity using coherence as a connectivity measure, the global efficiency and local efficiency in the AED poor responders (N = 4) decreased, whereas the small-worldness index increased. In the analysis of functional connectivity using phase locking value as a connectivity measure, the global efficiency and local efficiency in the AED poor responders decreased. However, in the AED good responders (N = 34), none of the network measures were different from those in healthy controls. Conclusions: We newly found that there were significant differences of functional connectivity based on initial EEG according to AED response in the patients with JME. This suggests that brain connectivity could play a role as a new biomarker predicting AED response in patients with JME.