Vagus Nerve Stimulation differentially modulates P3b in responders and non-responders: toward a biomarker of therapeutic efficacy

IntroductionAlthough vagus nerve stimulation (VNS) is an established therapy for drug-resistant epilepsy, its mechanisms of action remain unresolved, resulting in variable clinical efficacy. Given the strong anatomical and functional coupling between vagal afferents and the locus coeruleus, noradrenergic system, this study investigated whether VNS directly impacts an electrophysiological marker of this system, the P3b event-related potential, and how such modulation relates to therapeutic outcomes.MethodsFifteen adults who had undergone long-term VNS implantation performed an auditory oddball task with the device disabled (OFF) and enabled (ON), with ON separated into electrical pulse-train (ON HIGH) and inter-burst break (ON LOW) phases to investigate the direct impact of electrical stimulation on the P3b.ResultsLinear mixed-effects modelling revealed a significant interaction between VNS condition and clinical response: responders showed reduced P3b amplitude (p = 0.023) and prolonged latency during ON (p = 0.007), whereas non-responders exhibited increased amplitude (p = 0.009) and trending shortened latency (p = 0.078). These VNS-induced changes correlated monotonically with a continuous clinical response score (r_amplitude = −0.62, r_latency = 0.48). In addition, a simple classification approach based on a composite amplitude-to-latency index was included to illustrate the potential of P3b modulation as a biomarker for distinguishing responders from non-responders, showing an overall accuracy of 86.7%. No pulse-locked modulation was observed between ON HIGH and ON LOW.DiscussionThese findings demonstrate that VNS elicits group-specific acute effects on cognitive, electrophysiological markers and support P3b modulation as a promising biomarker for predicting therapeutic efficacy.