519. Baseline EEG connectivity predicts differential clinical response to donepezil and memantine in Alzheimer’s disease
Abstract
Abstract Background Alzheimer’s disease (AD) is a progressive neurodegenerative disorder characterized by cognitive decline and widespread disruption of large-scale brain networks. Pharmacological treatments for moderate to severe AD primarily include high-dose acetylcholinesterase (AChE) inhibitors and non-competitive antagonists of the N-methyl-D-aspartate (NMDA) receptor. Despite their widespread clinical use, individual responses to these medications vary substantially, making it difficult to predict long-term treatment outcomes at the time of medication initiation. Currently, there are no reliable neurophysiological biomarkers to guide medication selection in this population. Resting-state electroencephalography (EEG), which reflects functional brain connectivity and neurotransmitter-related neural oscillatory activity, has emerged as a promising non-invasive tool for identifying biomarkers of disease progression and treatment response in AD. Aims & Objectives This study aimed to investigate whether baseline EEG-based brain connectivity measures could serve as potential biomarkers for predicting long-term treatment response in patients with Alzheimer’s disease. Specifically, we examined differences in EEG connectivity patterns between patients who responded well to donepezil monotherapy and those who responded well to memantine in combination with high-dose donepezil. Method Resting-state EEG was recorded from a total of 102 patients with Alzheimer’s disease (mean age 76.28 ± 7.04 years; 35 males [73.75 ± 8.61], 67 females [77.34 ± 6.04]) at treatment initiation and after longitudinal follow-up. The mean treatment duration was 202.98 ± 123.37 weeks. Based on longitudinal changes in Mini-Mental State Examination (MMSE) and Clinical Dementia Rating–Sum of Boxes (CDR-SB) scores, patients were classified into two groups: good responders to donepezil monotherapy (n = 65) and good responders to memantine combined with high-dose donepezil (n = 47). Functional brain connectivity among 246 cortical regions defined by the Brainnetome atlas was analyzed using spectral coherence across five frequency bands (delta, theta, alpha, beta, and gamma) using the FieldTrip toolbox. Group differences in connectivity were assessed using analysis of covariance (ANCOVA), controlling for age, sex, and baseline MMSE scores. Results Compared with the donepezil monotherapy responder group, patients treated with memantine in combination with high-dose donepezil exhibited significantly reduced and disrupted regional connectivity in the higher-frequency bands, particularly the theta and gamma bands, involving left fronto–occipito–parietal networks (p < 0.001). In the alpha frequency band, significant reductions in regional connectivity were also observed between the left frontal and medial temporal regions (p < 0.001). Discussion & Conclusions Patients in the memantine combined with high-dose donepezil group demonstrated increased baseline gamma-band EEG connectivity between right anterior and posterior cortical regions. Gamma-band activity is known to increase during selective attention and is closely associated with cholinergic function in healthy individuals. These findings suggest that altered baseline gamma-band EEG connectivity may reflect underlying differences in cholinergic network integrity and could serve as a potential biomarker for predicting long-term pharmacological treatment response in Alzheimer’s disease.