Longitudinal resting-state EEG in amyloid-positive patients along the Alzheimer’s disease continuum: considerations for clinical trials

Abstract

Background To enable successful inclusion of electroencephalography (EEG) outcome measures in Alzheimer’s disease (AD) clinical trials, we retrospectively mapped the progression of resting-state EEG measures over time in amyloid-positive patients with mild cognitive impairment (MCI) or dementia due to AD. Methods Resting-state 21-channel EEG was recorded in 148 amyloid-positive AD patients (MCI, n = 88; dementia due to AD, n = 60). Two or more EEG recordings were available for all subjects. We computed whole-brain and regional relative power (i.e., theta (4-8 Hz), alpha1 (8-10 Hz), alpha2 (10-13 Hz), beta (13-30 Hz)), peak frequency, signal variability (i.e., theta permutation entropy), and functional connectivity values (i.e., alpha and beta corrected amplitude envelope correlation, theta phase lag index, weighted symbolic mutual information, inverted joint permutation entropy). Whole-group linear mixed effects models were used to model the development of EEG measures over time. Group-wise analysis was performed to investigate potential differences in change trajectories between the MCI and dementia subgroups. Finally, we estimated the minimum sample size required to detect different treatment effects (i.e., 50% less deterioration, stabilization, or 50% improvement) on the development of EEG measures over time, in hypothetical clinical trials of 1- or 2-year duration. Results Whole-group analysis revealed significant regional and global oscillatory slowing over time (i.e., increased relative theta power, decreased beta power), with strongest effects for temporal and parieto-occipital regions. Disease severity at baseline influenced the EEG measures’ rates of change, with fastest deterioration reported in MCI patients. Only AD dementia patients displayed a significant decrease of the parieto-occipital peak frequency and theta signal variability over time. We estimate that 2-year trials, focusing on amyloid-positive MCI patients, require 36 subjects per arm (2 arms, 1:1 randomization, 80% power) to detect a stabilizing treatment effect on temporal relative theta power. Conclusions Resting-state EEG measures could facilitate early detection of treatment effects on neuronal function in AD patients. Their sensitivity depends on the region-of-interest and disease severity of the study population. Conventional spectral measures, particularly recorded from temporal regions, present sensitive AD treatment monitoring markers.

Reliability of Spectral Features of Resting-State Brain Activity: A Magnetoencephalography Study

“…MCI showed a smaller value than CH at 0.01 [4] (below MDCs). MCI due to AD exhibited a smaller value than CH at 0.02 [21], exceeding MDC80.…”

“…DAD showed smaller values than CH at frequencies of 0.69 [22], 1.00 [23], 1.32 [24], and 1.43 Hz [16], all surpassing MDC95. Vascular dementia (VaD), Parkinson's disease dementia (PDD), and dementia with Lewy bodies (DLB) displayed smaller values than CH at 0.80 [21], 1.70 [23], and 1.80 Hz [23], respectively, each exceeding MDC95. MCI showed smaller values than CH at 0.05 [4] (below MDCs) and 0.63 Hz [25] (above MDC95).…”

“…Only one prior longitudinal study was identified, reporting that IAF (the parieto-occipital median frequency at 4-13 Hz) exhibited a decrease of 0.55 Hz over a 24-month follow-up period in the dementia due to Alzheimer's disease (DAD) group [21], a reduction that surpassed the MDC95.…”

“…Preferably, the estimated withinparticipant variabilities of the spectral parameters (e.g., MDC95) should be evaluated against withinparticipant changes (i.e., longitudinal changes) in different cognitive statuses (e.g., healthy and with dementia). However, we evaluated most parameters in light of group-level differences reported in previous studies because longitudinal studies were very sparse on this research topic, and to our knowledge, there is only one for IAF [21]. This may have biased our interpretation of the results.…”

EEG in Down Syndrome—A Review and Insights into Potential Neural Mechanisms

“…Some studies have shown increased beta activity in people with DS, and some studies have had the opposite results. Several factors may have contributed to this, such as beta-amyloid load (amyloidpositive patients with AD show a decrease in [37] or an increase in beta power [38]). It is not known how beta power relates to amyloid load in the context of DS.…”