P3-110: preclinical alzheimer's disease: which eeg biomarkers for precision medicine?

Resting state eyes-closed electroencephalographic (rsEEG) rhythms at delta (< 4 Hz) and alpha (8-12 Hz) bands characterized groups of Alzheimer's disease (AD) patients with dementia and mild cognitive impairment (Babiloni et al., 2017, PMID: 2845484; Babiloni et al., 2018, PMID: 29439335). In the INSIGHT-preAD study, we evaluated whether these rhythms may be abnormal even in individuals with subjective memory complaints (SMC). Scalp rsEEG rhythms were evaluated in 163 SMC subjects with raw mini mental stage evaluation score ≥ 28 and artifact-free rsEEG data after detrend fluctuation, independent component analysis, and “artifact-free” rsEEG power density spectra as judged by two experts. After this pruning procedure, 55 SMC subjects negative to the amyloid PET diagnostic markers (SMC neg) and 51 SMC subjects positive to those markers (SMC pos as preclinical AD) were selected. Individual alpha frequency peak was used to determine the delta, theta, alpha1, alpha2, and alpha3 frequency bands. Fixed beta1 (13-20 Hz), beta2 (20-30 Hz), and gamma (30-40 Hz) bands were also considered. An ANOVA design (p < 0.05) used rsEEG power density as dependent variable the factors Group (SMC neg, SMC pos), Electrode (Fz, Pz, O1, O2), and Frequency Band (delta, theta, alpha1, alpha2, alpha3, beta1, beta2, gamma). The ANOVA showed a statistically significant 3-way interaction (p < 0.01; Figure 1). Duncan post-doc indicated that compared with the SMC neg, the SMC pos exhibited lower alpha2 (about 8–10 Hz) power density at the occipital electrodes (p < 0.05), while the delta band showed no significant differences between the groups (p > 0.05).