Public Health.

Background: Poor sleep is associated with cognitive decline, and ∼45% of Alzheimer's disease (AD) patients experience sleep disturbances. Emerging evidence suggests that reduced non-rapid eye movement (NREM) slow wave sleep (SWS) is linked to amyloid accumulation and APOE ε4-related genetic vulnerability to AD. Here, we investigate the effects of amyloid and APOE status on SWS, and their interaction, in a cohort of cognitively unimpaired (CU) individuals at higher risk of AD.

Method: We analyzed data from 80 sleep medication free participants (65.26 years old on average, 55% females, Table 1) from the AlfaSleep project, which involves characterization with polysomnography, APOE genotyping, CSF amyloid-β (Aβ) 40 and Aβ42 (Roche NeuroToolKit), and [F]flutemetamol-PET, among other measures. Participants were classified as amyloid positive (A+) using either CSF Aβ 42/40 ratio (<0.071, N = 75), or amyloid PET (>12 Centiloids, N = 5) thresholds for early amyloid deposition. We computed average power spectral densities (PSD) from NREM sleep 30-s epochs in frontal electroencephalogram channels, and power of slow oscillations (SO), delta and slow wave activity (SWA) as the area under the PSD curve between 0.5-1Hz, 1-4Hz and 0.5-4Hz, respectively. We log-transformed SWS metrics due to non-normal distribution. We used linear regression models to test the main effect of amyloid or APOE status on SWS metrics in separated models, and next we entered both in the same model. Finally, we the tested interaction between APOE and amyloid status on SWS metrics. Models were adjusted by age, sex, and apnea-hypopnea index (AHI).

Result: Both A+ status and APOE-ε4 carriership were associated with significantly lower power in SO and SWA in separated models (Figure 1A). When combining both in the same model, only the effect of APOE on SO power remained significant, as well as a trend for lower SWA (Figure 1B). Interaction analyses yielded no significance.

Conclusion: Early amyloid accumulation and APOE-ε4 carriership are associated with SWS disruption in CU adults, with APOE-ε4 carriers probably being particularly vulnerable to SWS disruption. Longitudinal studies spanning younger age individuals are needed to characterize the contribution of APOE-ε4 to SWS disruption and to test whether preventive interventions targeting SWS could help to reduce AD dementia risk.