Decrease in sleep depth is associated with higher cerebrospinal fluid neurofilament light levels in patients with Alzheimer's disease

STUDY OBJECTIVES: The majority of studies investigating the association between sleep and Alzheimer's disease (AD) biomarkers have been performed in healthy participants. Our objective was to investigate the association between sleep and several biomarkers that reflect distinct aspects of AD physiopathology. METHODS: The cohort included 104 individuals with mild-moderate AD. The participants were submitted to one-night polysomnography, and cerebrospinal fluid was collected in the following morning to measure the selected biomarkers associated with amyloid deposition, tau pathology, neurodegeneration, axonal damage, synaptic integrity, neuroinflammation, and oxidative damage. RESULTS: There was a positive correlation between neurofilament light (NF-L) and the time spent in stage 1 of non-rapid eyes movement (NREM) (N1) sleep and a negative correlation between this marker and the time spent in stage 3 of NREM (N3) sleep. Accordingly, we observed that deep sleep was associated with lower levels of NF-L, whereas light sleep increased the probability of having higher levels of this marker. Furthermore, chitinase-3-like-1 (YKL-40) was negatively correlated with sleep efficiency, the time spent in stage 2 of NREM (N2) sleep, and the time spent in N3 sleep. Conversely, there was a positive correlation between N3 sleep and the oxidative protein damage markers N-ε-(carboxyethyl)lysine and N-ε-(malondialdehyde)lysine. CONCLUSIONS: There were significant correlations between sleep parameters and AD biomarkers related to axonal damage and neuroinflammation, such as NF-L and YKL-40. A lack of deep sleep was associated with higher levels of NF-L. This highlights a potential role for NF-L as a biomarker of sleep disruption in patients with mild-moderate AD in addition to its role in predicting neurodegeneration and cognitive decline.This study was supported by the Generalitat of Catalonia, Department of Health (PERIS 2019 SLT008/18/00050) and “Fundació La Marató TV3” (464/C/2014) to G.P.R.; by the Spanish Ministry of Economy and Competitiveness, Institute of Health Carlos III (grant number P114/00328), the Spanish Ministry of Science, Innovation and University (RTI 2018–099) of Catalonia, and Agency for Management of University and Research grants (2017 SGR696) to R.P. This study has been co-financed by FEDER funds from the European Union (“A way to build Europe”). IRBLleida is a CERCA Programme/Generalitat of Catalonia. F.D. was supported by the Agency for Management of University and Research grants (FI_B100153)