Associations of the Neurological Institute of Athens Sleep-mediated Neuropsychological Battery to Biomarkers of Neurodegeneration
Elissaios Karageorgiou1, Nikolaos Andronas1, Konstantina Sykara1, Athanasia Alexoudi1, Zoi Devenopoulou1, Margarita Maimani1, Lei Zhang2, Luiz Eduardo Mateus-Brandao2, Jonathan Cedernaes2, Klimentini Karageorgiou1
1Neurological Institute of Athens, 2University of Uppsala
Objective:
We examined whether the Neurological Institute of Athens (NIA) Sleep-Mediated Neuropsychological Battery (SM-NP) can predict neurodegenerative brain burden.
Background:
Sleep-mediated cognitive processes, especially episodic memory and executive function, are associated with brain neurodegeneration. Leveraging fast and easy sleep-mediated cognitive tests in clinic may lead to improved diagnosis, prognosis, and treatment response assessments.
Design/Methods:
We recruited 45 older patients with either cognitive or sleep disorders and non-impaired controls (Age 66.5 ± 9.9; F:M 24:21) through the Sleep & Memory Center at the NIA as part of the MES-CoBraD Horizon2020 project. Participants underwent structured clinical examinations, within-day neuropsychological testing, actigraphy, polysomnography, pre- and post-sleep plasma analyses (p-tau217, total Tau, Aβ40, Aβ42, GFAP, NFL, insulin, cortisol), and the 12-minute-long SM-NP containing a 9-item learning task and a phonemic fluency task. We explored associations between pre- and post-sleep plasma biomarkers and SM-NP metrics through parametric correlations and partial correlations, and applied 5-fold linear discriminant analysis using SM-NP metrics to classify people into high vs. low likelihood of Alzheimer’s disease (AD) pathology using p-tau217 as the gold standard (<0.4 or >0.63 pg/mL). Multiple comparisons were addressed with Holm–Bonferroni correction.
Results:
Learning, recall, and recognition metrics were associated with pre- and post-sleep p-tau217 and GFAP (r = -0.53 - -0.69; p<0.0001), whereas only pre-sleep phonemic fluency was associated with pre-sleep p-tau217 (r=-0.54; p=0.0001), suggesting that executive difficulties worsen at night in neurodegeneration. The SM-NP classification accuracy was 80.5%, with episodic memory metrics having the highest loadings (-0.66 – -0.79), age somewhat lower (0.55), and sex and phonemic fluency the lowest (-0.13 – 0.09). Pre- to post-sleep p-tau217-based AD grouping changed in 14.3%, with morning values being larger.
Conclusions:
The SM-NP is easy-to-administer and fast and is both correlated to and accurate in predicting neurodegeneration-associated plasma biomarkers. Furthermore, timing of blood collection modifies the sensitivity and specificity of plasma p-tau217.
Disclaimer: Abstracts were not reviewed by Neurology® and do not reflect the views of Neurology® editors or staff.