Age-adjusted CSF t-tau and NfL do not improve diagnostic accuracy for prodromal Alzheimer’s Disease
Knudtzon, Stephanie; Nordengen, Kaja; Grøntvedt, Gøril Rolfseng; Jarholm, Jonas Alexander; Eliassen, Ingvild Vøllo; Selnes, Per; Pålhaugen, Lene; Espenes, Johan Jacob; Gisladottir, Berglind; Waterloo, Knut K; Fladby, Tormod; Kirsebom, Bjørn-Eivind Seljelid
Journal article, Peer reviewed
Published version
Date
2024Metadata
Show full item recordCollections
Original version
Neurobiology of Aging Volume 141 , September 2024, Pages 74-84 10.1016/j.neurobiolaging.2024.05.016Abstract
Cerebrospinal fluid total-tau (t-tau) and neurofilament light chain (NfL) are biomarkers of neurodegeneration and are increased in Alzheimer’s disease (AD). In order to adjust for age-related increases in t-tau and NfL, cross-sectional age-adjusted norms were developed based on amyloid negative cognitively normal (CN) adults aged 41–78 years (CN, n = 137). The age-adjusted norms for t-tau and NfL did not improve receiver operating curve based diagnostic accuracies in individuals with mild cognitive impairment (MCI) due to AD (AD-MCI, n = 144). Furthermore, while NfL was correlated with higher age in AD-MCI, no significant correlation was found for t-tau. The cox proportional hazard models, applied in 429 participants with baseline t-tau and NfL, showed higher hazard ratio of progression to MCI or dementia without age-adjustments (HR = 3.39 for t-tau and HR = 3.17 for NfL), as compared to using our norms (HR = 2.29 for t-tau and HR = 1.89 for NfL). Our results indicate that utilizing normative reference data could obscure significant age-related increases in these markers associated with neurodegeneration and AD leading to a potential loss of overall diagnostic accuracy.