What do elevated cerebrospinal fluid p‐tau levels signify in the absence of beta‐amyloid pathology?

• Published in: Alzheimer's & dementia Vol. 19; no. S15
• Main Authors: Wisse, Laura, Binette, Alexa Pichet, Strandberg, Olof, Palmqvist, Sebastian, Stomrud, Erik, Janelidze, Shorena, Mattsson‐Carlgren, Niklas, Blennow, Kaj, Hansson, Oskar
• Format: Journal Article
• Language: English
• Published: 01.12.2023
• ISSN: 1552-5260; 1552-5279
• DOI: 10.1002/alz.078132

Background The introduction of the A/T/(N) (β‐amyloid/tau/neurodegeneration) framework has revealed an A‐T+ (β‐amyloid‐negative/tau‐positive) group which has been named suspected non‐Alzheimer’s pathophysiology. Cerebrospinal fluid (CSF) markers are often utilized to divide individuals in the A and T categories, but it is unclear what elevated CSF p‐tau levels signify in the absence of β‐amyloid. We aim to better understand A‐T+ as defined by CSF markers in two independent samples of cognitively unimpaired and impaired individuals by assessing baseline and longitudinal change in CSF, magnetic resonance (MRI), positron emission tomography (PET) and cognition measures. Method Cognitively normal (CN) and Mild Cognitive Impairment subjects from BioFINDER‐1 and 2 were included. A status was based on two CSF measures of β‐amyloid 42/40 (Aβ42/40) where A‐ individuals had to be negative on both measures using previously established cut offs. T status was determined by the 75th, 80th and 85th percentile of A‐ CN subjects for Roche Elecsys p‐tau181 (Table 1). Analyses of covariance and linear mixed models were performed to assess baseline and change differences in CSF (p‐tau181, Aβ42/40 ratio, neurofilament light (NFL)), MRI (hippocampal volume, entorhinal cortex and temporal meta‐region of interest thickness), cognition (modified Preclinical Alzheimer’s Cognitive Composite and delayed recall), and PET measures (entorhinal cortex Tau‐PET and early and global Alzheimer’s Disease composite β‐amyloid‐PET uptake) over 8 years (10 for cognition) for BioFINDER‐1 and 4 years for BioFINDER‐2. Result In A‐T+ versus A‐T‐, higher baseline NFL levels were found. No other baseline differences were found except for p‐tau181 and Aβ42/40 ratio by design. A‐T+ did not show significant change compared to A‐T‐ on any measure. Using different cut offs for p‐tau181 did not notably change the results. Comparisons with the A+ groups are shown in Figures 1‐2. Conclusion Our results indicate that A‐T+ as determined by p‐tau181 does not seem to be clinically meaningful as this group shows no decline on any of the biomarkers or cognitive measures over several years. The high NFL levels in the A‐T+ group suggest the presence of a latent factor that affects both NFL and p‐tau levels. Analyses will be repeated with other p‐tau isotopes and assays.