Plasma p-tau181/Aβ1-42 ratio predicts Aβ-PET status and correlates with CSF-p-tau181/Aβ1-42 and future cognitive decline

Abstract

Background: In Alzheimer's disease (AD), plasma amyloid beta (Aβ)1-42 and phosphorylated tau (p-tau) predict high amyloid status from Aβ positron emission tomography (PET); however, the extent to which combination of these plasma assays can predict remains unknown.

Methods: Prototype Simoa assays were used to measure plasma samples from participants who were either cognitively normal (CN) or had mild cognitive impairment (MCI)/AD in the Australian Imaging, Biomarkers and Lifestyle (AIBL) study.

Results: The p-tau181/Aβ1-42 ratio showed the best prediction of Aβ-PET across all participants (area under the curve [AUC] = 0.905, 95% confidence interval [CI]: 0.86-0.95) and in CN (AUC = 0.873; 0.80-0.94), and symptomatic (AUC = 0.908; 0.82-1.00) adults. Plasma p-tau181/Aβ1-42 ratio correlated with cerebrospinal fluid (CSF) p-tau181 (Elecsys, Spearman's ρ = 0.74, P < 0.0001) and predicted abnormal CSF Aβ (AUC = 0.816; 0.74-0.89). The p-tau181/Aβ1-42 ratio also predicted future rates of cognitive decline assessed by AIBL Preclinical Alzheimer Cognitive Composite or Clinical Dementia Rating Sum of Boxes (P < 0.0001).

Discussion: Plasma p-tau181/Aβ1-42 ratio predicted both Aβ-PET status and cognitive decline, demonstrating potential as both a diagnostic aid and as a screening and prognostic assay for preclinical AD trials.

Keywords: Alzheimer's disease; amyloid beta amyloid imaging; blood biomarkers; blood diagnostic for Alzheimer's disease; cerebrospinal fluid; phosphorylated tau; plasma amyloid beta; plasma phosphorylated tau181; positron emission tomography.

Conflict of interest statement

Eugeen Vanmechelen is cofounder and shareholder of ADx NeuroSciences. Erik Stoops is employee and shareholder of ADx NeuroSciences. Jeroen Vanbrabant, Nele Dewit, and Kimberley Mauroo are employees of ADx NeuroSciences. Paul Maruff is a full‐time employee of Cogstate Ltd. Christopher C. Rowe has received research grants from NHMRC, Enigma Australia, Biogen, Eisai, and Abbvie; he is on the scientific advisory board for Cerveau Technologies; and consulted for Prothena, Eisai, Roche, and Biogen Australia. The other authors did not report any conflict of interest. Author disclosures are available in the supporting information

© 2022 The Authors. Alzheimer's & Dementia: Diagnosis, Assessment & Disease Monitoring published by Wiley Periodicals, LLC on behalf of Alzheimer's Association.

Figures

FIGURE 1

Box and whisker plot for Aβ1‐42, p‐tau181 and p‐tau231, Aβ1‐42/1‐40 ratio, p‐tau181/Aβ1‐42 ratio, and p‐tau231/Aβ1‐42 ratio at Assessment 1. Individual markers (Aβ1‐40, Aβ1‐42 p‐tau181, and p‐tau231) are measured in pg/ml. A, Aβ1‐42. B, Aβ1‐42/40. C, pTau181. D, pTau181/Aβ1‐42. E, pTau231. F, pTau231/Aβ1‐42. Plots represent the five‐number summary, with the median, first, and third quartiles shown in each box. Aβ, amyloid beta; NEG, negative; PET, positron emission tomography; POS, positive; p‐tau, phosphorylated tau

FIGURE 2

Correlation between plasma and CSF biomarkers. Sample size for correlation plots at Assessment 1 was N = 155. Linear fit lines are drawn irrespective of Aβ‐PET status. Red points represent participants who were Aβ‐PET+; blue points represent participants who were Aβ‐PET−. Circle points represent those participants who were CN, square points represent those participants with MCI, triangle points represent those participants with AD. A, Plasma versus CSF Aβ1‐42. B, Plasma versus CSF Aβ1‐42/1‐40. C, Plasma versus CSF pTau181. D, Plasma versus CSF pTau181/Aβ1‐42. Aβ, amyloid beta; AD, Alzheimer's disease; CN, cognitively normal; CSF, cerebrospinal fluid; MCI, mild cognitive impairment; PET, positron emission tomography; p‐tau, phosphorylated tau

FIGURE 3

Agreement between Aβ‐PET and plasma p‐tau181/Aβ1‐42 ratio at Assessment 1. Threshold line for Aβ‐PET was set at 20 CL. Threshold for the p‐tau181/Aβ1‐42 ratio was set using the Youden's Index (1.48) from the ROC model for Aβ‐PET status with the CL threshold set at 20 CL at Assessment 1. Red points represent participants who were Aβ‐PET+; blue points represent participants who were Aβ‐PET−. Circle points represent those participants who were CN, square points represent those participants with MCI, triangle points represent those participants with AD. Aβ, amyloid beta; AD, Alzheimer's disease; CL, Centiloid; CN, cognitively normal; CSF, cerebrospinal fluid; MCI, mild cognitive impairment; PET, positron emission tomography; p‐tau, phosphorylated tau; ROC, receiver operating characteristic

FIGURE 4

Cognitive decline in CDR‐SoB and the AIBL PACC score using the plasma between CN/CI groups and the p‐tau181/Aβ1‐42 ratio as measured by linear mixed effects models. Time on the x‐axis refers to the first AIBL assessment whereby the plasma was collected. Sample sizes for cognitive collection points for the plasma sample set are shown in Table S1. The binary p‐tau181/Aβ1‐42 ratio was created using the Youden's Index (1.48) created from the ROC model using plasma p‐tau181/Aβ1‐42 versus Aβ‐PET using a CL threshold at 20 CL. Threshold for the plasma p‐tau181/Aβ1‐42 ratio was 1.483. Aβ, amyloid beta; AD, Alzheimer's disease; AIBL, Australian Imaging, Biomarkers and Lifestyle study; CDR‐SoB, Clinical Dementia Rating Sum of Boxes; CL, Centiloid; CN, cognitively normal; CSF, cerebrospinal fluid; MCI, mild cognitive impairment; PACC, Preclinical Alzheimer Cognitive Composite; PET, positron emission tomography; p‐tau, phosphorylated tau; ROC, receiver operating characteristic