Accuracy of plasma biomarkers to detect Alzheimer's disease proteinopathy prior to dementia

Abstract

Introduction: Plasma biomarkers sensitive to Alzheimer's disease (AD) proteinopathy prior to the onset of dementia have significant implications for early detection.

Methods: In 304 individuals without dementia, we investigated whether C₂N Diagnostics' mass spectrometry (MS)-based plasma biomarkers (amyloid beta 42/40, %phosphorylated tau [p-tau]181, and %p-tau217) and amyloid probability scores (APS, PrecivityAD and APS2, PrecivityAD2) are associated with brain amyloid, brain tau, or preclinical cognitive decline.

Results: In this cohort study, %p-tau217 and the APS2 had high discriminative accuracy (area under the curve > 0.93) for identifying elevated brain amyloid and tau and were associated with faster preclinical cognitive decline. Using %p-tau217 or the APS2 in a theoretical AD trial screening scenario reduced amyloid positron emission tomography imaging costs up to 41% or 45%, respectively.

Discussion: These findings suggest that C₂N Diagnostics' MS-based plasma biomarkers can detect brain amyloid and tau with high accuracy prior to dementia and could aid in identifying candidates for clinical trials or therapeutic intervention.

Highlights: C₂N plasma biomarkers differentiated Alzheimer's disease proteinopathy status prior to dementia. Plasma %phosphorylated tau (p-tau)217 and the C₂N Diagnostics PrecivityAD2 (APS2) were concordant with amyloid and tau positron emission tomography status. Plasma %p-tau217 and the APS2 were associated with preclinical cognitive decline.

Keywords: biomarkers; mass spectrometry; plasma; positron emission tomography; preclinical Alzheimer's disease.

FIGURE 1

ROC analysis of plasma biomarkers for distinguishing brain amyloid and brain tau. The diagnostic performance for predicting amyloid PiB PET positivity (A) and tau MK‐6240 PET positivity (B) is shown as ROC curves for plasma biomarkers, individually and in combination, with the inset showing the corresponding AUC and 95% confidence intervals for each plasma biomarker. Aβ, amyloid beta; APS, amyloid probability score; AUC, area under the curve; CI, confidence interval; MCI, mild cognitive impairment; PET, positron emission tomography; PiB, Pittsburgh compound B, p‐tau, phosphorylated tau; ROC, receiver operating characteristic; SUVR, standardized uptake value ratio; TMR, temporal meta‐ROI.

FIGURE 2

Concordance between plasma biomarkers and PET biomarkers of amyloid and tau. Quadrant plots illustrating the association between plasma biomarkers and brain amyloid as measured by equivalent Centiloids (CL; A‐E) and meta‐temporal tau as measured by MK‐6240 SUVR (F‐J) with the Youden‐based thresholds for each plasma marker shown as vertical dotted lines and the thresholds for amyloid PET positivity (PiB DVR > 1.19, Equivalent CL > 21.6) and tau PET positivity (TMR MK‐6240 SUVR > 1.30) shown as horizontal dashed lines. Individual scatter points are shaped by the diagnosis at the plasma assessment (cognitively unimpaired = circles; MCI = triangles) and shaded by amyloid PET positivity (PiB+ filled shape; PiB– empty shape). Youden‐based thresholds for each plasma biomarker predicting amyloid PET positivity (Aβ42/40 > 0.093, p‐tau181 ratio > 18.68, p‐tau217 ratio > 2.66, APS > 32.5, APS2 > 22.5; vertical dotted lines, A‐E) and tau PET positivity (Aβ42/40 > 0.092, p‐tau181 ratio > 18.36, p‐tau217 ratio > 4.90, APS > 47.5, APS2 > 31.5; vertical dotted lines, F‐J). Aβ, amyloid beta; APS, amyloid probability score; CU, cognitively unimpaired; DVR, distribution volume ratio; MCI, mild cognitive impairment; PET, positron emission tomography; PiB, Pittsburgh compound B; p‐tau, phosphorylated tau; SUVR, standardized uptake value ratio; TMR, temporal meta‐ROI.

FIGURE 3

Associations of plasma biomarkers with longitudinal cognitive decline. This figure shows retrospective longitudinal PACC‐3 trajectories derived from linear mixed‐effects models with longitudinal PACC‐3 as the outcome and age, age², sex, education, cohort, practice, plus each plasma biomarker and it's interactions with age and age² included separately from each other as predictors. Each plot shows the estimated PACC‐3 trajectory for a plasma biomarker‐negative (gray) or biomarker‐positive (pink/purple/green/blue) initially cognitively unimpaired participant with average education, average practice, and female sex. Vertical dotted lines indicate the estimated age when PACC‐3 trajectories diverged between the average plasma biomarker‐negative and biomarker‐positive individual for each plasma biomarker. Plasma biomarker positivity was determined using the optimal Youden‐based cutoffs for predicting amyloid PET positivity. Shaded areas represent 95% confidence intervals of the regression lines. Aβ, amyloid beta; APS, amyloid probability score; PACC‐3, three test Preclinical Alzheimer's Cognitive Composite, PET, positron emission tomography; p‐tau, phosphorylated tau.