학술논문
A blood-based biomarker workflow for optimal tau-PET referral in memory clinic settings
Document Type
article
Author
Wagner S. Brum; Nicholas C. Cullen; Joseph Therriault; Shorena Janelidze; Nesrine Rahmouni; Jenna Stevenson; Stijn Servaes; Andrea L. Benedet; Eduardo R. Zimmer; Erik Stomrud; Sebastian Palmqvist; Henrik Zetterberg; Giovanni B. Frisoni; Nicholas J. Ashton; Kaj Blennow; Niklas Mattsson-Carlgren; Pedro Rosa-Neto; Oskar Hansson
Source
Nature Communications, Vol 15, Iss 1, Pp 1-11 (2024)
Subject
Language
English
ISSN
2041-1723
Abstract
Abstract Blood-based biomarkers for screening may guide tau positrion emissition tomography (PET) scan referrals to optimize prognostic evaluation in Alzheimer’s disease. Plasma Aβ42/Aβ40, pTau181, pTau217, pTau231, NfL, and GFAP were measured along with tau-PET in memory clinic patients with subjective cognitive decline, mild cognitive impairment or dementia, in the Swedish BioFINDER-2 study (n = 548) and in the TRIAD study (n = 179). For each plasma biomarker, cutoffs were determined for 90%, 95%, or 97.5% sensitivity to detect tau-PET-positivity. We calculated the percentage of patients below the cutoffs (who would not undergo tau-PET; “saved scans”) and the tau-PET-positivity rate among participants above the cutoffs (who would undergo tau-PET; “positive predictive value”). Generally, plasma pTau217 performed best. At the 95% sensitivity cutoff in both cohorts, pTau217 resulted in avoiding nearly half tau-PET scans, with a tau-PET-positivity rate among those who would be referred for a scan around 70%. And although tau-PET was strongly associated with subsequent cognitive decline, in BioFINDER-2 it predicted cognitive decline only among individuals above the referral cutoff on plasma pTau217, supporting that this workflow could reduce prognostically uninformative tau-PET scans. In conclusion, plasma pTau217 may guide selection of patients for tau-PET, when accurate prognostic information is of clinical value.