Discriminative accuracy of the A/T/N scheme to identify cognitive impairment due to Alzheimer's disease

Tharick A. Pascoal, Antoine Leuzy, Joseph Therriault, Mira Chamoun, Firoza Lussier, Cecile Tissot, Olof Strandberg, Sebastian Palmqvist, Erik Stomrud, Pamela C.L. Ferreira, João Pedro Ferrari-Souza, Ruben Smith, Andrea Lessa Benedet, Serge Gauthier, Oskar Hansson, Pedro Rosa-Neto

Research output: Contribution to journalArticlepeer-review

Abstract

Introduction: The optimal combination of amyloid-β/tau/neurodegeneration (A/T/N) biomarker profiles for the diagnosis of Alzheimer's disease (AD) dementia is unclear. Methods: We examined the discriminative accuracy of A/T/N combinations assessed with neuroimaging biomarkers for the differentiation of AD from cognitively unimpaired (CU) elderly and non-AD neurodegenerative diseases in the TRIAD, BioFINDER-1 and BioFINDER-2 cohorts (total n = 832) using area under the receiver operating characteristic curves (AUC). Results: For the diagnosis of AD dementia (vs. CU elderly), T biomarkers performed as well as the complete A/T/N system (AUC range: 0.90–0.99). A and T biomarkers in isolation performed as well as the complete A/T/N system in differentiating AD dementia from non-AD neurodegenerative diseases (AUC range; A biomarker: 0.84–1; T biomarker: 0.83–1). Discussion: In diagnostic settings, the use of A or T neuroimaging biomarkers alone can reduce patient burden and medical costs compared with using their combination, without significantly compromising accuracy.

Original languageEnglish
Article numbere12390
JournalAlzheimer's and Dementia: Diagnosis, Assessment and Disease Monitoring
Volume15
Issue number1
DOIs
Publication statusPublished - 2023

Subject classification (UKÄ)

  • Neurosciences

Free keywords

  • A/T/N system
  • Alzheimer's disease
  • biomarker
  • diagnose
  • PET

Fingerprint

Dive into the research topics of 'Discriminative accuracy of the A/T/N scheme to identify cognitive impairment due to Alzheimer's disease'. Together they form a unique fingerprint.

Cite this