The AT(N) Research Framework — Why Multi-Analyte Profiling Matters

The AT(N) framework, developed by the National Institute on Aging and Alzheimer's Association (NIA-AA) research community, organizes Alzheimer's disease biomarkers into three pathophysiological categories. No single analyte adequately captures the complexity of AD pathology — amyloid deposition, Tau tangle formation, and neurodegeneration each contribute distinct information about disease stage, progression rate, and therapeutic target engagement.

AT(N) Biomarker Categories Covered by This Panel

  • "A" — Amyloid Pathology: Aβ40 and Aβ42, with the Aβ42/Aβ40 ratio reflecting cerebral amyloid deposition (concordant with amyloid PET).
  • "T" — Tau Pathology: p-Tau181, p-Tau217, p-Tau231, and total Tau — covering multiple phospho-epitopes with distinct temporal trajectories across the AD continuum.
  • "N" — Neurodegeneration: NfL (neurofilament light, axonal injury) and GFAP (glial fibrillary acidic protein, astrocyte reactivity/neuroinflammation).
  • Exploratory Markers: Additional analytes providing complementary information on synaptic function, neuroinflammation, and vascular contributions.

Research Applications

  • Therapeutic trials — comprehensive pharmacodynamic biomarker panel for amyloid-targeting, Tau-targeting, and multi-modal investigational therapies in a single assay run.
  • Longitudinal observational cohorts — track AT(N) biomarker trajectories across the AD continuum from preclinical to symptomatic stages, with internally consistent data from a single platform.
  • Cohort enrichment — multi-marker screening strategy to identify research participants with specific biomarker profiles (e.g., A+T+N+ vs. A−T−N−) for interventional studies.
  • Disease staging research — investigate transitions between AT(N) biomarker categories as predictors of cognitive decline and disease progression.
  • Differential research profiling — distinguish Alzheimer's pathology from other neurodegenerative conditions using multi-analyte patterns rather than single-marker cutoffs.