Scalable biological-cognitive profiling for Alzheimer’s disease in the population

Abstract Plasma phosphorylated tau217 has been suggested as a core biomarker for establishing a biological Alzheimer’s disease diagnosis. This blood biomarker has not been studied together with scalable cognitive assessment tools in population-based samples. We investigated the prevalence of cognitive and Alzheimer’s disease biomarker abnormalities and associations between plasma phosphorylated tau217 and remotely measured cognitive function in individuals without dementia. We used a population-based cross-sectional sample of 65–85-year-olds (n = 691, 57% females), excluding those with previously diagnosed Alzheimer’s disease or other dementia-causing neurodegenerative disease. Cognition was measured with a telephone-administered word list recall task (episodic memory) and animal naming (semantic fluency). Plasma phosphorylated tau217 was determined with the ALZpath assay. The prevalence of individuals with abnormalities in tests measuring episodic memory, semantic fluency, and plasma phosphorylated tau217 was 10–13%. Higher plasma phosphorylated tau217 levels were associated with lower scores on telephone-administered cognitive tests. We found a substantial minority of a population-based sample of individuals without a clinical diagnosis of Alzheimer’s disease to have cognitive and plasma phosphorylated tau217 profiles suggesting underlying Alzheimer’s disease. Combining plasma phosphorylated tau217 with remote cognitive assessment could be a scalable, accessible, and cost-effective protocol for screening individuals with undiagnosed or at risk for Alzheimer’s disease.