Plasma proteomic Alzheimer's risk score: Biological age clock and pseudotime trajectory.
Shiwei Liu, Yen-Ning Huang, Tamina Park, Eun Hye Lee, Soumilee Chaudhuri, Nicholas Adzibolosu, Paula J Bice, Jeffrey L Dage, Jared R Brosch, Sujuan Gao, Liana G Apostolova, Donna M Wilcock, Shannon L Risacher, Andrew J Saykin, Kwangsik Nho
Abstract
Open AccessINTRODUCTION: Alzheimer's disease (AD) involves progressive cognitive decline. Plasma proteomics data may provide insights into disease risk and mechanisms. METHODS: Plasma proteomics data (N = 498) by the SomaScan 7k platform from the Indiana AD Research Center was used to calculate proteomics-based Alzheimer's risk scores: organ-specific aging clocks and pseudotime trajectory. The Alzheimer's risk scores were investigated for associations with plasma- and imaging-based biomarkers for AD and diagnosis and evaluated for classification performance for diagnosis and biomarker positivity status. RESULTS: Cognition-optimized brain and liver age acceleration showed significant associations with AD diagnosis and biomarkers. Pseudotime showed a molecular trajectory from cognitively normal to AD individuals, in association with plasma and imaging biomarkers. Previously known plasma biomarkers yielded better classification performance with inclusion of proteomics-based Alzheimer's risk scores. DISCUSSION: The findings highlight proteomics-derived biological aging clocks and pseudotime trajectory as potential biomarkers to complement current biomarker frameworks and identify disease mechanisms. HIGHLIGHTS: Plasma proteomic aging clocks and pseudotime provide potential markers for Alzheimer's disease (AD) risk. Cognition-optimized brain aging and pseudotime link to diagnosis and AD biomarkers. Brain aging clock and pseudotime improve the diagnosis classification performance. Pseudotime outperformed established plasma biomarkers for diagnosis classification.