A multianalyte plasma protein signature associated with cognitive function in a healthy ageing population

Abstract

AbstractBackgroundSeveral studies have been conducted towards the identification of biomarkers for Alzheimer’s disease (AD); however, there are limited biomarker studies investigating the preclinical phase prior to the incidence of AD. The present study aims to identify a multianalyte panel of blood biomarkers associated with overall cognitive function in a healthy ageing male population.MethodA multiple reaction monitoring (MRM)‐based targeted proteomic analysis was conducted using plasma samples of 448 male participants. The abundance levels of 269 plasma proteins known to participate in several physiological processes such as lipid transport, oxidative stress, inflammation, cytoskeleton signalling and complement pathway were measured. Protein concentration values outside the limit of quantitation were assigned as missing and proteins with greater than 10% missing values were removed from the analysis. Consequently, 125 proteins were included in the statistical analysis. In order to build a multimarker model, the least absolute shrinkage and selection operator (LASSO) regression analysis was conducted. This was followed by pathway mapping of the proteins selected by the model.ResultPreliminary analysis revealed a 12‐marker model (R‐squared = 32%) comprising age and the following proteins: Apolipoprotein A‐I, Apolipoprotein C‐I, Alpha‐1‐acid glycoprotein 1, Carboxypeptidase N catalytic chain, Coagulation factor XI, Complement C1q subcomponent subunit B, Glutathione peroxidase 3, Neuropilin‐2, Insulin‐like growth factor‐binding protein complex acid labile subunit, Protein AMBP and Lysozyme C.ConclusionA multianalyte panel consisting of plasma proteins was built for predicting cognitive function using cross‐sectional analysis methods. Furthermore, to predict long‐term cognitive changes, follow‐up assessments will be required.