Plasma Phospholipid Biomarkers Related to the Risk of Cognitive Decline in the Elderly: Results from a Cohort Study

TL;DR

This study identifies blood phospholipid biomarkers that predict cognitive decline in elderly individuals, improving early detection and risk assessment.

Contribution

The study introduces a novel panel of eight phospholipid biomarkers that enhance prediction of cognitive decline when added to traditional risk factors.

Findings

Eight phospholipid species were identified as biomarkers for cognitive decline using LASSO regression.

Four specific phospholipids (PE(O-40:5), LPC(18:3), PI(38:2), PA(39:4)) showed significant associations with cognitive decline risk.

Adding these biomarkers improved the predictive model's AUC from 0.743 to 0.866.

Abstract

Phospholipids provide both structural and functional varieties for neuro cells, and their dysregulation in brain has been related to pathogenesis of cognitive impairment. The reflection of these phospholipid alterations in the blood might serve as biomarkers for the early recognition of cognitive decline risk preceding clinical symptoms and provide potential targets for intervention. In this cohort study, detailed phospholipid molecular profiles including 229 species were quantified. A total of 209 participants aged 60–80 years (including 138 women and 73 men) were followed for one year, during which 32 participants developed significant cognitive decline, defined as a decrease of three or more points in the Montreal Cognitive Assessment score. A biomarker panel of eight phospholipid molecular species related to cognitive decline was identified by Least Absolute Shrinkage and Selection Operator (LASSO) logistic regression between cases and non-cases. Among these, four biomarkers, including PE(O-40:5), LPC(18:3), PI(38:2) and PA(39:4), were further proved to be significantly associated with the risk of cognitive decline through a logistic regression model, indicating that the degradation of phospholipids and the accumulation of ether phospholipid and PI might participate in the process of cognitive decline in early stage. By adding the eight phospholipid biomarkers to a reference model that included demographics, lifestyle, hypertension, fasting blood glucose and blood lipid parameters, the AUC value of the predictive model improved from 0.743 to 0.866, which provided a possible auxiliary screening tool for the early detection of cognitive impairment in the elderly.

Figures