Metabolomics analysis of the Finnish Geriatric Intervention Study to Prevent Cognitive Impairment and Disability (FINGER)

Abstract

Background: The Finnish Geriatric Intervention Study to Prevent Cognitive Impairment and Disability (FINGER) is a randomized controlled trial that showed significant benefits on cognition for a 2‐year multidomain lifestyle intervention vs regular health advice in 1260 at‐risk older individuals from the general population. The aim of this study was to investigate the metabolic effects of the intervention and their association with cognitive outcomes. Method: We acquired metabolomics data of blood plasma on two time points (baseline and 2 years) in a set of 1089 participants (547 intervention and 542 control). A multi‐technology platform consisting of using ultra‐high performance liquid chromatography ‐ mass spectrometry (UHPLC‐MS) and 1H nuclear magnetic resonance spectroscopy (NMR) was used to generate extensive measurement coverage of the metabolome. The data was acquired using both reversed‐phase chromatography (RPC) tailored for the retention and separation of both complex and neutral lipids and hydrophilic interaction liquid chromatography (HILIC) for the retention and separation of small polar metabolites. The UHPLC‐MS untargeted data contains a total of 10,825 metabolomic features, 1244 of them currently annotated. The NMR data contains measurements of both small and macromolecules, including 112 lipoprotein fractions (derived using Bruker lipoprotein subclass analysis: BI‐LISA). Data analysis was conducted using linear mixed models. Result: Several baseline metabolite levels had a statistically significant association with cognition. Results will also be presented from ongoing analyses to investigate intervention effects on change in metabolites. Conclusion: Complex metabolomics analyses are new in the context of multidomain lifestyle dementia prevention trials. While causality is yet to be investigated further, some of the metabolite classes and pathways observed (e.g. sphingolipids) have been previously implicated in cognitive decline and neurological disorders, suggesting early interpretable leads in the rich dataset at hand.