Abstract
The relationship between fatty acids and Alzheimer's Disease (AD) risk has been an area of growing interest but remains insufficiently understood. This study aimed to develop and validate a fatty acid score (FAS) derived from blood fatty acid levels and explore its association with AD risk. We analyzed 148,308 UK Biobank participants (age 37-73; mean 55.96 years) with a mean follow-up of 12.3 years (maximum 16), and 1193 ADNI subjects (age 55-90; mean 73.50 years) with a mean follow-up of 4.2 years (maximum 8). Lasso regression was used to construct the FAS based on UKB, and Cox regression and linear regression was employed to assess the relationships of FAS with AD risk, cognition, hippocampal volume, and/or cerebrospinal fluid markers in both cohorts. Stratified effects by APOE ε4 status were examined. Causal mediation, proteomic, and bioinformatic analyses were performed to reveal potential mechanisms. Higher FAS was associated with increased AD risk in both cohorts (UKB: HR = 1.298, 95% CI 1.183-1.423, P < 0.001; ADNI: HR = 1.413, 95% CI 1.105-1.808, P = 0.006). In UKB, higher FAS was linked to reduced hippocampal volume (P < 0.001), and in ADNI, it was associated with faster hippocampal atrophy (P = 0.002) and cognitive decline (P < 0.001). These associations were stronger in APOE ε4 carriers. Hippocampal volume partly mediated the link between FAS and cognitive decline. Proteomic analyses demonstrated that the protein expression levels of Adhesion G protein-coupled receptor G1 (ADGRG1), Chitinase-3-like protein 1 (CHI3L1), RNA-binding FOX-1 homolog 3 (RBFOX3), and Growth differentiation factor 15 (GDF15) could mediate the effect of FAS on AD risk. The enriched pathways include cytokine activity, neurotrophic signaling, and pathways related to nervous system development. Blood levels of fatty acid could aid in AD prediction, but further research is needed to confirm causality.