Metabolomic Interactions Between Alzheimer’s and Metabolic Diseases: Insights from the Nightingale–Mass General Brigham Biobank
Ezgi Sila Toklucu1, Haoqi Sun2, Jessica Lasky Su3, Elizabeth Karlson4, Can Zhang5
1Acibadem University School of Medicine, 2Department of Neurology, Beth Israel Deaconess Medical Center, 3Channing Laboratory, Brigham and Women's Hospital, 4Department of Rheumatology, Brigham and Women’s Hospital, 5Department of Neurology, Massachusetts General Hospital
Objective:
To characterize how Alzheimer’s disease (AD) interacts metabolically with dyslipidemia (DL) and type 2 diabetes (T2D) and to evaluate the mediating role of APOE genotype using large-scale metabolomic data.
Background:
Neurodegenerative diseases often coexist with metabolic disorders, yet their biochemical overlap remains poorly defined. In this study, we investigated overlaps, main similarities and differences between significant metabolic alterations found in comorbidities such as DL and T2D, and metabolic alterations observed in AD.
Design/Methods:
We analyzed 249 metabolites quantified by Nightingale NMR metabolomics in 45,514 participants from the Mass General Brigham Biobank. Comprehensive linear interaction models (metabolite ~ AD + DL + T2D + interactions + covariates) were adjusted for age, sex, and APOE genotype. Bonferroni correction was applied using 24 principal components (α = 0.002). Visualization included volcano, Venn diagram, bubble, and UMAP plots; pathway enrichment identified altered metabolic routes.
Results:
Dyslipidemia showed specific metabolic changes, especially becoming more prominent by increased omega-3 fatty acids, reduced HDL size, and decreased ω6/ω3 ratio (792/996 significant associations). Alzheimer's disease demonstrated 72 synergistic AD:DL interactions with negative coefficients (β = –0.29 to –0.42) affecting LDL and VLDL particles, cholesteryl esters, and ApoB metabolism predominantly. In contrast, T2D displayed substantially different interaction profiles, specifically on amino acid and protein metabolism. When analyses were stratified by APOE genotype, APOE4 carriers exhibited stronger and more consistent lipid and fatty acid disturbances, whereas APOE2 carriers demonstrated partial reversal of these effects. Pathway enrichment detected disruptions in ammonia recycling, branched-chain amino acid degradation, phenylalanine/tyrosine metabolism, and ketone body pathways.
Conclusions:
Alzheimer's disease demonstrates unique metabolic vulnerability to dyslipidemia, with additional T2D-related metabolic shifts and strong APOE-dependent changes. Metabolic overlaps and genotype-dependent differences across AD, DL, and T2D suggest that personalized, APOE-guided lipid and metabolic management strategies may improve understanding and treatment of neurodegeneration.
Disclaimer: Abstracts were not reviewed by Neurology® and do not reflect the views of Neurology® editors or staff.