Résumé
Metabolic biomarkers can potentially be used for early diagnosis, prognostic risk stratification and/or early treatment and prevention of individuals at risk to develop Alzheimer’s disease (AD). Our goal is to evaluate changes in metabolite concentration levels associated with AD to identify biomarkers that could support early and accurate diagnosis and therapeutic interventions by using targeted mass spectrometry and machine learning approaches. Serum samples collected from a total of 107 individuals, including 55 individuals diagnosed with AD and 52 healthy controls (HC) enrolled previously to ADDIA cohort were analyzed using the Biocrates® 400 metabolite panel. Several machine learning models including Lasso, Random Forest, and XGBoost were trained to classify AD and HC. Repeated cross-validation was used to ensure performance evaluation. We identified 18 metabolites with nominal differences (p<0.05; AUC>0.60) between AD and HC. These included alterations in acylcarnitines, phosphatidylcholines, sphingomyelins, triglycerides, and amino acids, suggesting disruptions in lipid metabolism, mitochondrial function, and oxidative stress. The best model achieved an average AUC of 0.88 on the train set and 0.73 on the test set. Classification performance was further improved by combining multiple metabolites in a single panel and adding APOE genotyping (AUC=0.902). These results highlight important metabolic signatures that could help to reduce misdiagnosis and support the development of metabolomic panels to detect AD. The combination of multiple serum metabolic biomarkers and APOE genotyping can significantly improve classification accuracy and potentially assist in making non-invasive, cost-effective diagnostic approach.