EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Integrating large-scale serum baseline metabolomics and APOE ε4 genotype status for the non-invasive detection of Alzheimer’s disease in the Alzheimer’s Disease Neuroimaging Initiative (ADNI) cohort

Dany Mukesha, Hüseyin Firat, Guillaume Sacco

PLOS One · 17.11.2026 · 21 (11) · e0356137

DOI: 10.1371/journal.pone.0356137 ↗

Kurz gefasst

Fachartikel in der Zeitschrift PLOS One, online erschienen am 17.11.2026. Verfasst von Dany Mukesha und 2 weiteren Autor:innen. Die Arbeit nennt 42 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: CombiDiag, HORIZON – MSCA Doctoral Networks 2021 program.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Introduction

Accurate Alzheimer’s disease (AD) detection remains challenging and often requires invasive or costly procedures. Blood-based metabolomic signatures offer a promising non-invasive comparative approach for disease stratification. This study aimed to evaluate whether large-scale serum metabolite panel combined with apolipoprotein E ( APOE ) ε4 genotype status could distinguish AD from cognitively normal (CN) individuals within the Alzheimer’s Disease Neuroimaging Initiative (ADNI) cohort.

Methods

Baseline data from 594 participants in ADNI (237 AD, 357 CN) were analyzed. High-resolution serum metabolomics (Biocrates MxP ® Quant 500) and APOE genotype data were integrated into a machine learning framework. Feature selection was performed using LASSO regression followed by model training and evaluation using repeated cross-validation and a held-out test set. Sensitivity analyses included models adjusted for potential cofounders, and evaluation of progressively smaller metabolites.

Results

A panel of 151 metabolites distinguished AD from CN with high accuracy (test-set AUC = 0.90). Addition of APOE to the panel further improved model performance (AUC = 0.91 versus AUC = 0.75 for APOE alone; DeLong test p < 0.001), achieving strong sensitivity (0.92), specificity (0.84), and negative predictive value (0.94). Formal DeLong testing confirmed significant improvement of the combined model relative to APOE -alone classification, while the incremental gain over metabolomics alone was modest but consistent. Key predictive metabolites included bile acids, ether-linked phosphatidylcholines, and acylcarnitines, which are associated with pathways related to lipid metabolism, mitochondrial function, and the gut–liver–brain axis. Sensitivity analyses demonstrated overall stability of the metabolomic signature across covariate adjustments and subgroup analyses.

Conclusion

Our findings reinforce and extend prior observations that systemic metabolic dysregulation accompanies Alzheimer’s disease. The integration of standardized targeted serum metabolomics with APOE ε4 genotype provides complementary predictive information beyond established genetic risk factors and may serve as a complementary tool alongside state-of-the-art plasma biomarkers such as p-tau217 and GFAP. Although external validation in independent cohorts and longidinal prediction studies remain essential prerequisites, the present results demonstrate the potential utility of integrated metabolomics and genomics approaches for non-invasive AD risk stratification. ClinicalTrials.gov Identifier: NCT00106899 and related ADNI phases.

Förderung

  • CombiDiag, HORIZON – MSCA Doctoral Networks 2021 program

Zitieren

APA

Mukesha, D., Firat, H., & Sacco, G. (2026). Integrating large-scale serum baseline metabolomics and APOE ε4 genotype status for the non-invasive detection of Alzheimer’s disease in the Alzheimer’s Disease Neuroimaging Initiative (ADNI) cohort. PLOS One, 21(11), e0356137. https://doi.org/10.1371/journal.pone.0356137

Vancouver

Mukesha D, Firat H, Sacco G. Integrating large-scale serum baseline metabolomics and APOE ε4 genotype status for the non-invasive detection of Alzheimer’s disease in the Alzheimer’s Disease Neuroimaging Initiative (ADNI) cohort. PLOS One. 2026;21(11):e0356137. doi:10.1371/journal.pone.0356137

BibTeX
@article{Mukesha2026,
  author = {Mukesha, Dany and Firat, Hüseyin and Sacco, Guillaume},
  title = {Integrating large-scale serum baseline metabolomics and APOE ε4 genotype status for the non-invasive detection of Alzheimer’s disease in the Alzheimer’s Disease Neuroimaging Initiative (ADNI) cohort},
  journal = {PLOS One},
  year = {2026},
  volume = {21},
  number = {11},
  pages = {e0356137},
  doi = {10.1371/journal.pone.0356137}
}
Kostenlos starten →