FDG-PET brain glucose hypometabolism predicts Alzheimer's disease progression pathways in cognitively normal adults: A longitudinal competing risks modeling Article

Full Text via DOI: 10.1016/j.metop.2025.100400 Web of Science: 001588435600001

Cited authors

  • Alhasan MS, Alhasan AS, Milburn J, Khalil M, Almaghraby A, Alharthi O, Hamoud S, Essibayi MA, Elhassan YH, Feltrin F, Singh S, Azzam AY

Abstract

  • Introduction: Alzheimer's disease progression follows distinct pathways in cognitively normal individuals: direct conversion to dementia versus sequential decline through mild cognitive impairment (MCI). The metabolic determinants of pathway selection remain unclear, limiting personalized intervention strategies. Methods: We analyzed 1136 cognitively normal participants from the Alzheimer's Disease Neuroimaging Initiative with baseline fluorodeoxyglucose positron emission tomography (FDG-PET) and longitudinal outcomes over ten years. Competing risks regression modeled pathway-specific transitions, while multinomial logistic regression predicted pathway membership using brain glucose metabolism. Cross-validation assessed pathway classification accuracy across temporal splits. Results: Four progression pathways were concluded from our analyses, cognitive stability (32.8 %), sequential MCI-only decline (34.9 %), accelerated MCI-to-dementia progression (15.8 %), and rapid direct conversion (16.5 %). Brain glucose hypometabolism determined pathway selection with significant effects: participants with severe hypometabolism (FDG z-score < -0.5) demonstrated 7.4-fold acceleration in direct conversion velocity compared to preserved metabolism (17.12 vs 2.31 per 100 person-years, P-value<0.001). Pathway prediction models achieved excellent discrimination for direct conversion (AUC = 0.994) and acceptable performance for sequential pathways (AUC = 0.680). Metabolic phenotyping demonstrated peculiar vulnerability profiles, cognitive stability maintained metabolic reserve (FDG +0.57 +/- 0.58), while rapid converters demonstrated metabolic failure patterns (FDG -0.18 +/- 0.88). Conclusions: Based on our modeling findings, we observed that brain glucose metabolism could serve as a pathway determinant rather than simply a decline predictor, which could play a promising role in precision medicine approaches to Alzheimer's disease prevention. FDG-PET biomarkers can stratify individuals for pathway-specific interventions, transforming reactive dementia care into proactive pathway-guided management.

Authors

Publication date

  • 2025

Number of pages

  • 16

Volume

  • 28