A multidimensional ODE-based model of Alzheimer's disease progression

Area of Science:

• Neuroscience
• Biostatistics
• Computational Biology

Background:

• Alzheimer's disease (AD) progression models often rely on the amyloid cascade hypothesis, assuming linear pathological events.
• Observed heterogeneity in AD patient populations challenges linear models, necessitating more flexible approaches.
• Current models struggle to capture the diverse trajectories of biomarkers and cognitive decline in sporadic AD.

Purpose of the Study:

• To develop a flexible statistical model for the temporal evolution of biomarkers and cognitive tests in Alzheimer's disease.
• To enable diverse biomarker paths and improve predictions for individuals with heterogeneous disease progression.
• To provide a data-driven tool for clinical prediction, disease mechanism understanding, and clinical trial design in AD.

Main Methods:

• A multivariate dynamic model using ordinary differential equations to jointly model biomarker and cognitive test changes.
• An ordinal logistic model for clinical diagnosis prediction based on forecasted biomarker trajectories.
• Utilized longitudinal data from the Alzheimer's Disease Neuroimaging Initiative (ADNI).

Main Results:

• The model successfully captures heterogeneous biomarker dynamics in sporadic Alzheimer's disease.
• Demonstrated the ability to predict time to conversion from Mild Cognitive Impairment (MCI) to dementia.
• Illustrated interpretable patterns of biomarker rates of change.

Conclusions:

• The proposed dynamical model offers an interpretable, flexible alternative to linear AD progression hypotheses.
• It can accommodate individual variability in biomarker trajectories, crucial for understanding sporadic AD.
• The model enhances prognostic capabilities and aids in personalized medicine approaches for Alzheimer's disease.