Connectome-Based Propagation Model in Amyotrophic Lateral Sclerosis.

Area of Science:

• Neuroscience
• Computational Biology
• Medical Imaging

Background:

• Amyotrophic lateral sclerosis (ALS) clinical trials predominantly use survival or functional scales as endpoints.
• Quantitative biomarkers, including neuroimaging, are emerging but disease spread patterns remain unclear.
• Cerebral magnetic resonance imaging (MRI) can detect in vivo ALS pathology.

Purpose of the Study:

• To simulate disease propagation in ALS using network analyses of cerebral MRI data.
• To predict disease progression in ALS patients through computational modeling.
• To validate simulated disease patterns against empirical clinical and imaging data.

Main Methods:

• Network-based statistics were applied to longitudinal brain MRI data from 208 ALS patients and controls.
• A computational model simulated progressive network degeneration originating from the motor cortex.
• Simulated aggregation levels were validated against white matter integrity and clinical decline in internal and external datasets.

Main Results:

• Computer-simulated aggregation levels closely mimicked true disease patterns in ALS patients.
• Simulated cortical involvement overlapped significantly with empirically impaired brain regions at group and individual levels.
• Findings were further validated using an external dataset of 30 ALS patients.

Conclusions:

• Computational models effectively predict ALS disease progression, aligning with pathological staging systems.
• Simulated disease patterns offer potential as prognostic biomarkers for ALS.
• These models may enhance patient stratification in future ALS clinical trials.