White Matter Functional Dysregulation in Amyotrophic Lateral Sclerosis: Machine Learning-Based Biomarkers and...

Amyotrophic lateral sclerosis (ALS) is a progressive neurodegenerative disorder characterized by motor system degeneration, yet its white matter (WM) functional pathophysiology remains underexplored.

This study utilized resting-state functional magnetic resonance imaging to decode WM functional abnormalities in 50 ALS patients and 55 healthy controls. Next, machine learning analysis was applied to evaluate the utility of these WM functional patterns in diagnosing ALS and predicting disease progression, and their pathophysiological mechanisms were preliminary explored through neurotransmitter mapping and imaging transcriptomics.

ALS patients exhibited reduced activity in central WM regions (including bilateral corticospinal tracts), accompanied by elevated activity in anterior and posterior WM territories. The aberrant topological properties and disrupted functional connectivity are predominantly localized within bilateral precentral/postcentral WM networks. A support vector machine model incorporating these features achieved 75.24% classification accuracy and predicted the rate of disease progression (r = 0.56, p = 0.001). The spatial pattern of WM dysfunction in ALS was associated with both the spatial distribution of disease-related neurotransmitters and the expression profiles of specific genes.

Our findings reveal distinct WM functional dysfunction patterns in ALS and their molecular-genetic underpinnings, providing novel insights into the pathophysiological mechanisms of ALS.

ALS involves specific patterns of WM dysfunction, and these WM-centric biomarkers may facilitate the development of therapeutic monitoring frameworks for this devastating disease.