Graph theoretical approach to brain remodeling in multiple sclerosis

Area of Science:

• Neuroscience
• Medical Imaging
• Network Science

Background:

• Multiple sclerosis (MS) is a neuroinflammatory disease impacting the brain's structural connectivity.
• While the brain has natural remodeling capabilities, effective biomarkers for assessing this in MS are lacking.
• Understanding network changes is crucial for evaluating disease progression and recovery.

Purpose of the Study:

• To investigate graph theory metrics, specifically modularity, as potential biomarkers for neural remodeling and cognitive function in MS.
• To compare network organization between individuals with MS and healthy controls.
• To explore the relationship between graph metrics, lesion load, and cognitive performance in MS.

Main Methods:

• Recruited 60 relapsing-remitting MS patients and 26 healthy controls.
• Acquired structural MRI, diffusion MRI, cognitive assessments, and disability evaluations.
• Calculated graph theory metrics (modularity, global efficiency) from tractography-derived brain connectivity matrices.
• Utilized general linear models to analyze associations, adjusting for relevant covariates.

Main Results:

• Individuals with MS exhibited significantly higher modularity and lower global efficiency compared to healthy controls.
• In the MS cohort, increased modularity correlated inversely with cognitive performance.
• Higher modularity in MS patients was positively associated with T2 lesion load.

Conclusions:

• Elevated modularity in MS appears to result from lesion-induced disruption of intermodular connections.
• The observed increase in modularity does not correlate with improved or preserved cognitive function in MS.
• Modularity shows promise as a biomarker for evaluating neural network disruption and impaired remodeling in multiple sclerosis.