"Brain age" predicts disability accumulation in multiple sclerosis.

Area of Science:

• Neuroimaging and computational neuroscience
• Clinical neurology and neuroimmunology

Background:

• Neurodegenerative diseases like multiple sclerosis (MS) often show signs of premature aging on medical imaging.
• While MS lesion biomarkers are established, reliable measures for neurodegeneration are still developing.
• Machine learning-based 'brain age' analysis quantifies apparent aging from structural MRI, offering a potential neurodegeneration metric.

Purpose of the Study:

• To evaluate the utility of brain age analysis as a biomarker for neurodegeneration in MS.
• To determine the explanatory and predictive power of brain age for disability in a large MS cohort.
• To assess how brain age relates to MS disease course, lesion burden, and treatment efficacy.

Main Methods:

• Utilized a large, real-world longitudinal dataset of over 13,000 imaging sessions from more than 6,000 MS patients.
• Compared the performance of three distinct brain age algorithms.
• Employed linear mixed-effects models to analyze associations with MS, disease course, disability, lesion burden, and disease-modifying therapy (DMT) efficacy.

Main Results:

• All three brain age algorithms indicated that MS is associated with advanced predicted brain age cross-sectionally and accelerated brain aging longitudinally.
• Patient disability was the strongest correlate of advanced brain age, more so than MS disease course (relapsing vs. progressive).
• Advanced brain age at the study's start predicted greater disability accumulation over time, while a younger brain age correlated with decreased disability.

Conclusions:

• Brain age analysis is a feasible and clinically significant biomarker for MS pathology.
• Advanced brain age in MS patients strongly correlates with and predicts increasing disability.
• This neuroimaging biomarker holds promise for monitoring disease progression and predicting outcomes in multiple sclerosis.