Prediction of Lesion Growth in Multiple Sclerosis: A longitudinal Study with High-resolution Whole-brain MRSI

Motivation: Detecting diffuse inflammation can help predict lesion growth in multiple sclerosis (MS), but conventional MRI/MRSI lacks the necessary sensitivity and efficacy. Goal(s): To investigate whether neuroinflammatory and neuronal biomarkers measured using high-resolution whole-brain MRSI could predict lesion growth in MS patients. Approach: High-resolution neurometabolite mapping was performed in a longitudinal cohort of 34 MS patients over six months using the SPICE technology. Results: Lac/NAA and mI/NAA levels within baseline lesions were significantly correlated with increasing lesion volumes. Combining all MRSI biomarkers yielded the best performance in predicting enlarging lesion volumes and identifying patients with higher risk of lesion growth. Impact: High-resolution whole-brain MRSI provides a promising tool for predicting lesion growth and stratifying patients with higher risk of enlarging lesions during short-term follow-up in MS.