Periventricular magnetisation transfer ratio abnormalities in multiple sclerosis improve after alemtuzumab

• Published in: Multiple sclerosis 2020-08, Vol.26 (9), p.1093-1101
• Main Authors: Brown, J William L, Prados Carrasco, Ferran, Eshaghi, Arman, Sudre, Carole H, Button, Tom, Pardini, Matteo, Samson, Rebecca S, Ourselin, Sebastien, Wheeler-Kingshott, Claudia AM Gandini, Jones, Joanne L, Coles, Alasdair J, Chard, Declan T
• Format: Article
• Language: English
• Subjects: Immunotherapy; Lesions; Magnetic resonance imaging; Monoclonal antibodies; Multiple sclerosis; Neuroimaging; Substantia alba
• ISSN: 1352-4585; 1477-0970
• DOI: 10.1177/1352458519852093

Background: In multiple sclerosis (MS), disease effects on magnetisation transfer ratio (MTR) increase towards the ventricles. This periventricular gradient is evident shortly after first symptoms and is independent of white matter lesions. Objective: To explore if alemtuzumab, a peripherally acting disease-modifying treatment, modifies the gradient’s evolution, and whether baseline gradients predict on-treatment relapses. Methods: Thirty-four people with relapsing-remitting MS underwent annual magnetic resonance imaging (MRI) scanning (19 receiving alemtuzumab (four scans each), 15 untreated (three scans each)). The normal-appearing white matter was segmented into concentric bands. Gradients were measured over the three bands nearest the ventricles. Mixed-effects models adjusted for age, gender, relapse rate, lesion number and brain parenchymal fraction compared the groups’ baseline gradients and evolution. Results: Untreated, the mean MTR gradient increased (+0.030 pu/band/year) but decreased following alemtuzumab (−0.045 pu/band/year, p = 0.037). Within the alemtuzumab group, there were no significant differences in baseline lesion number (p = 0.568) nor brain parenchymal fraction (p = 0.187) between those who relapsed within 4 years (n = 4) and those who did not (n = 15). However, the baseline gradient was significantly different (p = 0.020). Conclusion: Untreated, abnormal periventricular gradients worsen with time, but appear reversible with peripheral immunotherapy. Baseline gradients – but not lesion loads or brain volumes – may predict on-treatment relapses. Larger confirmatory studies are required.