Cortical Lesions Form Predominantly in Early Multiple Sclerosis

Background and Objectives: Cortical lesions are common in multiple sclerosis (MS) and associated with disability, but their characterization in early MS has been limited. Here, we aimed to characterize cortical lesions in newly diagnosed MS with 7 tesla (T) brain MRI. Methods: T1w images. Lesion burden in newly diagnosed MS was compared with a previously analyzed NIH cohort with longer time since diagnosis, imaged using a similar protocol. Results: 61 individuals were included in the newly diagnosed MS cohort (mean age 34 ± 4 years; 72% female; median time since diagnosis 5 months, interquartile range IQR 6). Cortical lesions were identified in 50/61 (81%) individuals, and subpial cortical lesions were identified in 46 (75%). Median cortical lesion number was 5 (IQR 11), median volume 319 μl (IQR 1049). Cortical lesions constituted a median of 14% of total brain lesion volume (IQR 43%), and in 21% of individuals, cortical lesions constituted >50% of total brain lesion volume. Cortical lesion number was associated with worse 9-hole peg test (ρ=0.33, p=0.008) and Symbol Digit Modalities Test performance (ρ=-0.29, p=0.02). When pooled with the NIH cohort (n=60, median time since diagnosis 12 years, IQR 17), non-cortical lesion volume was ~3.5 times higher in people with time since diagnosis >36 months (median 4.7 ml, IQR 8.7) vs ≤36 months (median 1.2 ml, IQR 2.4, p36 months (median 416 μl, IQR 1013) vs ≤36 months (median 318 μl, IQR 925, p=0.04). Non-cortical lesion volume was moderately associated with time since diagnosis (ρ=0.54, p<0.001) vs ρ=0.27 (p<0.001) for cortical lesions. Discussion: Cortical lesions are prevalent in newly diagnosed MS and constitute a substantial portion of total lesion burden. Cortical lesion volume is similar in early vs established MS, suggesting most cortical lesions form early in disease.