Abstract
The first genome-wide significant multiple sclerosis severity locus, rs10191329, has been pathologically linked to cortical lesion load and brain atrophy. However, observational cohorts such as MSBase have not replicated associations with disability outcomes, instead finding other loci. We evaluated rs10191329 and MSBase loci in a unique cohort of 53 people followed for 30 years after a clinically isolated syndrome, with deep clinical phenotyping and MRI measures of inflammation and neurodegeneration. After 30 years, 26 had developed relapsing-remitting multiple sclerosis, 15 secondary progressive multiple sclerosis and 12 remained diagnosed with a clinically isolated syndrome. Genetic associations with disease severity (age-related multiple sclerosis severity score and Expanded Disability Status Scale), disease course and brain MRI features (white matter lesions, cortical lesions and grey matter fraction) were investigated using regression models and survival analyses. rs10191329 was not associated with multiple sclerosis severity, secondary progressive multiple sclerosis diagnosis or brain MRI features at 30 years. Similarly, MSBase loci were not associated with 30-year disease severity, although rs73091975 was significantly associated with lower 14-year age-related multiple sclerosis severity score in those developing multiple sclerosis. Given that effect sizes for both rs10191329 and rs73091975 were greatest between 14 and 20 years, these findings suggest genetic effects on multiple sclerosis severity may interact non-linearly with disease duration.