White matter microstructural differences in children and genetic risk for multiple sclerosis: A population-based study

Background: MS patients show abnormalities in white matter (WM) on brain imaging, with heterogeneity in the location of WM lesions. The "pothole" method can be applied to diffusion-weighted images to identify spatially distinct clusters of divergent brain WM microstructure. Objective: To investigate the association between genetic risk for MS and spatially independent clusters of decreased or increased fractional anisotropy (FA) in the brain. In addition, we studied sex- and age-related differences. Methods: 3 Tesla diffusion tensor imaging (DTI) data were collected in 8- to 12-year-old children from a population-based study. Global and tract-based potholes (lower FA clusters) and molehills (higher FA clusters) were quantified in 3047 participants with usable DTI data. A polygenic risk score (PRS) for MS was calculated in genotyped individuals (n = 1087) and linear regression analyses assessed the relationship between the PRS and the number of potholes and molehills, correcting for multiple testing using the False Discovery Rate. Results: The number of molehills increased with age, potholes decreased with age, and fewer potholes were observed in girls during typical development. The MS-PRS was positively associated with the number of molehills (beta = 0.9, SE = 0.29, p = 0.002). Molehills were found more often in the corpus callosum (beta = 0.3, SE = 0.09, p = 0.0003). Conclusion: Genetic risk for MS is associated with spatially distinct clusters of increased FA during childhood brain development.