BERLIN—Research supports the previously observed association of periventricular lesions with cortical thinning in relapsing-remitting multiple sclerosis (MS) and extends this association to clinically isolated syndrome (CIS) and primary progressive MS, according to results presented at ECTRIMS 2018. “Beyond supporting the role of CSF-related factors for disease-related damage, a periventricular lesion MS phenotype associated with reduced cortical thickness might have the potential to be explored regarding response to different disease-modifying treatments,” said Lukas Pirpamer, a doctoral student at Graz University of Technology in Austria.
Neurologists previously had hypothesized that subpial cortical pathology in MS partly results from CSF factors. In an earlier single-center study, Mr. Pirpamer and colleagues investigated whether tissue changes in intracerebral areas close to CSF were related to cortical pathology. They found that the percentage of periventricular lesion occupancy was correlated with cortical thinning in relapsing-remitting MS, but not in CIS. To corroborate these findings and examine other forms of MS, Mr. Pirpamer and colleagues conducted a multicenter study of the MRI in MS (MAGNIMS) study group, which included patients with primary progressive MS and those with secondary progressive MS.
The investigators analyzed MRI scans of 564 patients from centers of the MAGNIMS network that provided 1-mm isotropic T1-magnetization prepared rapid acquisition gradient echo (MPRAGE) and T2-fluid-attenuated inversion recovery (FLAIR) sequences and manually segmented lesion masks. They used FreeSurfer 6.0 to measure cortical thickness and perform brain segmentation. To classify periventricular lesions, Mr. Pirpamer’s group used a region growing algorithm starting from the ventricle mask implemented in MATLAB. They defined the periventricular lesion load percentage as the proportion of periventricular to nonperiventricular lesions.
Periventricular lesion load percentage correlated with cortical thinning in the entire cohort (R = –0.23), but nonperiventricular lesion volume did not. The analysis of different disease phenotypes revealed significant correlations in the group of 81 patients with CIS (R = –0.24), the 280 patients with relapsing-remitting MS (R = –0.24), and the 73 patients with primary progressive MS, who had the strongest correlation coefficient (R = –0.32). The investigators found no correlation in the 99 patients with secondary progressive MS, however.