Abstracts

Rationale: Juvenile myoclonic epilepsy (JME) is a highly heritable epilepsy syndrome. Traits of the wider JME phenotype, such as frontal lobe impairment, are also present in healthy first-degree relatives. A genetically determined neurodevelopmental pathomechanism is assumed, but so far mainly corroborated by functional studies showing thalamo-fronto-cortical alterations. Given the strong link between genetic variations, neurodevelopment, and sulcation patterns, MRI markers of cortical morphology and folding have recently gained substantial attention in developmental and genetic studies. To assess in vivo biomarkers of genetic risk in JME, we conducted a multi-dimensional morphometric MRI analysis in JME patients and their unaffected siblings. Methods: We studied 29 JME patients, 16 unaffected siblings (related to 11 patients), and 20 healthy controls. Based on cortical surface models automatically extracted from high-resolution MRI data, we computed three markers of neurodevelopment: curvature, regional surface area, and average geodesic distance of a region to the remaining cortical areas. Multivariate Hoteling's t test aggregated these features and mapped joint abnormalities in patients and siblings compared to controls. This analysis was contrasted to a vertex-wise group comparison of cortical thickness, a feature more closely associated with overall disease load. Resulting statistical maps were corrected for multiple comparisons using random field theory, controlling the family-wise error to PFWE < 0.05 on a cluster level. Results: All groups were comparable for age and gender (age: JME median (interquartile range) 32 (15) years, siblings 42 (20), controls 32 (7), P=.29; male gender: JME 12, siblings 6, controls 6, P=.72). In the multivariate analysis, patients and siblings shared increased surface area, curvature and geodesic distance in the bilateral anterior prefrontal cortex when compared to controls (Figure 1). Separate analysis of each imaging marker showed atypical features in patients and siblings in similar regions. On the other hand, cortical thickness was only decreased in patients, within bilateral central motor and medial premotor cortices and in the right medial occipital cortex (Figure 2). Conclusions: MRI-based biomarkers of brain development co-segregate JME patients and siblings independently of disease activity, supporting a genetically determined disease mechanism. Affected regions lie within critical cognitive network hubs and may explain frontal lobe impairment in both groups. Funding: The study was funded through the Wellcome Trust (Project Grant No 079474) and the Henry Smith Charity (ref 20133416). BW was funded by a fellowship of the German Research Foundation (Deutsche Forschungsgemeinschaft; WA3135/1-1). BB, AB, NB were supported by the Canadian Institute of Health Research.