Abstracts

The Network Neuropsychology of Juvenile Myoclonic Epilepsy

Abstract number : 3.472
Submission category : 11. Behavior/Neuropsychology/Language / 11B. Pediatrics
Year : 2024
Submission ID : 1384
Source : www.aesnet.org
Presentation date : 12/9/2024 12:00:00 AM
Published date :

Authors :
Presenting Author: CAMILLE GARCIA-RAMOS, PhD – UW-Madison

Bruce Hermann, PhD – University of Wisconsin
Aaron Struck, MD – University of Wisconsin-Madison

Rationale: While there is considerable understanding of the neuropsychology of Juvenile Myoclonic Epilepsy (JME), unknown is the impact of JME on the interrelationship of specific cognitive abilities among each other including their patterns of integration and segregation, that is, the underlying network neuropsychology of JME. Whether network perturbations extend to unaffected siblings, raising the possibility of familial susceptibility, is similarly unknown.

Methods: Our cohort consisted of 78 JME participants (19.8 (3.7) years), 19 unaffected siblings of JME (15.9 (3.3) years), and 43 unrelated controls (20.2 (3.2) years). Administered neuropsychological tests included measures of intelligence, language, visuoperception, verbal/visual learning and memory, executive function and speeded processing. From this battery 15 specific measures were extracted (Table 1) as nodes for graph theory analyses that included calculations of global metrics (normalized global efficiency (GE), normalized average clustering coefficient (CC), modularity index (Q)), and landmark “hubs” by calculating betweenness centrality.

Results: JME participants exhibited significantly lower normalized CC compared to both unrelated controls (UC) and sibling controls (SC), however, they were similar to UC regarding normalized GE (Figure 1A, left). Q was lower in JME participants compared to both UC and SC groups (Figure 1A, right). Therefore, JME participants showed similar integration of information but with less segregation or marked subgroups (i.e., modules) within the network, indicative of less specialization in their cognitive processes. In contrast, SC has the highest normalized CC (global segregation) and Q (although not significantly different from UC when performing post-hoc group comparisons), separating them from JME more strongly than UC; however, they presented the lowest normalized GE (significantly different from both JME and UC when performing post-hoc p-values), indicating abnormal topological configuration. Regarding community structure (Figure 1B), UC and SC (both with 3 modules) showed cognitive landmarks with more apparent organization than JME (2 modules). In terms of hubs, UC presented 2, and SC and JME presented 3 hubs. Word knowledge (IQVOCS) was present in UC and JME, while delayed verbal memory (LLRNTDRS) was present in both UC and SC, and verbal working memory (NUMLETSS) was in common between SC and JME. Therefore, although there are some similarities between groups, there exist fundamental differences in the interplay between the various cognitive abilities.

Conclusions: In summary, significant differences exist in the inter-relationship between discrete cognitive abilities and the general configuration of cognitive networks in JME and unaffected siblings compared to unrelated controls. These differences in the configuration of cognitive landmarks could be contributing to the lower test scores observed in JME compared to unrelated controls (Table 1). Altogether, JME present a less complex and dysmature configuration of their cognitive landmarks compared to both UC and SC, while SC are also presenting significant discrepancies from UC.

Funding: NIH NINDS 5R01NS111022 (JME Connectome Project).

Behavior