Introduction: Brain arteriovenous malformations (bAVMs) are a major cause of morbidity in pediatric patients, with seizures as the second most common presentation. The mechanisms underlying seizures in this population remain poorly understood. Circulating plasma protein biomarkers may reflect neuroinflammatory or neurovascular changes associated with seizures. We hypothesized that plasma levels of inflammatory biomarkers differ between bAVM patients with and without seizures. Methods: Blood samples were collected from 155 enrolled bAVM study participants (<18 years) at multiple care visits at our institution between 2005–2022. We used the Olink Explore Inflammation I panel to assess circulating plasma levels of 384 proteins in 244 blood samples. Normalized protein levels were log 2 -transformed prior to analysis. Seizure was defined as a documented history of at least one clinical seizure (with or without associated hemorrhage) prior to blood sample collection. Multivariable linear regression analysis tested for the association of plasma biomarker levels with risk of seizure, adjusting for age, sex, batch effect, and prior intracranial hemorrhage. We applied cluster robust standard errors to account for within-person correlation and false discovery rate (FDR) correction to address multiple comparisons. Exponentiated coefficients are reported as Proportional Increase (PI). Results: Our cohort of 155 bAVM cases (51% male, 49% female, median age 13.3 years) included 37 (24%) cases who had seizure(s) prior to blood collection (median 84 days, range 0-1733 days). A total of 368 proteins were successfully measured. Two biomarkers were significantly associated with seizures at FDR p<0.10: NFASC, neurofascin (PI=1.20, 95% CI: 0.15-0.38, FDR p =0.0033) and HLA-DRA, HLA class II histocompatibility antigen, DR alpha chain (PI=1.23, 95% CI: 0.13-0.47, FDR p=0.0797). NFASC is a neurodevelopmental-associated epilepsy gene, while HLA-DRA expression has been associated with seizure duration and frequency. Conclusion: These findings highlight inflammatory and immune-related biomarkers associated with seizures in pediatric bAVM. Validation in external cohorts is warranted to assess their potential for seizure risk stratification and targeted interventions.
Nguyen et al. (2026) studied this question.