Brain-specific immune hub activation in pediatric focal lesional epilepsy: a case-control study

Abstract There are still some gaps in understanding the role of inflammation in focal lesional epilepsy. Surgical specimens often contain activated leukocytes, usually thought to originate from the systemic circulation. However, the cerebrospinal fluid, meninges, and skull form a distinct brain-specific immune hub that also reacts to local signals. This study assessed immune hub activation adjacent to epileptogenic lesions in children with focal lesional epilepsy using 2-[18F]-fluoro-2-deoxy-D-glucose (FDG)-PET. Children with focal lesional epilepsy who underwent brain FDG-PET for presurgical assessment were compared to age- and sex-matched controls who had total-body FDG-PET for other clinical indications. Spherical regions of interest (ROIs) were placed on the skull, with a cubic ROI on the pons used as a reference. In epilepsy patients, skull ROIs were positioned adjacent to the perilesional hypometabolic defect and contralaterally. In non-epilepsy patients, ROIs mirrored those of epilepsy patients. Metabolic activity was measured as maximum and mean standardized uptake values (SUVmax and SUVmean). Corrected SUV (cSUV), normalized to the pons, and asymmetry index (AI), comparing lesional and contralateral ROIs, were calculated. Twenty-nine epilepsy and 29 non-epilepsy patients (16 boys, 55%) were included. The median age at the time of the scan was 9.0 years (IQR: 3.0-14.0). Across all patients, the median cSUVmax and cSUVmean were 0.36 (IQR: 0.27-0.48) and 0.23 (IQR: 0.16-0.32), respectively. In epilepsy patients, cSUVmax and cSUVmean were 0.33 (IQR: 0.26-0.40) and 0.23 (IQR: 0.18-0.31), and in non-epilepsy patients, 0.44 (IQR: 0.29-0.55) and 0.23 (IQR: 0.13-0.33). Neither cSUVmax nor cSUVmean differed by side (W=800, p: 0.84; W=682, p: 0.25), confirming comparability. However, in epilepsy patients, both cSUVmax and cSUVmean were higher on the lesional side than on the contralateral side (W=43 and W=49, p<0.001 for both). In non-epilepsy patients, mirrored ROIs showed no significant difference (W=142, p: 0.17; W=154, p: 0.27). AI values for SUVmax and SUVmean were higher in epilepsy than in non-epilepsy patients (t=-4.36 and t=-3.58, both p<0.001), a difference that remained significant after covariate adjustment, demonstrating metabolic asymmetry relative to the epileptogenic lesion.In children with focal lesional epilepsy, we observed increased metabolic activity in the brain-specific immune hub adjacent to the epileptogenic lesion. This local immune activation likely plays a role in the disease mechanism, which may clarify why immune-modulating treatments can be effective and point the way toward new therapeutic approaches.