Amygdala Subnuclear Correlates of Neuroinflammation in Humans with Epilepsy Using [18F]FEPPA PET/MR
Abstract number :
2.307
Submission category :
5. Neuro Imaging / 5A. Structural Imaging
Year :
2024
Submission ID :
385
Source :
www.aesnet.org
Presentation date :
12/8/2024 12:00:00 AM
Published date :
Authors :
Presenting Author: Theodore Imhoff-Smith, BA, MCS – University of Wisconsin-Madison
Mariel Kalkach Aparicio, MD, MBE – University of Wisconsin-Madison
Brinda Sevak, BS – UC Davis
Rosario Ciliento, MD – University of Wisconsin-Madison
Bruce Hermann, PhD – University of Wisconsin
Veena Nair, PhD – University of Wisconsin-Madison
Vivek Prabhakaran, MD PhD – University of Wisconsin-Madison
Samuel Hurley, PhD – University of Wisconsin-Madison
Alan McMillan, PhD – University of Wisconsin-Madison
Nagesh Adluru, PhD – University of Wisconsin-Madison
Aaron Struck, MD – University of Wisconsin-Madison
Rationale: Studies across the epilepsies reveal widespread reductions in gray matter volume, including amygdala. By contrast, amygdala enlargement (AE) has been described as a subtype of temporal lobe epilepsy (TLE) in patients without hippocampal sclerosis (HS). In parallel, epilepsy is characterized by widespread neuroinflammation, a potential causative factor and consequence of seizures. Translocator Protein (TSPO), which predominantly reflects neuroinflammation through microglial activation, can be measured noninvasively by PET neuroimaging. In this study, we test group differences in amygdala subnuclear volumes and employ [18F]FEPPA PET/MR, a third generation TSPO radioligand, to test the effect of neuroinflammation on amygdala substructures in patients with AE.
Methods: Clinical data, T1-weighted MR, and PET images were acquired from 15 patients (PT; age=35+/-12.4, 5F, 11 without HS) and 12 typical controls (TC; age=39.5+/-12.0, 6F) on a GE HealthCare Signa 3T PET/MR system. Total brain volume (TBV) and amygdala subnuclear volumes were extracted using FreeSurfer. Subnuclei were combined into 3 regions of interest (ROIs) and normalized by TBV: basolateral (BLA), centromedial (CM), and cortical (Co). PETSurfer was used to derive anatomical parcellations of PET images using the Desikan Killiany atlas for ROI analysis. FEPPA binding was quantified with standard uptake values (SUV) for partial volume- and motion- corrected static PET images reconstructed between 50-70 minutes post injection. Participants were classified as high, mixed, or low-affinity binders based on TSPO polymorphism (rs6971). Analysis excluded participants with a low allele (n=2) or missing imaging data (n=4), resulting in 12 PT (age=38.3+/-13.1, 8 high allele, 4F) and 9 TC (age=35.2+/-10.2; 5 high allele, 4F). Group differences in each ROI volume and their interaction with amygdala SUV were tested. Analyses used ROIs and SUVs from the ipsilateral side of seizure or the bilateral mean (for TC and PT with bilateral seizures). Models were run with and without n=3 PT who exhibited HS. All models included covariates for age and sex, and allele status (for analyses including PET). Outliers were excluded based on Cook’s D > 4/(N-P).
Results: Results indicated a significant interaction effect such that increased ipsilateral amygdala SUV (increased microglial activation) was associated with a greater reduction in amygdala volume in PT (t(12)=-2.2, p=.048) and in no HS (t(10)=-2.6,p=.027) compared to TC. The interaction was also significant for BLA (t(10)=-2.47, p=.03), CM (t(10)=-2.77, p=.02), and Co (t(10)=-2.6, p=.03) in no HS compared to TC. Greater volume was revealed specifically in CM of no HS compared to TC (t(10)=2.76, p=.02; Fig. 1).
Conclusions: Findings demonstrate that increased microglial activation is associated with reduced amygdala volume in patients with AE. These results underscore the potential role of neuroinflammation in structural alterations of the amygdala in epilepsy, particularly in the centromedial subnuclei.
Funding: T32NS105602, R01NS111022, R01NS117568, R01NS123378, P50HD105353
Neuro Imaging