Rationale:
Neuroimaging studies have characterized temporal lobe epilepsy (TLE) as widespread gray matter atrophy rather than being confined to the medial temporal lobe and this atrophy pattern progresses over time. However, few studies have explored the underlying pathophysiological processes and whether the atrophy pattern varies among individuals. This study aims to apply a network-based atrophy model in presurgical evaluated TLE patients to identify putative epicenters of gray matter atrophy and to examine the relationship between different epicenter patterns and clinical features.
Methods:
We recruited 126 patients with unilateral TLE who underwent comprehensive presurgical evaluation, along with 60 age- and gender- matched healthy controls (HC). Gray matter volume (GMV) alterations were evaluated between patients and HC using linear regression model. Group-level epicenters were identified by combining TLE related GMV alterations and normative functional network through data-driven approach and simulated approach that models the spread of atrophy. Additionally, we applied k-means clustering algorithm to identify homogeneous epicenter subtypes at the individual-level and investigated differences in surgical outcomes and clinical characteristics across these subtypes.
Results:
The most prominent atrophy was observed in the ipsilateral hippocampus, thalamus, temporal lobe, and parahippocampal cortex in TLE patients. We further identified DMN related cortices as well as bilateral somatomotor cortex and ipsilateral thalamus as optimal putative epicenters of atrophy at the group level. The k-means clustering analysis identified two distinct subtypes: Subtype 1, with epicenter in the bilateral somatomotor cortex and right thalamus, and subtype 2, with epicenters in the bilateral lateral and medial temporal lobe and thalamus. Kaplan-Meier analysis showed a significantly higher seizure-free rate for subtype 1 compared to Subtype 2 (p = 0.05). Additionally, subtype 1 experienced more frequent seizure events (p = 0.008).
Conclusions:
Our findings suggest that gray matter atrophy in TLE may originate from specific epicenters and is constrained by brain network architecture. Individualized epicenter modeling reveals significant inter-individual variability in gray matter atrophy and could be instrumental for personalized prognostics and targeted therapeutic approaches.
Funding:
This study was supported by the National Natural Science Foundation of China (NSFC Grants No. 82171443, 81771402,82471482) and Natural Science Foundation of Sichuan Province (Grants No. 2022NSFSC1483).