# Systemic T Cell Receptor Profiling Reveals Adaptive Immune Activation and Potential Immune Signatures of Diagnosis and Brain Atrophy in Epilepsy

## TL;DR

This study shows that immune system changes in the blood, measured through T cell receptors, are linked to epilepsy severity and brain shrinkage, offering potential for non-invasive diagnosis and treatment guidance.

## Contribution

The study introduces peripheral TCR repertoire profiling as a novel immune signature for epilepsy diagnosis and monitoring.

## Key findings

- Epilepsy patients, especially those with drug-resistant or neuroinflammation-associated epilepsy, show reduced TCR diversity and distinct V/J gene usage.
- Machine learning models using TCR features can distinguish epilepsy patients from controls with 80% accuracy.
- Brain atrophy in regions like the thalamus correlates with TCR repertoire changes and specific V/J gene patterns.

## Abstract

Epilepsy is increasingly associated with immune dysregulation and inflammation. The T cell receptor (TCR), a key mediator of adaptive immunity, shows repertoire alterations in various immune‐mediated diseases. The unique TCR sequence serves as a molecular barcode for T cells, and clonal expansion accompanied by reduced overall TCR repertoire diversity reflects adaptive immune activation. We investigated peripheral TCR repertoire changes in epilepsy and their association with disease severity and brain atrophy.

We profiled TCR α/β chain repertoires from peripheral blood mononuclear cells of 100 individuals, including 45 patients with epilepsy (14 with well‐controlled epilepsy, 22 with drug‐resistant epilepsy [DRE], and 9 with neuroinflammation‐associated epilepsy [NIE]) and 55 unmatched healthy controls. NIE included new‐onset epilepsy following possible autoimmune or infectious neuroinflammation. We comprehensively evaluated clonotype distribution, diversity, interindividual sharing, and V/J gene usage. Machine learning models evaluated the diagnostic potential of TCR repertoire features. Brain volumes were measured by MRI and correlated with TCR repertoire characteristics.

Patients with epilepsy showed significantly reduced TCR diversity, particularly in DRE or NIE. They also showed distinct patterns of V and J gene usage and decreased interindividual sharing of epilepsy‐associated clonotypes. Machine learning models incorporating V/J usage and public clonotypes distinguished patients with epilepsy from controls with a mean classification accuracy of 0.80 (95% bias‐corrected and accelerated bootstrap confidence interval (BCa CI), 0.69–0.86) and the area under the curve of 0.80 (95% BCa CI, 0.70–0.87). TCR diversity correlated with seizure frequency among patients without daily seizures or clinical evidence of neuroinflammation. Brain atrophy, notably in the thalamus and basal ganglia, was also associated with TCR repertoire alterations and specific V/J gene usage patterns.

Peripheral TCR repertoire profiling reveals that systemic immune dysregulation is present in epilepsy and is associated with neurodegeneration. Our findings highlight the peripheral TCR repertoire as a disease‐relevant immune signature with the potential to non‐invasively interrogate epilepsy status and guide therapeutic interventions.