Lighting up the Lesion: TSPO-PET Advances the Hunt for Hidden Cortical Dysplasia

Abstract

Emerging translocator protein–positron emission tomographic imaging improves detection of focal cortical dysplasia

Zichen Qiao, Yingxue Yang, Yuanhong Chen, Linai Guo, Qing Xue, Lehong Gao, Jia Chen, Bixiao Cui, Jinghui Liu, Haoxun Yang, Xueyuan Wang, Yihe Wang, Huaqiang Zhang, Cuiping Xu, Yicong Lin, Tao Yu, Yuping Wang, Yueshan Piao, Jie Lu, Liankun Ren

Objective: The identification of epileptic lesions is crucial for improving surgical outcomes. Nevertheless, substantial focal cortical dysplasia (FCD) may be invisible on magnetic resonance imaging (MRI). We aimed to characterize the expression pattern of 18-kDa translocator protein (TSPO) in FCD and to evaluate the effectiveness of this inflammation-reflective molecular imaging technique for detecting FCD. Methods: Patients clinically diagnosed with FCD, based on clinical features, interictal electroencephalographic (EEG) findings, and MRI characteristics, underwent positron emission tomography (PET) imaging using ¹⁸F-DPA714 and ¹⁸F-fluorodeoxyglucose (FDG) tracers. TSPO-PET activation patterns were qualitatively evaluated. Semiquantitative analysis using the highlight index (HI) was further performed to investigate its correlation with clinical characteristics. For patients who underwent stereo-EEG (SEEG) monitoring, the site of high-level TSPO-PET activation was compared with the seizure onset zone identified by SEEG. For patients who underwent resection surgery, the relationship between TSPO-PET uptake and histopathological findings was studied. Results: Twenty-four patients were enrolled. Three groups were identified: MRI-positive with visible high-level TSPO-PET activation (6 patients), MRI-negative with visible high-level TSPO-PET activation (13 patients), and MRI-positive with invisible low-level TSPO-PET activation (5 patients). Regions of high-level TSPO-PET activation showed concordance with ictal discharges in five patients who underwent SEEG. Compared with FDG-PET, TSPO-PET exhibited a more prominent signal against the background (p = .0158). HI was correlated with seizure frequency (p = .0362) and the occurrence of focal to bilateral tonic–clonic seizures (p = .0294), and shorter interval between the TSPO-PET scan and the last seizure was associated with higher TSPO-PET HI (R = −.4323, p = .0349). Postoperative histopathological examination confirmed high-level TSPO-PET activation rates of 3/3 for FCD type IIb and 1/3 for FCD type IIa. Significance: TSPO-PET activation patterns offer clinical significance for improving surgical outcomes by enhancing FCD detection during presurgical evaluation. Also, our observations offer new insights into the histopathological basis for increased TSPO uptake in humans.

Positron emission tomography (PET) is a metabolic imaging modality that has markedly advanced our ability to identify targets for invasive electroencephalography (EEG) and guide tailored resections in patients with drug-resistant epilepsy. Previously, 2-deoxy-2-[¹⁸F]fluoro-β-D-glucose ([¹⁸F]FDG), a glucose analog, was the only PET tracer used to assess abnormal metabolic activity during presurgical epilepsy evaluations. While interictal [¹⁸F] FDG-PET demonstrated a sensitivity approaching 74% to 90% for localizing epileptogenic zones (EZ) in the mesial temporal lobe,^1,2 its sensitivity was much lower for other brain regions or in cases with magnetic resonance imaging (MRI)-negative lesions, ² thereby motivating the development of novel imaging agents. After nearly 3 decades of development and screening over 50 candidate compounds, [¹⁸F]DPA-714 (N, N-diethyl-2-(2-(4-(2-[¹⁸F]fluoroethoxy)phenyl)-5,7-dimethylpyrazolo[1,5-a]pyrimidin-3-yl) acetamide) has emerged as a promising second-generation PET tracer. ³ It binds selectively to the 18 kDa translocator protein (TSPO), which is markedly upregulated in activated microglia and astrocytes, key cellular mediators of neuroinflammation. ⁴ Due to its high binding specificity and enhanced brain penetration, [¹⁸F]DPA-714 enables noninvasive monitoring of regions with elevated TSPO expression, offering a putative marker of active neuroinflammation. ⁴ Recent studies have demonstrated that [¹⁸F]DPA-714 PET outperforms [¹⁸F]FDG-PET in detecting metabolic anomalies and refining localization of seizure onset zone, particularly in patients with MRI-negative epilepsy or neocortical foci. ⁵ These findings support the expanding use of TSPO-targeted imaging for investigating the inflammatory mechanisms underlying epilepsy, including the search for cortical dysplasia, which often remains undetectable with conventional imaging techniques.

Focal cortical dysplasia (FCD) represents a cytoarchitectural malformation of the cerebral cortex that is strongly associated with seizure refractoriness. Prominent activation of major inflammatory pathways has been demonstrated in resected FCD tissue ⁶ ; however, the relationship between local immune activation and clinical or electrographic epileptiform activity generated by dysplastic tissue remains poorly understood. A recent study by Qiao et al made important initial strides in addressing this gap. ⁷ The authors prospectively enrolled adult patients with drug-resistant focal epilepsy and FCD, defined based on clinical characteristics, MRI criteria, or scalp EEG biomarkers previously associated with cortical dysplasia. Patients underwent sequential PET imaging with [¹⁸F]FDG and [¹⁸F]DPA-714 tracers on separate days, enabling control of key clinical and technical variables. MRI, FDG-PET, and TSPO-PET were interpreted both visually and semiquantitatively using the highlight index (HI), which measures the relative difference in tracer uptake between the presumed FCD region and whole-brain gray matter. In a subset of patients, seizure onset was confirmed using stereo-EEG (sEEG). Cellular markers of FCD and local inflammation were assessed in resected tissue specimens from patients referred for epilepsy surgery.

The study found that TSPO-PET was highly effective in detecting presumed FCD lesions, identifying abnormal activation in an additional 25% of patients compared to FDG-PET alone. Moreover, TSPO-PET revealed a high HI in more than half of the patients whose FDG-PET results were inconclusive. Notably, there was a strong concordance between regions of elevated TSPO uptake and the ictal onset zone in patients who underwent sEEG, supporting the potential of this imaging modality to aid in localizing the EZ. The authors also compared PET activation patterns between patients with identifiable MRI features of FCD and those without. TSPO-PET detected lesions in over 80% of MRI-negative presumed FCD cases. Collectively, these findings demonstrate that incorporating TSPO-PET into presurgical assessments can significantly enhance the localization of FCD lesions, especially when MRI and FDG-PET are nondiagnostic.

The investigators next examined the relationship between TSPO-PET activation patterns and key clinical characteristics. A significant positive correlation emerged between high HI patterns and markers of seizure severity, including high seizure frequency and shorter latency to the last seizure. Conversely, there was a negative correlation between HI and the occurrence of focal to bilateral tonic–clonic seizures. These observations are remarkable and suggest that TSPO-PET may be particularly useful for patients with active daily or weekly seizures and should ideally be administered shortly after ictal events. While high activation was the most common TSPO-PET pattern, some patients exhibited low-level activation in presumed FCD regions. This may reflect a distinct FCD subtype with different underlying pathophysiology or result from remote seizure activity or the effects of antiseizure medications (eg, benzodiazepines and valproate). The affinity of TSPO ligands varies significantly in healthy individuals with single-nucleotide polymorphism in the TSPO gene (rs6971). ⁸ This could have resulted in low uptake activation in some epilepsy patients in this study. Future studies are needed to determine whether low-activation TSPO-PET patterns exhibit specific spatial distributions or correlate with distinct clinical, genetic, or histopathological features.

The most compelling aspect of the study was the histopathological analysis of lesions corresponding to increased TSPO-PET signal. The qualitative assessment of these specimens revealed dysmorphic neurons and balloon cells surrounded or infiltrated by microglia. The authors also explored differences in TSPO activation patterns between patients with FCD Type IIa and Type IIb, though the small sample size precluded definitive conclusions. Overall, the study supports a link between frequent seizures and local increases in TSPO-PET uptake, indicating the presence of an inflammatory milieu in FCD lesions. Nevertheless, future studies incorporating quantitative analyses of tissue markers are needed to clarify the relationship between focal inflammation in different types of FCD and TSPO-PET activation.

Despite these impactful findings, the study had limitations. First, histopathological confirmation of FCD was not feasible for all regions showing abnormal TSPO activation. Additionally, reliance on scalp EEG biomarkers for lesion localization, rather than more definitive methods, weakens the anatomical correlations. Studies allowing the alignment between TSPO-PET abnormalities and histopathological findings in resected specimens would greatly enhance our understanding of the role of inflammation in FCD. Such insights could also be critical in decisions regarding the use of TSPO-PET in presurgical evaluations.