Abstract
Zhang S, Liu Y, Ma L, Yuan J, Wu C, Wang S. Epilepsia. 2025 Jan 13. doi: 10.1111/epi.18263. Epub ahead of print. PMID: 39804066. Objective: To evaluate iron deposition patterns in patients with cerebral cavernous malformation-related epilepsy (CRE) using quantitative susceptibility mapping (QSM) for detailed analysis of iron distribution associated with a history of epilepsy and severity. Methods: This study is part of the Quantitative Susceptibility Biomarker and Brain Structural Property for Cerebral Cavernous Malformation-Related Epilepsy (CRESS) cohort, a prospective multicenter study. QSM was used to quantify iron deposition in patients with sporadic cerebral cavernous malformation (CCMs). Lesions were segmented into intralesional, perilesional, and extralesional areas, with mean susceptibility values calculated for each subregion and analyzed in relation to epilepsy severity and duration. Results: Among the 46 patients studied, those with a history of epilepsy had significantly higher iron deposition values in the perilesional (P = .012) and extralesional areas (P = .01), as well as a greater extent of iron deposition (P < .001) compared to those without epilepsy. The extent of iron deposition effectively distinguished patients with and without epilepsy, with an area under the curve (AUC) of 0.901 (95% confidence interval [CI]: 0.816-0.985). Among patients with epilepsy, iron deposition in the extralesional area was positively correlated with the severity of epilepsy (r2 = 0.181, P = .043), and the extent of iron deposition was positively correlated with the duration of epilepsy (r2 = .214, P = .026). Significance: This study highlights QSM as a noninvasive tool for assessing iron deposition in CRE, identifying distinct subregional iron deposition patterns linked to epilepsy status and severity.
Commentary
Cerebral cavernous malformations (CCMs) are vascular abnormalities characterized by clusters of dilated capillaries (caverns) lined by an endothelial layer and filled with various stages of stagnant blood without intervening brain parenchyma within the lesion. 1 They affect approximately 1% of the population, with the majority (85%) being single, sporadic lesions caused by somatic mutations. 2 Multiple CCMs are often associated with familial and radiation-induced CCMs.1,2 In addition to the potential for intracranial hemorrhage, seizures are a common initial symptom of supratentorial CCMs, affecting around 50% of patients. 3 The 5-year risk of developing epilepsy after a first unprovoked seizure is 94%, with approximately 64% of patients progressing to drug-resistant epilepsy (DRE).3-5
CCM lesions contain hemosiderin, a byproduct of chronic and recurrent hemorrhage, which has been linked to epileptiform activity and seizures, resulting in signal decay and hypointensity on T2*-weighted MRI. 1 While susceptibility-weighted imaging offers higher sensitivity than conventional T2*-weighted MRI in detecting CCM lesions, it is a qualitative technique that tracks changes in lesion count over time but does not assess temporal changes in iron deposition. Quantitative susceptibility mapping (QSM) is an emerging MRI technique that estimates brain iron deposition by measuring local tissue magnetic susceptibility, showing potential as a monitoring biomarker for CCM-related epilepsy (CRE). 6 However, it remains unclear how QSM changes correlate with epilepsy status and severity in patients with CRE. To address this, Zang et al investigated the correlation between iron deposition and epilepsy in patients with sporadic CCMs using the QSM Biomarker and Brain Structural Property for the CRE cohort. 7 Unlike prior lesion-centered QSM analyses, 6 they segmented CCM lesions into intralesional, perilesional, and extralesional areas to assess localized iron deposition linked to epileptogenic potential. The study cohort included 46 patients: 23 with sporadic CCM-related epilepsy and 23 with sporadic CCM without epilepsy. 7
Three key findings emerged from this publication. 7 First, CCM patients with epilepsy showed elevated susceptibility values in the perilesional rim and extralesional deposition areas, with a larger extralesional deposition diameter compared to those without epilepsy. However, no correlation was found between intralesional deposition area, contralateral normal brain tissue susceptibility values, and epilepsy status. Second, receiver operating characteristic analysis revealed that the extralesional deposition diameter could serve as a reliable marker for identifying CCMs associated with prior epilepsy, with high sensitivity. Finally, positive correlations were observed between susceptibility values in the extralesional area and epilepsy severity, as well as between extralesional deposition diameter and the duration of epilepsy.
The findings of the current study have several clinical implications. First, QSM imaging could become an essential tool for assessing the extent of iron deposition in patients with CRE. Elevated susceptibility in the perilesional and extralesional areas could serve as a biomarker for identifying individuals at higher risk for epilepsy, facilitating early intervention and optimized management. This application of QSM is particularly valuable for patients without a confirmed epilepsy diagnosis or those with dual pathology. (eg, CCM and focal cortical dysplasia). Additionally, QSM, as a noninvasive diagnostic tool, has the potential to help clinicians monitor disease progression and treatment response, especially if novel drugs that prevent hemorrhage, remove hemosiderin, or act as antioxidants to reduce seizure burden become available.
Second, although the highlighted study focused on sporadic CCM cases with single lesions, QSM's ability to identify highly epileptogenic lesions could have significant implications for patients with multiple CCMs. It may help detect lesions with elevated iron deposition that carry a higher risk for seizures, allowing clinicians to monitor these lesions more closely or consider surgical intervention. This is particularly important in cases with multiple CCMs, where a single epileptogenic lesion can be identified through noninvasive video-EEG monitoring, which often predicts favorable surgical outcomes. 8
Third, the positive correlation between epilepsy severity and increased susceptibility in the extralesional area suggests that higher iron concentrations may exacerbate epilepsy severity. Besides, the positive correlation between the diameter of extralesional deposition and epilepsy duration may indicate that a longer disease course leads to the continued accumulation and spread of hemosiderin deposition over time. These findings support the consideration of early surgical intervention for some patients with highly epileptogenic CCMs, given the risk of bleeding and the negative correlation between epilepsy duration and postoperative seizure outcomes, even when the criteria for DRE have not been met. 8 Indeed, early resection of CCMs associated with seizures, compared to later resection, has been linked to better long-term seizure control in some series. 4
Finally, by segmenting CCM lesions into intralesional, perilesional, and extralesional zones, the authors were able to quantify iron levels more precisely, potentially explaining why not all patients with visible hemosiderin deposition develop epilepsy. The larger extralesional deposition diameter suggests a broader spread of hemosiderin into normal brain tissue, likely contributing to cortical irritability and seizures. This supports the idea that seizures originate not from the CCM itself (which lacks neurons), but from the surrounding gliotic cortical tissue, often stained by hemosiderin. 8 This observation may justify the resection of the surrounding gliotic, hemosiderin-stained cortical tissue when feasible, offering a curative surgical approach for CRE patients with DRE.3,4
Despite its promise, the study has notable limitations. QSM, an advanced MRI technique, is often limited in availability and clinical implementation. Although the study shows a correlation between iron deposition and seizures, causality remains unproven, as hemosiderin deposits may simply reflect prior damage rather than being a primary contributor to epileptogenesis. Other factors, such as neurovascular coupling, inflammation, and gliosis, may significantly drive epileptogenesis in CRE.1,8 Future longitudinal studies with repeated QSM assessments could offer deeper insights into the dynamics of iron accumulation and its role in epileptogenesis. Crucially, while elevated susceptibility values were observed in CRE patients, some nonepileptic patients also exhibited overlapping values, complicating predictions for individual cases. Additionally, the study's use of the LSSS as an epilepsy severity scale is problematic, as it struggles to accurately assess the impact of seizures on quality of life. 9 Furthermore, while QSM was performed at 3 T in this study, phase-based measurements like QSM benefit from higher contrast and resolution at 7 T, making this study a valuable step toward exploring QSM's potential at higher field strengths. Finally, current QSM methods, including those used in this study, face challenges, such as the need for standardized processing algorithms to ensure consistent results across studies and integration with other diagnostic tools to enhance diagnostic and predictive capabilities. 10
The article reviewed in this commentary deepens our understanding of iron deposition in CCMs, highlighting QSM's potential for mapping iron accumulation, tracking disease progression, and identifying individuals at risk for CRE due to sporadic CCMs. QSM's ability to visualize in vivo changes in iron deposition could also inspire future clinical trials in CRE aimed at preventing CCM lesion development and hemosiderin deposition. However, addressing technical challenges, such as the need for standardized processing algorithms and validating the technique through larger prospective studies, is essential before QSM can be integrated into clinical practice for CRE.
