Network-Imaging in Genetic Epilepsies

大约30%的癫痫患者有可疑或已证实的主要遗传病因。在罕见的所谓单基因癫痫中，可以确定原因突变，这为详细研究从遗传/分子基础到临床表型的病理生理学级联提供了独特的机会。对于这些突变中的几个，神经生理学缺陷已经被阐明，即SCN1A功能丧失突变的抑制性中间神经元钠电流减少和STX1B突触前递质释放缺陷。然而，这些分子变化对体内大规模人类神经网络的影响尚未被研究。在一项小型的先导性研究中，我们可以显示带有SCN1A功能缺失突变的受试者的结构和功能网络的差异。与这些单基因形式的癫痫相比，最常见的遗传性癫痫，即所谓的遗传性全身性癫痫(IGE/GGE)的遗传学基础鲜为人知。然而，基于自己的结果和最近发表的研究，网络成像仍然可以检测到IGE/GGE的结构和功能差异。在这项建议中，我们将评估常见的IGE/GGE以及两个临床相似的单基因癫痫及其兄弟姐妹的网络成像，以首次深入了解不同形式的遗传性癫痫中人类大脑的大规模网络变化以及已知特定基因突变的后果。为此，我们将在50名确诊的单基因癫痫患者中获得一个全面的功能和结构网络成像范例：即SCN1A和STX1B功能丧失，每个患者都有不同的、明确的病理生理机制。两者都与发热相关癫痫综合征(GEFS+)有关，包括未受影响的突变携带者。我们还将从我们的临床和科学数据库中分析60名IGE/GGE患者和30名未受影响的同胞，并从普通人群和家庭内对照中获得匹配的对照数据集。我们将使用图论分析来生成功能和结构网络结构的客观度量，微观结构完整性的扩散张量成像度量，以及基于工作记忆任务的fMRI/MEG。我们将使用包括临床协变量在内的组比较来区分潜在遗传原因的影响和癫痫的影响。临床严重程度将按顺序考虑，从未受影响的基因携带者到严重癫痫。作为下一步，我们将使用机器学习来测试是否可以在组内和跨组比较(单基因癫痫、Ige/GGE和兄弟姐妹)中识别泛发性遗传性癫痫的网络模式。这可以使基于成像的IGE/GGE分层成为可能。

Approximately 30% of epilepsies have a suspected or proven primarily genetic etiology. In rare, so called monogenic epilepsies a causal mutation can be identified, offering the unique opportunity to study the pathophysiological cascade from the genetic/molecular basis to the clinical phenotype in detail. For several of these mutations, the neurophysiological deficits could already be clarified, i.e. a reduction of sodium currents in inhibitory interneurons for SCN1A loss-of-function mutations and a defect of pre-synaptic transmitter release for STX1B. However, the consequences of these molecular alterations on large-scale human neuronal networks in-vivo have not been investigated. In a small pilot study, we could show structural and functional network differences in subjects with SCN1A loss-of-function mutations. In contrast to these monogenic forms of epilepsy, the genetic underpinnings of the most common genetic epilepsies, the so called genetic generalized epilepsy (IGE/GGE), are poorly understood. However, based on own results and recently published studies, network imaging can nonetheless detect structural and functional differences in IGE/GGE. Within this proposal, we will evaluate network imaging in common IGE/GGE as well as in two clinically similar monogenic epilepsies and their siblings to get a first insight on large-scale network alterations in human brains in different forms of genetic epilepsies and the consequences of known specific genetic mutations. To this end, we will acquire a comprehensive paradigm of functional and structural network imaging in a cohort of 50 subjects with established, monogenic epilepsy: namely SCN1A and STX1B loss-of function, each with a different, well-defined pathophysiological mechanism. Both are associated with a fever-associated epilepsy syndrome (GEFS+) including un-affected mutation carriers. We will also analyze 60 IGE/GGE patients and 30 unaffected siblings from our clinical and scientific database and acquire a matched control data set from the general population as well as intrafamilial controls. We will use graph theory analysis to generate objective measures of functional and structural network architecture, diffusion-tensor imaging measures of micro-structural integrity and working-memory task-based fMRI/MEG. We will use group comparisons including clinical covariables to distinguish effects of the underlying genetic causes from those of the epilepsy. Clinical severity will be taken into account in an ordinal scale from unaffected gene carriers to severe epilepsy. As next step, we will use machine learning to test whether network patterns for generalized genetic epilepsies can be identified in within- and across group comparisons (monogenic epilepsy, IGE/GGE and siblings). This could enable imaging-based sub-stratification of IGE/GGE.