Neurophysiological and neuroimaging markers of neurological outcome in Tuberous Sclerosis Complex

Tuberous sclerosis complex (TSC) is a rare genetic, multisystem disorder, characterized by hamartoma formation in various organs, including the brain, where they are referred to as tubers. The neurologic presentation may include epilepsy, intellectual disability (ID) and autism spectrum disorder (ASD), but the phenotypical spectrum is wide and unpredictable. Genetic markers are insufficiently accurate for prognostication. Thus, early markers of neurological outcome in TSC are needed to guide early therapeutic decisions, with the potential to improve outcome. In addition, targeted experimental treatments, for example with mTOR inhibitors, may have disease-modifying effects, and trials could be justified in patients at high risk for adverse neurological outcome. This thesis explores neurophysiological (electroencephalography, EEG) and neuroimaging (Diffusion Tensor Imaging, DTI) markers of neurological outcome. The first part of the thesis reviews molecular mechanism underlying TSC, and how aberrant neural connectivity can lead to ASD symptoms. Next, DTI is explained to the non-radiologist, and how it can noninvasively characterize microstructural properties of the brain. The limitations of the technique are highlighted, and novel diffusion models are discussed. The putative role of DTI as a neurological biomarker in TSC is reviewed. The second part of the thesis discusses three papers which investigate whether the white matter DTI measures correlate cross-sectionally with overall neurological outcome in TSC, and with ASD in particular. With regards to DTI and epilepsy in TSC, next the foundation is laid for study of the development of seizures (“epileptogenesis”) by exploring the longitudinal evolution of DTI measures in various neuroanatomical areas in TSC. We investigate the continuum of neuropathology from tuber to perituber tissue, extending also to deeper white matter. In addition, we quantify and model the longitudinal diffusion changes of these tissue types. The findings are discussed in the context of apparent controversies in the literature on tuber vs. perituber onset of seizures. The third part of the thesis studies how conventional and computational EEG can be used in the prediction of neurological outcome in TSC. First, early results of a prospective observational study of early EEG markers for epilepsy in TSC are presented. With conventional EEG interpretation, epileptiform EEG abnormalities precede the onset of clinical seizures in TSC, and serial routine clinical EEG is may be a feasible strategy for early detection of epilepsy in TSC. After studying the aforementioned disconnection model of ASD in patients with TSC structurally with DTI in part two of the thesis, we study it functionally in the third part. First, through analysis of EEG coherence measures. We found an increased short-range connectivity at the cost of long-range connectivity, in patients with ASD – regardless of the etiology. Important limitations of the interpretation of EEG connectivity measures in the context of the broad, non-converging literature on the neurobiology and neurophysiology of ASD are discussed. Next, we apply graph theoretical network analysis to probe the design and performance of the functional EEG network as a whole. In conclusion, both DTI and EEG measures are feasible and biologically relevant markers of neurological outcome in TSC. Further validation through prospective studies is currently underway.