Network topology in brain tumor patients with and without structural epilepsy- a prospective MEG-study

Abstract

It has been proposed that functional connectivity (FC) and network topology (NT) are altered in patients with glial brain tumors. So far there is no consensus on the pattern of these changes, furthermore data on FC in patients with brain metastases (BM) as well as on the presence and absence of tumor-related epilepsy is lacking. We aimed to analyze preoperative NT of newly diagnosed, singular, supratentorial glial brain tumors (GBT) and BMs with and without structural epilepsy. FC and NT derived from resting state magnetoencephalography (MEG) were compared between patients (PAs) and matched healthy controls (HCs), between GBTs and BMs as well as between patients with and without structural epilepsy. We investigated all patients, who met our inclusion criteria from February 2019 to March 2021. Clinical data was collected from patients electronic medical charts. We analyzed whole brain (wb) connectivity in six frequency bands, calculated three different network topological parameters (node degree (ND), shortest path length (L), clustering coefficient (CC)) and performed a stratification, where differences in the power were to be found. For data analysis we used Fieldtrip, Brain Connectivity Matlab toolboxes and in-house built scripts. We consecutively included 41 patients (21 men), mean age 60.1ys (range 23- 82), who were operated on at our institution. Tumor histology included high-grade gliomas (n=18), low-grade gliomas (n=4), dysembryoplastic neuroepithelial tumor (DNET, n=1), BMs (n=14) and others (n=4). Statistical analysis revealed a significant decrease for wb ND in patients compared to healthy controls in every frequency range analyzed at the descriptive and corrected level (p1-30Hz=0.002, p(gamma)=0.002, p(beta)=0.002, p(alpha)=0.002, p(theta)=0.024, p(delta)=0.002). Furthermore, at the descriptive level, we found a significant augmentation for wb CC (p1-30Hz=0.031, p(delta)=0.013) in PAs compared to HCs, which did not persist the false discovery rate (FDR) correction. There were no differences in the networks of glial brain tumors and metastases identified. However, we found a significant increase in wb CC in patients with structural epilepsy (p(theta)= 0.048), and significantly lower wb ND (p(alpha)= 0.039) at the uncorrected level. Our data suggests that network topology is altered in brain tumor patients, which is in line with previous studies. Tumor histology per se might not influence the brain s functional network, however, tumor-related epilepsy seems to do so. Longitudinal studies and in-depth analysis of possible factors and confounders might be helpful to further substantiate these findings.