Decreasing Shortest Path Length of the Sensorimotor Network Induces Frontal Glioma-Related Epilepsy

Shengyu Fang; Lianwang Li; Shimeng Weng; Yuhao Guo; Zhong Zhang; Lei Wang; Xing Fan; Yinyan Wang; Tao Jiang

doi:10.3389/fonc.2022.840871

Decreasing Shortest Path Length of the Sensorimotor Network Induces Frontal Glioma-Related Epilepsy

Front Oncol. 2022 Feb 16:12:840871. doi: 10.3389/fonc.2022.840871. eCollection 2022.

Authors

Shengyu Fang^{1

2}, Lianwang Li², Shimeng Weng², Yuhao Guo², Zhong Zhang¹, Lei Wang², Xing Fan², Yinyan Wang¹, Tao Jiang^{1

2

3}

Affiliations

¹ Department of Neurosurgery, Beijing Tiantan Hospital, Capital Medical University, Beijing, China.
² Beijing Neurosurgical Institute, Capital Medical University, Beijing, China.
³ Research Unit of Accurate Diagnosis, Treatment, and Translational Medicine of Brain Tumors, Chinese Academy of Medical Sciences, Beijing, China.

Abstract

Background: Glioma-related epilepsy (GRE) is a common symptom in patients with prefrontal glioma. Epilepsy onset is associated with functional network alterations. This study investigated alterations of functional networks in patients with prefrontal glioma and GRE.

Methods: Sixty-five patients with prefrontal lobe gliomas were retrospectively assessed and classified into GRE and non-GRE groups. Additionally, 25 healthy participants were enrolled after matching for general information. Imaging data were acquired within 72 h in pre-operation. The sensorimotor network was used to delineate alterations in functional connectivity (FC) and topological properties. One-way analysis of variance and post-hoc analysis with Bonferroni correction were used to calculate differences of FC and topological properties.

Results: All significant alterations were solely found in the sensorimotor network. Irrespective of gliomas located in the left or right prefrontal lobes, the edge between medial Brodmann area 6 and caudal ventrolateral Brodmann area 6 decreased FC in the GRE group compared with the non-GRE group [p < 0.0001 (left glioma), p = 0.0002 (right glioma)]. Moreover, the shortest path length decrease was found in the GRE group compared with the non-GRE group [p = 0.0292 (left glioma) and p = 0.0129 (right glioma)].

Conclusions: The reduction of FC between the medial BA 6 (supplementary motor area) and caudal ventrolateral BA 6 in the ipsilateral hemisphere and the shortening of the path length of the sensorimotor network were characteristics alterations in patients with GRE onset. These findings fill in the gap which is the relationship between GRE onset and the alterations of functional networks in patients with prefrontal glioma.

Significance statement: Glioma related epilepsy is the most common symptom of prefrontal glioma. It is important to identify characteristic alterations in functional networks in patients with GRE. We found that all significant alterations occurred in the sensorimotor network. Moreover, a decreased FC in the supplementary motor area and a shortening of the path's length are additional characteristics of glioma-related epilepsy. We believe that our findings indicate new directions of research that will contribute to future investigations of glioma-related epilepsy onset.

Keywords: functional network; glioma; glioma-related epilepsy; graph theory; human.