Introduction

Alteration of structural-functional coupling in the connectivity and network topology is associated with the functional impairment in diseased brain.^1,2 However, alterations of the functional-structural coupling in brains with glioma remain less investigated. In this study, we aimed to explore the structure-function coupling in the connectivity and network topology in brains with glioma by quantifying intermodal correlation in the contralesional hemisphere using the diffusion tract imaging (DTI) and resting-state functional MRI (rs-fMRI) data, and contemplate its association with the malignancy grades of gliomas (Fig. 1).

Subjects and Methods

This study was approved by local institutional review board. A total of 41 subjects with histologically confirmed glioma in the left hemisphere were consecutively included. Out of the 41 tumors, 13 tumors were classified as WHO II ( 8 females, age = 36.80 ± 14.20 years), 10 as WHO III ( 5 females, age = 37.88 ± 12.86 years), and 18 as WHO IV ( 7 females, age = 49.19 ± 11.73 years). All MRI scans were performed on a 3.0T magnetic scanner (Siemens Verio, Germany) with a 12 channel phased array head coil. DTI was obtained using a echo-planar imaging (EPI) sequence with TR/TE 7,600/91 ms, FOV 230mm × 230 mm, matrix size 128 × 128, slice thickness 3 mm, 20 diffusion gradient encoding directions at b value of 1,000s/mm2. Resting-state BOLD fMRI was acquired using a gradient-echo EPI sequence with TR/TE 2,000/35ms, FOV 210mm × 210mm, matrix 64 × 64, 33 slices with thickness of 4.0mm, 240 volumes. In addition, diffusion and function MRI data of 18 healthy subjects was acquired as controls (HC, 9 females, aged 50.78 ± 6.65 years). PANDA was utilized to preprocess DTI images, conduct the probabilistic fiber tracking, and construct the structural network. The brain was segmented into 392 regions of interest (ROIs) based on a fine-grained atlas (CC400).³ Only the 162 ROIs in the contralesional hemisphere were taken into account. The normalized structural connectivity (SC) between ROIs was defined based on the following fomula²: $$$SC_{ij}=\frac{log(p_{ij})-min[log(p_{mn})]}{max[log(p_{mn})]-min[log(p_{mn})]}$$$, where P_ij represents the connection probability between ROI-i and ROI-j, 1 ≤ m ≠ n ≤ N, and N = 162. DPABI was used to preprocess fMRI data and calculate the pairwise functional connectivity (FC) which was defined as the Pearson’s correlation coefficient between the contralesional ROIs with Fisher’s Z transformation. Then, the integrals of the global (Eg) and local efficiency (Eloc)⁴ across a wide sparsity range (0.1 ≤ sparsity ≤ 0.34, interval = 0.01) were used to quantify the integration and segregation capability of the unsigned structural (SN) and functional networks (FN). Subsequently, the Spearman-rank correlation between non-zero SCs and corresponding FCs was calculated at the individual level. The effects of left-brain gliomas on the contralesional SC-FC coupling and network topological features were estimated with the one-way ANOVA analysis. The significance level was set at P < 0.05 with false discovery rate correction.

Results

Enhanced SC-FC coupling and network topology were identified in the contralesional hemisphere. The contralesional SC-FC coupling was significantly increased in brains with left-brain WHO III and IV gliomas, as compared to HCs (Fig. 2) (P < 0.05). Brains with WHO III showed significantly higher SC-FC coupling than those with WHO II (P = 0.03). Higher global and local efficiency of SNs was identified for all patients (P < 0.05) (Fig. 3A-B), but only in FNs for patients with WHO II and III, as compared to HCs (P < 0.05) (Fig. 3C-D).

Discussion and Conclusions

Increased structural and functional coupling was associated with deteriorated cognitive impairment.² The enhanced SC-FC coupling of the contralesional hemisphere in this study suggests a functional dysregulation of the brain in addition to the structural damage as a result of tumor growth, emphasizing the rationality to take the SC-FC coupling into account for the investigation of plastic mechanism in brains with glioma. Increased structural and functional network topological efficiency in patients with low-grade gliomas may indicate an intrinsic mechanism featuring enhanced funcitonal integration and segregation to compensate for the tissue loss and functional disturbance in the tumor hemisphere.⁵ As compared with the structural plasticity, the functional remodeling was limited for patients with high-grade glioma, suggesting the different plasticity potential of structural and functional networks.¹ The diversity in the structural-funcitonal coupling provide new insights into the plastic remodling of diseased brain, and may serve as a new biomarker for better characterization of glioma.

Acknowledgements

This study was partially supported by 2021A1515010193, NSFC81627901 and 2020B1212060051.