Object

Multicentric gliomas are uncommon lesions of the central nervous system. Till now, the evolution of multicentric gliomas is poorly understood. Due to the white matter infiltration of glioma, it is hard to exclude the existence of a connection in cases deemed to be multicentric. As yet there remains lack of methods to elucidate the lesions are totally separated unless surgical biopsy. A non-invasive method involving a probabilistic approach may help us to identify the white matter dissemination routes which can provide some novel insights into multicentric gliomas. In the current study, we use probabilistic fiber tracking to depict pathway and quantify the probability connection of two lesions in patients with multicentric glioma and patients with diffuse glioma.

Method

Study type; Retrospective. Subjects; twenty patients with multicentric gliomas and 21 patients with diffuse gliomas were included in our study based on the criteria suggested by previous study [1]. Acquisition of neuroimaging data; All patients underwent a diffusion tensor imaging (DTI) sequence and enhanced 3D T1-weighted MRI preoperatively (Siemens skyra 3.0T, 30 diffusion directions , b = 1000 s/mm2 , 30 slices , thickness 3 mm), High-resolution contrast-enhanced T1 weighted images are acquired with a 3D MPRAGE sequence (TR/TE/ Flip angle =1540 ms/2.4 ms/8°, matrix =256 ×256, 1 mm thickness). Data processing; Firstly, the lesions served as the seeds and target for probabilistic tractography were segmented using ITK-SNAP software (www.itksnap.org) from each patient. Secondly, the DTI images for all patients were pre-processed using the FSL 5.0.9 Diffusion Toolbox (FDT) (http://fsl.fmrib.ox.ac.uk/fsl/fslwiki/FDT). Diffusion parameters at each voxel in the whole brain were first estimated using latest version of BEDPOST. The resulting distributions were then used for probabilistic fiber tracking using the latest version of PROBTRACK. The algorithm evaluates connectivity values between the seed and the target masks defined a priori. Finally, the probabilistic value and fractional anisotropy (FA) between lesions was extracted from the probabilistic fiber tracking(fig.1). Statistical analyses; Intergroup comparison of clinical characteristics was performed with T-statistics and Chi square test. The connectivity value and the probabilistic value were assessed using the Mann-Whitney test.

Results

In group of multicentric gliomas, patients were 12 glioblastomas, 1 oligodendroglioma, 2 diffuse astrocytomas,2 H3K27 mutant diffuse midline gliomas, 3 anaplastic astrocytoma. In group of diffuse gliomas, patients were 4 glioblastomas, 5 anaplastic oligodendroglioma, 1 anaplastic astrocytoma,8 diffuse astrocytoma, 2 oligodendroglioma, 1 H3K27 mutant diffuse midline glioma. There was no difference between multicentric gliomas and diffuse gliomas in gender, age, disease duration, status of IDH mutation and the Ki67 expression. However, the connectivity value and probabilistic value of the lesions in multicentric gliomas were significantly lower than the lesions in diffuse gliomas (fig.2-4).

Acknowledgements

This work was supported by the National Natural Science Foundation of China (Grant nos. 81371528, 81820108018).

References

[1] Batzdorf U, Malamud N: The problem of multicentric glioma. J Neurosurg 1963; 20;30-126. [2] Behrens TEJ, Johansen-Berg H, Woolrich MW, Smith SM, Wheeler Kingshott CAM, Boulby PA, et al. Non-invasive mapping of connections between human thalamus and cortex using diffusion imaging. Nat Neurosci 2003; 6: 750–7.