Introduction: Preoperative accurate brain tumor diagnosis plays an essential role in the selection of the optimum treatment strategy, as their management and prognosis are different. Advanced MRI techniques, such as DWI, PWI, MRS and DKI, have been wildly used to grading gliomas and the diagnostic efficiency for grading gliomas has been gradually improved.1-3 Newly developed neurite orientation dispersion and density imaging (NODDI) is an advanced diffusion weighted imaging and it assumes a three-compartment biophysical tissue model including intracellular, extracellular, and cerebrospinal fluids in a single voxel, which enables the inference and quantification of the direction and structure of neurites (axons and dendrites).4 Hence, NOODI could apply more information with the changes of the tumor environment and representative parameters of NODDI are intracellular volume fraction (ficvf) and orientation dispersion index (ODI). Nowadays, NODDI has been applied to analyze multiple sclerosis5, focal cerebral corital dysplasia6 and central nervous system degenerative disease7 while, barely, NOODI was used to evaluate and grade gliomas. Therefore, the purpose of our study is to quantitatively evaluate the diagnostic efficiency of NODDI in tumor parenchyma (TP) and peritumoral (PT) area for grading gliomas. Methods: 29 patients (male: 18, female: 11, mean age: 45.4 y) were prospectively recruited and they underwent conventional, whole-brain diffusion-weighted images which were collected at three b value (0, 1000 and 2000 s/mm2) and 1000 and 2000 s/mm2 with 30 directions. Both the b = 1000 s/mm2 and b = 2000 s/mm2 data were used for the NODDI analysis. Neurite volume fraction (ficvf) and orientation dispersion index (ODI) maps were generated by post procession software (matlab toolbox). With each tumor, 6-10 regions of interest (ROIs) were manually placed on TP, PT and the contralateral normal brain area (CNBA) by Image J. The minimum (min), mean and maximum (max) values of ficvf and odi and their ratio of TP/CNBA were calculated and their diagnostic efficiency was assessed by Mann-Whitney test and ROC analysis. Results: The mean values of min, mean and max values of the ficvf and ODI in normal brain were (0.41±0.15), (0.53±0.13), (0.69±0.18) and (0.23±0.13), (0.37±0.13), (0.55±0.19) respectively. Compared with CNBA, in TP area, the mean value of ficvf was significantly lower in glioma (irrespective of glioma grade). The mean value of ODI was significantly higher in high grade glioma (HGG) than in CNBA while ODI mean value was significantly lower in low grade glioma (LGG). Compared with LGG, in TP area, the min, mean, max values of ficvf, ODI and their ratios of TP/CNBA are significantly higher in HGG (p<0.007) and the mean value of ficvf (0.40HGG vs. 0.23LGG) showed the highest diagnostic value (AUC=0.81, cut-off value: 0.33) and specificity (82%). Further, in PT area, the max values of ficvf and ODI and the mean value of ODI are significantly higher in HGG (p<0.038) and the max value of ficvf (0.48HGG vs. 0.36LGG) demonstrated the highest diagnostic value (AUC=0.62) with the highest specificity (87%) and relatively lower sensitivity (44%). Dissusion: Glioma is a kind of neoplasm and characterized by varying degrees of hypercellularity, nuclear pleomorphism, endothelial proliferation and microvascular density. All those transforms in replace of the normal brain tissue could modify the brain mircroenviroment, hence, would change the freedom of water molecules to diffuse within the tissue. We found that the ficvf and odi were significantly higher in HGG than in LGG. The higher tumor cellularity, microvascular desity and much complicated endothelial proliferation of HGG8,9 might cause those differences. The higher tumor cellularity would accompany with higher axon density, therefore, the ficvf was higher. Since the more restriction by the tumor cell, to some extent, might induce higher index of orientation dispersion. The normal brain tissue with blood brain barrier (BBB) usually have higher ficvf, therefore our study showed that, irrespective of glioma grades, the ficvf was lower in glioma. However, for ODI, the ODI in HGG was higher than in norm brain tissue. This might due to the higher tumor cellularity and the serious destroy of the BBB and the LGG with relatively less tumor cellularity and less BBB destroy, therefore, the ODI was significantly lower than in normal brain. For PT area, compared with LGG, the signal abnormality in HGG is not only caused by the altered interstitial water but also by the infiltration of the scattered tumor cells10,11and this might cause the differences in ficvfmax、odimean、odimax. Conclusions: Quantitative parameters from NODDI in TP and PT area can aid in gliomas grading and the mean value of ficvf showed the highest diagnostic power.

Acknowledgements

No acknowledgement found.

References

1. Sadeghi N, D'Haene N, Decaestecker C, et al. Apparent diffusion coefficient and cerebral blood volume in brain gliomas: relation to tumor cell density and tumor microvessel density based on stereotactic biopsies[J]. AJNR Am J Neuroradiol,2008,29(3):476-482. 2. Calvar J A, Meli F J, Romero C, et al. Characterization of brain tumors by MRS, DWI and Ki-67 labeling index[J]. J Neurooncol,2005,72(3):273-280. 3. Zonari P, Baraldi P, Crisi G. Multimodal MRI in the characterization of glial neoplasms: the combined role of single-voxel MR spectroscopy, diffusion imaging and echo-planar perfusion imaging[J]. Neuroradiology,2007,49(10):795-803. 4. Hui Z, Torben S, Claudia A, et al. NODDI: Practical in vivo neurite orientation dispersion and density imaging of the human brain. NeuroImage 61 (2012) 1000–1016. 5. Magnollay L, Grussu F, Wheelerkingshott C,et al. An investigation of brain neurite density and dispersion in multiple sclerosis using single shell diffusion imaging. In: (Proceedings) Joint Annual Meeting ISMRM-ESMRMB. (pp. 2048). 6.Winston G, Micallef C, Symms M, et al. Advanced diffusion imaging sequences could aid assessing patients with focal cortical dysplasia and epilepsy. Epilepsy Res. 2014 Feb;108(2):336. 7. Zaja-Milatovic S, Milatovic D, Schantz AM, et al. Dendritic degeneration in neostriatal medium spiny neurons in Parkinson disease. Neurology. 2005;64(3):545547. 8. Kono K, Inoue Y, Nakayama K, Shakudo M, Morino M, Ohata K et al. The role of diffusion-weighted imaging in patients with brain tumors. AJNR Am J Neuroradiol. 2001; 22:1081–88. PMID: 11415902 Synopsis 9.Toh CH, Wei KC, Ng SH, et al. Differentiation of tumefaetive demyelinating lesions from high-grade gliomas with the use of diffusion tensor imaging [J]. AJNR, 2012, 33(5): 846-851. 10. Burger P. Classification, grading, and patterns of spread of malignant gliomas. In: Apuzzo ML, ed. Neurosurgical topics: malignant cerebral glioma. Park Ridge, Ill: American Association of Neurological Surgeons.1990;3–17. 11. Burger PC, Vogel FS, Green SB, Strike TA. Glioblastoma multiforme and anaplastic astrocytoma: pathologic criteria and prognostic implications. Cancer. 1985; 56: 1106–1111.