Prediction of diffuse glioma grade and tumor cell proliferative activity by synthetic MRI combined with three dimensional arterial spin labeling imaging
10.3760/cma.j.cn112149-20210617-00569
- VernacularTitle:合成MRI联合三维动脉自旋标记成像预测弥漫性胶质瘤分级及肿瘤细胞增殖活性的研究
- Author:
Xin GE
1
;
Shengyu SUN
;
Wenxiao LIU
;
Jianguo ZHAO
;
Ying SHEN
;
Ruirui LYU
;
Xueying HUANG
;
Xiaodong WANG
Author Information
1. 宁夏医科大学临床医学院，银川 750004
- Keywords:
Glioma;
Magnetic resonance imaging;
Arterial spin labeling
- From:
Chinese Journal of Radiology
2022;56(5):524-529
- CountryChina
- Language:Chinese
-
Abstract:
Objective:To evaluate the value of synthetic MRI combined with three dimensional-arterial spin labeling (3D-ASL) imaging in the grading of diffuse glioma and its correlation with tumor cell proliferative activity (Ki-67).Methods:This study was prospective. The clinical and imaging manifestations of 66 patients with diffuse glioma who underwent synthetic MRI combined with 3D-ASL imaging from August 2020 to June 2021 in General Hospital of Ningxia Medical University were analyzed. Among 66 patients, there were 36 males and 30 females, aged 4-76 years, and divided into low grade glioma (LGG) group ( n=25) (WHO Ⅱ) and high grade glioma (HGG) group ( n=41) (WHO Ⅲ and vⅣ). T 1, T 2, proton density (PD) and cerebral blood flow (CBF) of tumor parenchyma were measured by GE ADW4.7 postprocessing software. The Ki-67 label index (Ki-67 LI) in postoperative pathological sections was detected by immunohistochemistry. Independent sample t test or Mann-Whitney U test was used to compare the differences of quantitative parameters between HGG group and LGG group. Receiver operating characteristic (ROC) curve was used to analyze the diagnostic efficacy of T 1, PD, CBF and the combination. Spearman test was used to analyze the correlation between the parameters and Ki-67 label index (LI). Results:T 1[(1 573±173)ms], PD[(86.2±2.4)pu] and CBF[(129±48)ml·100 g -1·min -1] in HGG group were significantly higher than those in LGG group [(1 376±134)ms, (83.0±2.5)pu and (77±49)ml·100g -1·min -1 respectively], and difference had statistical significance ( t=-4.86, -5.08, -4.24, P<0.01). ROC confirmed that the area under curve (AUC) of T 1, PD and CBF in differentiating HGG from LGG were 0.847, 0.843 and 0.777, respectively. In multi-parameter analysis, the combination of three parameters had the best diagnostic efficiency (AUC=0.973) and the sensitivity and specificity were 87.8% and 100%, respectively. In LGG and HGG groups, there was no correlation between T 1, T 2, PD, CBF and Ki-67 LI. In the overall cohort, T 1, PD and CBF had slight positive correlation with Ki-67 LI ( r=0.394, 0.411 and 0.406, respectively, all P<0.01). There was no correlation between T 2 and Ki-67 LI ( r=-0.100, P=0.423). Conclusion:Synthetic MRI and 3D-ASL can noninvasively evaluate the pathological grade of glioma and predict the expression of Ki-67, among which T 1 and PD are novel imaging marks.
