MinADC Values Predict Tumor Behaviour And Prognosis In Gliomas By 3.0T MR Imaging

Part one The application of minADC value for predicting tumor behaviour in low grade gliomasObjective Our objective was to retrospectively evaluate the application value of diffusion-weighted imaging (DWI) and the minimum apparent diffusion coefficient (minADC) for predicting tumor behaviour and prognosis in low grade gliomas (LGGs) at 3.0-T magnetic resonance imaging (MRI).Methods We collected the clinical, image and pathological dates of the 21 LGGs, retrospectly. According to the European Organization for Research and Treatment of Cancer (EORTC), LGGs were split into low-risk group and high-risk group. There were several unfavorable prognostic factors for survival:age≥ 40 years, astrocytoma histology subtype, largest diameter of the tumor≥ 6 cm, tumor crossing the midline, and presence of neurologic deficit before surgery. Presence of up to two of these factors identifies the low-risk group, whereas a higher score identifies high-risk group. Everyone in the study did not do any puncture, radiotherapy, chemotherapy and γ-knife treatment. There were 21 LGGs underwent head MRI examination, who were examined using conventional MRI, enhanced MRI and DWI at 3.0-T MR (Verio; Siemens, Erlangen, Germany) with head coil before surgery. DW imaging was performed with a spin shot echo-planar imaging sequence (repetition time/echo time in ms,6600/100 ms; section thickness,6.5 mm; intersection gap,2 mm; 20 sections; matrix,200×200; field of view,220×220 mm) with three orthogonal directional motion-probing gradients (b=1000 sec/mm2), followed by automatic generation of isotropic DW images. Images without motion-probing gradients (b=0 sec/mm2) were also simultaneously obtained. ADC maps were automatically generated on a Syngo workstation (SIEMENS, Germany). Two neuroradiologist with more than 10 years of experience with head MRI reviewed the conventional MR with contrast-enhanced and DW images. They put several regions of interest (ROI) on the solid tumor components. We choose the minimum ADC value as the minADC value. These regions were placed one by one, avoiding cystic, necrotic, or hemorrhagic tumor components. We put all minADC values into SPSS 1.60 software, and tested the normal distribution and variance homogeneity of the two group dates. We calculated the mean and standard deviation of low-risk group and high-risk group. The statistically difference was calculated by independent-samples t test.Results The two groups’ minADC values obeyed normal distribution and were variance homogeneity. The mean minADC values of low-risk group and high-risk group were (1.11±0.31)×10-3mm2/s and (0.72±0.19)x10-3mm2/s, respectively. The former was significant higher than the latter, and the difference was significant (P=0.008).Conclusion In this study, with lower minADC value, the LGGs had worse tumor behaviour. On the level of molecule, the minADC value is of great significance for preoperatively evaluating tumor behaviour and prognosis in LGGs.Part two MinADC values predict prognosis in low-grade and high-grade gliomas by 3.0-T MRIObjective To retrospectively evaluate diffusion-weighted imaging (DWI) and the minimum apparent diffusion coefficient (minADC) value for predicting the malignancy and prognosis of gliomas.Methods There were 38 gliomas in our study. No patients make remedies before MRI, including puncture, radiotherapy, chemotherapy and γ-knife treatment. All patients underwent 3.0-T magnetic resonance (MR) (Verio; Siemens, Erlangen, Germany) with head coil, including conventional MRI, enhanced MRI and diffusion-weighted imaging (DWI). The ADC map was automatically generated by DWI (b=0,1000s/mm2) on a Syngo workstation (SIEMENS, Germany). Two neuroradiologist with more than 10 years of experience reviewed all the MR images. They make sure the solid tumor components without cystic, necrotic, or hemorrhagic. They put several regions of interest in the solid tumor component. We chose the minimum ADC value as the minADC value. All patients underwent intracranial tumor resection. The seasoned neuropathologist analyzed the pathological section and calculated the Immunohistochemical Ki-67 staining. All patients were followed-up at last 1 year. According to the WHO classification, the gliomas were segmented into low grade gliomas (LGGs) and high grade gliomas (HGGs). There were 16 LGGs and 22 HGGs. We divided the low-grade gliomas into LGGs with low KI-67 LIs (< or=4% in diffuse astrocytoma,< or=5% in oligodendrogliomas, and< or=6% in oligoastrocytomas) and LGGs with high KI-67 LIs (>4% in diffuse astrocytoma,>5% in oligodendrogliomas, and>6% in oligoastrocytomas). They included 9 LGGs with low Ki-67 labeling index (LI) and 7 LGGs with a high Ki-67 LI. We used the SPSS 16.0 software analyze all dates. The normal distribution and variance homogeneity were teasted. (1) The difference in progression-free survival (PFS) was evaluated among tumor groups with Fisher’s exact test. (2) The Spearman rank correlation coefficient was calculated to establish the relationship between the minADC values and Ki-67 LIs in LGGs, HGGs and in all. (3) Student’s t-test was performed to determine the minADC value differences between tumor groups (LGGs with low Ki-67 LIs, LGGs with high Ki-67 LIs and HGGs).Results1..2 patients were withdrawn. Progression-free survival (PFS) was 100% in LGGs with a low Ki-67 LI,42.9% in LGGs with a high Ki-67 LI and 35.0% in HGGs. The significant difference consisted in each two group.2、The minADC glioma value was negatively correlated with the Ki-67 LI (r=-0.688, P=0.000) in all gliomas. An inverse association between these parameters was also noted in LGGs (r=-0.529, P=0.035) and HGGs (r=-0.483, P=0.023).3, The mean minADC values (10"3mm2/s) of LGGs with high Ki-67 Lls and low Ki-67 LIs were (0.76±0.18) and (1.12±0.32), respectively. The former was significantly lower than the latter (P=0.008). The minADC values of LGGs with high Ki-67 LIs and HGGs were (0.76±0.18) and (0.69±0.25), respectively. This difference was not significant (P=0.559).Conclusion Our results indicate that the minADC value can predict the malignancy and prognosis of glioma, noninvasively and preoperatively. With low minADC value, patient always had high grade glioma and worse prognosis. In low-grade gliomas, patient with low Ki-67 LI had latent malignancy and worse prognosis. Therefore, the minADC value could be valuable for predicting the malignancy and prognosis of gliomas.