CT Perfusion On Brain Gliomas With Pre-and Post-irradiation By X-knife Treatment And The Correlation Between MRI And Expression Of VEGF,MMP-2,MMP-9 Of The Tumors: Experimental Study.

IntroductionRat C6 brain gliomas have been used widely to simulate II grade astrocyto-mas in animal experiments. CT perfusion was seldom applied to study the radiation therapeutic effects and angiogenesis. There was no discussion about the correlation between the expression of VEGF, MMP — 2 and MMP - 9 and MRI characteristics of rat C6 brain gliomas. So the purpose was to discuss the feasibility of using CT perfusion evaluating the microcirculation of rat brain gliomas with pre - and post - irradiation by X - knife treatment and to determine the correlation between expression of VEGF, MMP - 2, MMP - 9 and MR characteristics on rat C6 brain gliomas.Material and Method1. Intracerebral Glioma Model in Rat brainAll experiments were carried out on female Wistar rats ( body weight range 250 ~300g). The rats were anesthetized and placed on a stereotactic headhold-er. A middle scalp incision was made. The cell suspension (106 cells in 1μl) was slowly injected using a Hamilton syringe through a burr hole in right caudate nucleus (lmm posterior to suture, 2mm lateral to the midline, 5mm depth from the dura).2. X - knife treatmentOn the 16 th day after implantation of C6 cells, some bearing -tumor rats were treated with 5 mm collimater with the adhesive restraints on the organic glass plate, which was matched and fixed with the leksell Model G stereotactic frame. The rats with the frame were performed with CT scan. Following the un-enhanced CT scan with the large field for placement, the contrast - enhanced scan was performed with the small field for describing the radiation area. Because two scans were applied with the same protocal on the same scout image, the images were fused to sketch the target positon on the 3D workstation. According to the radiosurgical treatment planning, some rats each received maximal doses of 20Gy using a single isocenter of irradiation delivered with the 5mm col-limator helmet and five converging arcs of the X - knife unit by using a 6 - MeV linear accelerator.3. CT perfusionSome bearing - tumor animals were used to study CT perfusion, which was performed on GE lightspeed 16 multislice spiral CT on 1 week, 2 weeks, 3 weeks after tumor implantation as well as 5 days after X - knife treatment. The scan started at the same time with the injection of contrast agent through the tail vein, and 792 images were acquired and transferred to 3D workstation. The CBF, CBV, MTT and PS images were reconstructed through Perfusion 3 software. The values in the center, periphery of the tumors and tumor - brain districts and contralateral brain tissueswere detected (the value of the tumors and contralateral brain tissues only were detected in 1 - week tumors).4.MRIOther rats were performed on 1 week, 2 weeks, 3 weeks after tumor implantation with 3 inch and/or 5 inch surface coil on 1. 5T Magnet. The images were acquired using a classical spin - echo (SE) sequence, predominantly T2 -weighted. The contrast - enhanced T1 - weighted images including three directions were acquired with the administration of Gd - DTPA. The volume of the tumors and total volume including peritumor edema were detected on the contrast - enhanced T1 - and T2 - weighted images separatively, so the index of peritumor edema was expressed as the differrence between the tumor volume and total volume individed by the tumor volume. Furthermore, the increased signal inten-sity was calculated on the unenhanced and contrast - enhanced T\ - weighted images.5. Pathology and immunohistochemistryThe animals were sacrificed less than 1 hour after CT or MRI scan. The brains were carefully excised and fixed in 10% formalin for 1 or 2 weeks and embedded in paraffin. Axial sections (4ujn) of the tumors were obtained. Two different staining methods were used: hematoxylin and eosin ( H&E ) and immu-nohistochemical staining for GFAP, S - 100, CD34, VEGF, MMP - 2 and MMP-9.MVD was defined as the number of microvessels per unit area in histologi-cal sections stained with CD34. The Vessels were counted at high magnification in 5 separate fields selected from the most vascular regions in the central non -necrotic areas and peripheral areas of the tumors. The micro vessel density was expressed the average as a number of counts/mm . Microvessel density was also determined in the regions of interest defined in the contralateral areas.The VEGF — immunoreactive brown - colored deposits were mainly located in vascular endothelial and tumor cells. According to the staining intensity, the results were individed into three levels: - similar to the staining intensity in the contralateral brain. + slightly stronger than that in the contralateral brain, + + obvious stronger than that in the contralateral brains. MMP - 2 and MMP - 9 were also acquired in tumor cells and vascular endothelial cells. MMP -2 and MMP -9 expression were determined on random 5 visual fields (X200). Quantitative evaluation was performed by counting of the pecentage of positively stained cells.6. Statistical AnalysisResults were expressed as the mean ± SD. SPSS 11.5 for windows was used for all statistics analysis such as one - way ANOVA, two - tailed independent T - test, Chi - square test and correlation of Spearman. Statistical significance was declared when p <0.05.ResultsThe sixty - four rats were implanted successfully. Several rats appeared symptoms of slightly intracranial hypertension represented for decreased weight and motion on 3 weeks. More rats developed abnormalities in feeding behavior or external appeareances even motor deficits. Rats in treatment group with no significant neurological deficits, were sacrified 5 days, except 1 rat died 1 hour after treatment.The results of histopathological analysis of rat C6 brain gliomas in this study were consistent with previous descriptions of this brain tumor model. There were not tumor tissues in the gross on 1 - week group. The fish - like neoplasm was not watched in the right caudate nucleus until 2 weeks later. The areas of bleeding and necrosis were larger in 2 cases of the treatment group than in the 3 -week group. Hypocellular regions and tumor edema were seen more frequently in the treatment group than 3 - week one. The tumors demonstrated as ⅡorⅢ grade astrocytoma and invaded into the adjacent normal brain tissues. All the tumors expressed glial fibrillary acidic protein and/or S - 100 protein, which confirmed the origin of gliomas. Micro vessel density in the tumors of 2 - week and 3 - week groups, which appeared no difference between them, was higher than that in the contralateral brain and 1 - week group with a distinct absence of angiogenesis. Microvessel density decreased significantly in the tumor of the treatment group compared with in 3 - week one. Tumor angiogenesis was observed more in the peripheral areas than in the central non - necrotic areas of the tumors. There were no expression of VEGF, MMP -2 and MMP -9 in the normal brain tissues. The expression of VEGF did not appear until 2 weeks after tumor implantation, while the tumors in 1 week group expressed MMP - 2 and MMP -9. There was more expression of VEGF, MMP -2 and MMP -9 around the vessels in the peripheral area of the tumors.The tumor shapes in PS images were accordance with the contrast - enhanced areas on CT. The tumor margin appeared well demacrated from the adjacent normal brain tissues, which were similar to what the tumors displayed oncontrast - enhanced Tj - weighted images. The area of increased MTT was smaller significantly than that of increased PS in the tumor, even the higher MTT district in corresponding to the lower PS area. The CBF and CBV images demacrat-ed the tumor worse than the PS images and the distribution of their higher value was not also correspondent.All the values of paramenters in all groups except MTT in 1 - week and treatment groups increased compared with on the contralateral brain. The difference of PS values in the tumors between in 2 - week group and 1 - week group was found statistically significant. There was significant difference of PS values in the tumor - brain areas between in 3 - week group and in 2 - week group. The CBV and PS values showed difference in the periphery of the tumors between in the treatment group and in the 3 - week group. There was negative correlation between CBF and the counts of MVD in the peripheral areas of the tumors in tumor groups. Furthermore, the negative correlation was also observed between the decrease of CBF values and increase of the counts of microvessel density in the center of the tumors in the treatment and 3 - week groups.MRI demonstrated the expected, predictable increase in tumor size and gross morphology over the 3 - week postimplantantion period. The most consistent conventional imaging finding in all tumors was enhencement on postcontrast T, -weighted images. Indead, contrast — enhencement was observed even in the tumors at 1 week after implantation. The tumors showed slight low signal on T, - weighted images and slight high signal on T2 - weighted images. Tumors both at 2 and 3 week's after implantation showed an increase of the volume, peritu-moral edema and extensive contrast - enhancement. There was heterogeneity of the tumors on both Tj - and T2 weighted images with clear evidence of central necrosis and mass effect at the 3 rd week. The parameters of MRI such as the volume, increased signal intensity and El demonstrate significantly positive correlation with the expression of VEGF and MMP -9, and the expression of MMP -2 showed correlation with the increased percent of T1 signal intensity and El.