For retrospective analysis, consecutive 6 patients who were pathologically diagnosed as epithelioid glioblastoma (eGBM) were enrolled. 4 cases were male, 2 were female, aged from 18-year to 52-year, the median age was 35.5 years. Durations of the clinical symptoms varied from several days to 1 year. Prior to any treatment, all the cases performed conventional MR scanning with or without contrast agent, diffusion-weighted imaging (DWI), susceptible-weighted imaging (SWI) and 1H-MR spectroscopy (1H-MRS). Moreover, 4 cases undertook dynamic susceptibility contrast-enhanced perfusion weighted imaging (DSC-PWI). 3 cases accepted plaint and enhanced CT scan.
This study was approved ethically by the institutional review board at Daping Hospital, Army Medical University (Chongqing, China) (IRB #201879). All the patients with suspected glioma had previously informed the possibility of using their medical imaging for research and obtained authorization for utilization of their medical records.
2.1 Radiological scanning protocol
All patients performed conventional and advanced MR scanning on a 3.0T MR scanner (Verio, Siemens, Erlangen, Germany), with an 8-channel head coil. Conventional MR imaging included following sequence: turbo spin-echo T2-weighted images (T2WI, TR/TE 4900/96ms, number of excitations, NEX 1), T2-weighted fluid-attenuated inversion-recovery (FLAIR, TR/TE 8000/94ms, NEX 1), pre-contrast gradient-echo T1-weighted images (T1WI, TR/TE 440/2.48ms, NEX 1) and post-contrast T1WI with identical parameters used in pre-contrast T1WI scanning. All the images were acquired with a field of view (FOV) of 230×230mm2 and 24 axial slices of 5mm with intersection space of 1mm. The contrast agent Gd-DTPA (magnevist solution, Guangzhou Consun Pharmaceutical Co., Ltd, Guangzhou, China) was injected at a dose of 0.2mmol/kg of body weight and rate of 3-4ml/s. DWI was performed prior to T1WI with contrast, using an echo planar imaging (EPI) sequence, TR/TE 6600/100ms, b-value (0, 600, 1000), FOV230×230mm2. Apparent diffusion coefficient (ADC) maps were acquired using an images post-processing software (Siemens syngo, Siemens, Erlangen, Germany). DSC-PWI was performed using a SE-EPI sequence with following parameters: TR/TE 1872/30ms, flip angle 90°, NEX 1.0, matrix 128×128, FOV 230×230mm2. For each section, 20 images were obtained. After acquiring the 6th image, a bonus of Gd-DTPA at a dose of 0.2mmol/kg was injected at a rate of 3-4ml/s, immediately following by a bonus of equivalent dose of saline. The parameters of 1H-MRS (2D multi-voxel chemical shift imaging, CSI) were as follows: TR/TE 1700/135ms, FOV 160×160mm2, voxel thickness 10mm, NEX 3.0, FWHM<20HZ. The regions of interest (ROI) covered tumorous solid area, periphery region of tumor and contralateral normal region. SWI was performed using a 3D multi-echo flow compensated GRE sequence with the following parameters: TR/TE 27/20ms, flip angle 10°, FOV 200 × 200 mm2, matrix 256 × 182, slice thickness 1.5 mm, scan range 135 mm.
CT was performed on a 64-slice CT scanner (LightSpeed VCT, GE healthcare, Milwaukee, WI). The parameters were as follows: tube voltage 120KV, tube current 350mA, slice thickness 5mm, matrix 512×512, FOV 9.6cm. After plain CT, the iodine contrast agent (Ultravist, Bayer Healthcare, Leverkusen, Germany; 370mg/ml) at a dose of 300mg/kg was administrated via an automatic switching injector (Ulrich medical, UIm, Germany) at a speed of 3.5ml/s.
2.2 Imaging post-processing and analysis
All the DSC-PWI and DWI original data were processed with a commercial post-processing software package (SygnoMMWP VE36A, Siemens, Erlangen, Germany) to obtain cerebral blood volume (CBV) and ADC maps. The imaging analysis was performed by two experienced, board-certified neuro-radiologist with more than 5-year work experience. The ROIs with identical size were placed on the solid component and contralateral normal appearance white matter (NAWM), CBVmax and ADCmin was determined. Then, the rCBVmax and rADCmin were calculated by normalizing CBVmax and ADCmin to the value of contralateral NAWM. On 1H-MRS, Lip/Lac value was stratified: -, totally negative or can’t be measured; 1+, detectable; 2+, the second peak; 3+, the largest peak [5]. According to the method provided by Park et al. [6], intratumoral susceptibility signal intensity (ITSS) on SWI was scored as follows: 0, absent; 1, number of ITSS ≤ 5; 2, ITSS ≤ 10; 3, ITSS ≥ 11.
2.3 HE and immunohistological staining
HE staining was performed on 4μm sections of formalin-fixed paraffin-embedded (FFPE) specimens. Immunohistological staining was done following EnVision kit specification and using antibodies against glial fibrillary acidic protein (GFAP), Nestin, IDH-1, Olig-2, p53, NeuN, S-100, integrase interactor 1 (INI-1), Syn, CD34 and Ki-67. All the primary antibodies were brought from MXB (MXB Biotechnology Co, Fuzhou, China).
2.4 Mutation analysis of BRAF V600E
DNA was extracted from the paraffin-embedded specimens by using DNA extraction Kit (Amoy Diagnostics, Xiamen, China) and following the manufactural protocol. BRAF V600E ARMS-PCR Kit (Amoy Diagnostics, Xiamen, China) was applied to determine the mutation statue of the specimens, all the procedures followed the kit specification.
The total reaction system was 45μL, including 5μL DNA template and 40μL reaction mixture (consist of Taq DNA polymerase, scorpions primer, reaction buffer and so on). Each reaction contained a positive and a negative control. The PCR reaction was performed as follows: Stage i 95℃for 5min; Stage ii 95℃for 25s followed 72℃for 20s, repeated 15 cycles; Stage iii 93℃ for 25s, 60℃ for 35s, 72℃ for 20s, repeated 31 cycles. Data was automatically collected when the reaction was completed. Mutation result was concluded based on the manufacture instruction.