Non-Invasive Estimation of Glioma IDH1 Mutation and VEGF Expression by Histogram Analysis of Dynamic Contrast-Enhanced MRI

Objectives To investigate whether glioma isocitrate dehydrogenase (IDH) 1 mutation and vascular endothelial growth factor (VEGF) expression can be estimated by histogram analysis of dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI). Methods Chinese Glioma Genome Atlas (CGGA) database was wined for differential expression of VEGF in gliomas with different IDH genotypes. The VEGF expression and IDH1 genotypes of 56 glioma samples in our hospital were assessed by immunohistochemistry. Preoperative DCE-MRI data of glioma samples were reviewed. Regions of interest (ROIs) covering tumor parenchyma were delineated. Histogram parameters of volume transfer constant (Ktrans) and volume of extravascular extracellular space per unit volume of tissue (Ve) derived from DCE-MRI were obtained. Histogram parameters of Ktrans, Ve and VEGF expression of IDH1 mutant type (IDH1mut) gliomas were compared with the IDH1 wildtype (IDH1wt) gliomas. Receiver operating characteristic (ROC) curve analysis was performed to differentiate IDH1mut from IDH1wt gliomas. The correlation coefficients were determined between histogram parameters of Ktrans, Ve and VEGF expression in gliomas. Results In CGGA database, VEGF expression in IDHmut gliomas was lower as compared to wildtype counterpart. The immunohistochemistry of glioma samples in our hospital also confirmed the results. Comparisons demonstrated statistically significant differences in histogram parameters of Ktransand Ve [mean, standard deviation (SD), 50th, 75th, 90th. and 95th percentile] between IDH1mutand IDH1wtgliomas (P < 0.05, respectively). ROC curve analysis revealed that 50th percentile of Ktrans (0.019 min−1) and Ve (0.039) provided the perfect combination of sensitivity and specificity in differentiating gliomas with IDH1mutfrom IDH1wt. Irrespective of IDH1 mutation, histogram parameters of Ktransand Ve were correlated with VEGF expression in gliomas (P < 0.05, respectively). Conclusions VEGF expression is significantly lower in IDH1mut gliomas as compared to the wildtype counterpart, and it is non-invasively predictable with histogram analysis of DCE-MRI.


INTRODUCTION
Glioma is the most common primary intracranial tumor. According to the 2016 World Health Organization (WHO) classification criteria, mutation in the gene encoding isocitrate dehydrogenase (IDH) enzyme has been identified in the substratification of glioma (1). Emerging evidence has shown that IDH mutation can convert a-ketoglutarate (a-KG) to 2-hydroxyglutaate (2-HG). Excessive 2-HG can activate prolyl hydroxylase (PHD) and then promote the degradation of hypoxia induced factor (HIF), which can weaken tumor microvascular proliferation (2,3). Moreover, a study has reported that IDH mutation inhibited PI3K/Akt signaling and reduced the HIF expression level (4). Vascular endothelial growth factor (VEGF) regulated by HIF is one of the angiogenesis-related genes, and its over-expression correlates with poor prognoses in gliomas (5,6). So, anti-VEGF therapy increases progression-free survival and improves quality of life in patients with glioma (7,8). On account of these findings, we assumed that the expression level of VEGF was lower in IDH mutant type (IDH mut ) gliomas, which fitted with the indolent clinical course of IDH mut gliomas. IDH genotypes and the expression level of VEGF can be detected by surgery or biopsy, but it suffers from several drawbacks such as sampling error, tumor heterogeneities, and risk of surgical complications. Therefore, it is urgent to find a non-invasive technique. The pharmacokinetic parameters derived from dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) have the ability to predict the microvascular features of glioma in a non-invasive manner (9,10). Based on the value of DCE-MRI in gliomas and the hypothetical correlation between glioma IDH mutation and VEGF expression, we aimed to investigate whether glioma IDH1 mutation and VEGF expression can be estimated by histogram analysis of DCE-MRI.

Chinese Glioma Genome Atlas Database Analysis
The data of VEGF expression and IDH genotypes in 641 unique samples with glioma (286 samples with IDH mut and 355 with IDH wildtype (IDH wt )) were available from the CGGA database. Wilcox test implemented in R language was utilized to analyze the differential expression of VEGF in gliomas with different IDH genotypes.

Glioma Samples
The inclusion criteria for case selection of this study were as follows: (1) pathologically confirmed gliomas; (2) treatment-naive before MRI examination and surgery; (3) underwent preoperative DCE-MRI. The final study population included 56 patients. All the data were retrospectively reviewed. The clinical cohort consisted of 33 males and 23 females, aged 55 ± 14 years (range, 22-75 years). This retrospective data evaluation was approved by the local institutional review board and written informed consent was obtained from each patient.
DCE-MRI was done using dynamic scan of a T1-fast field echo (T1-FFE; RF-spoiled gradient echo) sequence and setting the following parameters: repetition time (TR), 5.1 ms; echo time (TE), 1.4 ms; slice thickness, 2.8 mm; matrix, 256 × 210; field of view (FOV), 250 mm × 250 mm; axial scanning. Precontrast images with multiple flip angles 3, 6, 9, 12, and 15°were acquired for the T1 maps. Then, the contrast agent (Ommiscan, GE Healthcare, Oslo, Norway) was administered (0.1 mmol/kg of body weight) through the antecubital vein via a power injector at 4 ml/s, followed by a flush of 15 ml saline. A series of 1,000 images at 50 time points for 20 axial sections were acquired with a temporal resolution approximately of 7 s for each time point. Finally, the postcontrast T1-weighted imaging was conducted in the same axial geometry.

Image Analysis
A software module (OmniKinetics, GE Healthcare, China) was applied for post-processing. The volume transfer constant (K trans ) and volume of extravascular extracellular space per unit volume of tissue (V e ) were obtained based on the extended Tofts and Kermode pharmacokinetic model. The arterial input function was located on the superior sagittal sinus. The postcontrast T1-weighted imaging functioned as reference. When the tumor was without enhancement, the T2-Flair images were registered in the DCE images, and the registration images were taken as reference. Regions of interest (ROIs) covering tumor parenchyma were placed after consensus was reached between two experienced radiologists. The ROIs placed on the reference images can automatically transfer onto the parameters maps, and then histogram parameters of K trans and V e were generated automatically.

Pathology
All clinical glioma specimens were fixed in 4% formalin and embedded in paraffin. The paraffin blocks were cut into 5-mm sections and stained with hematoxylin and eosin (H&E).
The IDH1 mutation was detected with a mutation-specific antibody. Staining was interpreted as positive when ≥10% of the tumor cells exhibited an intense cytoplasmic staining, whereas staining was deduced as negative when positive cells <10% (11).
Tissue blocks from each specimen were stained for VEGF rabbit anti-human polyclonal antibody (Beijing Bioss Biotechnology Co., Ltd.). All specimens were analyzed with Motic Images Advanced software (version 3.2, Motic China

Statistical Analysis
Statistical computations were carried out with the SPSS 22.0 (SPSS Inc, Chicago, IL, USA). Histogram parameters (mean, standard deviation (SD), 50th, 75th, 90 th , and 95th percentile) of K trans , V e and VEGF expression of IDH1 mut gliomas were compared with the IDH1 wt gliomas using the Mann-Whitney U test. Receiver operating characteristic (ROC) curve analysis was plotted to differentiate IDH1 mut from IDH1 wt gliomas. The Spearman test was performed to calculate correlation coefficients between histogram parameters of K trans , V e and VEGF expression in gliomas. P < 0.05 was considered statistically significant.

Differential Expression of VEGF in Gliomas With Different IDH Genotypes in CGGA Database
The CGGA database analysis showed the expression level of VEGF was lower in IDH mut gliomas ( Figure 1).
| Graphs showed the differential VEGF expression in gliomas with different IDH genotypes from CGGA database (A, B). CGGA, Chinese Glioma Genome Atlas; IDH, isocitrate dehydrogenase; VEGF, vascular endothelial growth factor.

Glioma Samples' Demographics
A total of 56 patients were included. Immunohistochemistry revealed 17 patients with IDH1 mut and 39 patients with IDH1 wt gliomas ( Table 1).
Comparisons of Histogram Parameters of K trans , V e and VEGF Expression Between IDH1 mut and IDH1 wt Gliomas Comparisons demonstrated that the mean, SD, 50th, 75th, 90th, and 95th percentile of K trans , V e and VEGF expression were significantly lower in IDH1 mut than IDH1 wt gliomas (P < 0.05, respectively) ( Table 2 and Figures 2, 3).

ROC Curve Analysis of Histogram
Parameters of K trans and V e for Differentiating IDH1 mut From IDH1 wt Gliomas ROC curve analysis revealed that 50th percentile of K trans (0.019 min −1 ) and V e (0.039) provided the perfect combination of sensitivity (0.872, 0.846) and specificity (0.882, 0.824) in distinguishing IDH1 mut with IDH1 w t gliomas (P < 0.05, respectively). Area under the curve (AUC) for K trans and V e were 0.899 and 0.880 (Table 3 and Figure 4).   Correlations Between Histogram Parameters of K trans , V e and VEGF

Expression in Gliomas
Irrespective of IDH1 mutation, histogram parameters of K trans and V e were positively correlated with VEGF expression in gliomas (P < 0.05, respectively) ( Table 4).

DISCUSSION
IDH mutation indicates a favorable clinical prognosis as compared to IDH without mutation in gliomas (12,13). A previous study found that patients with IDH1 wt anaplastic astrocytomas even exhibited worse prognosis than those with IDH1 mut glioblastomas (14). Therefore, there are several differences between the two  genotypes of glioma. Analysis of CGGA database and clinical practice in our hospital indicated that VEGF expression was significantly lower in IDH1 mut as compared to the wildtype counterpart. Some studies demonstrated that the expression of VEGF was inhibited in the IDH mut gliomas (15,16). These findings supported our results indirectly. In recent years, several MRI techniques provide insight into exploring the link between glioma IDH mutation and microvascular characteristics. Conventional MRI showed the frequency and degree of enhancing were lower in gliomas with IDH mutation (17,18). Moreover, radiomics served as predictive markers in assessing the IDH mutation in gliomas (19,20). Dynamic susceptibility contrast-enhanced magnetic resonance imaging (DSC-MRI) and arterial spin labeling (ASL) revealed that blood flow or blood volume was significantly lower in IDH mut than IDH wt gliomas (3,(21)(22)(23). The current results implied a discriminant function for differentiating between IDH1 mut and IDH1 wt gliomas was constructed using DCE-MRI. Hilario et al. and Zhang et al. found that K trans was lower in gliomas with IDH mutation, which corroborates our results (24,25).
Hitherto, there is no study focusing on the combined estimation of glioma IDH1 mutation and VEGF expression with histogram analysis of DCE-MRI. In our study, we found that the signal intensities of lesions on K trans and V e maps were lower in IDH1 mut than IDH1 wt gliomas. Simultaneously, the   (31,32). One reason might be that glioma samples were not exactly same. More glioblastomas were recruited in our study. The other might be that the methods used were different. The studies mentioned above chose partial areas of solid tumor as ROI, while entire tumor parenchyma was included in our study. More importantly, histogram analysis can assess the heterogeneity of tumors and improve the accuracy of diagnosis. However, we have recognized a few limitations in the current study. First, the small sample size of glioma specimens restricts the accuracy of statistical analysis, and more samples are needed to verify the results in the future. Second, the IDH2 mutation has not been considered due to the lack of genomic sequence analysis in our hospital. Owing to the low frequency of IDH2 mutation in gliomas, the immunohistochemistry technique for detecting IDH1 mutation could represent the maximum number of IDH mutation in gliomas (33). Nonetheless, gene sequencing should substantiate these results in following studies.

CONCLUSIONS
VEGF expression is significantly lower in IDH1 mut gliomas as compared to that in the wildtype counterpart, and it is noninvasively predictable with histogram analysis of DCE-MRI.

DATA AVAILABILITY STATEMENT
The raw data supporting the conclusions of this article will be made available by the authors without undue reservation.

ETHICS STATEMENT
The studies involving human participants were reviewed and approved by the Ethics Committee of Affiliated Hospital of Nantong University. The patients/participants provided their written informed consent to participate in this study. Written informed consent was obtained from the individual(s) for the publication of any potentially identifiable images or data included in this article.