Preoperative Contrast-Enhanced MRI in Differentiating Glioblastoma From Low-Grade Gliomas in The Cancer Imaging Archive Database: A Proof-of-Concept Study

PurposeThis study aimed to develop a repeatable MRI-based machine learning model to differentiate between low-grade gliomas (LGGs) and glioblastoma (GBM) and provide more clinical information to improve treatment decision-making.MethodsPreoperative MRIs of gliomas from The Cancer Imaging Archive (TCIA)–GBM/LGG database were selected. The tumor on contrast-enhanced MRI was segmented. Quantitative image features were extracted from the segmentations. A random forest classification algorithm was used to establish a model in the training set. In the test phase, a random forest model was tested using an external test set. Three radiologists reviewed the images for the external test set. The area under the receiver operating characteristic curve (AUC) was calculated. The AUCs of the radiomics model and radiologists were compared.ResultsThe random forest model was fitted using a training set consisting of 142 patients [mean age, 52 years ± 16 (standard deviation); 78 men] comprising 88 cases of GBM. The external test set included 25 patients (14 with GBM). Random forest analysis yielded an AUC of 1.00 [95% confidence interval (CI): 0.86–1.00]. The AUCs for the three readers were 0.92 (95% CI 0.74–0.99), 0.70 (95% CI 0.49–0.87), and 0.59 (95% CI 0.38–0.78). Statistical differences were only found between AUC and Reader 1 (1.00 vs. 0.92, respectively; p = 0.16).ConclusionAn MRI radiomics-based random forest model was proven useful in differentiating GBM from LGG and showed better diagnostic performance than that of two inexperienced radiologists.

Contribution of advanced neuro-imaging (MR diffusion, perfusion and proton spectroscopy) in differentiation between low grade gliomas GII and MR morphologically similar non neoplastic lesions

Background Non-neoplastic brain lesions can be misdiagnosed as low-grade gliomas. Conventional magnetic resonance (MR) imaging may be non-specific. Additional imaging modalities such as spectroscopy (MRS), perfusion and diffusion imaging aid in diagnosis of such lesions. However, contradictory and overlapping results are still present. Hence, our purpose was to evaluate the role of advanced neuro-imaging in differentiation between low-grade gliomas (WHO grade II) and MR morphologically similar non-neoplastic lesions and to prove which modality has the most accurate results in differentiation. Results All patients were classified into two main groups: patients with low-grade glioma (n = 12; mean age, 38.8 ± 16; 8 males) and patients with non-neoplastic lesions (n = 27; mean age, 36.6 ± 15; 19 males) based on the histopathological and clinical–radiological diagnosis. Using ROC curve analysis, a threshold value of 0.93 for rCBV (AUC = 0.875, PPV = 92%, NPV = 71.4%) and a threshold value of 2.5 for Cho/NAA (AUC = 0.829, PPV = 92%, NPV = 71.4%) had 85.2% sensitivity and 83.3% specificity for predicting neoplastic lesions. The area under the curve (AUC) of ROC analysis was good for relative cerebral blood volume (rCBV) and Cho/NAA ratios (> 0.80) and fair for Cho/Cr and NAA/Cr ratios (0.70–0.80). When the rCBV measurements were combined with MRS ratios, significant improvement was observed in the area under the curve (AUC) (0.969) with improved diagnostic accuracy (89.7%) and sensitivity (88.9%). Conclusions Evaluation of rCBV and metabolite ratios at MRS, particularly Cho/NAA ratio, may be helpful in differentiating low-grade gliomas from non-neoplastic lesions. The combination of dynamic susceptibility contrast (DSC) perfusion and MRS can significantly improve the diagnostic accuracy and can help avoiding the need for an invasive biopsy.

Circumferential sulcus-guided resection technique for improved outcomes of low-grade gliomas

OBJECTIVE Many neurosurgeons resect nonenhancing low-grade gliomas (LGGs) by using an inside-out piecemeal resection (PMR) technique. At the authors’ institution they have increasingly used a circumferential, perilesional, sulcus-guided resection (SGR) technique. This technique has not been well described and there are limited data on its effectiveness. The authors describe the SGR technique and assess the extent to which SGR correlates with extent of resection and neurological outcome. METHODS The authors identified all patients with newly diagnosed LGGs who underwent resection at their institution over a 22-year period. Demographics, presenting symptoms, intraoperative data, method of resection (SGR or PMR), volumetric imaging data, and postoperative outcomes were obtained. Univariate analyses used ANOVA and Fisher’s exact test. Multivariate analyses were performed using multivariate logistic regression. RESULTS Newly diagnosed LGGs were resected in 519 patients, 208 (40%) using an SGR technique and 311 (60%) using a PMR technique. The median extent of resection in the SGR group was 84%, compared with 77% in the PMR group (p = 0.019). In multivariate analysis, SGR was independently associated with a higher rate of complete (100%) resection (27% vs 18%) (OR 1.7, 95% CI 1.1–2.6; p = 0.03). SGR was also associated with a statistical trend toward lower rates of postoperative neurological complications (11% vs 16%, p = 0.09). A subset analysis of tumors located specifically in eloquent brain demonstrated SGR to be as safe as PMR. CONCLUSIONS The authors describe the SGR technique used to resect LGGs and show that SGR is independently associated with statistically significantly higher rates of complete resection, without an increase in neurological complications, than with PMR. SGR technique should be considered when resecting LGGs.

High Expression of H2BC12 Predicts Poor Survival Outcome of Low-Grade Gliomas: A Study Based on TCGA Data

Purpose Low-grade gliomas (LGG) have highly variable clinical behaviors, with a high incidence of disease progression as 70% within ten years. Regardless of treatment combining surgery and radiotherapy or chemotherapy, LGG is still associated with adverse survival outcomes. Therefore, our study was performed to satisfy the increasing demand of novel sensitive biomarkers and therapeutic targets in treatment and diagnosis of LGG. Methods The TCGA data set was used to examine the relationship between H2BC12 expression and clinical pathologic characteristics. The significance of H2BC12 expression in prognosis was also investigated. In addition, H2BC12 expression-related pathways were enriched by gene set enrichment analysis (GSEA). Association analysis of H2BC12 gene expression and immune infiltration was performed by single sample gene set enrichment analysis (ssGSEA). Results Significantly up-regulated expression of H2BC12 mRNA was found in LGG tissue when compared to normal tissue and was proven to be diagnostic (have diagnostic significance) for LGG. In the meantime, high H2BC12 levels were associated with WHO grade, IDH status, 1p/19q codeletion, primary therapy outcome and histological type of LGG, and additionally, prognostic for adverse survival outcomes. In the multivariate analysis, high H2BC12 levels were identified to be an independent predictor for poor survival outcomes of LGG patients. Pathways in cancer, signaling by Wnt or PI3K-AKT signaling pathway, DNA repair, cellular senescence and DNA double strand break repair were differentially activated in the phenotype that positively associated with H2BC12. Conclusion H2BC12 is a promising biomarker for the diagnosis and prognosis of LGG.

The Effect of Necroptosis-related Gene Signature on the Progression, Prognosis, and Immune Microenvironment of Low-grade Glioma

BackgroundDespite the incorporation of various clinical and molecular criteria in the diagnosis and prognosis prediction of low-grade glioma, individual variation and risk stratification have not been completely explored. Necroptosis is considered closely related to different types of cancers, including low-grade gliomas. In this study, we obtained the necroptosis genes from the Kyoto Encyclopedia of Genes and Genomes website, extracted necroptosis genes from The Cancer Genome Atlas, and established a necroptosis-related gene signature (NECSig) through hazard analyses. Then we established a prognostic risk model consisting of four NECSig (BID, H2AFY2, MAPK9, and TNFRSF10B).ResultBased on the model, the high-risk group is significantly associated with poorer overall survival. The accuracy of this model is further supported by the receiver operating characteristic curve. Then, we constructed a prognostic nomogram combining NECSig and clinical features, which shows good predictive power for stratification of survival risk. We discovered variations in the kind of immune infiltration, immune cells, and functions between the high-risk and low-risk groups using this risk model. We also showed that drug therapy is more sensitive in high-risk populations.ConclusionThe results revealed a prognostic indicator of NECSig, which may provide information for immunological research and treatment of low-grade gliomas.