Glioblastoma Multiforme Dataset Research Articles

Adaptive fine-tuning based transfer learning for the identification of MGMT promoter methylation status.

Background.Glioblastoma Multiforme (GBM) is an aggressive form of malignant brain tumor with a generally poor prognosis.O6-methylguanine-DNA methyltransferase (MGMT) promoter methylation has been shown to be a predictive bio-marker for resistance to treatment of GBM, but it is invasive and time-consuming to determine methylation status. There has been effort to predict the MGMT methylation status through analyzing MRI scans using machine learning, which only requires pre-operative scans that are already part of standard-of-care for GBM patients.Purpose.To improve the performance of conventional transfer learning in the identification of MGMT promoter methylation status, we developed a 3D SpotTune network with adaptive fine-tuning capability. Using the pretrained weights of MedicalNet with the SpotTune network, we compared its performance with a randomly initialized network for different combinations of MR modalities.Methods.Using a ResNet50 as the base network, three categories of networks are created: (1) A 3D SpotTune network to process volumetric MR images, (2) a network with randomly initialized weights, and (3) a network pre-trained on MedicalNet. These three networks are trained and evaluated using a public GBM dataset provided by the University of Pennsylvania. The MRI scans from 240 patients are used, with 11 different modalities corresponding to a set of perfusion, diffusion, and structural scans. The performance is evaluated using 5-fold cross validation with a hold-out testing dataset.Results.The SpotTune network showed better performance than the randomly initialized network. The best performing SpotTune model achieved an area under the Receiver Operating Characteristic curve (AUC), average precision of the precision-recall curve (AP), sensitivity, and specificity values of 0.6604, 0.6179, 0.6667, and 0.6061 respectively.Conclusions.SpotTune enables transfer learning to be adaptive to individual patients, resulting in improved performance in predicting MGMT promoter methylation status in GBM using equivalent MRI modalities as compared to a randomly initialized network.

HSPA5 and FGFR1 genes in the mesenchymal subtype of glioblastoma can improve a treatment efficacy

ABSTRACT Tyrosine kinase inhibitors (TKIs) have emerged as a potential treatment strategy for glioblastoma multiforme (GBM). However, their efficacy is limited by various drug resistance mechanisms. To devise more effective treatments for GBM, genetic characteristics must be considered in addition to pre-existing treatments. We performed an integrative analysis with heterogeneous GBM datasets of genomic, transcriptomic, and proteomic data from DepMap, TCGA and CPTAC. We found that poor prognosis was induced by co-upregulation of heat shock protein family A member 5 (HSPA5) and fibroblast growth factor receptor 1 (FGFR1). Co-up regulation of these two genes could regulate the PI3K/AKT pathway. GBM cell lines with co-upregulation of these two genes showed higher drug sensitivity to PI3K inhibitors. In the mesenchymal subtype, the co-upregulation of FGFR1 and HSPA5 resulted in the most malignant subtype of GBM. Furthermore, we found this newly discovered subtype was correlated with homologous recombination deficiency (HRD) In conclusion, we discovered novel druggable candidates within the group exhibiting co-upregulation of these two genes in GBM, suggest potential strategies for combination therapy.

Antioxidant network-based signatures cluster glioblastoma into distinct redox-resistant phenotypes.

Aberrant reactive oxygen species (ROS) production is one of the hallmarks of cancer. During their growth and dissemination, cancer cells control redox signaling to support protumorigenic pathways. As a consequence, cancer cells become reliant on major antioxidant systems to maintain a balanced redox tone, while avoiding excessive oxidative stress and cell death. This concept appears especially relevant in the context of glioblastoma multiforme (GBM), the most aggressive form of brain tumor characterized by significant heterogeneity, which contributes to treatment resistance and tumor recurrence. From this viewpoint, this study aims to investigate whether gene regulatory networks can effectively capture the diverse redox states associated with the primary phenotypes of GBM. In this study, we utilized publicly available GBM datasets along with proprietary bulk sequencing data. Employing computational analysis and bioinformatics tools, we stratified GBM based on their antioxidant capacities and evaluated the distinctive functionalities and prognostic values of distinct transcriptional networks in silico. We established three distinct transcriptional co-expression networks and signatures (termed clusters C1, C2, and C3) with distinct antioxidant potential in GBM cancer cells. Functional analysis of each cluster revealed that C1 exhibits strong antioxidant properties, C2 is marked with a discrepant inflammatory trait and C3 was identified as the cluster with the weakest antioxidant capacity. Intriguingly, C2 exhibited a strong correlation with the highly aggressive mesenchymal subtype of GBM. Furthermore, this cluster holds substantial prognostic importance: patients with higher gene set variation analysis (GSVA) scores of the C2 signature exhibited adverse outcomes in overall and progression-free survival. In summary, we provide a set of transcriptional signatures that unveil the antioxidant potential of GBM, offering a promising prognostic application and a guide for therapeutic strategies in GBM therapy.

Modular Hub Genes in DNA Microarray Suggest Potential Signaling Pathway Interconnectivity in Various Glioma Grades.

Gliomas have displayed significant challenges in oncology due to their high degree of invasiveness, recurrence, and resistance to treatment strategies. In this work, the key hub genes mainly associated with different grades of glioma, which were represented by pilocytic astrocytoma (PA), oligodendroglioma (OG), anaplastic astrocytoma (AA), and glioblastoma multiforme (GBM), were identified through weighted gene co-expression network analysis (WGCNA) of microarray datasets retrieved from the Gene Expression Omnibus (GEO) database. Through this, four highly correlated modules were observed to be present across the PA (GSE50161), OG (GSE4290), AA (GSE43378), and GBM (GSE36245) datasets. The functional annotation and pathway enrichment analysis done through the Database for Annotation, Visualization, and Integrated Discovery (DAVID) showed that the modules and hub genes identified were mainly involved in signal transduction, transcription regulation, and protein binding, which collectively deregulate several signaling pathways, mainly PI3K/Akt and metabolic pathways. The involvement of several hub genes primarily linked to other signaling pathways, including the cAMP, MAPK/ERK, Wnt/β-catenin, and calcium signaling pathways, indicates potential interconnectivity and influence on the PI3K/Akt pathway and, subsequently, glioma severity. The Drug Repurposing Encyclopedia (DRE) was used to screen for potential drugs based on the up- and downregulated hub genes, wherein the synthetic progestin hormones norgestimate and ethisterone were the top drug candidates. This shows the potential neuroprotective effect of progesterone against glioma due to its influence on EGFR expression and other signaling pathways. Aside from these, several experimental and approved drug candidates were also identified, which include an adrenergic receptor antagonist, a PPAR-γ receptor agonist, a CDK inhibitor, a sodium channel blocker, a bradykinin receptor antagonist, and a dopamine receptor agonist, which further highlights the gene network as a potential therapeutic avenue for glioma.

Abstract 4333: Identification of glucose-6-phosphate dehydrogenase dependency in IDH1 mutant glioma cells using functional genomics

Abstract Astrocytoma and oligodendroglioma are most commonly initiated by missense mutations at the arginine 132 codon of isocitrate dehydrogenase 1 (IDH1). Biochemically the mutant IDH1 protein acquires neomorphic activity: reductive NADPH-dependent catalysis of α-ketoglutarate (αKG) to (R)-2-hydroxyglutarate [(R)-2-HG]. (R)-2-HG interferes with the function of αKG-dependent epigenetic modifiers, resulting in extensive epigenetic remodeling that impairs cellular differentiation and promotes gliomagenesis. Strategies for selectively targeting IDH1-mutant cells offer promise for improved treatment over the standard modalities of surgery, chemotherapy and radiotherapy. Whole-genome CRISPR/Cas9 knockout screens were conducted to identify gene dependencies in isogenic IDH1-mutant and wild-type (WT) U-87 MG cells. Cells were transduced with the Brunello single guide RNA (sgRNA) library, grown for 21 days, then sgRNA distributions were assessed by amplicon sequencing. Gene dependencies were identified using CRISPRcleanR and BAGELR pipelines. We discovered that glucose-6-phosphate dehydrogenase (G6PD), a pentose phosphate pathway (PPP) enzyme, was needed for viability of IDH1-mutant cells, but was dispensable for U-87 MG WT cells. The PPP is the main producer of cytosolic NADPH that is required for de novo lipogenesis, maintenance of antioxidants and production of (R)-2-HG by mutant IDH1. Previous reports demonstrated that IDH1-mutant cells increase PPP flux to support these processes. We have validated this dependency using G6PDi-1, a cell-active inhibitor of G6PD. Growth inhibition (IC50) assays demonstrated that IDH1-mutant U-87 MG cells were threefold more sensitive to G6PDi-1 with IC50 of 23 μM and enhanced cell death (increased propidium iodide uptake), compared with WT U-87 MG cells with IC50 of 67 μM. G6PD knockout was then conducted to further validate this dependency. Nucleofection of IDH1 WT U-87 MG cells with G6PD multi guide RNA-Cas9 ribonucleoproteins resulted in highly efficient knockout of G6PD at day 5 (99-100%), which was maintained at day 21 (96-98%). In contrast, efficient G6PD knockout in IDH1-mutant U-87 MG cells at day 5 (98-100%) was lost following 21 days culture (0-5%), indicating death of almost all edited cells within the culture. Finally, Kaplan-Meier analysis of IDH1-mutant patients in the TCGA lower grade glioma dataset demonstrated that higher G6PD expression correlated with worse median overall survival (7.3 versus 9.5 years; log-rank p=0.01), but was not prognostic in the IDH1 WT TCGA glioblastoma multiforme dataset. These findings demonstrate that a cell line model of IDH1-mutant glioma is highly dependent on G6PD for survival. This observation builds on the recognized importance of the PPP in the metabolic biology of IDH1-mutant malignancy by establishing G6PD as a targetable dependency in this disease. Citation Format: Kim M. Paras, Pooja Patel, Paul Choi, William R. Wilson, Tet-Woo Lee, Stephen M. Jamieson, Dean C. Singleton. Identification of glucose-6-phosphate dehydrogenase dependency in IDH1 mutant glioma cells using functional genomics [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 4333.

Comprehensive somatic mutational analysis in glioblastoma: Implications for precision medicine approaches.

Glioblastoma multiforme (GBM), a malignant neoplasm originating from glial cells, remains challenging to treat despite the current standard treatment approach that involves maximal safe surgical resection, radiotherapy, and adjuvant temozolomide chemotherapy. This underscores the critical need to identify new molecular targets for improved therapeutic interventions. The current study aimed to explore the somatic mutations and potential therapeutic targets in GBM using somatic mutational information from four distinct GBM datasets including CGGA, TCGA, CPTAC and MAYO-PDX. The analysis included the evaluation of whole exome sequencing (WES) of GBM datasets, tumor mutation burden assessment, survival analysis, drug sensitivity prediction, and examination of domain-specific amino acid changes. The results identified the top ten commonly altered genes in the aforementioned GBM datasets and patients with mutations in OBSCN and AHNAK2 alone or in combination had a more favorable overall survival (OS). Also, the study identified potential drug sensitivity patterns in GBM patients with mutations in OBSCN and AHNAK2, and evaluated the impact of amino acid changes in specific protein domains on the survival of GBM patients. These findings provide important insights into the genetic alterations and somatic interactions in GBM, which could have implications for the development of new therapeutic strategies for this aggressive malignancy.

IMAGGS: a radiogenomic framework for identifying multi-way associations in breast cancer subtypes

Investigating correlations between radiomic and genomic profiling in breast cancer (BC) molecular subtypes is crucial for understanding disease mechanisms and providing personalized treatment. We present a well-designed radiogenomic framework image–gene–gene set (IMAGGS), which detects multi-way associations in BC subtypes by integrating radiomic and genomic features. Our dataset consists of 721 patients, each of whom has 12 ultrasound (US) images captured from different angles and gene mutation data. To better characterize tumor traits, 12 multi-angle US images are fused using two distinct strategies. Then, we analyze complex many-to-many associations between phenotypic and genotypic features using a machine learning algorithm, deviating from the prevalent one-to-one relationship pattern observed in previous studies. Key radiomic and genomic features are screened using these associations. In addition, gene set enrichment analysis is performed to investigate the joint effects of gene sets and delve deeper into the biological functions of BC subtypes. We further validate the feasibility of IMAGGS in a glioblastoma multiforme dataset to demonstrate the scalability of IMAGGS across different modalities and diseases. Taken together, IMAGGS provides a comprehensive characterization for diseases by associating imaging, genes, and gene sets, paving the way for biological interpretation of radiomics and development of targeted therapy.

The RNA polymerase II subunit B (RPB2) functions as a growth regulator in human glioblastoma

Glioblastoma multiforme (GBM) is the most common and aggressive brain tumor with a poor prognosis. The growth of GBM cells depends on the core transcriptional apparatus, thus rendering RNA polymerase (RNA pol) complex as a candidate therapeutic target. The RNA pol II subunit B (POLR2B) gene encodes the second largest subunit of the RNA pol II (RPB2); however, its genomic status and function in GBM remain unclear. Certain GBM data sets in cBioPortal were used for investigating the genomic status and expression of POLR2B in GBM. The function of RPB2 was analyzed following knockdown of POLR2B expression by shRNA in GBM cells. The cell counting kit-8 assay and PI staining were used for cell proliferation and cell cycle analysis. A xenograft mouse model was established to analyze the function of RPB2 in vivo. RNA sequencing was performed to analyze the RPB2-regulated genes. GO and GSEA analyses were applied to investigate the RPB2-regulated gene function and associated pathways. In the present study, the genomic alteration and overexpression of the POLR2B gene was described in glioblastoma. The data indicated that knockdown of POLR2B expression suppressed tumor cell growth of glioblastoma in vitro and in vivo. The analysis further demonstrated the identification of the RPB2-regulated gene sets and highlighted the DNA damage-inducible transcript 4 gene as the downstream target of the POLR2B gene. The present study provides evidence indicating that RPB2 functions as a growth regulator in glioblastoma and could be used as a potential therapeutic target for the treatment of this disease.

Quantitative susceptibility mapping using plug-and-play alternating direction method of multipliers

Quantitative susceptibility mapping employs regularization to reduce artifacts, yet many recent denoisers are unavailable for reconstruction. We developed a plug-and-play approach to QSM reconstruction (PnP QSM) and show its flexibility using several patch-based denoisers. We developed PnP QSM using alternating direction method of multiplier framework and applied collaborative filtering denoisers. We apply the technique to the 2016 QSM Challenge and in 10 glioblastoma multiforme datasets. We compared its performance with four published QSM techniques and a multi-orientation QSM method. We analyzed magnetic susceptibility accuracy using brain region-of-interest measurements, and image quality using global error metrics. Reconstructions on glioblastoma data were analyzed using ranked and semiquantitative image grading by three neuroradiologist observers to assess image quality (IQ) and sharpness (IS). PnP-BM4D QSM showed good correlation (β = 0.84, R2 = 0.98, p < 0.05) with COSMOS and no significant bias (bias = 0.007 ± 0.012). PnP-BM4D QSM achieved excellent quality when assessed using structural similarity index metric (SSIM = 0.860), high frequency error norm (HFEN = 58.5), cross correlation (CC = 0.804), and mutual information (MI = 0.475) and also maintained good conspicuity of fine features. In glioblastoma datasets, PnP-BM4D QSM showed higher performance (IQGrade = 2.4 ± 0.4, ISGrade = 2.7 ± 0.3, IQRank = 3.7 ± 0.3, ISRank = 3.9 ± 0.3) compared to MEDI (IQGrade = 2.1 ± 0.5, ISGrade = 2.1 ± 0.6, IQRank = 2.4 ± 0.6, ISRank = 2.9 ± 0.2) and FANSI-TGV (IQGrade = 2.2 ± 0.6, ISGrade = 2.1 ± 0.6, IQRank = 2.7 ± 0.3, ISRank = 2.2 ± 0.2). We illustrated the modularity of PnP QSM by interchanging two additional patch-based denoisers. PnP QSM reconstruction was feasible, and its flexibility was shown using several patch-based denoisers. This technique may allow rapid prototyping and validation of new denoisers for QSM reconstruction for an array of useful clinical applications.

A hierarchical machine learning model based on Glioblastoma patients' clinical, biomedical, and image data to analyze their treatment plans

Aim of studyGlioblastoma Multiforme (GBM) is an aggressive brain cancer in adults that kills most patients in the first year due to ineffective treatment. Different clinical, biomedical, and image data features are needed to analyze GBM, increasing complexities. Besides, they lead to weak performances for machine learning models due to ignoring physicians' knowledge. Therefore, this paper proposes a hierarchical model based on Fuzzy C-mean (FCM) clustering, Wrapper feature selection, and twelve classifiers to analyze treatment plans. Methodology/ApproachThe proposed method finds the effectiveness of previous and current treatment plans, hierarchically determining the best decision for future treatment plans for GBM patients using clinical data, biomedical data, and different image data. A case study is presented based on the Cancer Genome Atlas Glioblastoma Multiforme dataset to prove the effectiveness of the proposed model. This dataset is analyzed using data preprocessing, experts' knowledge, and a feature reduction method based on the Principal Component Analysis. Then, the FCM clustering method is utilized to reinforce classifier learning. Outcomes of studyThe proposed model finds the best combination of Wrapper feature selection and classifier for each cluster based on different measures, including accuracy, sensitivity, specificity, precision, F-score, and G-mean according to a hierarchical structure. It has the best performance among other reinforced classifiers. Besides, this model is compatible with real-world medical processes for GBM patients based on clinical, biomedical, and image data.

Machine learning and bioinformatics approaches for classification and clinical detection of bevacizumab responsive glioblastoma subtypes based on miRNA expression

For the precise treatment of patients with glioblastoma multiforme (GBM), we classified and detected bevacizumab (BVZ)-responsive subtypes of GBM and found their differential expression (DE) of miRNAs and mRNAs, clinical characteristics, and related functional pathways. Based on miR-21 and miR-10b expression z-scores, approximately 30% of GBM patients were classified as having the GBM BVZ-responsive subtype. For this subtype, GBM patients had a significantly shorter survival time than other GBM patients (p = 0.014), and vascular endothelial growth factor A (VEGF) methylation was significantly lower than that in other GBM patients (p = 0.005). It also revealed 14 DE miRNAs and 7 DE mRNAs and revealed functional characteristics between GBM BVZ subgroups. After comparing several machine learning algorithms, the construction and cross-validation of the SVM classifier were performed. For clinical use, miR-197 was optimized and added to the miRNA panel for better classification. Afterwards, we validated the classifier with several GBM datasets and discovered some key related issues. According to this study, GBM BVZ subtypes can be classified and detected by a combination of SVM classifiers and miRNA panels in existing tissue GBM datasets. With certain modifications, the classifier may be used for the classification and detection of GBM BVZ subtypes for future clinical use.

Non-negative matrix factorization and differential expression analyses identify hub genes linked to progression and prognosis of glioblastoma multiforme

One of the most prevailing primary brain tumors in adult human male is glioblastoma multiforme (GBM), which is categorized by rapid cellular growth. Even though the combination therapy comprises surgery, chemotherapy, and adjuvant therapies, the survival rate, on average, is 14.6 months. Glioma stem cells (GSCs) have key roles in tumorigenesis, progression, and defiance against chemotherapy and radiotherapy. In our study, firstly, the gene expression dataset GSE124145 was retrieved; the non-negative matrix factorization (NMF) method was applied on GBM dataset, and differentially expressed genes analysis (DEGs) was performed. After which, overlapping genes between metagenes and DEGs were detected to examine the Gene Ontology (GO) categories in the biological process (BP) in the stemness of GBM. The common hub genes were used to construct protein–protein interaction (PPI) network and further GO, while Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway was utilized to pinpoint the real hub genes. The analysis of hub genes particular for the same GO categories demonstrated that specific hub genes triggered distinct features of the same biological processes. After utilizing GSE124145 and The Cancer Genome Atlas (TCGA) dataset for survival analysis, we screened five real hub genes: GUCA1A, RFC2, GNG11, MMP19, and NRG1, which are strongly associated with the progression and prognosis of GBM. The DEGs analysis revealed that all real hub genes were overexpressed in GBM and TCGA datasets, which further validates our results. The constructed study of PPI, GO, and KEGG pathway on common hub genes was performed. Finally, the KEGG pathways performed on the top 15 candidate hub genes (including six real hub genes) of the PPI network in the GBM gene expression dataset study found mitogen-activated protein kinase (Mapk) signaling pathway to be the most significant pathway. The rest of the hub genes reviewed throughout the analysis might be favorable targets for diagnosing and treating GBM and lower-grade gliomas.

MRI-based Identification and Classification of Major Intracranial Tumor Types by Using a 3D Convolutional Neural Network: A Retrospective Multi-institutional Analysis.

To develop an algorithm to classify postcontrast T1-weighted MRI scans by tumor classes (high-grade glioma, low-grade glioma [LGG], brain metastasis, meningioma, pituitary adenoma, and acoustic neuroma) and a healthy tissue (HLTH) class. In this retrospective study, preoperative postcontrast T1-weighted MR scans from four publicly available datasets-the Brain Tumor Image Segmentation dataset (n = 378), the LGG-1p19q dataset (n = 145), The Cancer Genome Atlas Glioblastoma Multiforme dataset (n = 141), and The Cancer Genome Atlas Low Grade Glioma dataset (n = 68)-and an internal clinical dataset (n = 1373) were used. In all, a total of 2105 images were split into a training dataset (n = 1396), an internal test set (n = 361), and an external test dataset (n = 348). A convolutional neural network was trained to classify the tumor type and to discriminate between images depicting HLTH and images depicting tumors. The performance of the model was evaluated by using cross-validation, internal testing, and external testing. Feature maps were plotted to visualize network attention. The accuracy, positive predictive value (PPV), negative predictive value, sensitivity, specificity, F1 score, area under the receiver operating characteristic curve (AUC), and area under the precision-recall curve (AUPRC) were calculated. On the internal test dataset, across the seven different classes, the sensitivities, PPVs, AUCs, and AUPRCs ranged from 87% to 100%, 85% to 100%, 0.98 to 1.00, and 0.91 to 1.00, respectively. On the external data, they ranged from 91% to 97%, 73% to 99%, 0.97 to 0.98, and 0.9 to 1.0, respectively. The developed model was capable of classifying postcontrast T1-weighted MRI scans of different intracranial tumor types and discriminating images depicting pathologic conditions from images depicting HLTH.Keywords MR-Imaging, CNS, Brain/Brain Stem, Diagnosis/Classification/Application Domain, Supervised Learning, Convolutional Neural Network, Deep Learning Algorithms, Machine Learning Algorithms Supplemental material is available for this article. © RSNA, 2021.

Identification of an IFN-β-associated gene signature for the prediction of overall survival among glioblastoma patients.

BackgroundBrain glioblastoma multiforme (GBM) is the most common primary malignant intracranial tumor. The prognosis of this disease is extremely poor. While the introduction of β-interferon (IFN-β) regimen in the treatment of gliomas has significantly improved the outcome of patients; The mechanism by which IFN-β induces increased TMZ sensitivity has not been described. Therefore, the main objective of the study was to elucidate the molecular mechanisms responsible for the beneficial effect of IFNβ in GBM.MethodsMessenger RNA expression profiles and clinicopathological data were downloaded from The Cancer Genome Atlas (TCGA) GBM and GSE83300 dataset from the Gene Expression Omnibus. Univariate Cox regression analysis and lasso Cox regression model established a novel 4-gene IFN-β signature (peroxiredoxin 1, Sec61 subunit beta, X-ray repair cross-complementing 5, and Bcl-2-like protein 2) for GBM prognosis prediction. Further, GBM samples (n=50) and normal brain tissues (n=50) were then used for real-time polymerase chain reaction experiments. Gene set enrichment analysis (GSEA) was performed to further understand the underlying molecular mechanisms. Pearson correlation was applied to calculate the correlation between the long non-coding RNAs (lncRNAs) and IFN-β-associated genes. An lncRNA with a correlation coefficient |R2|>0.3 and P<0.05 was considered to be an IFN-β-associated lncRNA.ResultsPatients in the high-risk group had significantly poorer survival than patients in the low-risk group. The signature was found to be an independent prognostic factor for GBM survival. Furthermore, GSEA revealed several significantly enriched pathways, which might help explain the underlying mechanisms. Our study identified a novel robust 4-gene IFN-β signature for GBM prognosis prediction. The signature might contain potential biomarkers for metabolic therapy and treatment response prediction for GBM patients.ConclusionsIn the present study, we established a novel IFN-β-associated gene signature to predict the overall survival of GBM patients, which may help in clinical decision making for individual treatment.

Identification of Core Genes and Screening of Potential Targets in Glioblastoma Multiforme by Integrated Bioinformatic Analysis.

Glioblastoma multiforme is the most common primary intracranial malignancy, but its etiology and pathogenesis are still unclear. With the deepening of human genome research, the research of glioma subtype screening based on core molecules has become more in-depth. In the present study, we screened out differentially expressed genes (DEGs) through reanalyzing the glioblastoma multiforme (GBM) datasets GSE90598 from the Gene Expression Omnibus (GEO), the GBM dataset TCGA-GBM and the low-grade glioma (LGG) dataset TCGA-LGG from the Cancer Genome Atlas (TCGA). A total of 150 intersecting DEGs were found, of which 48 were upregulated and 102 were downregulated. These DEGs from GSE90598 dataset were enriched using the overrepresentation method, and multiple enriched gene ontology (GO) function terms were significantly correlated with neural cell signal transduction. DEGs between GBM and LGG were analyzed by gene set enrichment analysis (GSEA), and the significantly enriched Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways involved in synapse signaling and oxytocin signaling pathways. Then, a protein-protein interaction (PPI) network was constructed to assess the interaction of proteins encoded by the DEGs. MCODE identified 2 modules from the PPI network. The 11 genes with the highest degrees in module 1 were designated as core molecules, namely, GABRD, KCNC1, KCNA1, SYT1, CACNG3, OPALIN, CD163, HPCAL4, ANK3, KIF5A, and MS4A6A, which were mainly enriched in ionic signaling-related pathways. Survival analysis of the GSE83300 dataset verified the significant relationship between expression levels of the 11 core genes and survival. Finally, the core molecules of GBM and the DrugBank database were assessed by a hypergeometric test to identify 10 drugs included tetrachlorodecaoxide related to cancer and neuropsychiatric diseases. Further studies are required to explore these core genes for their potentiality in diagnosis, prognosis, and targeted therapy and explain the relationship among ionic signaling-related pathways, neuropsychiatric diseases and neurological tumors.

Receptor Tyrosine Kinase Signaling and Targeting in Glioblastoma Multiforme.

Glioblastoma multiforme (GBM) is amongst the deadliest of human cancers, with a median survival rate of just over one year following diagnosis. Characterized by rapid proliferation and diffuse infiltration into the brain, GBM is notoriously difficult to treat, with tumor cells showing limited response to existing therapies and eventually developing resistance to these interventions. As such, there is intense interest in better understanding the molecular alterations in GBM to guide the development of more efficient targeted therapies. GBM tumors can be classified into several molecular subtypes which have distinct genetic signatures, and they show aberrant activation of numerous signal transduction pathways, particularly those connected to receptor tyrosine kinases (RTKs) which control glioma cell growth, survival, migration, invasion, and angiogenesis. There are also non-canonical modes of RTK signaling found in GBM, which involve G-protein-coupled receptors and calcium channels. This review uses The Cancer Genome Atlas (TCGA) GBM dataset in combination with a data-mining approach to summarize disease characteristics, with a focus on select molecular pathways that drive GBM pathogenesis. We also present a unique genomic survey of RTKs that are frequently altered in GBM subtypes, as well as catalog the GBM disease association scores for all RTKs. Lastly, we discuss current RTK targeted therapies and highlight emerging directions in GBM research.

Integrating cullin2-RING E3 ligase as a potential biomarker for glioblastoma multiforme prognosis and radiosensitivity profiling

Background and purposePositive epidermal growth factor receptor (EGFR) immunoreactivity in glioblastoma multiforme (GBM) often predicts poor radiation response. Meanwhile, all attempts to target EGFR pharmaceutically have been unsuccessful, mainly due to molecular heterogeneity of EGFR expression in GBM. A molecular biology-based and efficient way to access cellular protein levels of EGFR is urgently needed. EGFR, together with HIF-1α and Cyclin B1, is degraded via cullin2-RING E3 ligase (CRL2). It is worthwhile to investigate the possible involvement of CRL2 on GBM survival and radiosensitivity. Materials and methodsClinicopathological features, radiographic images, survival data, RNA-Seq, copy number variations (CNVs), and other genetic changes were studied on over 3800 glioma and GBM cases, which are derived from 5 independent cohorts. These datasets include the Chinese Glioma Genome Atlas, The Cancer Genome Atlas, Ivy Glioblastoma Atlas Project, Clinical Proteomic Tumor Analysis Consortium, and the 2008 Parson’s GBM dataset. ResultsExpression of CUL2, which encodes the scaffold protein cullin2 in the CRL2 E3 ligase, can predict GBM progression and survival rate. Cullin2 protein levels are inversely related to those of HIF-1α, VEGF-A, Cyclin B1, and EGFR. Elevated CUL2 expression predicts increased radiosensitivity and dampened signal intensities in perfusion imaging. CUL2 expression are regulated via CNVs, which are inheritable structural DNA variations. ConclusionCUL2 expression levels and CNVs can be integrated as potential biomarkers in facilitating GBM and prognosis and radiosensitivity profiling.

DC-AL GAN: Pseudoprogression and true tumor progression of glioblastoma multiform image classification based on DCGAN and AlexNet.

Pseudoprogression (PsP) occurs in 20-30% of patients with glioblastoma multiforme (GBM) after receiving the standard treatment. PsP exhibits similarities in shape and intensity to the true tumor progression (TTP) of GBM on the follow-up magnetic resonance imaging (MRI). These similarities pose challenges to the differentiation of these types of progression and hence the selection of the appropriate clinical treatment strategy. To address this challenge, we introduced a novel feature learning method based on deep convolutional generative adversarial network (DCGAN) and AlexNet, termed DC-AL GAN, to discriminate between PsP and TTP in MRI images. Due to the adversarial relationship between the generator and the discriminator of DCGAN, high-level discriminative features of PsP and TTP can be derived for the discriminator with AlexNet. We also constructed a multifeature selection module to concatenate features from different layers, contributing to more powerful features used for effectively discriminating between PsP and TTP. Finally, these discriminative features from the discriminator are used for classification by a support vector machine (SVM). Tenfold cross-validation (CV) and the area under the receiver operating characteristic (AUC) were applied to evaluate the performance of this developed algorithm. The accuracy and AUC of DC-AL GAN for discriminating PsP and TTP after tenfold CV were 0.920 and 0.947. We also assessed the effects of different indicators (such as sensitivity and specificity) for features extracted from different layers to obtain a model with the best classification performance. The proposed model DC-AL GAN is capable of learning discriminative representations from GBM datasets, and it achieves desirable PsP and TTP classification performance superior to other state-of-the-art methods. Therefore, the developed model would be useful in the diagnosis of PsP and TTP for GBM.

RADIO-IBAG: RADIOMICS-BASED INTEGRATIVE BAYESIAN ANALYSIS OF MULTIPLATFORM GENOMIC DATA.

Technological innovations have produced large multi-modal datasets that include imaging and multi-platform genomics data. Integrative analyses of such data have the potential to reveal important biological and clinical insights into complex diseases like cancer. In this paper, we present Bayesian approaches for integrative analysis of radiological imaging and multi-platform genomic data, wherein our goals are to simultaneously identify genomic and radiomic, i.e., radiology-based imaging markers, along with the latent associations between these two modalities, and to detect the overall prognostic relevance of the combined markers. For this task, we propose Radio-iBAG: Radiomics-based Integrative Bayesian Analysis of Multiplatform Genomic Data, a multi-scale Bayesian hierarchical model that involves several innovative strategies: it incorporates integrative analysis of multi-platform genomic data sets to capture fundamental biological relationships; explores the associations between radiomic markers accompanying genomic information with clinical outcomes; and detects genomic and radiomic markers associated with clinical prognosis. We also introduce the use of sparse Principal Component Analysis (sPCA) to extract a sparse set of approximately orthogonal meta-features each containing information from a set of related individual radiomic features, reducing dimensionality and combining like features. Our methods are motivated by and applied to The Cancer Genome Atlas glioblastoma multiforme data set, where-in we integrate magnetic resonance imaging-based biomarkers along with genomic, epigenomic and transcriptomic data. Our model identifies important magnetic resonance imaging features and the associated genomic platforms that are related with patient survival times.

Prognostic significance of contactin 3 expression and associated genes in glioblastoma multiforme.

Contactin 3 (CNTN3) is a member of the contactin family that is primarily expressed in the nervous system. However, to the best of our knowledge, expression of contactin and its role in the development and progression of brain tumours has not been studied. Although glioblastoma multiforme (GBM) is the most common malignant brain tumour, advances in therapeutic options for patients with GBM have been modest due to an incomplete understanding of the molecular mechanisms underlying development and progression. The aim of the present study was to examine the correlation between CNTN3 and its associated genes and the clinical outcome in patients with GBM. CNTN3 and the expression levels of associated genes were analysed in GBM datasets obtained from the SAGE Anatomical viewer website, Gene Expression Omnibus, Oncomine and The Cancer Genome Atlas. CNTN3 was significantly downregulated in patients with GBM. Subsequently, the expression of CNTN3 was further validated using immunohistochemistry in a cohort of GBM specimens. The immunohistochemistry results were consistent with the in silico analyses. Kaplan-Meier analysis indicated that patients with lower expression levels of CNTN3 had a significantly shorter overall survival (OS) time compared with patients with higher levels of CNTN3 expression. Univariate and multivariate Cox regression analyses demonstrated that CNTN3 expression was an independent prognostic indicator in patients with GBM. Furthermore, gene set enrichment analysis revealed that CNTN3 was associated with the receptor tyrosine-protein kinase (ErbB) signalling pathway. In the ErbB signalling pathway, epidermal growth factor receptor (EGFR) was negatively correlated with CNTN3. Taken together, these data suggest that lower expression levels of CNTN3 may be an independent biomarker that predicts poor OS time in patients with GBM, and that EGFR expression in the ErbB pathway may be associated with CNTN3 expression.