Type Of Glioma Research Articles

Diffusion kurtosis imaging-based habitat analysis identifies high-risk molecular subtypes and heterogeneity matching in diffuse gliomas.

High-risk types of diffuse gliomas in adults include isocitrate dehydrogenase (IDH) wild-type glioblastomas and grade 4 astrocytomas. Achieving noninvasive prediction of high-risk molecular subtypes of gliomas is important for personalized and precise diagnosis and treatment. We retrospectively collected data from 116 patients diagnosed with adult diffuse gliomas. Multiple high-risk molecular markers were tested, and various habitat models and whole-tumor models were constructed based on preoperative routine and diffusion kurtosis imaging (DKI) sequences to predict high-risk molecular subtypes of gliomas. Feature selection and model construction utilized Least absolute shrinkage and selection operator (LASSO) and support vector machine (SVM). Finally, the Wilcoxon rank-sum test was employed to explore the correlation between habitat quantitative features (intra-tumor heterogeneity score，ITH score) and heterogeneity, as well as high-risk molecular subtypes. The results showed that the habitat analysis model based on DKI performed remarkably well (with AUC values reaching 0.977 and 0.902 in the training and test sets, respectively). The model's performance was further enhanced when combined with clinical variables. (The AUC values were 0.994 and 0.920, respectively.) Additionally, we found a close correlation between ITH score and heterogeneity, with statistically significant differences observed between high-risk and non-high-risk molecular subtypes. The habitat model based on DKI is an ideal means for preoperatively predicting high-risk molecular subtypes of gliomas, holding significant value for noninvasively alerting malignant gliomas and those with malignant transformation potential.

HGG-17. DECIPHERING GENETICS AND EPIGENETICS OF HIGH-GRADE GLIOMAS WITH BRAF ACTIVATING ALTERATION IS SUPPORTING THE DISCOVERY OF A NEW INTRACRANIAL SARCOMA-LIKE BRAF POSITIVE ENTITY

Abstract BACKGROUND The current WHO classification of Central Nervous System tumors recognizes several glioma types driven by the BRAFV600E mutation. These tumors mostly encompass pediatric low-grade gliomas. Nevertheless, a subset of them behaves as anaplastic gliomas and, currently, the unique predictive criterion linked to anaplasia is an elevated mitotic index (more than 2.5 mitoses/mm2). Histologically, the high-grade gliomas harboring BRAFV600E alteration mostly correspond to WHO grade 3 pleomorphic xanthoastrocytoma, IDH wild-type glioblastoma, entering the WHO grade 4 subset like the epithelioid subtype, and the rarer high-grade gliomas with piloïd features. These tumor types share their histomolecular features making differential diagnoses difficult. In this setting, DNA methylation profiling may be very useful to precise the integrative diagnosis. However, some of those high-grade forms failed in our experience to be classified in any methylation class with v11b4 version of the Heidelberg classifier. METHODS To understand the reasons explaining this standard classification failure, we investigated with multi-omics approaches (epigenetics, next-generation sequencing, RNAseq, proteomics and metabolomics) a well clinical characterized cohort of 17 high-grade pediatric intracranial tumors harboring this mutation. RESULTS Interestingly, almost all tumors show similar morphological features and associated passenger molecular alterations mostly CDKN2A homozygous deletion. When using the new v12.5 brain tumor classifier, the methylation profiling of the 2/17 unclassified tumors was clustering with a new MPNST (Malignant Peripheral Nerve Sheath Tumor) methylation entity. The RNAseq and proteomics analyses were showing a different molecular pattern from standard BRAF-mutated high-grade gliomas, even the clinical, radiological and histomolecular characteristics were comparable. CONCLUSION So, methylation profiling should be part of the mandatory molecular assessments to perform when a BRAF-mutated tumor is seen by neuropathologists. Furthermore, the multiomics analyses helped us to define a specific molecular pattern of this new BRAF-driven entity, where frequency and prognosis need to be determined on another larger series.

The molecular signature and prognosis of glioma with preoperative intratumoral hemorrhage: a retrospective cohort analysis

BackgroundIntratumoral hemorrhage, though less common, could be the first clinical manifestation of glioma and is detectable via MRI; however, its exact impacts on patient outcomes remain unclear and controversial. The 2021 WHO CNS 5 classification emphasised genetic and molecular features, initiating the necessity to establish the correlation between hemorrhage and molecular alterations. This study aims to determine the prevalence of intratumoral hemorrhage in glioma subtypes and identify associated molecular and clinical characteristics to improve patient management.MethodsIntegrated clinical data and imaging studies of patients who underwent surgery at the Department of Neurosurgery at Peking Union Medical College Hospital from January 2011 to January 2022 with pathological confirmation of glioma were retrospectively reviewed. Patients were divided into hemorrhage and non-hemorrhage groups based on preoperative magnetic resonance imaging. A comparison and survival analysis were conducted with the two groups. In terms of subgroup analysis, we classified patients into astrocytoma, IDH-mutant; oligodendroglioma, IDH-mutant, 1p/19q-codeleted; glioblastoma, IDH-wildtype; pediatric-type gliomas; or circumscribed glioma using integrated histological and molecular characteristics, according to WHO CNS 5 classifications.Results457 patients were enrolled in the analysis, including 67 (14.7%) patients with intratumoral hemorrhage. The hemorrhage group was significantly older and had worse preoperative Karnofsky performance scores. The hemorrhage group had a higher occurrence of neurological impairment and a higher Ki-67 index. Molecular analysis indicated that CDKN2B, KMT5B, and PIK3CA alteration occurred more in the hemorrhage group (CDKN2B, 84.4% vs. 62.2%, p = 0.029; KMT5B, 25.0% vs. 8.9%, p = 0.029; and PIK3CA, 81.3% vs. 58.5%, p = 0.029). Survival analysis showed significantly worse prognoses for the hemorrhage group (hemorrhage 18.4 months vs. non-hemorrhage 39.1 months, p = 0.01). In subgroup analysis, the multivariate analysis showed that intra-tumoral hemorrhage is an independent risk factor only in glioblastoma, IDH-wildtype (162 cases of 457 overall, HR = 1.72, p = 0.026), but not in other types of gliomas. The molecular alteration of CDK6 (hemorrhage group p = 0.004, non-hemorrhage group p < 0.001), EGFR (hemorrhage group p = 0.003, non-hemorrhage group p = 0.001), and FGFR2 (hemorrhage group p = 0.007, non-hemorrhage group p = 0.001) was associated with shorter overall survival time in both hemorrhage and non-hemorrhage groups.ConclusionsGlioma patients with preoperative intratumoral hemorrhage had unfavorable prognoses compared to their nonhemorrhage counterparts. CDKN2B, KMT5B, and PIK3CA alterations were associated with an increased occurrence of intratumoral hemorrhage, which might be future targets for further investigation of intratumoral hemorrhage.

Convolutional neural network and glioma genotyping: A systematic review.

e14038 Background: Gliomas are the most common primary intracranial tumors. Overall, there are 23 different glioma types, and the grading and prognosis of each type are determined by their specific mutations. Several identified mutations are IDH gene status, CDKN2A/B homozygote delegation, PTEN mutation, p53 mutation, TERT promotor mutation, H3K27M mutation, MGMT promotor methylation, trisomy 7/monozomy10, EGFR amplification, ATRX mutation, 1p/19q codeletion, and BRAF mutation. These specific mutations and genomic features can have their distinctive radiologic features seen in different magnetic resonance imaging (MRI) sequences. By using a Convolutional Neural Network (CNN) for feature extraction, the genotyping power of MRI for glioma could be increased. To address this, we systematically reviewed the use of different CNN-based models to detect the genetic context of glioma based on the MRI images. Methods: We conducted a systematic search in PubMed, Embase, Scopus, Web of Science, and Google Scholar. We included original peer-reviewed human studies (retrospective or prospective), focusing on the use of CNN in individuals with glioma. We excluded animal studies, studies with no CNN-based models, reviews, book chapters, conference abstracts, and case reports. JBI checklists were utilized for quality assessment. Results: The initial search resulted in 973 articles. after removing duplicates, 563 studies entered the screening process by title and abstract. 59 studies met our inclusion criteria (56 retrospective and 3 prospective cohort studies) with a total of 26434 glioma patients included (studies ranged from 21 – 2648 individuals). 31 studies used CNN methods only in different stages of image processing (tumor segmentation, feature extraction, image encoder) and 28 studies used CNN as a classifier and mutation predictor. ResNet was the most commonly used CNN-based model (n = 14). U-net (n = 7) and DenseNet models (n = 4) were other commonly used CNN-based models. Included studies targeted IDH status (n=44), 1p/19q codeletion (n=18), MGMT promotor methylation (n=11), ATRX mutation (n=5), TERT promotor mutation (n=5), CDKN2A/B homozygote deletion (n=3), PTEN mutation (n=3), p53 mutation (n=3), H3K27M mutation (n=2), trisomy 7/monozomy10 (n=2), EGFR amplification (n=2), and BRAF mutation (n=1). Highly severe methodological heterogeneity among these studies excluded the possibility of meta-analysis. Conclusions: Current studies have shown that, as a non-invasive method, using CNN-based models on multiple MRI sequences increases the power of early detection and prediction of glioma genetic types before pathological confirmation and has promising results. However, DL performance varies among mutations. Combining multiple DL models with clinical and radiomic features in different and larger databases can further increase the final results.

Development of Predicting Nomograms for Diffuse Astrocytoma and Anaplastic Astrocytoma: A Study Based on the Surveillance, Epidemiology, and End Results Database

Astrocytoma is a type of adult-type diffuse gliomas that includes diffuse astrocytoma (DA) and anaplastic astrocytoma (AA).However, comprehensive investigations into the risk assessment and prognosis of DA and AA using population-based studies remain noticeably scarce. In this study, we developed two predictive nomograms to evaluate the susceptibility and prognosis associated with DA and AA. The study cohort comprised 3,837 individuals diagnosed with DA or AA between 2010 and 2019 selected from the Surveillance, Epidemiology, and End Results (SEER) database. Independent predictors were identified and used to construct the nomograms for overall death (OD) and cancer-specific death (CSD) rates. The performance of the models was assessed using c-index, calibration curves, and receiver operating characteristic curve (ROC), and the clinical applicability was evaluated using decision curve analysis (DCA). The ROC curves in this study show excellent clinical applicability and predictive power. Notably, the area under the curves (AUCs) of the training and verification queues was higher than 0.80, thereby cementing the models' precision. Additionally, the calibration plots demonstrate that the anticipated mortality rates strikingly match with the measured values.This alignment of figures is sustained in the validation cohort. Furthermore, the decision curve analysis (DCA) corroborates the models' translational potential, reinforcing their relevance within real-world clinical settings. The presented nomograms have not only exhibited good predictive performance but also showcased pragmatic clinical utility in prognosticating patient outcomes. Significantly, which will undoubtedly serve as a valuable asset for oncologists, facilitating informed treatment decisions and facilitating meticulous follow-up planning.

Pediatric-type high-grade gliomas with PDGFRA amplification in adult patients with Li-Fraumeni syndrome: clinical and molecular characterization of three cases

Li-Fraumeni syndrome (LFS) is an autosomal dominant tumor predisposition syndrome caused by heterozygous germline mutations or deletions in the TP53 tumor suppressor gene. Central nervous system tumors, such as choroid plexus tumors, medulloblastomas, and diffuse gliomas, are frequently found in patients with LFS. Although molecular profiles of diffuse gliomas that develop in pediatric patients with LFS have been elucidated, those in adults are limited. Recently, diffuse gliomas have been divided into pediatric- and adult-type gliomas, based on their distinct molecular profiles. In the present study, we investigated the molecular profiles of high-grade gliomas in three adults with LFS. These tumors revealed characteristic histopathological findings of high-grade glioma or glioblastoma and harbored wild-type IDH1/2 according to whole exome sequencing (WES). However, these tumors did not exhibit the key molecular alterations of glioblastoma, IDH-wildtype such as TERT promoter mutation, EGFR amplification, or chromosome 7 gain and 10 loss. Although WES revealed no other characteristic gene mutations or copy number alterations in high-grade gliomas, such as those in histone H3 genes, PDGFRA amplification was found in all three cases together with uniparental disomy of chromosome 17p, where the TP53 gene is located. DNA methylation analyses revealed that all tumors exhibited DNA methylation profiles similar to those of pediatric-type high-grade glioma H3-wildtype and IDH-wildtype (pHGG H3-/IDH-wt), RTK1 subtype. These data suggest that high-grade gliomas developed in adult patients with LFS may be involved in pHGG H3-/IDH-wt. PDGFRA and homozygous alterations in TP53 may play pivotal roles in the development of this type of glioma in adult patients with LFS.

Can the metabolic characteristics of diffuse glioma on &lt;sup&gt;11&lt;/sup&gt;C-methionine PET/CT serve as a marker of its IDH status? Cross sectional study

INTRODUCTION: Since 2016, molecular markers, in particular, mutations in isocitrate dehydrogenase (IDH) 1 and 2, have been introduced as a classifying feature of cerebral gliomas that provided superior prognostication. The search for non-invasive biomarkers of the molecular profile of gliomas is necessary to improve the quality of preoperative diagnostics, identify patients with good and poor prognosis and determine treatment tactics.OBJECTIVE: Was to study the relationship between the IDH genotype of diffuse cerebral gliomas and metabolic biomarkers according to the results of PET/CT with [11C]methionine.MATERIALS AND METHOD: The results of PET/CT with 11C-methionine were identified to a retrospective analysis of 260 patients aged 18 to 75 years (median 40 years) with untreated cerebral glioma. Based on histological and molecular genetic studies of the surgical material including the determination of a mutation in the isocitrate dehydrogenase 1 (IDH1132H) gene, diffuse gliomas were classified according to the 2016 WHO classification of CNS tumors. Metabolic biomarkers included the calculation of tumor-to-brain ratio of 11С-methionine (TBRmax, TBRpeak and TBRmean) as well as the metabolic tumor volume (MTV). Statistics. Non-parametric tests were performed to compare the differences among patient groups. ROC curve analysis was performed to screen the optimal parameter and its best cutoff value for the discrimination of glioma genotype. All data analyses were performed using “Statistica 10,0” and “MedCalc” ststistical software. p-values less than 0.05 were considered statistically significant.RESULTS: According to the 2016 WHO classification astrocytic and oligodendroglial tumors of the adult type were divided into three groups: astrocytic gliomas with a mutation in the IDH1 gene (IDH1 mut) (n=95), astrocytic gliomas without a mutation in the IDH1 gene (IDH1 wild type — IDH1 wt) (n=103), and IDH1-mutant oligodendrogliomas (n=62). Significant differences in all ratios between the three molecular groups of gliomas were established. TBRmax cutoff of 2.27 differentiated between IDH1 wt and IDH1 mut gliomas with a sensitivity of 61% and a specificity of 77% (area under curve — AUC 0.752). When considering subgroups of gliomas that are homogeneous in terms of the IDH1 status or Grade, the dependence of TBR on the glioma histotype and grading was additionally established. In IDH1 mut oligodendrogliomas, TBR was significantly higher than in mutant astrocytomas, and in IDH1 wt astrocytomas, significant differences in TBR were established between Grade 2 and Grade 3–4. TBRmax was not a predictor of glioma type according to the WHO 2016 classification due to significant overlap of individual of TBR values. But TBRmax allowed diagnosing a cluster of malignant gliomas, including glioblastoma and astrocytoma Grade 3 IDH wt, as well as oligodendroglioma Grade 3 IDH1 mut, with a sensitivity of 65% and a specificity of 89% (AUC 0.848) at a cutoff of TBR=2.7. A strong correlation between the three tumor-to-brain ratios allows any ratio to be used in diagnostics. There were no significant differences in MTV between molecular types of gliomas.DISCUSSION: Distinguishing glioma types based on the 2016 WHO classification of the CNS tumors on the basis of 11Cmethionine uptake seems to be not reliable due to many factors that affect its uptake. In astrocytomas high TBR is associated with malignant grade and wild type IDH1 gene. However, the lack of differences in TBR between these astrocytomas and Grade 3 IDH1-mutant oligodendrogliomas does not allow one to predict the IDH1 status of the tumor in the absence of other radiological signs of the glioma histotype. The absence of differences in TBR between Grade 2 and Grade 3 astrocytomas IDH1 mut supports the view that they are considered as a single subgroup of lower grade gliomas. CONCLUSION: PET/CT with 11C-methionine has limited potential to assess the IDH status of diffuse gliomas. High TBR is associated with malignant glioma with wild-type IDH1 gene or oligodendroglial structure.

The relationships of platelet-derived growth factor, microvascular proliferation, and tumor cell proliferation in canine high-grade oligodendrogliomas: Immunohistochemistry of 45 tumors and an AFOB-01 xenograft mouse model.

High-grade oligodendroglioma (HGOG) is the most common type of glioma in dogs and expresses platelet-derived growth factor receptor-α (PDGFR-α). Microvascular proliferation is often observed in HGOG. Therefore, the present study investigated the functional relationships between PDGFR-α, microvascular proliferation, and tumor cell proliferation in canine HGOG. The expression of PDGFR-α and PDGF-subunit A (PDGF-A) in tumor cells, as well as endothelial cells and pericytes of tumor-associated microvascular proliferations, in 45 canine HGOGs were examined immunohistochemically. Microvascular proliferation was observed in 24/45 cases (53%). PDGFR-α expression in tumor cells and microvascular proliferations was observed in 45/45 (100%) and 2/24 cases (8%), respectively. Furthermore, PDGF-A expression in tumor cells and microvascular proliferations was detected in 13/45 (29%) and 24/24 cases (100%), respectively. In vitro, stimulation of the canine HGOG cell line AOFB-01 with PDGF-A showed that the doubling time of AOFB-01 cells was significantly shorter with PDGF-A than without PDGF-A. Crenolanib (a PDGFR inhibitor) inhibited AOFB-01 cell proliferation. In vivo, the AOFB-01 xenograft mouse model was treated with crenolanib. Tumor xenografts were smaller in crenolanib-treated mice than in untreated control mice. PDGFR-α expression in tumor cells and PDGF-A expression in microvascular proliferations and tumor cells suggest autocrine and paracrine effects of PDGF-A in canine HGOG. The results of in vitro assays indicate that canine HGOG expresses functional PDGFR-α, which responds to PDGF-A. Therefore, PDGF-A produced by microvascular proliferations and tumor cells may promote the proliferation of PDGFR-α-expressing tumor cells in canine HGOG. PDGFR-α signaling has potential as a therapeutic target.

The mechanism of extracellular CypB promotes glioblastoma adaptation to glutamine deprivation microenvironment

Glioblastoma, previously known as glioblastoma multiform (GBM), is a type of glioma with a high degree of malignancy and rapid growth rate. It is highly dependent on glutamine (Gln) metabolism during proliferation and lags in neoangiogenesis, leading to extensive Gln depletion in the core region of GBM. Gln-derived glutamate is used to synthesize the antioxidant Glutathione (GSH). We demonstrated that GSH levels are also reduced in Gln deficiency, leading to increased reactive oxygen species (ROS) levels. The ROS production induces endoplasmic reticulum (ER) stress, and the proteins in the ER are secreted into the extracellular medium. We collected GBM cell supernatants cultured with or without Gln medium; the core and peripheral regions of human GBM tumor tissues. Proteomic analysis was used to screen out the target-secreted protein CypB. We demonstrated that the extracellular CypB expression is associated with Gln deprivation. Then, we verified that GBM can promote the glycolytic pathway by activating HIF-1α to upregulate the expression of GLUT1 and LDHA. Meanwhile, the DRP1 was activated, increasing mitochondrial fission, thus inhibiting mitochondrial function. To explore the specific mechanism of its regulation, we constructed a si-CD147 knockout model and added human recombinant CypB protein to verify that extracellular CypB influenced the expression of downstream p-AKT through its cell membrane receptor CD147 binding. Moreover, we confirmed that p-AKT could upregulate HIF-1α and DRP1. Finally, we observed that extracellular CypB can bind to the CD147 receptor, activate p-AKT, upregulate HIF-1α and DRP1 in order to promote glycolysis while inhibiting mitochondrial function to adapt to the Gln-deprived microenvironment.

Clinical significance of immune-related antigen CD58 in gliomas and analysis of its potential core related gene clusters

BackgroundThe clinical significance of immune-related antigen CD58 in gliomas remains uncertain. The aim of this study was to examine the clinical importance and possible core related genes of CD58 in gliomas. MethodsPan-cancer analysis was to observe the association between CD58 and different tumors, glioma RNA sequencing data and clinical sample analyses were used to observe the relationship between CD58 and glioma, shRNA interference models were to observe the impact of CD58 on glioma cell function, and four glioma datasets and two online analysis platforms were used to explore the core related genes affecting the correlation between CD58 and glioma. ResultsHigh CD58 expression was associated with worse prognosis in various tumors and higher malignancy in glioma. Down regulation of CD58 expression was linked to decreased proliferation, increased apoptosis, and reduced metastasis in glioma cells. The pathways involved in CD58-related effects were enriched for immune cell adhesion and immune factor activation, and the core genes were CASP1, CCL2, IL18, MYD88, PTPRC, and TLR2. The signature of CD58 and its core-related genes showed superior predictive power for glioma prognosis. ConclusionHigh CD58 expression is correlated with more malignant glioma types, and also an independent risk factor for mortality in glioma. CD58 and its core-related genes may serve as novel biomarkers for diagnosing and treating glioma.

Metabolic models predict fotemustine and the combination of eflornithine/rifamycin and adapalene/cannabidiol for the treatment of gliomas.

Gliomas are the most common type of malignant brain tumors, with glioblastoma multiforme (GBM) having a median survival of 15months due to drug resistance and relapse. The treatment of gliomas relies on surgery, radiotherapy and chemotherapy. Only 12 anti-brain tumor chemotherapies (AntiBCs), mostly alkylating agents, have been approved so far. Glioma subtype-specific metabolic models were reconstructed to simulate metabolite exchanges, in silico knockouts and the prediction of drug and drug combinations for all three subtypes. The simulations were confronted with literature, high-throughput screenings (HTSs), xenograft and clinical trial data to validate the workflow and further prioritize the drug candidates. The three subtype models accurately displayed different degrees of dependencies toward glutamine and glutamate. Furthermore, 33 single drugs, mainly antimetabolites and TXNRD1-inhibitors, as well as 17 drug combinations were predicted as potential candidates for gliomas. Half of these drug candidates have been previously tested in HTSs. Half of the tested drug candidates reduce proliferation in cell lines and two-thirds in xenografts. Most combinations were predicted to be efficient for all three glioma types. However, eflornithine/rifamycin and cannabidiol/adapalene were predicted specifically for GBM and low-grade glioma, respectively. Most drug candidates had comparable efficiency in preclinical tests, cerebrospinal fluid bioavailability and mode-of-action to AntiBCs. However, fotemustine and valganciclovir alone and eflornithine and celecoxib in combination with AntiBCs improved the survival compared to AntiBCs in two-arms, phase I/II and higher glioma clinical trials. Our work highlights the potential of metabolic modeling in advancing glioma drug discovery, which accurately predicted metabolic vulnerabilities, repurposable drugs and combinations for the glioma subtypes.

Machine Learning-Assisted Classification of Paraffin-Embedded Brain Tumors with Raman Spectroscopy.

Raman spectroscopy (RS) has demonstrated its utility in neurooncological diagnostics, spanning from intraoperative tumor detection to the analysis of tissue samples peri- and postoperatively. In this study, we employed Raman spectroscopy (RS) to monitor alterations in the molecular vibrational characteristics of a broad range of formalin-fixed, paraffin-embedded (FFPE) intracranial neoplasms (including primary brain tumors and meningiomas, as well as brain metastases) and considered specific challenges when employing RS on FFPE tissue during the routine neuropathological workflow. We spectroscopically measured 82 intracranial neoplasms on CaF2 slides (in total, 679 individual measurements) and set up a machine learning framework to classify spectral characteristics by splitting our data into training cohorts and external validation cohorts. The effectiveness of our machine learning algorithms was assessed by using common performance metrics such as AUROC and AUPR values. With our trained random forest algorithms, we distinguished among various types of gliomas and identified the primary origin in cases of brain metastases. Moreover, we spectroscopically diagnosed tumor types by using biopsy fragments of pure necrotic tissue, a task unattainable through conventional light microscopy. In order to address misclassifications and enhance the assessment of our models, we sought out significant Raman bands suitable for tumor identification. Through the validation phase, we affirmed a considerable complexity within the spectroscopic data, potentially arising not only from the biological tissue subjected to a rigorous chemical procedure but also from residual components of the fixation and paraffin-embedding process. The present study demonstrates not only the potential applications but also the constraints of RS as a diagnostic tool in neuropathology, considering the challenges associated with conducting vibrational spectroscopic analysis on formalin-fixed, paraffin-embedded (FFPE) tissue.

Abstract 5200: Multiparametric MRI along with machine learning informs on molecular underpinnings, prognosis, and treatment response in pediatric low grade glioma

Abstract We report a comprehensive radiogenomic analysis, including multiparametric MRI and RNA sequencing of 502 tumors across 13 histological types of pediatric low-grade glioma (PLGG). Through an unsupervised analysis of treatment-naïve imaging, we were able to identify 3 distinct imaging clusters that were associated with different molecular pathways according to transcriptomic pathway analysis. Using xCell enrichment scores as input for clustering, we discovered three distinct immunological groups based on immune cell infiltration. The results suggested the presence of an ‘immune-hot” cluster within the tumors with a poorer prognosis, which may potentially benefit from emerging immunotherapies. We subsequently generated a radiomic signature that was able to predict immunological profiles. We also developed a clinicoradiomic predictive model that can assess patients’ risk of progression and predict treatment response. Finally, we evaluated the transcriptomic pathways that were most predictive of clinicoradiomic risk scores and identified the biological pathways most predictive of risk of progression in PLGG, when adjusting for molecular subtype, race, reported gender, age at diagnosis, and the anatomic location of the tumor. This is the first large-scale radiogenomic analysis in PLGG that can aid in prognostication, assessing patients’ risk of progression, predicting treatment response in conventional therapies, and early stratification of patients to identify potential candidates for targeted therapies. Citation Format: Ali Nabavizadeh, Adam Kraya, Komal Rathi, Meen Chul Kim, Nastaran Khalili, Ariana Familiar, Varun Kesherwani, Hannah Anderson, Deep Gandhi, Aria Mahtabfar, Sina Bagheri, Xiaoyan Huang, Yuankun Zhu, Alexander Sickler, Saksham Phul, Sabine Mueller, Mateusz P Koptyra, Philip B Storm, Jeff Ware, Yuanquan Song, Michael J Fisher, Arastoo Vossough, Christos Davatzikos, Adam C Resnick, Anahita Fathi Kazerooni. Multiparametric MRI along with machine learning informs on molecular underpinnings, prognosis, and treatment response in pediatric low grade glioma [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 5200.

Gene Expression of GABAA Receptor Subunits and Association with Patient Survival in Glioma.

Rapid neuronal inhibition in the brain is mediated by γ-aminobutyric acid (GABA) activation of GABAA receptors. The GABRA5 gene, which encodes the α5 subunit of the GABAA receptor, has been implicated in an aggressive subgroup of medulloblastoma (MB), a type of pediatric brain tumor. However, the possible role of GABAA receptor subunits in glioma remains poorly understood. Here, we examined the expression of genes encoding GABAA receptor subunits in different types of glioma, and its possible association with patient prognosis assessed by overall survival (OS). Data were obtained from the French and The Cancer Genome Atlas Brain Lower Grade Glioma (TCGA-LGG) datasets and analyzed for expression of GABAA receptor subunit genes. OS was calculated using the Kaplan-Meier estimate. We found that genes GABRA2, GABRA3, GABRB3, GABRG1, and GABRG2 showed a significant association with OS, with higher gene expression indicating better prognosis. In patients with GBM, high expression of GABRA2 was associated with shorter OS, whereas, in contrast, higher levels of GABRB3 were associated with better prognosis indicated by longer OS. In patients with lower grade gliomas, GABRA3, GABRB3, GABRG1, and GABRG2, were associated with longer OS. High GABRB3 expression was related to longer survival when low grade glioma types were analyzed separately. Our results suggest an overall association between higher expression of most genes encoding GABAA receptor subunits and better prognosis in different types of glioma. Our findings support the possibility that down-regulation of GABAA receptors in glioma contributes to promoting tumor progression by reducing negative inhibition. These findings might contribute to further evaluation of GABAA receptors as a therapeutic target in glioma.

Integral neuro-oncological diagnosis As the basis for personalized treatment Of brain tumors (on the example of gliomas In children)

Brain tumors are the most common group of neoplasms in children, which is in the top-3 causes of infant mortality from oncopathology. The aim of the study is to make a literature review of modern approaches to the personalization of treatment of CNS tumors in children based on the study of molecular genetics, immunohistochemical and imagingт characteristics. The article provides a description of the most common types of neoplasms of the nervous system — low grade gliomas (LGG), high grade gliomas (HGG), mixed glioneural tumors, etc. The molecular genetics, immunohistochemical, visualization characteristics of each type of gliomas are described. Modern information on prognosis and treatment of tumors is also given in the publication. The prognosis of the course of LGG/GNT in children depends more often on the patient’s age, histology, and location of the tumor, as well as its molecular profile. Older patients have a more favorable prognosis than younger children. Well-visualized tumors located superficially have a better outcome than diffuse gliomas, as well as deeply located tumors. Detection of BRAF mutations and FGFR rearrangements may indicate a better prognosis than in the presence of SNV. Local mutations of BRAF with concomitant deletion of CDKN2A, as well as mutations with H3.3 pK27 are the most dangerous.

Gene Expression of CSF3R/CD114 Is Associated with Poorer Patient Survival in Glioma.

Gliomas comprise most cases of central nervous system (CNS) tumors. Gliomas afflict both adults and children, and glioblastoma (GBM) in adults represents the clinically most important type of malignant brain cancer, with a very poor prognosis. The cell surface glycoprotein CD114, which is encoded by the CSF3R gene, acts as the receptor for the granulocyte colony stimulating factor (GCSF), and is thus also called GCSFR or CSFR. CD114 is a marker of cancer stem cells (CSCs), and its expression has been reported in several cancer types. In addition, CD114 may represent one among various cases where brain tumors hijack molecular mechanisms involved in neuronal survival and synaptic plasticity. Here, we describe CSF3R mRNA expression in human gliomas and their association with patient prognosis as assessed by overall survival (OS). We found that the levels of CSF3R/CD114 transcripts are higher in a few different types of gliomas, namely astrocytoma, pilocytic astrocytoma, and GBM, in comparison to non-tumoral neural tissue. We also observed that higher expression of CSF3R/CD114 in gliomas is associated with poorer outcome as measured by a shorter OS. Our findings provide early evidence suggesting that CSF3R/CD114 shows a potential role as a prognosis marker of OS in patients with GBM.

Use of artificial intelligence and radio genomics in neuroradiology and the future of brain tumour imaging and surgical planning in low- and middleincome countries.

Brain tumour diagnosis involves assessing various radiological and histopathological parameters. Imaging modalities are an excellent resource for disease monitoring. However, manual inspection of imaging is laborious, and performance varies depending on expertise. Artificial Intelligence (AI) driven solutions a non-invasive and low-cost technology for diagnostics compared to surgical biopsy and histopathological diagnosis. We analysed various machine learning models reported in the literature and assess its applicability to improve neuro-oncological management. A scoping review of 47 full texts published in the last 3 years pertaining to the use of machine learning for the management of different types of gliomas where radiomics and radio genomic models have proven to be useful. Use of AI in conjunction with other factors can result in improving overall neurooncological management within LMICs. AI algorithms can evaluate medical imaging to aid in the early detection and diagnosis of brain tumours. This is especially useful where AI can deliver reliable and efficient screening methods, allowing for early intervention and treatment.

Probing the glioma microvasculature: a case series of the comparison between perfusion MRI and intraoperative high-frame-rate ultrafast Doppler ultrasound

BackgroundWe aimed to describe the microvascular features of three types of adult-type diffuse glioma by comparing dynamic susceptibility contrast (DSC) perfusion magnetic resonance imaging (MRI) with intraoperative high-frame-rate ultrafast Doppler ultrasound.MethodsCase series of seven patients with primary brain tumours underwent both DSC perfusion MRI and intra-operative high-frame-rate ultrafast Doppler ultrasound. From the ultrasound images, three-dimensional vessel segmentation was obtained of the tumour vascular bed. Relative cerebral blood volume (rCBV) maps were generated with leakage correction and normalised to the contralateral normal-appearing white matter. From tumour histograms, median, mean, and maximum rCBV ratios were extracted.ResultsLow-grade gliomas (LGGs) showed lower perfusion than high-grade gliomas (HGGs), as expected. Within the LGG subgroup, oligodendroglioma showed higher perfusion than astrocytoma. In HGG, the median rCBV ratio for glioblastoma was 3.1 while astrocytoma grade 4 showed low perfusion with a median rCBV of 1.2. On the high-frame-rate ultrafast Doppler ultrasound images, all tumours showed a range of rich and organised vascular networks with visually apparent abnormal vessels, even in LGG.ConclusionsThis unique case series revealed in vivo insights about the microvascular architecture in both LGGs and HGGs. Ultrafast Doppler ultrasound revealed rich vascularisation, also in tumours with low perfusion at DSC MRI. These findings warrant further investigations using advanced MRI postprocessing, in particular for characterising adult-type diffuse glioma.Relevance statementOur findings challenge the current assumption behind the estimation of relative cerebral blood volume that the distribution of blood vessels in a voxel is random.Key points• Ultrafast Doppler ultrasound revealed rich vascularity irrespective of perfusion dynamic susceptibility contrast MRI state.• Rich and organised vascularisation was also observed even in low-grade glioma.• These findings challenge the assumptions for cerebral blood volume estimation with MRI.Graphical

Pediatric low-grade glioma models: advances and ongoing challenges.

Pediatric low-grade gliomas represent the most common childhood brain tumor class. While often curable, some tumors fail to respond and even successful treatments can have life-long side effects. Many clinical trials are underway for pediatric low-grade gliomas. However, these trials are expensive and challenging to organize due to the heterogeneity of patients and subtypes. Advances in sequencing technologies are helping to mitigate this by revealing the molecular landscapes of mutations in pediatric low-grade glioma. Functionalizing these mutations in the form of preclinical models is the next step in both understanding the disease mechanisms as well as for testing therapeutics. However, such models are often more difficult to generate due to their less proliferative nature, and the heterogeneity of tumor microenvironments, cell(s)-of-origin, and genetic alterations. In this review, we discuss the molecular and genetic alterations and the various preclinical models generated for the different types of pediatric low-grade gliomas. We examined the different preclinical models for pediatric low-grade gliomas, summarizing the scientific advances made to the field and therapeutic implications. We also discuss the advantages and limitations of the various models. This review highlights the importance of preclinical models for pediatric low-grade gliomas while noting the challenges and future directions of these models to improve therapeutic outcomes of pediatric low-grade gliomas.

Comparison of diagnostic performance of radiologist- and AI-based assessments of T2-FLAIR mismatch sign and quantitative assessment using synthetic MRI in the differential diagnosis between astrocytoma, IDH-mutant and oligodendroglioma, IDH-mutant and 1p/19q-codeleted

PurposeThis study aimed to compare assessments by radiologists, artificial intelligence (AI), and quantitative measurement using synthetic MRI (SyMRI) for differential diagnosis between astrocytoma, IDH-mutant and oligodendroglioma, and IDH-mutant and 1p/19q-codeleted and to identify the superior method.MethodsThirty-three cases (men, 14; women, 19) comprising 19 astrocytomas and 14 oligodendrogliomas were evaluated. Four radiologists independently evaluated the presence of the T2-FLAIR mismatch sign. A 3D convolutional neural network (CNN) model was trained using 50 patients outside the test group (28 astrocytomas and 22 oligodendrogliomas) and transferred to evaluate the T2-FLAIR mismatch lesions in the test group. If the CNN labeled more than 50% of the T2-prolonged lesion area, the result was considered positive. The T1/T2-relaxation times and proton density (PD) derived from SyMRI were measured in both gliomas. Each quantitative parameter (T1, T2, and PD) was compared between gliomas using the Mann–Whitney U-test. Receiver-operating characteristic analysis was used to evaluate the diagnostic performance.ResultsThe mean sensitivity, specificity, and area under the curve (AUC) of radiologists vs. AI were 76.3% vs. 94.7%; 100% vs. 92.9%; and 0.880 vs. 0.938, respectively. The two types of diffuse gliomas could be differentiated using a cutoff value of 2290/128 ms for a combined 90th percentile of T1 and 10th percentile of T2 relaxation times with 94.4/100% sensitivity/specificity with an AUC of 0.981.ConclusionCompared to the radiologists’ assessment using the T2-FLAIR mismatch sign, the AI and the SyMRI assessments increased both sensitivity and objectivity, resulting in improved diagnostic performance in differentiating gliomas.