IDH Mutation Status Research Articles

Identification of critical biomarkers and immune landscape patterns in glioma based on multi-database

PurposeGlioma is the most prevalent tumor of the central nervous system. The poor clinical outcomes and limited therapeutic efficacy underscore the urgent need for early diagnosis and an optimized prognostic approach for glioma. Therefore, the aim of this study was to identify sensitive biomarkers for glioma.Patients and methodsDifferentially expressed genes (DEGs) of glioma were downloaded from The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO) databases. The potential biomarkers were identified using weighted gene co-expression network analysis (WGCNA) and least absolute shrinkage and selection operator (LASSO) regression. The prognostic ability of the potential biomarkers was evaluated by Cox regression and survival curve. CellMiner was used to access the correlation between the expression of potential biomarkers and anticancer drug sensitivity. We then explored the association of potential biomarkers and tumor immune infiltration by single-sample GSEA (ssGSEA) and CIBERSORT. Immune staining in glioma patient samples and cell experiments of potential biomarkers further verified their expression and function.ResultsUltimately, we identified three potential biomarkers: SLC8A2, ATP2B3, and SRCIN1. These 3 genes were found significantly correlated with clinicopathological features (age, WHO grade, IDH mutation status, 1p19q codeletion status). Furthermore, the overall survival (OS), disease-specific survival (DSS), and progression-free survival (PFS) were found to be positively correlated with high expression of these 3 potential biomarkers. Besides, there was a substantial relationship between the sensitivity of anticancer drugs and these biomarkers expression. More importantly, the negative association between the 3 genes with most tumor immune cells was also established. Moreover, the decreased expression of the biomarkers was also verified in glioma patient samples. Finally, we confirmed that these 3 genes might promotes glioma proliferation and migration in vitro.ConclusionSLC8A2, ATP2B3, and SRCIN1 were identified as underlying biomarkers for glioma associated with prognosis assessments and personal immunotherapy.Graphical

Correlation of microRNAs-10b/21/34a expression levels with IDH1-mutation status in patients with glioblastoma

Introduction/Objective. Isocitrate dehydrogenase mutations play a significant role in gliomagenesis. Specific microRNAs such as miR-10b/21 act as oncogenic microRNAs, while miR-34a acts as tumor suppressors in glioblastoma. Our study aimed to investigate the mutation status of IDH correlate with microRNAs-10b/21/34a expression levels in patients with glioblastoma. Methods. The study included 43 patients diagnosed with glioblastoma. We examined microRNA-10b, microRNA-21 and microRNA-34a expression levels in peripheral blood mononuclear cells after surgery and prior to concurrent radiotherapy with temozolomide, at the 15th and 30th fractions of radiotherapy with temozolomide. The data on IDH1 mutation status were gathered from medical history and histopathology. Results. Two groups were created to assess the association of microRNAs-10b/21/34a expression levels: glio-blastoma IDH1-wild type and glioblastoma IDH1-mutant + Not Other Specified (NOS). The median microRNA-10b expression level before the initiation of concurrent radiotherapy with temozolomide was 130.44 (52.20 - 622.53) in the IDH1-wildtype glioblastoma group and 94.61 (2.13 - 816.89) in the IDH1-mutant + glioblastoma NOS group. The median microRNA-21 expression level was 57.16 (2.68 - 278.98) in the IDH1-wildtype glioblastoma group and 69.74 (4.60 - 825.43) in the IDH1-mutant + glioblastoma NOS group. The median microRNA-34a expression level was 13.52 (3.16 - 105.20) in the IDH1-wildtype glioblastoma group and 10.11 (1.00 - 210.55) in the IDH1-mutant + glioblastoma NOS group. The results showed no statistically significant difference in the expression levels of miR-10b/21/34a between the two groups (p > 0.05). Conclusion. Our results suggest that the IDH1 mutation status may not be a critical factor for altered expression of microRNAs-10b/21/34a in glioblastoma patients.

IDH Status in Brain Gliomas Can Be Predicted by the Spherical Mean MRI Technique.

Diffuse gliomas, a heterogeneous group of primary brain tumors, have traditionally been stratified by histology, but recent insights into their molecular features, especially the IDH mutation status, have fundamentally changed their classification and prognosis. Current diagnostic methods, still predominantly relying on invasive biopsy, necessitate the exploration of noninvasive imaging alternatives for glioma characterization. In this prospective study, we investigated the utility of the spherical mean technique (SMT) in predicting the IDH status and histologic grade of adult-type diffuse gliomas. Patients with histologically confirmed adult-type diffuse glioma underwent a multiparametric MRI examination using a 3T system, which included a multishell diffusion sequence. Advanced diffusion parameters were obtained using SMT, diffusional kurtosis imaging, and ADC modeling. The diagnostic performance of studied parameters was evaluated by plotting receiver operating characteristic curves with associated area under curve, specificity, and sensitivity values. A total of 80 patients with a mean age of 48 (SD, 16) years were included in the study. SMT metrics, particularly microscopic fractional anisotropy (μFA), intraneurite voxel fraction, and μFA to the third power (μFA3), demonstrated strong diagnostic performance (all AUC = 0.905, 95% CI, 0.835-0.976; P < .001) in determining IDH status and compared favorably with diffusional kurtosis imaging and ADC models. These parameters also showed a strong predictive capability for tumor grade, with intraneurite voxel fraction and μFA achieving the highest diagnostic accuracy (AUC = 0.937, 95% CI, 0.880-0.993; P < .001). Control analyses on normal-appearing brain tissue confirmed the specificity of these metrics for tumor tissue. Our study highlights the potential of SMT for noninvasive characterization of adult-type diffuse gliomas, with a potential to predict IDH status and tumor grade more accurately than traditional ADC metrics. SMT offers a promising addition to the current diagnostic toolkit, enabling more precise preoperative assessments and contributing to personalized treatment planning.

Deep learning radiomics nomograms predict Isocitrate dehydrogenase (IDH) genotypes in brain glioma: A multicenter study.

To explore the feasibility of Deep learning radiomics nomograms (DLRN) in predicting IDH genotype. A total of 402 glioma patients from two independent centers were retrospectively included, and the data from center I was randomly divided into a training cohort (n = 239) and an internal validation cohort (n = 103) on a 7:3 basis. Center II served as an independent external validation cohort (n = 60). We developed a DLRN for IDH classification of gliomas based on T2 images. This hybrid model integrates deep learning features, radiomics features, and clinical features most relevant to IDH genotypes and finally classifies them using multivariate logistic regression analysis. We used the area under the curve (AUC) of the receiver operating characteristic (ROC) to evaluate the performance of the model and applied the DLRN score to the survival analysis of some of the follow-up glioma patients. The proposed model had an area under the curve (AUC) of 0.98 in an externally validated cohort, and DLRN scores were significantly associated with the overall survival of glioma patients. Deep learning radiomics nomograms performed well in non-invasively predicting IDH mutation status in gliomas, assisting stratified management and targeted therapy for glioma patients.

Prognostic significance and gene co-expression network of CD16A and FGL2 in gliomas.

The CD16A protein encoding gene FcγRIIIa (FCGR3A) and its potential ligand Fibrinogen-like protein 2 (FGL2) are involved in various cell physiological activities on the extracellular surface. Aberrant expression of these genes has been linked to tumorigenesis. To assess the prognostic significance of FCGR3A and FGL2 transcription expression in glioma and explore their roles in glioma initiation and progression, we utilized multiple online databases, including TCGA, GEPIA, CGGA, cBioPortal, TISCH, LinkedOmics, Ivy Glioblastoma Atlas Project, and Human Protein Atlas. Our analysis revealed that FCGR3A and FGL2 expression was significantly correlated with clinical variables such as age, tumor type, WHO grade, histology, IDH-1 mutation, and 1p19q status. A strong correlation was also observed between the transcriptional expression levels of FCGR3A and FGL2. High expression of both genes predicted poor prognosis in primary and recurrent glioma patients, particularly those with lower grade gliomas. Cox regression analysis further confirmed that elevated expression of FCGR3A and FGL2 were independent prognostic factors for shorter overall survival in glioma patients. Gene co-expression network analysis suggested that FCGR3A, FGL2, and their co-expressed genes were involved in inflammatory activities and tumor-related signaling pathways. Additionally, tissue microarrays from glioma patients at Tiantan Hospital showed significantly higher FCGR3A protein expression in high-grade gliomas compared to low-grade gliomas. In conclusion, our findings suggest that FCGR3A and FGL2 could serve as promising prognostic biomarkers and potential therapeutic targets for glioma patients.

RETROSPECTIVE ANALYSIS OF GLIOBLASTOMA OUTCOMES

Background Glioblastoma (GBM) is one of the most aggressive brain cancers, with a median overall survival (OS) of about 15 months. This retrospective study focuses on 144 GBM cases treated over 12 years in our clinic in Romania, analyzing factors affecting progression-free survival (PFS) and OS, including genetic markers and treatment modalities. Methods This study involved patients treated between 2012 and 2024 at the National Institute of Neurology and Neurovascular Diseases in Bucharest. The cohort included 144 patients, with variables such as age, gender, tumor location, IDH mutation status, Karnofsky Performance Status (KPS), and genetic markers (MGMT methylation and EGFR amplification) analyzed. Survival outcomes were evaluated using Kaplan-Meier curves and log-rank tests. Results - Patient Demographics: The cohort included 64 males (44.4%) and 80 females (55.6%) with a mean age of 60.7 years. - Tumor Characteristics: The most common tumor location was the frontal lobe (31.9%), followed by multilobular GBM (26.3%). Right-side tumors were more prevalent (51.3%). - Genetic Markers: MGMT promoter was methylated in 46 patients (31.9%), and EGFR was amplified in 76 patients (52.7%). - Treatment Modalities: 105 patients (72.9%) received chemotherapy with temozolomide (TMZ), and 116 patients (80.5%) underwent radiotherapy. - Survival Outcomes: Median PFS was 5 months, and median OS was 8.5 months. Kaplan-Meier survival curves indicated significantly increased OS in patients with MGMT methylation undergoing chemotherapy (p&lt;0.005) and radiotherapy (p&lt;0.005). Similarly, OS was significantly higher in patients without EGFR amplification receiving chemotherapy (p&lt;0.005) and radiotherapy (p&lt;0.005). Conclusion The study emphasizes the importance of MGMT methylation and EGFR amplification in predicting survival outcomes in GBM patients. Immediate postoperative adjuvant therapy significantly improves PFS and OS. The findings underscore the need for coordinated care and timely access to adjuvant therapies to enhance survival rates in GBM patients. Second surgical interventions also showed a survival benefit, highlighting their potential role as salvage therapy in recurrent GBM cases.

RTID-04. A MULTICENTER, RANDOMIZED, CONTROLLED PHASE 2B TRIAL OF SURVIVIN VACCINE SURVAXM PLUS ADJUVANT TEMOZOLOMIDE FOR NEWLY-DIAGNOSED GLIOBLASTOMA (SURVIVE)

Abstract BACKGROUND Glioblastoma cells express high levels of survivin protein, whose presentation by MHC class I gets recognized by antibodies and cytotoxic T-lymphocytes (CTLs). A phase 2a trial (NCT02455557) demonstrated a median overall survival (OS) of 25.9 months in newly-diagnosed glioblastoma (nGBM) patients with 15-amino-acid-peptide-conjugate survivin vaccine (SurVaxM), which also stimulated production of survivin-directed antibodies and anti-tumor CTL. This phase 2b trial aims to further investigate efficacy and safety of SurVaxM plus adjuvant temozolomide (TMZ) for nGBM. METHODS SURVIVE is an ongoing, multicenter, randomized, placebo-controlled, investigator- and patient-blinded, phase 2b trial that randomizes nGBM patients receiving standard-of-care therapies to either SurVaxM (SurVaxM arm) or saline (control arm) in 3:2 ratio at 15 sites. Included patients have age ≥18 years, normal organ function, KPS≥70, and receive surgical resection with ≤1cm3 residual MRI contrast enhancement within 72 hours post-resection plus TMZ. Patients with recurrent, multicentric, brainstem- or cerebellar-origin GBM are excluded as well as patients on TTFields or other immunotherapies. All patients undergo 3 phases: (1) vaccine priming (VP) phase, starting ≤4 weeks after completing chemoradiation – 4 doses of SurVaxM or placebo administered q2weeks; (2) TMZ phase, started after ≥1 VP dose; and (3) vaccine maintenance phase, started 8 weeks after VP, with SurVaxM or placebo given q8weeks. TMZ phase overlaps with VP and VM phases. Primary endpoint is median OS. Assuming 1-year-OS of 75% in SurVaxM arm, based on prior trials, and 60% for control, the required sample size is N=228; 137 in SurVaxM and 91 in control arm. Accounting for dropouts, we aimed to enroll 239 patients. A log-rank test stratified by KPS (70-80 vs 90-100), MGMT methylation status, and IDH mutation status will be used for analysis. Secondary endpoints are OS and PFS at 15, 18, and 24 months, median PFS, and toxicity. A single sequential OS-driven interim analysis is planned when 50% of deaths occur. Trial is fully enrolled (NCT05163080).

CTNI-27. SYSTEMIC DELIVERY OF LOW ENERGY RADIOFREQUENCY ELECTROMAGNETIC FIELDS FOR THE TREATMENT OF PATIENTS WITH PRIMARY BRAIN TUMORS

Abstract BACKGROUND Tumor Treating Fields plus maintenance temozolomide (mTMZ) showed a survival advantage in the first-line glioblastoma (GBM) treatment. The survival benefit remains limited in patients with unmethylated MGMT gene promoter. The AutEMsys device emits low-energy amplitude modulated systemic electromagnetic fields (EMF) at patient specific frequencies. It is an outpatient procedure that can be integrated with mTMZ. METHODS This study involved a prospective compassionate access program targeting adult GBM patients using an intrabuccal EMF exposure via AutEMsys in addition to chemoRT+mTMZ. The program explored the effects on survival and patients’ reported quality-of-life (QOL) by EORTC-QL-C30. Patients underwent a 90-minute session of EMF exposure therapy repeated every 2 weeks until progression, deterioration in QOL or death. All patients had pathological specimens for diagnosis, IDH mutation status through NGS techniques, and MGMT promoter region methylation status. RESULTS 31 participants were enrolled, including 29 with GBM and 2 with diffuse midline glioma. 14 patients (45%) received EMF+chemoRT followed by EMF+mTMZ and 17 patients (55%) received EMF+mTMZ (post chemoRT). Survival time was measured from the start of mTMZ to the date of the last follow-up. The median overall survival (mOS) for all 31 patients was 26.0 months (95% CI: 14.4 – 37.2 months). Specifically for 21 patients with unmethylated MGMT, the mOS was 25.0 months (95% CI: 15.9 – 34.1 months). Similarly, for the 28 patients with IDH wildtype status, the mOS remained 26.0 months (95% CI: 9.6 – 42.4 months). The addition of AutEMsys with chemoRT+mTMZ demonstrated an improved role functioning in 79% of patients with 25% of these with &amp;gt;10-point score. CONCLUSION The AutEMsys device has shown promising initial results when added to chemoRT+mTMZ treatment of GBM. These findings warrant further prospective investigation and validation in a multicenter clinical trial setting.

QOL-31. RESULTS OF A TELEHEALTH-BASED PSYCHOLOGICAL INTERVENTION FOR CAREGIVERS OF PATIENTS WITH PRIMARY MALIGNANT BRAIN TUMORS: A MODERATION ANALYSIS OF A RANDOMIZED CONTROLLED TRIAL

Abstract BACKGROUND In a recent randomized controlled trial, caregivers of patients with primary malignant brain tumors (PMBT) who received NeuroCARE—a six-session, telehealth-based psychological intervention—demonstrated improved anxiety and depression symptoms, coping, and self-efficacy at 11-weeks compared to usual care (UC) participants. We now explore how sociodemographic variables, disease-specific factors, and baseline mood symptoms moderate intervention effects on anxiety and depression symptoms. METHODS We examined data from 120 enrolled caregivers (Mage=53 years; 83% female, 92% White, 70% spouses/partners) of PMBT patients (84% glioblastoma). Eligible caregivers had elevated anxiety (Generalized Anxiety Disorder-7 ≥ 5). Participants were randomized 1:1 to NeuroCARE or UC and completed baseline sociodemographic questionnaires and the Hospital Anxiety and Depression Scale at baseline and 11-weeks. We used linear regression to evaluate whether selected sociodemographic characteristics (gender, age, relationship to patient), tumor characteristics (tumor location/laterality, IDH mutation status), and baseline mood symptoms moderated intervention effects on anxiety and depression symptoms. We report the group-by-moderator (b) effects, using a 0.20 significance level. RESULTS The beneficial effects of NeuroCARE on reducing 11-week anxiety symptoms were greater in women compared to men (b=-2.63, CI: -6.17, 0.91, p=0.14) and greater in caregivers who reported higher baseline anxiety symptoms (b=-0.43, CI: -0.78, -0.08, p=0.02) and depression symptoms (b=-0.56, CI: -0.96, -0.17, p=0.01). The beneficial effects of NeuroCARE on reducing 11-week depression symptoms were greater in caregivers of patients with IDH-wildtype tumors compared to caregivers of patients with IDH-mutant tumors (b=2.49, CI: -0.26, 5.24, p=0.08). Caregiver age, relationship to patient, and tumor location/laterality did not significantly moderate NeuroCARE effects on mood. CONCLUSIONS The positive effects of NeuroCARE were particularly salient for caregivers with elevated anxiety and/or depression symptoms and those caring for patients with IDH-wildtype tumors. Content may be further tailored to account for the unique needs of men and caregivers of patients with IDH-mutant tumors.

PATH-62. CORRELATES OF NEURAL SIGNALING ELUCIDATE SURVIVAL BENEFIT IN GLIOMAS

Abstract Recent literature has highlighted neural signaling markers (SM) as drivers for synaptogenesis, invasion, and proliferation in gliomas. Further work on elucidating key mutations can provide improved prognostication and potential treatment targets. Our objective is to identify correlates of neural signaling in relation to glioma histopathology and patient outcomes. We used a combination of bulk RNA sequencing (RNA-seq) and a DNA gene panel (DPG) by Tempus next-generation sequencing. Differential gene expression (DGE) was performed using DESeq2 after subsetting genes for neural SM. Dimensionality reduction was applied through t-SNE projections. Variant calls were analyzed using EnsemblVEP and maftools. Survival analysis was determined through log-rank tests. All analyses were conducted with R. 176 and 184 patient samples were analyzed for RNA-seq and DPG, respectively. t-SNE of DGE highlights two distinct functional clusters with no correlation to WHO classification status (Chi-squared p = 0.39) The median survival for the big cluster was observed at 957 days, while no median survival was reached even after 4,420 days of follow-up in the small cluster (p = 0.01). This contrast remains consistent when examining subsets of gliomas with unmethylated MGMT promoter (p = 0.02) and glioblastomas (p = 0.02), but not others such as IDH mutation status and low-grade gliomas (p &amp;gt; 0.1). Genes upregulated in the small cluster had equal representation of synaptic and paracrine SM (45% and 55%), while the big cluster had a greater representation of paracrine SM (69% and 31%). DPG of low-grade gliomas highlighted 3 somatic signaling mutations (FGFR1, PTPRD, and SLIT2) that are more prevalent in the big cluster. Functional classification based on neural signaling markers can elucidate a novel set of biomarkers that may impact survival in glioma patients. Further studies are required to validate such findings and develop therapeutic targets.

Identification of a coagulation-related gene signature for predicting prognosis in recurrent glioma

BackgroundRecurrent gliomas rapidly progress and have high mortality and poor prognosis in the central nervous system. Therefore, further investigation of prognostic and therapeutic markers is needed.MethodsThe mRNA expression, clinical data, and coagulation-related genes (CRGs) associated with recurrent glioma were obtained and calculated from the Chinese Glioma Genome Atlas and Kyoto Encyclopedia of Genes and Genomes databases. The significant CRGs were calculated by Weighted gene co-expression network analysis and PPI network. A prediction model was constructed using the least absolute shrinkage and selection operator regression analysis. Recurrent gliomas were stratified into high and low-risk groups based on the median risk score (RS). The Kaplan–Meier curve was used to analyze the difference in overall survival (OS) between these groups, while the receiver operating characteristic (ROC) curve was used to evaluate the accuracy of the gene model at 1-, 3-, and 5-years. Clinical factors, including age, gender, MGMT methylation status, radiotherapy, chemotherapy, and RS, were included in the univariate and multivariate regression analysis. A prognostic nomogram and calibration curve were established based on these factors.ResultsOverall, seven CRGs associated with the prognosis were identified, including BTK, ITGB1, GNAI3, CFH, LYN, CFI, and F3. OS and survival rates were lower in the high-risk group compared with the low-risk group. The ROC curve demonstrated the area under the curve values >0.65 at 1-, 3-, and 5-years, confirming the reliability of the prognostic model. The univariate regression analysis indicated that tumor grade (grades 2, 3, and 4), histopathology (GBM or not), chemotherapy, IDH mutation, and 1p19q co-deletion status were independent prognostic indicators. Univariate and multivariate regression analyses indicated that RS was an independent prognostic factor for patients with recurrent glioma. Immune analysis revealed low infiltration of resting dendritic cells and high expression of programmed death receptor 1 in the high-risk group.ConclusionThis study comprehensively investigated the correlation between CRGs and recurrent glioma prognosis, offering crucial insights for further research into glioma recurrence mechanisms and treatment strategies.

EPID-28. TRENDS IN LOW GRADE GLIOMA PROGNOSIS AND THE EVOLVING IMPACT OF GROSS TOTAL RESECTION ACROSS FIFTY YEARS: A MULTICOHORT ANALYSIS

Abstract OBJECTIVE Significant advancements in neurosurgical planning, approach, and practice over the past five decades have improved the surgical management of diffuse low and high-grade gliomas. We sought to evaluate the extent to which LGG survival outcomes have evolved over the past fifty years using glioma registry datasets with histopathological groups defined by WHO diagnostic criteria and supplemented with next generation genome-sequencing. METHODS Three datasets (SEER, TCGA, and GENIE) were used to evaluate trends in prognosis following LGG diagnosis and resection between 1975 and present (GBM as a comparator cohort). The primary outcomes evaluated were longitudinal disease-specific survival and overall survival, with secondary fixed-timepoint outcomes evaluated at 12-months, 24-months, and 60-months. Subjects from SEER were categorized by histology, while subjects from TCGA and GENIE were included based IDH mutation status with available histopathology. RESULTS A total of 39,958 adult patients were included. Overall survival increased for both LGG and GBM between 1975 and 2010, however only LGG outcomes continued to improve following 2010 (2010-; HR=0.955, 95%CI 0.926 to 0.986). Evaluation of the interaction between resection extent and diagnosis year revealed a progressive increase in the survival benefit of GTR in recent years, particularly for LGG. Independent of other available clinical/molecular features and considering the interaction term between year of diagnosis and resection extent, the gap in independent prognostic benefit associated with GTR between LGG and GBM has significantly widened (2010-2020 vs 1990-2000, p&amp;lt;0.05). CONCLUSION Over the past three decades, consistent improvements in LGG outcomes were observed while prognosis following GBM diagnosis has remained largely unchanged following 2010. Notably, we observed increasing prognostic value of achieving GTR particularly in patients with LGG.

EPID-19. INTERACTION BETWEEN GERMLINE AND SOMATIC GENETICS TO ELUCIDATE RISK OF ADULT DIFFUSE GLIOMA

Abstract Acquired tumor alterations are part of the WHO diagnosis of glioma. However, germline predisposition to these alterations is unknown. GWAS in European and Asian populations identified germline variants in 30 regions, many in or near genes that also acquire alterations, suggesting that interactions between inherited and acquired genetics are involved in glioma development. To evaluate such interactions, we used 3686 glioma cases and 1889 controls. A case-control design and multinomial logistic regression were used to identify germline variants associated with development of TERT-promoter mutant gliomas, generating two interesting observations. First, TERT germline variants were associated with predisposition to IDHwt glioma that have a TERT promoter mutation, but not with IDH-mutant codeleted gliomas that have a TERT promoter mutation. Second, two novel regions were observed in this more homogeneous subgroup of TERT mutated tumors: variants in TXNDC11 (OR=2.66, p=5.7x10-8) and RUVBL2 (OR=3.1, p=5.8x10-8). High tumor expression of TXNDC11 is associated with poor prognosis in glioma and RUVBL2 interacts with TERT to affect telomerase activity. Similar analyses are being performed on additional clinically relevant acquired alterations. In parallel, case-case GWAS analyses were performed to evaluate germline variants associated with other tumor alterations. Comparing IDH-mutant versus IDHwt stratified by rs55705857 genotype, the most significant variants were PHLDB1 (p=1.2x10-13) and D2HGDH (p=2.1x10-10), highlighting the importance of these two variants independent of CCDC26 for the development of IDH-mutant glioma. As a parallel approach, we utilized recursive partitioning and regression trees (rpart) to classify IDH mutation status using the 30 known variants. The top four branches of the regression tree included CCDC26, D2HGDH, PHLDB1, and MAML2, as well as two Asian variants (RAB27A, CYP4F12). This highlights the importance of utilizing variants discovered in non-European populations in analyses. Overall, understanding these interactions are necessary to develop therapeutics, stratify clinical trials, and to develop relevant tumor models.

EPID-02. BLOOD-BASED CLASSIFICATION OF IDH-STATUS OF NON-ENHANCING GLIOMAS: A MACHINE LEARNING APPROACH

Abstract BACKGROUND Non-invasive determination of IDH mutational status in patients with glioma could offer significant therapeutic opportunities. While IDH wildtype (WT) tumors typically show enhancement on MRI, IDH mutant (MUT) tumors often lack this enhancement. However, relying solely on anatomic radiology may lead to misclassification, and currently, tissue acquisition is the primary method for assessing IDH-status in gliomas. These limitations hamper the development of neoadjuvant or intraoperative therapeutic strategies based on IDH-status; thus, more minimally invasive methods to determine IDH-status are needed. In this study, we assessed peripheral immune cell proportions, which vary by glioma subtype, from pre-surgery peripheral blood samples as an alternative method of classifying the IDH-status in gliomas without enhancement. MATERIALS AND METHODS We employed a highly accurate large language model (GPT-4-Turbo-128k) with over 99% accuracy to read radiology notes and exclude patients with enhancing gliomas. Then, we identified 12 immune cell subtypes from whole blood using deconvolution algorithms based on DNA methylation data. These immune cell subtypes, along with patient age, were integrated into a machine learning model (random forest) to predict IDH-status (WT vs. MUT), leveraging conditional Generative Adversarial Networks to generate synthetic data and mitigate bias in the datasets. Two independent datasets were included for training and validating the model. RESULTS The random forest model had an AUC of 0.90 and accurately identified IDH-status in 81% of a training set of 287 gliomas (65 WT and 222 MUT at varied time points after diagnosis). In an independent validation data set of 99 gliomas (6 WT and 93 MUT) from pre-surgery blood samples, the AUC was 0.93, and accurately predicted IDH-status in 92% of cases with a sensitivity of 93% and specificity of 83%. CONCLUSIONS We predicted the IDH-status of non-enhancing gliomas using patients’ age and immunomethylomic data from pre-surgical peripheral blood samples. This minimally invasive approach is a promising step toward reducing risk and enabling earlier therapies based on IDH-status.

Mitochondrial Iron Metabolism as a Potential Key Mediator of PD-L1 Thermal Regulation.

Glioblastoma (GBM) is the most common primary brain malignancy in the U.S. with a 5-year overall survival < 5% despite an aggressive standard of care. Laser interstitial thermal therapy (LITT) is a surgical approach to treating GBM that has gained traction, providing a safe option for reducing intracranial tumor burden. LITT is believed to potentially modulate GBM immune responses; however, the biochemical mechanisms underlying the modulation of immune checkpoints in GBM cells have been poorly characterized. The present study aimed to preliminarily evaluate the effects of thermal therapy and radiation on PD-L1 modulation in vitro, as a function of IDH mutational status. U87 cells and their IDH-mutant counterpart (U87R132H), which was generated using a crispr-cas9 knock-in approach, were utilized for this preliminary evaluation. Cell heating was achieved by harvesting with trypsin centrifugation where the cell pellets were treated on a heat block for the associated time and temperature. Following thermal therapy, cells were resuspended and irradiated using a 37-Cesium irradiator at 0.6 Gy min-1. Immediately following treatment, cells were either plated as single cells to allow colonies to form, and stained with Coomassie blue to be counted approximately 10-14 days later or harvested for Western blot analysis. Cell lysates were analyzed for PD-L1 expression with respect to various iron metabolic parameters (mortalin (HSPA9), transferrin receptor, and ferritin heavy chain) using a Western blotting approach. In both U87 and U87R132H cell lines, thermal therapy showed a temperature-dependent cell-killing effect, but U87R132H cells appeared more sensitive to thermal treatment when treated at 43 °C for 10 min. Moreover, thermal therapy had minimal effects on cell responses to 2 Gy irradiation. Treatment with thermal therapy downregulated PD-L1 expression in U87R132H cells, which was associated with increased expression of the mitochondrial iron metabolic enzyme, HSPA9. Thermal therapy reversed the radiation-induced overexpression of PD-L1, transferrin receptor, and ferritin heavy chain in U87R132H cells. No effects were observed in wild-type U87 cells. Moreover, Ga(NO3)3 depleted mitochondrial iron content which, in turn, significantly enhanced the sensitivity of U87R132H cells to thermal therapy and 2 Gy irradiation and caused a significant increase in PD-L1 expression. These results suggest that thermal therapy alone can modulate the immune checkpoint PD-L1. This effect was more pronounced when thermal therapy was combined with radiation. Mechanistically, mitochondrial iron trafficking through HSPA9 may coordinate the regulation of PD-L1 in the context of thermal therapy and ionizing radiation, which can be targeted with gallium-based therapy. These novel, preliminary findings warrant further mechanistic investigations in pre-clinical models of LITT.

Predicting IDH and ATRX mutations in gliomas from radiomic features with machine learning: a systematic review and meta-analysis.

This systematic review aims to evaluate the quality and accuracy of ML algorithms in predicting ATRX and IDH mutation status in patients with glioma through the analysis of radiomic features extracted from medical imaging. The potential clinical impacts and areas for further improvement in non-invasive glioma diagnosis, classification and prognosis are also identified and discussed. The review followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses of Diagnostic and Test Accuracy (PRISMA-DTA) statement. Databases including PubMed, Science Direct, CINAHL, Academic Search Complete, Medline, and Google Scholar were searched from inception to April 2024. The Quality Assessment of Diagnostic Accuracy Studies (QUADAS-2) tool was used to assess the risk of bias and applicability concerns. Additionally, meta-regression identified covariates contributing to heterogeneity before a subgroup meta-analysis was conducted. Pooled sensitivities, specificities and area under the curve (AUC) values were calculated for the prediction of ATRX and IDH mutations. Eleven studies involving 1,685 patients with grade I-IV glioma were included. Primary contributors to heterogeneity included the MRI modalities utilised (conventional only vs. combined) and the types of ML models employed. The meta-analysis revealed pooled sensitivities of 0.682 for prediction of ATRX loss and 0.831 for IDH mutations, specificities of 0.874 and 0.828, and AUC values of 0.842 and 0.948, respectively. Interestingly, incorporating semantics and clinical data, including patient demographics, improved the diagnostic performance of ML models. The high AUC in the prediction of both mutations demonstrates an overall robust diagnostic performance of ML, indicating the potential for accurate, non-invasive diagnosis and precise prognosis. Future research should focus on integrating diverse data types, including advanced imaging, semantics and clinical data while also aiming to standardise the collection and integration of multimodal data. This approach will enhance clinical applicability and consistency.

P10.09.A BLOOD-BASED CLASSIFICATION OF IDH-STATUS OF NON-ENHANCING GLIOMAS: A MACHINE LEARNING APPROACH

Abstract BACKGROUND Non-invasive determination of IDH mutational status in patients with glioma could offer significant therapeutic opportunities. While IDH wildtype tumors (e.g., GBM) typically show enhancement on MRI, IDH mutant tumors often lack this enhancement. However, relying solely on anatomic radiology may lead to misclassification, and currently, tissue acquisition is the primary method for assessing IDH status in gliomas. This limitation hampers the development of neoadjuvant or intraoperative therapeutic strategies based on IDH status; thus, more minimally invasive methods to determine IDH status are needed. In this study, we assessed immune cell proportions, which vary by glioma subtype, from pre-surgery peripheral blood samples as an alternative method of classifying the IDH status in gliomas without enhancement. MATERIAL AND METHODS First, we employed a highly accurate large language model (GPT-4-Turbo-128k) with over 99% accuracy to read radiology notes and exclude patients with enhancing gliomas. Then, we identified twelve immune cell subtypes from whole blood using deconvolution algorithms based on DNA methylation data. These immune cell subtypes, along with patient age, were integrated into a machine learning model (random forest) to predict IDH status (wildtype vs. mutant), leveraging conditional Generative Adversarial Networks to generate synthetic data and mitigate bias in the datasets. Two independent datasets were included for training and validating the model. RESULTS The random forest model had an area under the ROC curve (AUC) of 0.90 and accurately identified IDH status in 81% of a training set of 287 gliomas (65 wildtype and 222 mutant tumors at varied time points after diagnosis). In an independent validation data set of 99 gliomas (6 wildtype and 93 mutant) from pre-surgery blood samples, the AUC was 0.93, and accuracy predicted IDH status in 92% of cases with a sensitivity of 93% and specificity of 83%. CONCLUSION We could accurately predict IDH status in pre-surgical peripheral blood samples from patients with non-enhancing gliomas. This minimally invasive approach is a promising step toward reducing risk and exploring opportunities for earlier therapies based on IDH status.

Cognitive phenotypes: Unraveling the heterogeneity in cognitive dysfunction among patients with primary brain tumors receiving radiotherapy.

Patients with primary brain tumors demonstrate heterogeneous patterns of cognitive dysfunction, which we explore using latent profile analysis (LPA) to identify cognitive phenotypes and their trajectories in patients receiving radiotherapy (RT). Ninety-six patients completed neuropsychological testing before and post-RT (3, 6, 12-months) on a prospective longitudinal trial, including measures of processing speed, executive function, language, and verbal and visual memory. Models with 2-4 classes were examined. Demographic and clinical data were examined across phenotypes and post-RT cognitive change was evaluated. The optimal model identified three unique cognitive phenotypes including a group of patients with generalized impairments (11.5%), a group with isolated verbal memory impairments (21.9%), and a group with minimal impairments (66.7%). The Verbal Memory phenotype had fewer years of education (p=.007) and a greater proportion of males (p<.001); the Generalized group had a greater proportion of patients with IDH-wild type gliomas and showed greater symptoms of anxiety and poorer quality of life (p-values<.05); and the Minimal Impairment phenotype had higher rates of IDH-Mutant gliomas. Approximately 50% of patients declined on at least one cognitive domain with memory the most vulnerable. Patients that declined reported greater symptoms of depression (p=.007) and poorer quality of life (p=.025). We identified three distinct cognitive phenotypes in patients with primary brain tumors receiving RT, each associated with unique demographic and clinical (e.g., IDH mutational status) profiles, with mood symptoms associated with late cognitive decline. This patient-centered approach enhances our understanding of clinical profiles associated with cognitive dysfunction and treatment-related neurotoxicity.

Amino acid metabolism in glioma: in vivo MR-spectroscopic detection of alanine as a potential biomarker of poor survival in glioma patients.

Reprogramming of amino acid metabolism is relevant for initiating and fueling tumor formation and growth. Therefore, there has been growing interest in anticancer therapies targeting amino acid metabolism. While developing personalized therapeutic approaches to glioma, in vivo proton magnetic resonance spectroscopy (MRS) is a valuable tool for non-invasive monitoring of tumor metabolism. Here, we evaluated MRS-detected brain amino acids and myo-inositol as potential diagnostic and prognostic biomarkers in glioma. We measured alanine, glycine, glutamate, glutamine, and myo-inositol in 38 patients with MRI-suspected glioma using short and long echo-time single-voxel PRESS MRS sequences. The detectability of alanine, glycine, and myo-inositol and the (glutamate + glutamine)/total creatine ratio were evaluated against the patients' IDH mutation status, CNS WHO grade, and overall survival. While the detection of alanine and non-detection of myo-inositol significantly correlated with IDH wildtype (p = 0.0008, p = 0.007, respectively) and WHO grade 4 (p = 0.01, p = 0.04, respectively), glycine detection was not significantly associated with either. The ratio of (glutamate + glutamine)/total creatine was significantly higher in WHO grade 4 than in 2 and 3. We found that the overall survival was significantly shorter in glioma patients with alanine detection (p = 0.00002). Focusing on amino acids in MRS can improve its diagnostic and prognostic value in glioma. Alanine, which is visible at long TE even in the presence of lipids, could be a relevant indicator for overall survival.

Radiosensitizing Effect of PARP Inhibition on Chondrosarcoma and Chondrocyte Cells Is Dependent on Radiation LET.

Chondrosarcoma is a rare malignant tumor that forms in bone and cartilage. The primary treatment involves surgical removal of the tumor with a margin of healthy tissue. Especially if complete surgical removal is not possible, radiation therapy and chemotherapy are used in conjunction with surgery, but with a generally low efficiency. Ongoing researches are focused on understanding the genetic and molecular basis of chondrosarcoma following high linear energy transfer (LET) irradiation, which may lead to treatments that are more effective. The goal of this study is to evaluate the differential effects of DNA damage repair inhibitors and high LET irradiation on chondrosarcoma versus chondrocyte cells and the LET-dependency of the effects. Two chondrosarcoma cell lines with different IDH mutation status and one chondrocyte cell line were exposed to low LET (X-ray) and high LET (carbon ion) irradiation in combination with an Olaparib PARP inhibitor. Cell survival and DNA repair mechanisms were investigated. High LET irradiation drastically reduced cell survival, with a biological efficiency three times that of low LET. Olaparib significantly inhibited PARylation in all the tested cells. A significant reduction in cell survival of both chondrosarcoma and chondrocyte cells was observed following the treatment combining Olaparib and X-ray. PARP inhibition induced an increase in PARP-1 expression and a reduced effect on the cell survival of WT IDH chondrosarcoma cells. No radiosensitizing effect was observed in cells exposed to Olaparib paired with high LET irradiation. NHEJ was activated in response to high LET irradiation, neutralizing the PARP inhibition effect in both chondrosarcoma cell lines. When high LET irradiation is not available, PARP inhibition could be used in combination with low LET irradiation, with significant radiosensitizing effects on chondrosarcoma cells. Chondrocytes may be affected by the treatment combination too, showing the need to preserve normal tissues from radiation fields when this kind of treatment is suggested.