Despite mounting evidence for the use of re-irradiation (re-RT) in recurrent high grade glioma, optimal patient selection criteria for re-RT remain unknown. We present a novel scoring system based on radiobiology principles including target independent factors, the likelihood of target control, and the anticipated organ at risk (OAR) toxicity to allow for proper patient selection in the setting of recurrent glioma. Thirty one patients with recurrent glioma who received re-RT (2008–2016) at NCI – NIH were included in the analysis. A novel scoring system for overall survival (OS) and progression free survival (PFS) was designed to include:1) target independent factors (age, KPS (Karnofsky Performance Status), histology, presence of symptoms), 2) target control, and 3) OAR toxicity risk. Normal tissue complication probability (NTCP) calculations were performed using the Lyman model. Kaplan-Meier analysis was performed for overall survival (OS) and progression free survival (PFS) for comparison amongst variables. No patient, including those who received dose to OAR above the published tolerance dose, experienced any treatment related grade 3–5 toxicity with a median PFS and OS from re-RT of 4 months (0.5–103) and 6 months (0.7–103) respectively. Based on cumulative maximum doses the average NTCP was 25% (0–99%) for the chiasm, 21% (0–99%) for the right optic nerve, 6% (0–92%) for the left optic nerve, and 59% (0–100%) for the brainstem. The independent factor and target control scores were each statistically significant for OS and the combination of independent factors plus target control was also significant for both OS (p = 0.02) and PFS (p = 0.006). The anticipated toxicity risk score was not statistically significant. Our scoring system may represent a novel approach to patient selection for re-RT in recurrent high grade glioma. Further validation in larger patient cohorts including compilation of doses to tumor and OAR may help refine this further for inclusion into clinical trials and general practice.

Background: Glioblastoma (GBM) is the most common brain tumor with an overall survival (OS) of less than 30% at two years. Valproic acid (VPA) demonstrated survival benefits documented in retrospective and prospective trials, when used in combination with chemo-radiotherapy (CRT). Purpose: The primary goal of this study was to examine if the differential alteration in proteomic expression pre vs. post-completion of concurrent chemoirradiation (CRT) is present with the addition of VPA as compared to standard-of-care CRT. The second goal was to explore the associations between the proteomic alterations in response to VPA/RT/TMZ correlated to patient outcomes. The third goal was to use the proteomic profile to determine the mechanism of action of VPA in this setting. Materials and Methods: Serum obtained pre- and post-CRT was analyzed using an aptamer-based SOMAScan^® proteomic assay. Twenty-nine patients received CRT plus VPA, and 53 patients received CRT alone. Clinical data were obtained via a database and chart review. Tests for differences in protein expression changes between radiation therapy (RT) with or without VPA were conducted for individual proteins using two-sided t-tests, considering p-values of <0.05 as significant. Adjustment for age, sex, and other clinical covariates and hierarchical clustering of significant differentially expressed proteins was carried out, and Gene Set Enrichment analyses were performed using the Hallmark gene sets. Univariate Cox proportional hazards models were used to test the individual protein expression changes for an association with survival. The lasso Cox regression method and 10-fold cross-validation were employed to test the combinations of expression changes of proteins that could predict survival. Predictiveness curves were plotted for significant proteins for VPA response (p-value < 0.005) to show the survival probability vs. the protein expression percentiles. Results: A total of 124 proteins were identified pre- vs. post-CRT that were differentially expressed between the cohorts who received CRT plus VPA and those who received CRT alone. Clinical factors did not confound the results, and distinct proteomic clustering in the VPA-treated population was identified. Time-dependent ROC curves for OS and PFS for landmark times of 20 months and 6 months, respectively, revealed AUC of 0.531, 0.756, 0.774 for OS and 0.535, 0.723, 0.806 for PFS for protein expression, clinical factors, and the combination of protein expression and clinical factors, respectively, indicating that the proteome can provide additional survival risk discrimination to that already provided by the standard clinical factors with a greater impact on PFS. Several proteins of interest were identified. Alterations in GALNT14 (increased) and CCL17 (decreased) (p = 0.003 and 0.003, respectively, FDR 0.198 for both) were associated with an improvement in both OS and PFS. The pre-CRT protein expression revealed 480 proteins predictive for OS and 212 for PFS (p < 0.05), of which 112 overlapped between OS and PFS. However, FDR-adjusted p values were high, with OS (the smallest p value of 0.586) and PFS (the smallest p value of 0.998). The protein PLCD3 had the lowest p-value (p = 0.002 and 0.0004 for OS and PFS, respectively), and its elevation prior to CRT predicted superior OS and PFS with VPA administration. Cancer hallmark genesets associated with proteomic alteration observed with the administration of VPA aligned with known signal transduction pathways of this agent in malignancy and non-malignancy settings, and GBM signaling, and included epithelial–mesenchymal transition, hedgehog signaling, Il6/JAK/STAT3, coagulation, NOTCH, apical junction, xenobiotic metabolism, and complement signaling. Conclusions: Differential alteration in proteomic expression pre- vs. post-completion of concurrent chemoirradiation (CRT) is present with the addition of VPA. Using pre- vs. post-data, prognostic proteins emerged in the analysis. Using pre-CRT data, potentially predictive proteins were identified. The protein signals and hallmark gene sets associated with the alteration in the proteome identified between patients who received VPA and those who did not, align with known biological mechanisms of action of VPA and may allow for the identification of novel biomarkers associated with outcomes that can help advance the study of VPA in future prospective trials.

Glioblastomas (GBM) are rapidly growing, aggressive, nearly uniformly fatal, and the most common primary type of brain cancer. They exhibit significant heterogeneity and resistance to treatment, limiting the ability to analyze dynamic biological behavior that drives response and resistance, which are central to advancing outcomes in glioblastoma. Analysis of the proteome aimed at signal change over time provides a potential opportunity for non-invasive classification and examination of the response to treatment by identifying protein biomarkers associated with interventions. However, data acquired using large proteomic panels must be more intuitively interpretable, requiring computational analysis to identify trends. Machine learning is increasingly employed, however, it requires feature selection which has a critical and considerable effect on machine learning problems when applied to large-scale data to reduce the number of parameters, improve generalization, and find essential predictors. In this study, using 7k proteomic data generated from the analysis of serum obtained from 82 patients with GBM pre- and post-completion of concurrent chemoirradiation (CRT), we aimed to select the most discriminative proteomic features that define proteomic alteration that is the result of administering CRT. Thus, we present a novel rank-based feature weighting method (RadWise) to identify relevant proteomic parameters using two popular feature selection methods, least absolute shrinkage and selection operator (LASSO) and the minimum redundancy maximum relevance (mRMR). The computational results show that the proposed method yields outstanding results with very few selected proteomic features, with higher accuracy rate performance than methods that do not employ a feature selection process. While the computational method identified several proteomic signals identical to the clinical intuitive (heuristic approach), several heuristically identified proteomic signals were not selected while other novel proteomic biomarkers not selected with the heuristic approach that carry biological prognostic relevance in GBM only emerged with the novel method. The computational results show that the proposed method yields promising results, reducing 7k proteomic data to 7 selected proteomic features with a performance value of 93.921%, comparing favorably with techniques that do not employ feature selection.

Glioblastoma (GBM) is a fatal brain tumor with limited treatment options. O6-methylguanine-DNA-methyltransferase (MGMT) promoter methylation status is the central molecular biomarker linked to both the response to temozolomide, the standard chemotherapy drug employed for GBM, and to patient survival. However, MGMT status is captured on tumor tissue which, given the difficulty in acquisition, limits the use of this molecular feature for treatment monitoring. MGMT protein expression levels may offer additional insights into the mechanistic understanding of MGMT but, currently, they correlate poorly to promoter methylation. The difficulty of acquiring tumor tissue for MGMT testing drives the need for non-invasive methods to predict MGMT status. Feature selection aims to identify the most informative features to build accurate and interpretable prediction models. This study explores the new application of a combined feature selection (i.e., LASSO and mRMR) and the rank-based weighting method (i.e., MGMT ProFWise) to non-invasively link MGMT promoter methylation status and serum protein expression in patients with GBM. Our method provides promising results, reducing dimensionality (by more than 95%) when employed on two large-scale proteomic datasets (7k SomaScan^® panel and CPTAC) for all our analyses. The computational results indicate that the proposed approach provides 14 shared serum biomarkers that may be helpful for diagnostic, prognostic, and/or predictive operations for GBM-related processes, given further validation.

Glioblastoma (GBM) is a highly malignant and devastating brain cancer characterized by its ability to rapidly and aggressively grow, infiltrating brain tissue, with nearly universal recurrence after the standard of care (SOC), which comprises maximal safe resection followed by chemoirradiation (CRT). The metabolic triggers leading to the reprogramming of tumor behavior and resistance are an area increasingly studied in relation to the tumor molecular features associated with outcome. There are currently no metabolomic biomarkers for GBM. Studying the metabolomic alterations in GBM patients undergoing CRT could uncover the biochemical pathways involved in tumor response and resistance, leading to the identification of novel biomarkers and the optimization of the treatment response. The feature selection process identifies key factors to improve the model’s accuracy and interpretability. This study utilizes a combined feature selection approach, incorporating both Least Absolute Shrinkage and Selection Operator (LASSO) and Minimum Redundancy–Maximum Relevance (mRMR), alongside a rank-based weighting method (i.e., MetaWise) to link metabolomic biomarkers to CRT and the 12-month and 20-month overall survival (OS) status in patients with GBM. Our method shows promising results, reducing feature dimensionality when employed on serum-based large-scale metabolomic datasets (University of Florida) for all our analyses. The proposed method successfully identified a set of eleven serum biomarkers shared among three datasets. The computational results show that the utilized method achieves 96.711%, 92.093%, and 86.910% accuracy rates with 48, 46, and 33 selected features for the CRT, 12-month, and 20-month OS-based metabolomic datasets, respectively. This discovery has implications for developing personalized treatment plans and improving patient outcomes.

Glioblastoma (GBM) is a fatal brain cancer known for its rapid and aggressive growth, with some studies indicating that females may have better survival outcomes compared to males. While sex differences in GBM have been observed, the underlying biological mechanisms remain poorly understood. Feature selection can lead to the identification of discriminative key biomarkers by reducing dimensionality from high-dimensional medical datasets to improve machine learning model performance, explainability, and interpretability. Feature selection can uncover unique sex-specific biomarkers, determinants, and molecular profiles in patients with GBM. We analyzed high-dimensional proteomic and metabolomic profiles from serum biospecimens obtained from 109 patients with pathology-proven glioblastoma (GBM) on NIH IRB-approved protocols with full clinical annotation (local dataset). Serum proteomic analysis was performed using Somalogic aptamer-based technology (measuring 7289 proteins) and serum metabolome analysis using the University of Florida’s SECIM (Southeast Center for Integrated Metabolomics) platform (measuring 6015 metabolites). Machine learning-based feature selection was employed to identify proteins and metabolites associated with male and female labels in high-dimensional datasets. Results were compared to publicly available proteomic and metabolomic datasets (CPTAC and TCGA) using the same methodology and TCGA data previously structured for glioma grading. Employing a machine learning-based and hybrid feature selection approach, utilizing both LASSO and mRMR, in conjunction with a rank-based weighting method (i.e., GLIO-Select), we linked proteomic and metabolomic data to clinical data for the purposes of feature reduction to identify molecular biomarkers associated with biological sex in patients with GBM and used a separate TCGA set to explore possible linkages between biological sex and mutations associated with tumor grading. Serum proteomic and metabolomic data identified several hundred features that were associated with the male/female class label in the GBM datasets. Using the local serum-based dataset of 109 patients, 17 features (100% ACC) and 16 features (92% ACC) were identified for the proteomic and metabolomic datasets, respectively. Using the CPTAC tissue-based dataset (8828 proteomic and 59 metabolomic features), 5 features (99% ACC) and 13 features (80% ACC) were identified for the proteomic and metabolomic datasets, respectively. The proteomic data serum or tissue (CPTAC) achieved the highest accuracy rates (100% and 99%, respectively), followed by serum metabolome and tissue metabolome. The local serum data yielded several clinically known features (PSA, PZP, HCG, and FSH) which were distinct from CPTAC tissue data (RPS4Y1 and DDX3Y), both providing methodological validation, with PZP and defensins (DEFA3 and DEFB4A) representing shared proteomic features between serum and tissue. Metabolomic features shared between serum and tissue were homocysteine and pantothenic acid. Several signals emerged that are known to be associated with glioma or GBM but not previously known to be associated with biological sex, requiring further research, as well as several novel signals that were previously not linked to either biological sex or glioma. EGFR, FAT4, and BCOR were the three features associated with 64% ACC using the TCGA glioma grading set. GLIO-Select shows remarkable results in reducing feature dimensionality when different types of datasets (e.g., serum and tissue-based) were used for our analyses. The proposed approach successfully reduced relevant features to less than twenty biomarkers for each GBM dataset. Serum biospecimens appear to be highly effective for identifying biologically relevant sex differences in GBM. These findings suggest that serum-based noninvasive biospecimen-based analyses may provide more accurate and clinically detailed insights into sex as a biological variable (SABV) as compared to other biospecimens, with several signals linking sex differences and glioma pathology via immune response, amino acid metabolism, and cancer hallmark signals requiring further research. Our results underscore the importance of biospecimen choice and feature selection in enhancing the interpretation of omics data for understanding sex-based differences in GBM. This discovery holds significant potential for enhancing personalized treatment plans and patient outcomes.