Dysregulated Alanine as a Potential Predictive Marker of Glioma—An Insight from Untargeted HRMAS-NMR and Machine Learning Data

Abstract

Metabolic alterations play a crucial role in glioma development and progression and can be detected even before the appearance of the fatal phenotype. We have compared the circulating metabolic fingerprints of glioma patients versus healthy controls, for the first time, in a quest to identify a panel of small, dysregulated metabolites with potential to serve as a predictive and/or diagnostic marker in the clinical settings. High-resolution magic angle spinning nuclear magnetic resonance spectroscopy (HRMAS-NMR) was used for untargeted metabolomics and data acquisition followed by a machine learning (ML) approach for the analyses of large metabolic datasets. Cross-validation of ML predicted NMR spectral features was done by statistical methods (Wilcoxon-test) using JMP-pro16 software. Alanine was identified as the most critical metabolite with potential to detect glioma with precision of 1.0, recall of 0.96, and F1 measure of 0.98. The top 10 metabolites identified for glioma detection included alanine, glutamine, valine, methionine, N-acetylaspartate (NAA), γ-aminobutyric acid (GABA), serine, α-glucose, lactate, and arginine. We achieved 100% accuracy for the detection of glioma using ML algorithms, extra tree classifier, and random forest, and 98% accuracy with logistic regression. Classification of glioma in low and high grades was done with 86% accuracy using logistic regression model, and with 83% and 79% accuracy using extra tree classifier and random forest, respectively. The predictive accuracy of our ML model is superior to any of the previously reported algorithms, used in tissue- or liquid biopsy-based metabolic studies. The identified top metabolites can be targeted to develop early diagnostic methods as well as to plan personalized treatment strategies.

Assessing Metabolic Markers in Glioblastoma Using Machine Learning: A Systematic Review

“…These articles were screened by title and abstract, which returned 46 articles for a full-text review. Thirty-one articles were excluded, leaving 14 final studies included in this review, ten of which were included in the analysis [ 32 , 33 , 34 , 35 , 36 , 37 , 38 , 39 , 40 , 41 ]. Four studies were not included in the analysis because they did not perform an ML algorithm on GBM; however, they did discuss the topic [ 21 , 42 , 43 , 44 , 45 , 46 ].…”

“…Supervised ML is broadly used in a predictive scenario where a “ground truth” value can be determined (e.g., a diagnosis of GBM) and the user wishes to identify similar data sets with an unknown “ground truth.” The supervised ML algorithms used by these studies were SVM, random forest, ANN, deep neural networks (e.g., PASnet), DT, NB, partial least-squares discriminant analysis (PLS-DA), logistic regression models, and LASSO-penalized Cox regression analysis [ 32 , 34 , 35 , 36 , 37 , 38 , 39 , 40 , 41 , 42 , 43 ]. A logistic regression model appears to outperform other ML algorithms in classification systems, in this case, the classification of the IDH mutation.…”

“…Various alternate metabolic markers of importance are the levels of dysregulated amino acids. These amino acids have been identified as a product of activated or deactivated metabolic pathways in GBM to increase nutrient availability for tumors [ 34 ]. Many of those amino acids were discovered in previous experiments and then analyzed by ML in these studies to differentiate patients’ glioma grading, thereby ascertaining a method for a more precise diagnosis [ 34 , 42 ].…”

“…These amino acids have been identified as a product of activated or deactivated metabolic pathways in GBM to increase nutrient availability for tumors [ 34 ]. Many of those amino acids were discovered in previous experiments and then analyzed by ML in these studies to differentiate patients’ glioma grading, thereby ascertaining a method for a more precise diagnosis [ 34 , 42 ]. In fact, Firdous et al found that their diagnostic study utilizing an extra tree classifier, logistic regression integration, and random forest algorithms had greater predictive accuracy than any other previous studies of ML algorithms on the identification of metabolic markers in tissue or liquid-based biopsies [ 34 ].…”

PiDeeL: Pathway-Informed Deep Learning Model for Survival Analysis and Pathological Classification of Gliomas

“…Through a case-control study, alanine was reported to be a potential biomarker for malignant gliomas [45]. Similarly, glutamate was also found to play a central role in malignant gliomas through multiple mechanisms [46].…”