Deep Learning Approaches for Biomarker Discovery in Glioblastoma Multiforme

Abstract

Glioblastoma multiforme (GBM) is a particularly aggressive and deadly brain tumor that presents major challenges for diagnosis, therapeutic intervention, and prediction of patient outcomes. The discovery of reliable biomarkers remains essential to improve therapeutic outcomes, as patients survive an average of 14–16 months despite intensive treatments. Although genomic and proteomic studies have led to important biomarker discoveries, these traditional approaches often lack sufficient predictive ability to support clinical treatment decisions.
Deep learning represents a transformative approach that enables effective molecular profiling, radiomic assessment, and integration of multi-omics data to improve biomarker discovery. Computational methods facilitate the identification of novel biomarkers, improve patient stratification, and support better treatment decisions. The combination of convolutional neural networks (CNNs) and graph neural networks (GNNs) allows researchers to investigate complicated biological data for advanced subtype analyses and phenotype-genotype relationship understanding of GBM.
The review evaluates modern deep learning techniques for GBM biomarker exploration through assessments of subtype detection and biomarker detection from imaging data and model assessments between drugs and biomarkers while providing explanations about AI systems. The promising nature of these techniques faces major obstacles which consist of GBM variability and the need for inter-dataset confirmation and the transfer of algorithm-generated findings to real-world medical practice The successful translation of deep learning discoveries into improved patient care and treatment strategies depends on addressing these current limitations.

Keywords

• Glioblastoma Multiforme (GBM)
• deep learning
• biomarker discovery
• Explainable AI (XAI)
• Convolutional Neural Networks (CNNs)

References

1. 1. Agosti E, Mapelli K, Grimod G, Piazza A, Fontanella MM, Panciani PP. MRI-Based Radiomics for Non-Invasive Prediction of Molecular Biomarkers in Gliomas. Cancers (Basel). 2026;18(3):491. doi:10.3390/cancers18030491.
2. 2. Echle A, Rindtorff NT, Brinker TJ, Luedde T, Pearson AT, Kather JN. Deep learning in cancer pathology: a new generation of clinical biomarkers. Br J Cancer. 2021;124(4):686-96. doi:10.1038/s41416-020-01122-x.
3. 3. Han W, Qin L, Bay C, Chen X, Yu KH, Li A, et al. Integrating deep transfer learning and radiomics features in glioblastoma multiforme patient survival prediction. In: Išgum I, Landman BA, editors. Medical Imaging 2020: Image Processing; 2020 Feb 15-20; Houston, Bellingham, WA: SPIE; 2020;11313:113132S. doi:10.1117/12.2549325.
4. 4. Shams A. Leveraging state-of-the-art AI algorithms in personalized oncology: from transcriptomics to treatment. Diagnostics (Basel). 2024;14(19):2174. doi:10.3390/diagnostics14192174.
5. 5. Ma T, Wang J. GraphPath: a graph attention model for molecular stratification with interpretability based on the pathway–pathway interaction network. Bioinformatics. 2024;40(4):btae165. doi:10.1093/bioinformatics/btae165.
6. 6. Munquad S, Si T, Mallik S, Das AB, Zhao Z. A deep learning–based framework for supporting clinical diagnosis of glioblastoma subtypes. Front Genet. 2022;13:855420. doi:10.3389/fgene.2022.855420.
7. 7. Capuozzo S, Gravina M, Gatta G, Marrone S, Sansone C. A multimodal knowledge-based deep learning approach for MGMT promoter methylation identification. J Imaging. 2022;8(12):321. doi:10.3390/jimaging8120321.
8. 8. Yu F, Kazerooni AF, Toorens E, Akbari H, Sako C, Mamourian E. NIMG-22: An AI-based coordinate system elucidates radiogenomic heterogeneity of glioblastoma via deep learning and manifold embeddings. Neuro-Oncol. 2022;24(Suppl 7):vii166. doi:10.1093/neuonc/noac209.640.

9. 9. Fidon L, Thiriez C, Grimaldi A, Domingues OD, Maussion C, Hoffmann C. Histology-based prognosis prediction using deep learning outperforms and is independent of the MGMT methylation status in patients with glioblastoma [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 7395. doi:10.1158/1538-7445.AM2024-7395.
10. 10. Tang Z, Xu Y, Jin L, Aibaidula A, Lu J, Jiao Z, et al. Deep learning of imaging phenotype and genotype for predicting overall survival time of glioblastoma patients. IEEE Trans Med Imaging. 2020;39(6):2100-9. doi:10.1109/TMI.2020.2964310.
11. 11. Maeser D, Gruener RF, Galvin R, Lee A, Koga T, Grigore FN, et al. Integration of computational pipeline to streamline efficacious drug nomination and biomarker discovery in glioblastoma. Cancers (Basel). 2024;16(9):1723. doi:10.3390/cancers16091723.
12. 12. Samara MN, Harry KD. (2025). Integrating Boruta, LASSO, and SHAP for clinically interpretable glioma classification using machine learning. BioMedInformatics. 2025;5(3):34. doi:10.3390/biomedinformatics5030034.
13. 13. Hayakawa J, Seki T, Kawazoe Y, Ohe K. Pathway importance by graph convolutional network and Shapley additive explanations in gene expression phenotype of diffuse large B-cell lymphoma. PLoS One. 2022;17(6):e0269570. doi:10.1371/journal.pone.0269570.
14. 14. Chereda H, Bleckmann A, Menck K, Perera-Bel J, Stegmaier P, Auer F, et al. Explaining decisions of graph convolutional neural networks: patient-specific molecular subnetworks responsible for metastasis prediction in breast cancer. Genome Med. 2021;13(1):42. doi:10.1186/s13073-021-00845-7.
15. 15. Ansari ZA, Tripathi MM, Ahmed R. The role of explainable AI in enhancing breast cancer diagnosis using machine learning and deep learning models. Discov Artif Intell. 2025;5:75. doi:10.1007/s44163-025-00307-8.
16. 16. Hakami MA. Harnessing machine learning potential for personalised drug design and overcoming drug resistance. J Drug Target. 2024;32(8):918-30. doi:10.1080/1061186X.2024.2365934.
17. 17. Liu X, Liu J. Aided diagnosis model based on deep learning for glioblastoma, solitary brain metastases, and primary central nervous system lymphoma with multi-modal MRI. Biology (Basel). 2024;13(2):99. doi:10.3390/biology13020099.
18. 18. Farahani S, Hejazi M, Moradizeyveh S, Di Ieva A, Fatemizadeh E, Liu S. Diagnostic accuracy of deep learning models in predicting glioma molecular markers: a systematic review and meta-analysis. Diagnostics (Basel). 2025;15(7):797. doi: 10.3390/diagnostics15070797.
19. 19. Lao J, Chen Y, Li ZC, Li Q, Zhang J, Liu J, et al. A deep learning-based radiomics model for prediction of survival in glioblastoma multiforme. Sci Rep. 2017;7(1):10353. doi:10.1038/s41598-017-10649-8.
20. 20. Ramirez AFG, Tang CC, Kong XT, Bota DA, Chang PD. Deep learning for 4D modeling of glioblastoma multiforme with tumor treating fields (TTFields) therapy. J Clin Oncol. 2024;42(16 Suppl):e14032. doi:10.1200/JCO.2024.42.16_suppl.e14032.
21. 21. Young JD, Cai C, Lu X. Unsupervised deep learning reveals prognostically relevant subtypes of glioblastoma. BMC Bioinformatics. 2017;18(Suppl 11):381. doi:10.1186/s12859-017-1798-2.