Enhancing Machine Learning Approach for MGMT Promoter Methylation Detection in Glioma from MRI Features
DOI:
https://doi.org/10.22399/ijcesen.3610Keywords:
Radiogenomics, Glioblastoma, MGMT methylation, Artificial intelligence, Machine learningAbstract
Glioblastoma (GBM) is the most aggressive primary brain tumor in adults, marked by high recurrence rates and poor prognosis. The methylation status of the O-6-methylguanine-DNA methyltransferase (MGMT) gene promoter plays a critical role in determining treatment response and patient survival. This work provides non-invasive machine learning (ML) solution for prediction of MGMT methylation status using features of magnetic resonance imaging (MRI) scans, aiming to support personalized therapeutic strategies. The method involves a three-step pipeline: First, extraction of image features from multi-modal MR. Second, Selection of the most important common features using light gradient boosting machine (LightGBM) algorithm and categorical gradient boosting (CatBoost). then, a voting ensemble of multiple ML models is trained on the selected features to classify MGMT methylation. The model was developed using Brain Tumor Segmentation (BraTS) 2021 dataset, which includes both segmentation masks and MGMT annotations. Its performance was evaluated using accuracy, precision, sensitivity, specificity, and area under the receiver operating characteristic (ROC) curve (AUC). The model achieved an accuracy of 92.86% and AUC of 96.84%, demonstrating strong alignment with clinical outcomes and surpassing conventional methods. These findings highlight the effectiveness of features extraction from multi-modal MRI analysis, and ML-based classification for biomarker prediction. The approach offers a promising step forward in precision medicine for GBM, enabling more accurate and individualized treatment planning.
References
[1] Louis, D. N., Perry, A., Wesseling, P., Brat, D. J., Cree, I. A., Figarella-Branger, D., Hawkins, C., Ng, H. K., Pfister, S. M., Reifenberger, G., Soffietti, R., von Deimling, A., & Ellison, D. W. (2021). The 2021 WHO classification of tumors of the central nervous system: A summary. Neuro-Oncology, 23(8), 1231–1251. https://doi.org/10.1093/neuonc/noab106
[2] Hegi, M. E., Diserens, A.-C., Gorlia, T., Hamou, M.-F., de Tribolet, N., Weller, M., Kros, J. M., Hainfellner, J. A., Mason, W., Mariani, L., Bromberg, J. E. C., Hau, P., Mirimanoff, R. O., Cairncross, J. G., Janzer, R. C., & Stupp, R. (2005). MGMT gene silencing and benefit from temozolomide in glioblastoma. New England Journal of Medicine, 352(10), 997–1003. https://doi.org/10.1056/NEJMoa043331
[3] Stupp, R., Mason, W. P., van den Bent, M. J., Weller, M., Fisher, B., Taphoorn, M. J. B., Belanger, K., Brandes, A. A., Marosi, C., Bogdahn, U., Curschmann, J., Janzer, R. C., Ludwin, S. K., Gorlia, T., Allgeier, A., Lacombe, D., Cairncross, J. G., Eisenhauer, E., & Mirimanoff, R. O. (2005). Radiotherapy plus concomitant and adjuvant temozolomide for glioblastoma. New England Journal of Medicine, 352(10), 987–996. https://doi.org/10.1056/NEJMoa043330
[4] Kickingereder, P., Burth, S., Wick, A., Götz, M., Eidel, O., Schlemmer, H.-P., Maier-Hein, K. H., Wick, W., Bendszus, M., Radbruch, A., & Bonekamp, D. (2016). Radiomic profiling of glioblastoma: Identifying an imaging predictor of patient survival with improved performance over established clinical and radiologic risk models. Radiology, 280(3), 880–889. https://doi.org/10.1148/radiol.2016160845
[5] Baid, U., Ghodasara, S., Mohan, S., Bilello, M., Calabrese, E., Colak, E., Farahani, K., Kalpathy-Cramer, J., Kitamura, F. C., Pati, S., Prevedello, L. M., Rudie, J. D., Sako, C., Shinohara, R. T., Wiestler, B., Flanders, A. E., Menze, B., & Bakas, S. (2021). The RSNA-ASNR-MICCAI BraTS 2021 benchmark on brain tumor segmentation and radiogenomic classification. arXiv. https://doi.org/10.48550/arXiv.2107.02314
[6] Sasaki, T., Kinoshita, M., Fujita, Y., Fukai, J., Hayashi, S., Uematsu, Y., Okita, Y., Nonaka, M., Tsuyuguchi, N., Moriuchi, S., Ueda, T., Ozaki, Y., Nakajima, Y., Fujinaka, T., Yoshimine, T., & Kishima, H. (2019). Radiomics and MGMT promoter methylation for prognostication of newly diagnosed glioblastoma. Scientific Reports, 9(1), Article 14435. https://doi.org/10.1038/s41598-019-50849-y
[7] Hajianfar, G., Shiri, I., Maleki, H., Oveisi, M., & Zaidi, H. (2023). Noninvasive O6-methylguanine-DNA methyltransferase status prediction in glioblastoma multiforme cancer using magnetic resonance imaging radiomics features. Journal of Neuroimaging, 33(1), 104–113. https://doi.org/10.1111/jon.13054
[8] Qian, J., Herman, M. G., Brinkmann, D. H., Laack, N. N., Kemp, B. J., Hunt, C. H., Lowe, V., & Pafundi, D. H. (2020). Prediction of MGMT status for glioblastoma patients using radiomics feature extraction from 18F-DOPA-PET imaging. International Journal of Radiation Oncology, Biology, Physics, 108(5), 1339–1346. https://doi.org/10.1016/j.ijrobp.2020.02.012
[9] Tasci, E., Zhuge, Y., Zhang, L., Ning, H., Cheng, J. Y., Miller, R. W., Camphausen, K., & Krauze, A. V. (2025). Radiomics and AI-based prediction of MGMT methylation status in glioblastoma using multiparametric MRI: A hybrid feature weighting approach. Diagnostics, 15(10), Article 1292. https://doi.org/10.3390/diagnostics15101292
[10] Crisi, G., & Filice, S. (2022). Predicting MGMT promoter methylation of glioblastoma from dynamic susceptibility contrast perfusion: A radiomic approach. Journal of Neuroimaging, 32(3), 448–457. https://doi.org/10.1111/jon.12962
[11] Korfiatis, P., Kline, T. L., Coufalova, L., Lachance, D. H., Parney, I. F., Carter, R. E., Buckner, J. C., Kaufmann, T. J., & Erickson, B. J. (2016). MRI texture features as biomarkers to predict MGMT methylation status in glioblastomas. Medical Physics, 43(6), 2835–2844. https://doi.org/10.1118/1.4948668
[12] Han, Y., Wang, S., Li, J., Zhang, Z., Yang, S., Liu, Y., Zhang, S., Xia, S., Shi, Z., Yan, L. F., & Wang, W. (2018). Non-invasive genotype prediction of chromosome 1p/19q co-deletion by development and validation of an MRI-based radiomics signature in lower-grade gliomas. Neuro-Oncology, 20(6), 808–816. https://doi.org/10.1093/neuonc/nox207
[13] Karabacak, M., Jagtiani, P., Carrasquilla, A., Germano, I. M., & Margetis, K. (2023). Prognosis individualized: Survival predictions for WHO grade II and III gliomas with a machine learning-based web application. npj Digital Medicine, 6(1), Article 200. https://doi.org/10.1038/s41746-023-00948-y
[14] Do, D. T., Yang, R. J., Le, N. Q. K., & Wu, Y. (2022). Improving MGMT methylation status prediction of glioblastoma through optimizing radiomics features using genetic algorithm-based machine learning approach. European Radiology, 32(7), 4560–4570. https://doi.org/10.1007/s00330-021-08503-8
[15] Yu, X., Zhou, J., Wu, Y., Bai, Y., Meng, N., Wu, Q., Jin, S., Liu, H., Li, P., & Wang, M. (2024). Assessment of MGMT promoter methylation status in glioblastoma using deep learning features from multi-sequence MRI of intratumoral and peritumoral regions. Cancer Imaging, 24(1), Article 104. https://doi.org/10.1186/s40644-024-00817-1
[16] Jiang, C., Kong, Z., Liu, S., Feng, S., Zhang, Y., Zhu, R., Chen, W., Wang, Y., Lyu, Y., You, H., Zhao, D., Wang, R., Wang, Y., Ma, W., & Feng, F. (2019). Fusion radiomics features from conventional MRI predict MGMT promoter methylation status in lower grade gliomas. European Journal of Radiology, 121, Article 108714. https://doi.org/10.1016/j.ejrad.2019.108714
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2025 International Journal of Computational and Experimental Science and Engineering
This work is licensed under a Creative Commons Attribution 4.0 International License.
