A Radiomics-Based Machine Learning Model for Prediction of Tumor Mutational Burden in Lower-Grade Gliomas
Abstract
:Simple Summary
Abstract
1. Introduction
2. Results
2.1. Patient Characteristics
2.2. Baseline Comparison and Radiomics Signature Building
2.3. Imbalance Solving
2.4. Model Interpretation
2.5. Validation Results
3. Discussion
4. Materials and Methods
4.1. Patient Cohort
4.2. MRI Segmentation and Radiomics Feature Extraction
4.3. Feature Selection and the Genetic Algorithm (GA)
4.4. Machine Learning
4.5. Statistical Analysis
5. Conclusions
Author Contributions
Funding
Institutional Review Board Statement
Informed Consent Statement
Data Availability Statement
Conflicts of Interest
References
- Louis, D.N. WHO Classification of Tumours of the Central Nervous System; WHO Regional Office Europe: Copenhagen, Denmark, 2007. [Google Scholar]
- Louis, D.N.; Perry, A.; Reifenberger, G.; Von Deimling, A.; Figarella-Branger, D.; Cavenee, W.K.; Ohgaki, H.; Wiestler, O.D.; Kleihues, P.; Ellison, D.W. The 2016 World Health Organization classification of tumors of the central nervous system: A summary. Acta Neuropathol. 2016, 131, 803–820. [Google Scholar] [CrossRef] [PubMed] [Green Version]
- The Cancer Genome Atlas Research Network. Comprehensive, integrative genomic analysis of diffuse lower-grade gliomas. N. Engl. J. Med. 2015, 372, 2481–2498. [Google Scholar] [CrossRef] [PubMed] [Green Version]
- Ballas, Z.K. The 2018 Nobel Prize in Physiology or Medicine: An exemplar of bench to bedside in immunology. J. Allergy Clin. Immunol. 2018, 142, 1752–1753. [Google Scholar] [CrossRef] [PubMed] [Green Version]
- Xu, S.; Tang, L.; Li, X.; Fan, F.; Liu, Z. Immunotherapy for glioma: Current management and future application. Cancer Lett. 2020, 476, 1–12. [Google Scholar] [CrossRef]
- Chan, T.A.; Yarchoan, M.; Jaffee, E.; Swanton, C.; Quezada, S.A.; Stenzinger, A.; Peters, S. Development of tumor mutation burden as an immunotherapy biomarker: Utility for the oncology clinic. Ann. Oncol. 2019, 30, 44–56. [Google Scholar] [CrossRef]
- Chalmers, Z.R.; Connelly, C.F.; Fabrizio, D.; Gay, L.; Ali, S.M.; Ennis, R.; Schrock, A.; Campbell, B.; Shlien, A.; Chmielecki, J. Analysis of 100,000 human cancer genomes reveals the landscape of tumor mutational burden. Genome Med. 2017, 9, 34. [Google Scholar] [CrossRef] [Green Version]
- Braun, D.A.; Burke, K.P.; Van Allen, E.M. Genomic approaches to understanding response and resistance to immunotherapy. Clin. Cancer Res. 2016, 22, 5642–5650. [Google Scholar] [CrossRef] [Green Version]
- Yin, W.; Jiang, X.; Tan, J.; Xin, Z.; Zhou, Q.; Zhan, C.; Fu, X.; Wu, Z.; Guo, Y.; Jiang, Z. Development and Validation of a Tumor Mutation Burden–Related Immune Prognostic Model for Lower-Grade Glioma. Front. Oncol. 2020, 10, 1409. [Google Scholar] [CrossRef]
- Wang, L.; Ge, J.; Lan, Y.; Shi, Y.; Luo, Y.; Tan, Y.; Liang, M.; Deng, S.; Zhang, X.; Wang, W. Tumor mutational burden is associated with poor outcomes in diffuse glioma. BMC Cancer 2020, 20, 213. [Google Scholar] [CrossRef]
- Ahmadi, A.; Kashefi, M.; Shahrokhi, H.; Nazari, M.A. Computer aided diagnosis system using deep convolutional neural networks for ADHD subtypes. Biomed. Signal Process. Control 2021, 63, 102–227. [Google Scholar] [CrossRef]
- Ahmadi, A.; Davoudi, S.; Daliri, M.R. Computer Aided Diagnosis System for multiple sclerosis disease based on phase to amplitude coupling in covert visual attention. Comput. Methods Programs Biomed. 2019, 169, 9–18. [Google Scholar] [CrossRef]
- Delgado-Ortet, M.; Molina, A.; Alférez, S.; Rodellar, J.; Merino, A. A deep learning approach for segmentation of red blood cell images and malaria detection. Entropy 2020, 22, 657. [Google Scholar] [CrossRef]
- Jain, M.S.; Massoud, T.F. Predicting tumour mutational burden from histopathological images using multiscale deep learning. Nat. Mach. Intell. 2020, 2, 356–362. [Google Scholar] [CrossRef]
- Shi, X.; Niu, Y.; Lixia, W.; Zhang, X.; Han, Y.; Yang, C.; Bai, H.; Huang, K.; Ren, C.; Tian, G. Predicting tumor mutational burden from Lung adenocarcinoma histopathological images using deep learning. Front. Oncol. 2022, 12, 2678. [Google Scholar]
- Shimada, Y.; Okuda, S.; Watanabe, Y.; Tajima, Y.; Nagahashi, M.; Ichikawa, H.; Nakano, M.; Sakata, J.; Takii, Y.; Kawasaki, T. Histopathological characteristics and artificial intelligence for predicting tumor mutational burden-high colorectal cancer. J. Gastroenterol. 2021, 56, 547–559. [Google Scholar] [CrossRef] [PubMed]
- Duffau, H. A new philosophy in surgery for diffuse low-grade glioma (DLGG): Oncological and functional outcomes. Neurochirurgie 2013, 59, 2–8. [Google Scholar] [CrossRef] [Green Version]
- Wijnenga, M.M.; French, P.J.; Dubbink, H.J.; Dinjens, W.N.; Atmodimedjo, P.N.; Kros, J.M.; Smits, M.; Gahrmann, R.; Rutten, G.-J.; Verheul, J.B. The impact of surgery in molecularly defined low-grade glioma: An integrated clinical, radiological, and molecular analysis. Neuro-Oncology 2018, 20, 103–112. [Google Scholar] [CrossRef] [Green Version]
- Kawaguchi, T.; Sonoda, Y.; Shibahara, I.; Saito, R.; Kanamori, M.; Kumabe, T.; Tominaga, T. Impact of gross total resection in patients with WHO grade III glioma harboring the IDH 1/2 mutation without the 1p/19q co-deletion. J. Neuro-Oncol. 2016, 129, 505–514. [Google Scholar] [CrossRef] [PubMed]
- Patel, S.; Bansal, A.; Young, E.; Batchala, P.; Patrie, J.; Lopes, M.; Jain, R.; Fadul, C.; Schiff, D. Extent of surgical resection in lower-grade gliomas: Differential impact based on molecular subtype. Am. J. Neuroradiol. 2019, 40, 1149–1155. [Google Scholar] [CrossRef]
- Liu, Z.; Zhang, T.; Jiang, H.; Xu, W.; Zhang, J. Conventional MR-based preoperative nomograms for prediction of IDH/1p19q subtype in low-grade glioma. Acad. Radiol. 2019, 26, 1062–1070. [Google Scholar] [CrossRef]
- Ren, Y.; Zhang, X.; Rui, W.; Pang, H.; Qiu, T.; Wang, J.; Xie, Q.; Jin, T.; Zhang, H.; Chen, H. Noninvasive prediction of IDH1 mutation and ATRX expression loss in low-grade gliomas using multiparametric MR radiomic features. J. Magn. Reson. Imaging 2019, 49, 808–817. [Google Scholar] [CrossRef] [PubMed]
- Tang, X.; Qian, W.-l.; Yan, W.-f.; Pang, T.; Gong, Y.-l.; Yang, Z.-g. Radiomic assessment as a method for predicting tumor mutation burden (TMB) of bladder cancer patients: A feasibility study. BMC Cancer 2021, 21, 823. [Google Scholar] [CrossRef] [PubMed]
- Liu, E.-T.; Zhou, S.; Li, Y.; Zhang, S.; Ma, Z.; Guo, J.; Guo, L.; Zhang, Y.; Guo, Q.; Xu, L. Development and validation of an MRI-based nomogram for the preoperative prediction of tumor mutational burden in lower-grade gliomas. Quant. Imaging Med. Surg. 2022, 12, 1684. [Google Scholar] [CrossRef] [PubMed]
- Le, N.Q.K.; Hung, T.N.K.; Do, D.T.; Lam, L.H.T.; Dang, L.H.; Huynh, T.-T. Radiomics-based machine learning model for efficiently classifying transcriptome subtypes in glioblastoma patients from MRI. Comput. Biol. Med. 2021, 132, 104–320. [Google Scholar] [CrossRef] [PubMed]
- Kha, Q.-H.; Le, V.-H.; Hung, T.N.K.; Le, N.Q.K. Development and Validation of an Efficient MRI Radiomics Signature for Improving the Predictive Performance of 1p/19q Co-Deletion in Lower-Grade Gliomas. Cancers 2021, 13, 5398. [Google Scholar] [CrossRef]
- Clark, K.; Vendt, B.; Smith, K.; Freymann, J.; Kirby, J.; Koppel, P.; Moore, S.; Phillips, S.; Maffitt, D.; Pringle, M. The Cancer Imaging Archive (TCIA): Maintaining and operating a public information repository. J. Digit. Imaging 2013, 26, 1045–1057. [Google Scholar] [CrossRef] [Green Version]
- Camp, R.L.; Dolled-Filhart, M.; Rimm, D.L. X-tile: A new bio-informatics tool for biomarker assessment and outcome-based cut-point optimization. Clin. Cancer Res. 2004, 10, 7252–7259. [Google Scholar] [CrossRef] [Green Version]
- Bakas, S.; Akbari, H.; Sotiras, A.; Bilello, M.; Rozycki, M.; Kirby, J.S.; Freymann, J.B.; Farahani, K.; Davatzikos, C. Advancing the cancer genome atlas glioma MRI collections with expert segmentation labels and radiomic features. Sci. Data 2017, 4, 170117. [Google Scholar] [CrossRef] [Green Version]
- Vallières, M.; Freeman, C.R.; Skamene, S.R.; El Naqa, I. A radiomics model from joint FDG-PET and MRI texture features for the prediction of lung metastases in soft-tissue sarcomas of the extremities. Phys. Med. Biol. 2015, 60, 5471. [Google Scholar] [CrossRef]
- Thibault, G.; Fertil, B.; Navarro, C.; Pereira, S.; Cau, P.; Levy, N.; Sequeira, J.; Mari, J.-L. Shape and texture indexes application to cell nuclei classification. Int. J. Pattern Recognit. Artif. Intell. 2013, 27, 1357002. [Google Scholar] [CrossRef]
- Macyszyn, L.; Akbari, H.; Pisapia, J.M.; Da, X.; Attiah, M.; Pigrish, V.; Bi, Y.; Pal, S.; Davuluri, R.V.; Roccograndi, L. Imaging patterns predict patient survival and molecular subtype in glioblastoma via machine learning techniques. Neuro-Oncology 2015, 18, 417–425. [Google Scholar] [CrossRef] [PubMed] [Green Version]
- Collewet, G.; Strzelecki, M.; Mariette, F. Influence of MRI acquisition protocols and image intensity normalization methods on texture classification. Magn. Reson. Imaging 2004, 22, 81–91. [Google Scholar] [CrossRef] [PubMed]
- Haralick, R.M.; Shanmugam, K.; Dinstein, I.H. Textural features for image classification. IEEE Trans. Syst. Man Cybern. 1973, 6, 610–621. [Google Scholar] [CrossRef] [Green Version]
- Dasarathy, B.V.; Holder, E.B. Image characterizations based on joint gray level—Run length distributions. Pattern Recognit. Lett. 1991, 12, 497–502. [Google Scholar] [CrossRef]
- Amadasun, M.; King, R. Textural features corresponding to textural properties. IEEE Trans. Syst. Man Cybern. 1989, 19, 1264–1274. [Google Scholar] [CrossRef]
- Bilello, M.; Akbari, H.; Da, X.; Pisapia, J.M.; Mohan, S.; Wolf, R.L.; O’Rourke, D.M.; Martinez-Lage, M.; Davatzikos, C. Population-based MRI atlases of spatial distribution are specific to patient and tumor characteristics in glioblastoma. NeuroImage Clin. 2016, 12, 34–40. [Google Scholar] [CrossRef] [Green Version]
- Hogea, C.; Davatzikos, C.; Biros, G. An image-driven parameter estimation problem for a reaction–diffusion glioma growth model with mass effects. J. Math. Biol. 2008, 56, 793–825. [Google Scholar] [CrossRef] [Green Version]
- Pedregosa, F.; Varoquaux, G.; Gramfort, A.; Michel, V.; Thirion, B.; Grisel, O.; Blondel, M.; Prettenhofer, P.; Weiss, R.; Dubourg, V. Scikit-learn: Machine learning in Python. J. Mach. Learn. Res. 2011, 12, 2825–2830. [Google Scholar]
- He, H.; Bai, Y.; Garcia, E.A.; Li, S. ADASYN: Adaptive synthetic sampling approach for imbalanced learning. In Proceedings of the 2008 IEEE International Joint Conference on Neural Networks (IEEE World Congress on Computational Intelligence), Hong Kong, China, 1–8 June 2008; pp. 1322–1328. [Google Scholar]
- Chawla, N.V.; Bowyer, K.W.; Hall, L.O.; Kegelmeyer, W.P. SMOTE: Synthetic minority over-sampling technique. J. Artif. Intell. Res. 2002, 16, 321–357. [Google Scholar] [CrossRef]
- Han, H.; Wang, W.-Y.; Mao, B.-H. Borderline-SMOTE: A new over-sampling method in imbalanced data sets learning. In Proceedings of the International Conference on Intelligent Computing, Hefei, China, 23–26 August 2005; Springer: Berlin/Heidelberg, Germany, 2005; pp. 878–887. [Google Scholar]
- Nguyen, H.M.; Cooper, E.W.; Kamei, K. Borderline over-sampling for imbalanced data classification. Int. J. Knowl. Eng. Soft Data Paradig. 2011, 3, 4–21. [Google Scholar] [CrossRef]
- Menardi, G.; Torelli, N. Training and assessing classification rules with imbalanced data. Data Min. Knowl. Discov. 2014, 28, 92–122. [Google Scholar] [CrossRef]
- Le, N.Q.K.; Kha, Q.H.; Nguyen, V.H.; Chen, Y.-C.; Cheng, S.-J.; Chen, C.-Y. Machine Learning-Based Radiomics Signatures for EGFR and KRAS Mutations Prediction in Non-Small-Cell Lung Cancer. Int. J. Mol. Sci. 2021, 22, 9254. [Google Scholar] [CrossRef] [PubMed]
Training | Validation | p | ||
---|---|---|---|---|
Age | 44.06 ± 14.00 | 49.76 ± 13.44 | 0.040399 * | |
Gender | Male | 26 | 23 | 0.088934 |
Female | 37 | 19 | ||
Histology | Astrocytoma | 21 | 11 | 0.127063 |
Oligoastrocytoma | 18 | 10 | ||
Oligodendroglioma | 24 | 21 | ||
Grade | II | 26 | 20 | 0.262543 |
III | 37 | 22 | ||
Subtype | Classic-like | 2 | 3 | 0.09987 |
Codel | 13 | 15 | ||
G-CIMP-high | 38 | 16 | ||
G-CIMP-low | 2 | |||
Mesenchymal-like | 6 | 7 | ||
PA-like | 2 | 1 | ||
Vital status | Dead | 13 | 13 | 0.117054 |
Alive | 29 | 50 | ||
IDH status | Mutant | 53 | 31 | 0.099583 |
Wildtype | 10 | 11 | ||
1p_19q codeletion status | Codel | 13 | 14 | 0.073916 |
Non-codel | 50 | 28 | ||
MGMT promoter status | Methylated | 51 | 39 | 0.045106 * |
Unmethylated | 12 | 3 | ||
TMB group | TMB high | 25 | 18 | 0.37498 |
TMB low | 38 | 24 |
Algorithm | GA Features | Sensitivity | Specificity | Precision | Accuracy | Running Time (s) |
---|---|---|---|---|---|---|
LR | 13 | 0.64 ± 0.265 | 0.9 ± 0.079 | 0.8367 ± 0.162 | 0.7936 ± 0.132 | 0.159057 |
SVM | 10 | 0.56 ± 0.15 | 0.8714 ± 0.177 | 0.7733 ± 0.228 | 0.7462 ± 0.133 | 0.050138 |
RF | 6 | 0.64 ± 0.15 | 0.9179 ± 0.064 | 0.8833 ± 0.108 | 0.8089 ± 0.041 | 0.817011 |
LDA | 4 | 0.56 ± 0.16 | 0.8714 ± 0.131 | 0.77 ± 0.131 | 0.7449 ± 0.056 | 0.125044 |
LGBM | 11 | 0.72 ± 204 | 0.8893 ± 0.131 | 0.8367 ± 0.131 | 0.8218 ± 0.1 | 0.094271 |
XGB | 7 | 0.6 ± 204 | 0.9 ± 0.009 | 0.8 ± 0.106 | 0.7808 ± 0.08 | 1.63018 |
Method | Sensitivity | Specificity | Precision | Accuracy |
---|---|---|---|---|
ADASYN | 0.72 ± 0.204 | 0.7571 ± 0.168 | 0.7010 ± 0.244 | 0.7449 ± 0.103 |
BorderlineSMOTE | 0.8 ± 204 | 0.7143 ± 0.151 | 0.6573 ± 0.092 | 0.7462 ± 0.083 |
RandomOversampler | 0.72 ± 0.219 | 0.8143 ± 0.148 | 0.7262 ± 0.129 | 0.7782 ± 0.127 |
RandomUndersampler | 0.8 ± 0.4 | 0.7393 ± 0.15 | 0.6810 ± 0.169 | 0.7641 ± 0.069 |
SMOTE | 0.8 ± 0.32 | 0.7714 ± 0.19 | 0.7219 ± 0.258 | 0.7808 ± 0.11 |
SVMSMOTE | 0.76 ± 0.126 | 0.8107 ± 0.068 | 0.7952 ± 0.112 | 0.7936 ± 0.081 |
Study | Method Summary | Kind of Cancer | Result |
---|---|---|---|
Jain et al. [14] | Machine learning algorithm, Image2TMB, integrated three deep learning models. | Lung cancer | auPRC = 0.92 Precision = 0.89 |
Shi et al. [15] | Deep learning model is based on the ResNet18 architecture. | Lung cancer | AUC = 0.64 |
Shimada et al. [16] | Convolutional neural network (CNN)-based algorithm. | Colorectal cancer | AUC = 0.934 |
Tang et al. [23] | LASSO regression selected features. Nomogram model predicted TMB. | Bladder cancer | AUC = 0.853 |
Liu et al. [24] | Nomogram model predicted TMB. | Lower-grade glioma | AUC = 0.736 |
The proposed study | The genetic algorithm selected radiomics signatures. LGBM algorithm predicted TMB. | Lower-grade glioma | AUC = 0.7875 auPRC = 0.7556 |
Algorithm | Optimal Hyperparameters |
---|---|
Logistic Regression | solver = ‘saga’, C = 2.015990003658406, penalty = ‘l1’ |
Random Forest | ‘n_estimators’ = 5, ‘min_samples_split’ = 6, ‘min_samples_leaf’ = 3, ‘max_features’ = ‘auto’, ‘max_depth’ = 30, ‘bootstrap’ = False |
Support Vector Machine | kernel = ’rbf’, gamma = 1 × 10−4, C = 10 |
Linear Discriminant Analysis | solver = ‘svd’ |
Light GBM | learning_rate = 0.005, num_leaves = 15, max_depth = 25, min_data_in_leaf = 15, feature_fraction = 0.6, bagging_fraction = 0.6 |
XGBoost | max_depth = 1, gamma = 9, colsample_bytree = 0.5, min_child_weight = 1 |
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. |
© 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
Lam, L.H.T.; Chu, N.T.; Tran, T.-O.; Do, D.T.; Le, N.Q.K. A Radiomics-Based Machine Learning Model for Prediction of Tumor Mutational Burden in Lower-Grade Gliomas. Cancers 2022, 14, 3492. https://doi.org/10.3390/cancers14143492
Lam LHT, Chu NT, Tran T-O, Do DT, Le NQK. A Radiomics-Based Machine Learning Model for Prediction of Tumor Mutational Burden in Lower-Grade Gliomas. Cancers. 2022; 14(14):3492. https://doi.org/10.3390/cancers14143492
Chicago/Turabian StyleLam, Luu Ho Thanh, Ngan Thy Chu, Thi-Oanh Tran, Duyen Thi Do, and Nguyen Quoc Khanh Le. 2022. "A Radiomics-Based Machine Learning Model for Prediction of Tumor Mutational Burden in Lower-Grade Gliomas" Cancers 14, no. 14: 3492. https://doi.org/10.3390/cancers14143492