Clinical measures, radiomics, and genomics offer synergistic value in AI-based prediction of overall survival in patients with glioblastoma

• Published in: Scientific reports Vol. 12; no. 1; pp. 8784 - 13
• Main Authors: Fathi Kazerooni, Anahita, Saxena, Sanjay, Toorens, Erik, Tu, Danni, Bashyam, Vishnu, Akbari, Hamed, Mamourian, Elizabeth, Sako, Chiharu, Koumenis, Costas, Verginadis, Ioannis, Verma, Ragini, Shinohara, Russell T., Desai, Arati S., Lustig, Robert A., Brem, Steven, Mohan, Suyash, Bagley, Stephen J., Ganguly, Tapan, O’Rourke, Donald M., Bakas, Spyridon, Nasrallah, MacLean P., Davatzikos, Christos
• Format: Journal Article
• Language: English
• Published: London Nature Publishing Group UK 24.05.2022; Nature Publishing Group; Nature Portfolio
• Subjects: 631/114/1305; 631/114/2413; 692/4028/67/1922; 692/4028/67/2321; 692/4028/67/69; Artificial Intelligence; Brain cancer; Brain Neoplasms - diagnostic imaging; Brain Neoplasms - genetics; Brain Neoplasms - pathology; Deep learning; Genomics; Glioblastoma; Glioblastoma - diagnostic imaging; Glioblastoma - genetics; Glioblastoma - pathology; Humanities and Social Sciences; Humans; Isocitrate dehydrogenase; Magnetic resonance imaging; Magnetic Resonance Imaging - methods; multidisciplinary; Patients; Predictions; Radiomics; Replication; Reproducibility of Results; Retrospective Studies; Risk assessment; Science; Science (multidisciplinary); Survival; Tumors
• ISSN: 2045-2322; 2045-2322
• DOI: 10.1038/s41598-022-12699-z

Multi-omic data, i.e., clinical measures, radiomic, and genetic data, capture multi-faceted tumor characteristics, contributing to a comprehensive patient risk assessment. Here, we investigate the additive value and independent reproducibility of integrated diagnostics in prediction of overall survival (OS) in isocitrate dehydrogenase (IDH)-wildtype GBM patients, by combining conventional and deep learning methods. Conventional radiomics and deep learning features were extracted from pre-operative multi-parametric MRI of 516 GBM patients. Support vector machine (SVM) classifiers were trained on the radiomic features in the discovery cohort (n = 404) to categorize patient groups of high-risk (OS < 6 months) vs all, and low-risk (OS ≥ 18 months) vs all. The trained radiomic model was independently tested in the replication cohort (n = 112) and a patient-wise survival prediction index was produced. Multivariate Cox-PH models were generated for the replication cohort, first based on clinical measures solely, and then by layering on radiomics and molecular information. Evaluation of the high-risk and low-risk classifiers in the discovery/replication cohorts revealed area under the ROC curves (AUCs) of 0.78 (95% CI 0.70–0.85)/0.75 (95% CI 0.64–0.79) and 0.75 (95% CI 0.65–0.84)/0.63 (95% CI 0.52–0.71), respectively. Cox-PH modeling showed a concordance index of 0.65 (95% CI 0.6–0.7) for clinical data improving to 0.75 (95% CI 0.72–0.79) for the combination of all omics. This study signifies the value of integrated diagnostics for improved prediction of OS in GBM.