Integrating Radiosensitivity Gene Signature Improves Glioma Outcome and Radiotherapy Response Prediction

Response to radiotherapy (RT) in gliomas varies widely between patients. It is necessary to identify glioma-associated radiosensitivity gene signatures for clinically stratifying patients who will benefit from adjuvant radiotherapy after glioma surgery. Methods: Chinese Glioma Genome Atlas (CGGA) and the Cancer Genome Atlas (TCGA) glioma patient datasets were used to validate the predictive potential of two published biomarkers, the radiosensitivity index (RSI) and 31-gene signature (31-GS). To adjust these markers for the characteristics of glioma, we integrated four new glioma-associated radiosensitivity predictive indexes based on RSI and 31-GS by the Cox analysis and Least Absolute Shrinkage and Selection Operator (LASSO) regression analysis. A receiver operating characteristic (ROC) curve, integrated discrimination improvement (IDI), and net reclassification improvement (NRI) were used to compare the radiosensitivity predictive ability of these six gene signatures. Subgroup analysis was used to evaluate the discriminative capacity of those gene signatures in identifying radiosensitive patients, and a nomogram was built to improve the histological grading system. Gene Ontology (GO) analysis and Gene Set Enrichment Analysis (GSEA) were used to explore related biological processes. Results: We validated and compared the predictive potential of two published predictive indexes. The AUC area of 31-GS was higher than that of RSI. Based on the RSI and 31-GS, we integrated four new glioma-associated radiosensitivity predictive indexes-PI10, PI12, PI31 and PI41. Among them, a 12-gene radiosensitivity predictive index (PI12) showed the most promising predictive performance and discriminative capacity. Examination of a nomogram created from clinical features and PI12 revealed that its predictive capacity was superior to the traditional WHO classification system. (C-index: 0.842 vs. 0.787, p ≤ 2.2 × 10^-16) The GO analysis and GSEA showed that tumors with a high PI12 score correlated with various aspects of the malignancy of glioma. Conclusions: The glioma-associated radiosensitivity gene signature PI12 is a promising radiosensitivity predictive biomarker for guiding effective personalized radiotherapy for gliomas.