Abstract
BACKGROUND Lower-grade gliomas (LGGs) vary widely in terms of the patient’s overall survival (OS). There is no current, valid method that could exactly predict the survival. The effects of intratumoral immune infiltration on clinical outcome have been widely reported. Thus, we aim to develop an immune infiltration signature to predict the survival of LGG patients.METHODS We analyzed 1216 LGGs from 5 public data sets, including 2 RNA sequencing data sets and 3 microarray data sets. Least absolute shrinkage and selection operator (LASSO) Cox regression was used to select an immune infiltration signature and build a risk score. The performance of the risk score was assessed in the training set (329 patients), internal validation set (140 patients), and 4 external validation sets (405, 118, 88, and 136 patients).RESULTS An immune infiltration signature consisting of 20 immune metagenes was used to generate a risk score. The performance of the risk score was thoroughly verified in the training and validation sets. Additionally, we found that the risk score was positively correlated with the expression levels of TGF-β and PD-L1, which were important targets of combination immunotherapy. Furthermore, a nomogram incorporating the risk score, patient’s age, and tumor grade was developed to predict the OS, and it performed well in all the training and validation sets (C-index: 0.873, 0.881, 0.781, 0.765, 0.721, and 0.753).CONCLUSION The risk score based on the immune infiltration signature has reliable prognostic and predictive value for patients with LGGs and is a potential biomarker for the cotargeting immunotherapy.FUNDING This work was supported by The National Natural Science Foundation of China (grant nos. 81472370 and 81672506), the Natural Science Foundation of Beijing (grant no. J180005), the National High Technology Research and Development Program of China (863 Program, grant no. 2014AA020610), and the National Basic Research Program of China (973 Program, grant no. 2014CB542006).
Authors
Lai-Rong Song, Jian-Cong Weng, Cheng-Bei Li, Xu-Lei Huo, Huan Li, Shu-Yu Hao, Zhen Wu, Liang Wang, Da Li, Jun-Ting Zhang
Figure 8
Predictive nomogram and calibration curves.
