Development and validation of a neutrophil extracellular traps-related gene signature for lower-grade gliomas

There is growing evidence linking neutrophil extracellular traps (NETs) to tumor genesis, growth, distant metastasis, and tumor-related thrombosis. However, the roles of NETs-related genes (NETRGs) on LGG prognosis remain unclear. The purpose of this study was to integrate multiple machine learning techniques and experiment validation to develop a reliable NETs-based signature that opens up novel approaches for assessing the prognosis and treatment response of LGG patients. Consensus clustering, k-means clustering and Nonnegative Matrix Factorization was used for the TCGA-LGG dataset and identified two NETs-related subgroups. The prognostic hallmark and nomogram for LGG were developed, which consist of five differentially expressed NETRGs (FPR1, PTAFR, SLC11A1, ICAM1, LTF) based on nine analytic approaches. The ROC curves and calibration curves of our NETRGs signature and nomogram exhibited strong and robust prognosis prediction abilities in both the TCGA-LGG training set and CGGA-325, CGGA-693 validation sets. The prognosis for LGG individuals in the low-risk category was better. The TISCH was used to examine the five NETRGs at the single-cell level. Common immunological checkpoints were expressed at greater levels in high-risk individuals. LGG individuals in the low-risk category posses a higher likelihood of being sensitive to Carmustine and Vincristine, as indicated by the drug sensitivity analysis. The qRT-PCR experiment and immunohistochemistry images confirmed that the expression of FPR1, PTAFR, SLC11A1 and ICAM1 are higher in low-grade oligodendroglioma. The NETRGs signature and nomogram can accurately and conveniently predict the LGG patients' prognosis, which can facilitate individualized treatment and the improvement of prognosis.