Inflammatory expression profiling defines the tumor microenvironment and immunity in lower-grade gliomas

Objective: Lower-grade glioma (LGG) arises from glial or precursor cells including the world health organization (WHO) grade glioma, which has a favorable prognosis compared with glioblastomas. However, LGG progression is inevitable due to drug resistance and tumor invasiveness. The inflammatory response has been a well-established hallmark associated with tumor occurrence and recurrence. It is necessary to explore the heterogeneity of the inflammatory response in LGG.Methods: The differential and prognostic inflammatory response-related genes(IRRG) of the cancer genome atlas(TCGA) database were used to perform consensus clustering to identify robust clusters. The Chinese Glioma Genome Atlas(CGGA) database was used to validate this inflammation subtype. The clinical characteristics, immune infiltration, metabolism, and somatic variation of each cluster were assessed. A risk score associated with inflammation was constructed.Results: We successfully observed and validated three distinct inflammatory subtypes in LGG. It is noteworthy that the B subtype has a shorter overall survival than the A and C subtypes. The B subtype had higher immune, stromal, and ESTIMATE scores than the A and C subtypes but lower tumor purity than the A and C subtypes. In addition, immune, stromal, and ESTIMATE scores of the A subtype were higher than the hose of the C subtype, while tumor purity was lower than that of the C subtype. Three inflammation subtypes correlated with diverse metabolic characteristics, drug sensitivity, and somatic alterations. Furthermore, we developed and validated an inflammatory signature with better performance of prognosis prediction.Conclusion: This study identified a new classification of LGG according to inflammatory gene profiling and emphasized inflammatory heterogeneity within LGG.