Extended Application of Genomic Selection to Screen Multi-Omics Data for the Development of Novel Pyroptosis-Immune Signatures and Predicting Immunotherapy of Glioma

Abstract

Glioma is one of the most human malignant diseases and the leading cause of cancer-related deaths worldwide. Nevertheless, the present stratification systems do not accurately predict the prognosis and treatment benefit of glioma patients. Currently, no comprehensive analyses of multi-omics data have been performed to better understand the complex link between pyroptosis and immune. In this study, we constructed four pyroptosis immune subgroups by pyroptosis regulators and obtained nine pyroptosis immune signatures by analyzing the differentially expressed genes between the four pyroptosis immune subgroups. Nine novel pyroptosis immune signatures were provided for assessing the complex heterogeneity of glioma by the analyses of multi-omics data. The pyroptosis immune prognostic model (PIPM) was constructed by pyroptosis immune signatures, and the PIPM risk score was established for glioma cohorts with a total of 1716 samples. Then, analyses of the tumor microenvironment revealed an unanticipated correlation of the PIPM risk score with stemness, immune checkpoint expression, infiltrating the immune system, and therapy response in glioma. The low PIPM risk score patients had a better response to immunotherapy and showed sensitivity to radio-chemotherapy. The results of the pan-cancer analyses revealed the significant correlation between the PIPM risk score and clinical outcome, immune infiltration, and stemness. Taken together, we conclude that pyroptosis immune signatures may be a helpful tool for overall survival prediction and treatment guidance for glioma and other tumors patients.