Development and validation of a personalized classifier to predict the prognosis and response to immunotherapy in glioma based on glycolysis and the tumor microenvironment

BACKGROUND: Glycolysis is closely associated with cancer progression and treatment outcomes. However, the role of glycolysis in the immune microenvironment, prognosis, and immunotherapy of glioma remains unclear. METHODS: This study investigated the role of glycolysis on prognosis and its relationship with the tumor microenvironment (TME). Subsequently, we developed and validated the glycolysis-related gene signature (GRS)-TME classifier using multiple independent cohorts. Furthermore, we also examined the prognostic value, somatic alterations, molecular characteristics, and potential benefits of immunotherapy based on GRS-TME classifier. Lastly, the effect of kinesin family member 20A (KIF20A) on the proliferation and migration of glioma cells was evaluated in vitro. RESULTS: Glycolysis was identified as a significant prognostic risk factor in glioma, and closely associated with an immunosuppressive microenvironment characterized by altered distribution of immune cells. Furthermore, a personalized GRS-TME classifier was developed and validated by combining the glycolysis (18 genes) and TME (seven immune cells) scores. Patients in the GRS(low)/TME(high) subgroup exhibited a more favorable prognosis compared to other subgroups. Distinct genomic alterations and signaling pathways were observed among different subgroups, which are closely associated with cell cycle, epithelial—mesenchymal transition, p53 signaling pathway, and interferon-alpha response. Additionally, we found that patients in the GRS(low)/TME(high) subgroup exhibit a higher response rate to immunotherapy, and the GRS-TME classifier can serve as a novel biomarker for predicting immunotherapy outcomes. Finally, high expression of KIF20A is associated with an unfavorable prognosis in glioma, and its knockdown can inhibit the proliferation and migration of glioma cells. CONCLUSIONS: Our study developed a GRS-TME classifier for predicting the prognosis and potential benefits of immunotherapy in glioma patients. Additionally, we identified KIF20A as a prognostic and therapeutic biomarker for glioma.