Abstract
Glioblastoma multiforme (GBM) is a highly malignant primary brain tumour with a poor prognosis in adults. Identifying biomarkers that can aid in the molecular classification and risk stratification of GBM is critical. Here, we conducted a transcriptional profiling analysis of T-cell immunity in the tumour microenvironment of GBM patients and identified two novel T cell exhaustion (TEX)-related GBM subtypes (termed TEX-C1 and TEX-C2) using the consensus clustering. Our multi-omics analysis revealed distinct immunological, molecular and clinical characteristics for these two subtypes. Specifically, the TEX-C1 subtype had higher infiltration levels of immune cells and expressed higher levels of immune checkpoint molecules than the TEX-C2 subtype. Functional analysis revealed that upregulated genes in the TEX-C1 subtype were significantly enriched in immune response and signal transduction pathways, and upregulated genes in the TEX-C2 subtype were predominantly associated with cell fate and nervous system development pathways. Notably, patients with activated T-cell activity status in the TEX-C1 subgroup demonstrated a significantly worse prognosis than those with severe T cell exhaustion status in the TEX-C2 subgroup. Finally, we proposed a machine-learning-derived novel gene signature comprising 12 TEX-related genes (12TexSig) to indicate tumour subtyping. In the TCGA cohort, the 12TexSig demonstrated the ability to accurately predict the prognosis of GBM patients, and this prognostic value was further confirmed in two independent external cohorts. Taken together, our results suggest that the TEX-derived subtyping and gene signature has the potential to serve as a clinically helpful biomarker for guiding the management of GBM patients, pending further prospective validation.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have