Cancer immune profiling unveils biomarkers, immunological pathways, and cell type score associated with glioblastoma patients’ survival

Abstract

Introduction: Glioblastoma (GBM), isocitrate dehydrogenase (IDH) wild-type (IDHwt), and grade 4 astrocytomas, IDH mutant (IDHmut), are the most common and aggressive primary malignant brain tumors in adults. A better understanding of the tumor immune microenvironment may provide new biomarkers and therapeutic opportunities. Objectives: We aimed to evaluate the expression profile of 730 immuno-oncology-related genes in patients with IDHwt GBM and IDHmut tumors and identify prognostic biomarkers and a gene signature associated with patient survival. Methods: RNA was isolated from formalin-fixed, paraffin-embedded sections of 99 tumor specimens from patients treated with standard therapy. Gene expression profile was assessed using the Pan-Cancer Immune Profiling Panel (Nanostring Technologies, Inc., Seattle, WA, USA). Data analysis was performed using nSolverSoftware and validated in The Cancer Genome Atlas. In addition, we developed a prognostic signature using the cox regression algorithm (Least Absolute Shrinkage and Selection Operator). Results: We found 88 upregulated genes, high immunological functions, and a high macrophage score in IDHwt GBM compared to IDHmut tumors. Regarding IDHwt GBM, we found 24 upregulated genes in short-term survivors (STS) and overexpression of CD274 (programmed death-ligand 1, PD-L1). Immune pathways, CD45, cytotoxic, and macrophage scores were upregulated in STS. Two different prognostic groups were found based on the 12-gene signature (CXCL14, PSEN2, TNFRSF13C, IL13RA1, MAP2K1, TNFSF14, THY1, CTSL, ITGAE, CHUK, CD207, and IFITM1). Conclusion: The elevated expression of immune-oncology-related genes was associated with worse outcome in IDHwt GBM patients. Increased immune functions, CD45, cytotoxic cells, and macrophage scores were associated with a more aggressive phenotype and may provide promising possibilities for therapy. Moreover, a 12 gene-based signature could predict patients’ prognosis.