10087-GGE-7 THE IMMUNE PROGNOSTIC MODEL FOR GLIOBLASTOMA BASED ON SSGSEA ENRICHMENT SCORE
Takanari Okamoto, Daisuke Muraoka, Ayako Okamura, Ryo Mizuta, Shota Nohira, Yoshinobu Takahashi, Ichita Taniyama, Takahiro Ogawa, Naoya Hashimoto, Hirokazu Matsushita- Surgery
- Oncology
- Neurology (clinical)
Abstract
BACKGROUND
A deep understanding of Tumor Microenvironment (TME) holds the potential to develop effective therapeutic strategies for glioblastoma. However, there have been few reports on the comprehensive evaluation of the interaction between cancer cells and the immune system within TME. This study aimed to construct an immune prognostic model for glioblastoma based on the gene expression profiles in the tumor.
METHODS
Using The Cancer Genome Atlas (TCGA) RNA-seq data (141 cases of glioblastoma), we performed comprehensive single-sample gene set enrichment analysis (ssGSEA) on gene sets from the Molecular Signatures Database (including Hallmark, Curated, and Gene Ontology). After evaluating gene sets associated with prognosis through univariate Cox regression, we extracted gene sets related to the biological process and tumor immunity of gliomas. Finally, we employed Lasso regression to refine the gene sets and constructed a multivariate Cox regression model (nomogram). This immune prognostic model was validated using the Chinese Glioma Genome Atlas (CGGA) dataset (183 cases).
RESULTS
The immune prognostic model consisted of three gene sets related to biological processes (sphingolipids, steroid hormones, and intermediate filaments) and one related to tumor immunity (chemokine mediated signaling pathway involved in suppressed immune system). Kaplan-Meier curves for the training cohort (TCGA) and the validation cohort (CGGA) demonstrated that the high-risk group had significantly worse overall survival compared to the low-risk group (p<0.01 and p=0.04, respectively). Additionally, in silico cytometry using CIBERSORTx to assess the immune cell infiltration profiles revealed a significant increase in the fraction of M2 macrophages, an immunosuppressive cell type, in the high-risk group (p<0.01) in both cohorts.
CONCLUSION
Comprehensive TME assessment using ssGSEA and constructing an immune prognostic model could potentially provide valuable insights into the prognosis and immune profiles of glioblastoma patients, guiding treatment strategies.