Identification of immunogenic cell death gene-related subtypes and risk model predicts prognosis and response to immunotherapy in ovarian cancer.

Abstract

Background: Immunogenic cell death (ICD) has been associated with enhanced anti-tumor immunotherapy by stimulating adaptive immune responses and remodeling the immune microenvironment in tumors. Nevertheless, the role of ICD-related genes in ovarian cancer (OC) and tumor microenvironment remains unexplored.

Methods: In this study, high-throughput transcriptomic data from The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO) databases as training and validation sets separately were obtained and proceeded to explore ICD-related clusters, and an ICD-related risk signature was conducted based on the least absolute shrinkage and selection operator (LASSO) Cox regression model by iteration. Multiple tools including CIBERSORT, ESTIMATE, GSEA, TIDE, and immunohistochemistry were further applied to illustrate the biological roles of ICD-related genes as well as the prognostic capacity of ICD risk signature in OC.

Results: Two ICD-related subtypes were identified, with the ICD-high subtype showing more intense immune cell infiltration and higher activities of immune response signaling, along with a favorable prognosis. Additionally, four candidate ICD genes (IFNG, NLRP3, FOXP3, and IL1B) were determined to potentially impact OC prognosis, with an upregulated expression of NLRP3 in OC and metastatic omental tissues. A prognostic model based on these genes was established, which could predict overall survival (OS) and response to immunotherapy for OC patients, with lower-risk patients benefiting more from immunotherapy.

Conclusion: Our research conducted a prognostic and prediction of immunotherapy response model based on ICD genes, which could be instrumental in assessing prognosis and assigning immunotherapeutic strategies for OC patients. NLRP3 is a promising target for prognosis in OC.