Abstract
Ovarian cancer (OC) is the second leading cause of death in gynecological cancer. Multiple study have shown that the efficacy of tumor immunotherapy is related to tumor immune cell infiltration (ICI). However, so far, the Immune infiltration landscape of tumor microenvironment (TME) in OC has not been elucidated. In this study, We organized the transcriptome data of OC in the Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO) databases, evaluated the patient's TME information, and constructed the ICI scores to predict the clinical benefits of patients undergoing immunotherapy. Immune-related genes were further used to construct the prognostic model. After clustering analysis of ICI genes, we found that patients in ICI gene cluster C had the best prognosis, and their tumor microenvironment had the highest proportion of macrophage M1 and T cell follicular helper cells. This result was consistent with that of multivariate cox (multi-cox) analysis. The prognostic model constructed by immune-related genes had good predictive performance. By estimating Tumor mutation burden (TMB), we also found that there were multiple genes with statistically different mutation frequencies in the high and low ICI score groups. The model based on the ICI score may help to screen out patients who would benefit from immunotherapy. The immune-related genes screened may be used as biomarkers and therapeutic targets.