loading page

Prognostic risk model of immune-related genes in ovarian cancer
  • +1
  • Bo Ding,
  • Wenjing Yan,
  • Shizhi Wang,
  • Yang Shen
Bo Ding
Southeast University
Author Profile
Wenjing Yan
Southeast University
Author Profile
Shizhi Wang
Southeast University
Author Profile
Yang Shen
Southeast University Zhongda Hospital

Corresponding Author:[email protected]

Author Profile

Abstract

The prognosis of patients with ovarian cancer (OC) is highly heterogeneous which hinders to make an appropriate clinical decision. The study aimed to stratify patients’ prognoses by establishing a risk assessment model in the context of mRNA levels of immune-related genes (IRGs). Comprehensive bioinformatics analyses were done using datasets from The UCSC Xena platform, ICGC Data Portal, The Cancer Genome Atlas (TCGA), and the Genotype‐Tissue Expression (GTEx) project. LASSO regression was done to determine the independence of associations of specific factors with overall survival (OS). Nomogram that combined the independent prognostic factors was constructed to predict the OS of OC patients. The tumor microenvironment and immune response were estimated by cell type identification via estimating the relative subset of known RNA transcripts (CIBERSORT) and immunophenoscore (IPS). Overall ten IRGs were significantly associated with the OS probability and were used for the prognostic model construction of OC patients. According to the prognostic model, ovarian cancer samples were identified as high- or low-risk group. A nomogram containing risk score, stage, histological grade and age could significantly predict 1-year, 3-years, and 5-years OS probability of OC patients, as well as a higher IPS score and a higher immunoreactivity phenotype, which were probably correlated with better immunotherapy response and good prognosis. In conclusion, we established a reliable IRGs-based risk model with potential prognostic value for patients with ovarian cancer. Further studies are needed to confirm these prognosis-associated biomarkers.