Identification of the role of endoplasmic reticulum stress genes in endometrial cancer and their association with tumor immunity

• Main Authors: Tang ansu Zhang (Author), Qian Zhang (Author), Jun Zhang (Author), Rong Zhao (Author), Rui Shi (Author), Sitian Wei (Author), Shuangge Liu (Author), Qi Zhang (Author), Hongbo Wang (Author)
• Format: Book
• Published: BMC, 2023-10-01T00:00:00Z.
• Subjects: article
• Online Access: Connect to this object online.
• Notes: 10.1186/s12920-023-01679-5; 1755-8794
• Keywords: Endometrial cancer; Endoplasmic reticulum stress; Bioinformatic analysis; Risk signature; Prognosis; Immune infiltration; Internal medicine; RC31-1245; Genetics; QH426-470
• Source: BMC Medical Genomics, Vol 16, Iss 1, Pp 1-19 (2023)

Abstract Background Endometrial cancer (EC) is one of the worldwide gynecological malignancies. Endoplasmic reticulum (ER) stress is the cellular homeostasis disturbance that participates in cancer progression. However, the mechanisms of ER Stress on EC have not been fully elucidated. Method The ER Stress-related genes were obtained from Gene Set Enrichment Analysis (GSEA) and GeneCards, and the RNA-seq and clinical data were downloaded from The Cancer Genome Atlas (TCGA). The risk signature was constructed by the Cox regression and the least absolute shrinkage and selection operator (LASSO) analysis. The significance of the risk signature and clinical factors were tested by time-dependent receiver operating characteristic (ROC) curves, and the selected were to build a nomogram. The immunity correlation was particularly analyzed, including the related immune cells, pathways, and immune checkpoints. Functional enrichment, potential chemotherapies, and in vitro validation were also conducted. Result An ER Stress-based risk signature, consisting of TRIB3, CREB3L3, XBP1, and PPP1R15A was established. Patients were randomly divided into training and testing groups with 1:1 ratio for subsequent calculation and validation. Based on risk scores, high- and low-risk subgroups were classified, and low-risk subgroup demonstrated better prognosis. The Area Under Curve (AUC) demonstrated a reliable predictive capability of the risk signature. The majority of significantly different immune cells and pathways were enriched more in low-risk subgroup. Similarly, several typical immune checkpoints, expressed higher in low-risk subgroup. Patients of the two subgroups responded differently to chemotherapies. Conclusion We established an ER Stress-based risk signature that could effectively predict EC patients' prognosis and their immune correlation.