Discover Oncology (Jun 2023)

M1A and m7G modification-related genes are potential biomarkers for survival prognosis and for deciphering the tumor immune microenvironment in esophageal squamous cell carcinoma

  • Ruixi Wang,
  • Xingyuan Cheng,
  • Dongmei Chi,
  • Shiliang Liu,
  • Qiaoqiao Li,
  • Baoqing Chen,
  • Mian Xi

DOI
https://doi.org/10.1007/s12672-023-00710-6
Journal volume & issue
Vol. 14, no. 1
pp. 1 – 16

Abstract

Read online

Abstract Background Esophageal squamous cell carcinoma (ESCC) is the most common esophageal malignancy, and RNA methylation has been reported to be involved in the tumorigenesis of ESCC. However, no study has explored methylation modifications in m1A and m7G as prognostic markers for survival prediction in ESCC. Methods Public gene-expression data and clinical annotation of 254 patients obtained from The Cancer Genome Atlas and the Gene Expression Omnibus databases were analyzed to identify potential consensus clusters of m1A and m7G modification-related genes. The RNA-seq of 20 patients in Sun Yat-Sen University Cancer Center was used as the validation set. Following screening for relevant differentially expressed genes (DEGs) and enrichment pathways were elucidated. DEGs were used to construct risk models using the randomForest algorithm, and the prognostic role of the models was assessed by applying Kaplan–Meier analysis. Extent of immune cell infiltration, drug resistance, and response to cancer treatment among different clusters and risk groups were also evaluated. Results Consensus clustering analysis based on m1A and m7G modification patterns revealed three potential clusters. In total, 212 RNA methylation-related DEGs were identified. The methylation-associated signature consisting of 6 genes was then constructed to calculate methylation-related score (MRScore) and patients were dived into MRScore-high and MRScore-low groups. This signature has satisfied prognostic value for survival of ESCC (AUC = 0.66, 0.67, 0.64 for 2-, 3-, 4- year OS), and has satisfied performance in the validation SYSUCC cohort (AUC = 0.66 for 2- and 3-year OS). Significant correlation between m1A and m7G modification-related genes and immune cell infiltration, and drug resistance was also observed. Conclusions Transcriptomic prognostic signatures based on m1A and m7G modification-related genes are closely associated with immune cell infiltration in ESCC patients and have important correlations with the therapeutic sensitivity of multiple chemotherapeutic agents.

Keywords