N6-Methyladenosine Modification Patterns and Tumor Microenvironment Immune Characteristics Associated With Clinical Prognosis Analysis in Stomach Adenocarcinoma

Background: N6-methyladenosine (m6A) modification is a part of epigenetic research that has gained increasing attention in recent years. m6A modification is widely involved in many biological behaviors of intracellular RNA by regulating mRNA, thus affecting disease progression and tumor occurrence. However, the effects of m6A modification on immune cell infiltration of the tumor microenvironment (TME) are uncertain in stomach adenocarcinoma (STAD). Methods: The Cancer Genome Map (TCGA) database was used to download transcriptome data, clinicopathological data, and survival data for m6A-regulated genes in 433 STAD tissues that meet the requirements of this study. GSE84437 data were obtained from the Gene Expression Omnibus (GEO) database. The correlation between 23 m6A regulated genes was analyzed using R software. Sample clustering analysis was carried out on the genes of the m6A regulatory factor, and survival analysis and differentiation comparison were made for patients in clustering grouping. Then, the Gene Set Enrichment Analysis (GSEA), the single-sample GSEA (ssGSEA), and other methods were conducted to assess the correlation among m6A modification patterns, TME cell infiltration characteristics, and immune infiltration markers. The m6A modification pattern of individual tumors was quantitatively evaluated using the m6A score scheme of the principal component analysis (PCA). Results: From the TCGA database, 94/433 (21.71%) samples were somatic cell mutations, and ZC3H13 mutations are the most common. Based on the consensus, matrix k-3 is an optimal clustering stability value to identify three different clusters. Three types of m6A methylation modification patterns were significantly different in immune infiltration. Thus, 1028 differentially expressed genes (DEGs) were identified. The survival analysis of the m6A score found that patients in the high m6A score group had a better prognosis than those in the low m6A score group. Further analysis of the survival curve combining tumor mutation burden (TMB) and m6A scores revealed that patients had a significantly lower prognosis in the low tumor mutant group and the low m6A score group (p = 0.003). The results showed that PD-L1 was significantly higher in the high m6A score group than in the low score group (p < 2.22e-16). The high-frequency microsatellite instability (MSI-H) subtype score was significantly different from the other two groups. Conclusions: This study systematically evaluated the modification patterns of 23 m6A regulatory factors in STAD. The m6A modification pattern may be a critical factor leading to inhibitory changes and heterogeneity in TME. This elucidated the TME infiltration characteristics in patients with STAD through the evaluation of the m6A modification pattern.