Comprehensive Analysis of N6-Methyladenosine (mA) RNA Methylation Regulators and Tumour Microenvironment Cell Infiltration Involving Prognosis and Immunotherapy in Gastroesophageal Adenocarcinomas.

OBJECTIVE

Gastroesophageal adenocarcinoma (GEA) is a high deadly and heterogeneous cancer. RNA N6-methyladenosine (mA) modification plays a non-negligible role in shaping individual tumour microenvironment (TME) characterizations. However, the landscape and relationship of mA modification patterns and TME cell infiltration features remain unknown in GEA.

METHODS

In this study, we examined the TME of GEA using assessments of the RNA-sequencing data focusing on the distinct mA modification patterns from the public databases. Intrinsic patterns of mA modification were evaluated for associations with clinicopathological characteristics, underlying biological pathways, tumour immune cell infiltration, oncological outcomes, and treatment responses. The expression of key mA regulators and module genes was validated by qRT-PCR analysis.

RESULTS

We identified two distinct mA modification patterns of GEA (cluster 1/2 subgroup), and correlated two subgroups with TME cell-infiltrating characteristics. Cluster 2 subgroup correlates with a poorer prognosis, downregulated PD-1 expression, higher risk scores, and distinct immune cell infiltration. In addition, PPI and WGCNA network analysis were integrated to identify key module genes closely related to immune infiltration of GEA to find immunotherapy markers. COL4A1 and COL5A2 in the brown module were significantly correlated to the prognosis, PD-1/L1 and CTLA-4 expression of GEA patients. Finally, a prognostic risk score was constructed using mA regulator-associated signatures that represented an independent prognosis factor for GEA. Interestingly, COL5A2 expression was linked to the response to anti-PD-1 immunotherapy, mA regulator expression, and risk score.

CONCLUSION

Our work identified mA RNA methylation regulators as an important class of players in the malignant progression of GEA and were associated with the complexity of the TME. COL5A2 may be the potential biomarker which contributes to predicting the response to anti-PD-1 immunotherapy and patients' prognosis.