Identification of genes modified by N6-methyladenosine in patients with colorectal cancer recurrence

Background: Recent studies demonstrate that N6-methyladenosine (m(6)A) methylation plays a crucial role in colorectal cancer (CRC). Therefore, we conducted a comprehensive analysis to assess the m(6)A modification patterns and identify m(6)A-modified genes in patients with CRC recurrence. Methods: The m(6)A modification patterns were comprehensively evaluated by the NMF algorithm based on the levels of 27 m(6)A regulators, and tumor microenvironment (TME) cell-infiltrating characteristics of these modification patterns were systematically assessed by ssGSEA and CIBERSORT algorithms. The principal component analysis algorithm based on the m(6)A scoring scheme was used to explore the m(6)A modification patterns of individual tumors with immune responses. The weighted correlation network analysis and univariable and multivariable Cox regression analyses were applied to identify m(6)A-modified gene signatures. The single-cell expression dataset of CRC samples was used to explore the tumor microenvironment affected by these signatures. Results: Three distinct m(6)A modification patterns with significant recurrence-free survival (RFS) were identified in 804 CRC patients. The TME characterization revealed that the m(6)A modification pattern with longer RFS exhibited robust immune responses. CRC patients were divided into high- and low-score subgroups according to the m(6)A score individually, which was obtained from the m(6)A-related signature genes. The patients with low m(6)A scores had both longer RFS and overall survival (OS) with altered immune cell infiltration. Notably, m(6)A-modified genes showed significant differences related to the prognosis of CRC patients in the meta-GEO cohort and TCGA cohort. Single-cell expression indicated that ALVRL1 was centrally distributed in endothelial tip cells and stromal cells. Conclusion: The m(6)A modification plays an indispensable role in the formation of TME diversity and complexity. Importantly, the signatures (TOP2A, LRRC58, HAUS6, SMC4, ACVRL1, and KPNB1) were identified as m(6)A-modified genes associated with CRC recurrence, thereby serving as a promising predictive biomarker or therapeutic target for patients with CRC recurrence.