Construction of a novel prognostic risk model based on m6A-related miRNAs for acute myeloid leukemia

Abstract

Abstract N6-methyladenosine (m6A) modification and miRNAs were important in tumorigenesis and development. We aim to identify prognostic markers and construct a risk prediction model for AML patients. First, 17 prognostic m6A-related miRNAs were filtrated, whose expression profiles were included to cluster patients into 2 subtypes. The OS of cluster1 had worse prognosis. GSEA analysis showed cluster1 enriched in tumor-related pathways, including Toll like receptor signaling pathway, PPAR signaling pathway and apoptosis. Next, 10 miRNAs filtered by LASSO regression analysis were used to construct a risk model. Patients in high-risk groups had unfavorable prognosis and risk score might could act as the independent prognostic factors in AML. The expression of immune checkpoints (PD-L1, LAG-3 and CTLA4) were higher in high-risk group. Finally, we built a regulatory network of m6A regulators- m6A-related miRNAs- target mRNAs. The GO function analysis showed the target genes were enriched in the biological process related with leukemia, including tissue morphogenesis, regulation of leukocyte migration, positive regulation of cell adhesion and so on. The KEGG pathway analysis indicated that these genes were mainly enriched in Ras signaling pathway and signaling pathways regulating pluripotency of stem cells. The finding provided novel implication for efforts to improve the treatment of AML.