BACKGROUND:Myelodysplastic syndrome has worse hazards of acute myeloid leukemia transformation,and some studies have revealed that immune infiltration plays a vital part in the two.Nevertheless,more studies are required to confirm the relationship between immune infiltration and related differentially expressed gene regulation. OBJECTIVE:To screen the differentially expressed genes with prognostic significance between myelodysplastic syndrome and acute myeloid leukemia by bioinformatics analysis and explore the possible roles and mechanisms among these differentially expressed genes and immune infiltration mechanisms in the occurrence and progression of diseases. METHODS:The differentially expressed genes were screened for bioinformatics analysis using the GEO datasets,and analyzed by DO,GO,KEGG and GSEA.The TCGA prognostic database was used to plot the K-M curves of differentially expressed genes and receiver operating characteristic curve analysis was applied to evaluate the clinical diagnostic performance.Finally,CIBERSORT analysis was used to intuitively demonstrate the correlation between critical prognostic genes and the distribution of immuno-infiltrated cells.RT-qPCR was employed to detect peripheral blood samples from healthy controls,myelodysplastic syndrome and acute myeloid leukemia patients so as to verify the crucial genes preliminarily. RESULTS AND CONCLUSION:(1)A total of 150 differentially expressed genes were obtained between myelodysplastic syndrome and acute myeloid leukemia,among which 16 genes were up-regulated and 134 were down-regulated.(2)The results of DO,GO,KEGG and GSEA analysis suggested that differentially expressed genes might promote the development of myelodysplastic syndrome to acute myeloid leukemia by regulating the immune response.CIBERSORT revealed the differences in immune infiltration between myelodysplastic syndrome and acute myeloid leukemia.The distribution of CD4+ T cells,monocytes,neutrophils and M1 macrophages decreased in acute myeloid leukemia patients.In contrast,the distribution of inflammatory suppressor cells M2 macrophages increased,suggesting that it may be related to the immunosuppression of acute myeloid leukemia.(3)K-M curve and receiver operating characteristic curve analysis of 150 differentially expressed genes screened out four genes relevant to immunity and prognosis with good diagnostic performance:MANSC1,FLT3,BMX and CXCR2.(4)The results of RT-qPCR exhibited that MANSC1,BMX and CXCR2 were low expressed,while FLT3 was highly expressed in acute myeloid leukemia patients.These findings verify that the differential expression of MANSC1,FLT3,BMX and CXCR2 in patients with myelodysplastic syndrome and acute myeloid leukemia is not only significantly correlated with the prognosis of patients but may also affect the occurrence and development of myelodysplastic syndrome and acute myeloid leukemia by regulating the immune infiltration of patients.They can be used as potential biomarkers and therapeutic targets of the transformation from myelodysplastic syndrome to acute myeloid leukemia,providing a new direction for clinical diagnosis and treatment of the transformation of myelodysplastic syndrome.