Construction and validation of key genes-related prognosis model in children with acute myeloid leukaemia.

INTRODUCTION: To identify the differentially expressed genes of acute myeloid leukaemia (AML) and construct and verify a survival prognosis model combined with patient survival information.

METHODS: The TARGET database was searched to identify differentially expressed peripheral blood genes in children with AML and healthy children. A gene set functional analysis and pathway analysis were performed using gene ontology and the KEGG pathway. A prognostic model for children with AML was constructed using univariate Cox, LASSO Cox regression and multivariate Cox regression analyses. Time-dependent receiver operating characteristic (ROC) curves were adopted to assess the predictive capacity of the prognostic models.

RESULTS: In total, 1640 differentially expressed genes were screened (1119 upregulated and 521 downregulated genes). The differentially expressed genes were mainly involved in nutrient metabolism and cytochrome P450 metabolism. Six key genes related to the prognosis of AML, FAM157A, GPR78, IRX5, RP4-800G7.1, RP11-179H18.5 and RP11-61N20.3, were identified. Kaplan-Meier curves indicated that 3-year and 5-year overall survival was significantly higher in the low-risk group than in the high-risk group. The area under the ROC curve was 0.722. At different stages of AML, FAM157A and RP4-800G7.1 exhibited significant differences in expression. The expression levels of FAM157A were significantly decreased in AML, whereas the expression levels of GPR78, IRX5, RP4-800G7.1, RP11-179H18.5 and RP11-61N20.3 were significantly increased in AML.

CONCLUSION: A prognosis-related gene model of AML was successfully constructed, and the expression levels of the model genes varied with AML stage.