The Metabolic Gene Prognosis Model Of Non-M3 Acute Myeloid Leukemia Was Constructed Based On TCGA Database

Objective: The prognosis of non-M3 acute myeloid leukemia is highly heterogeneous.At present,studies have shown that tumor metabolism plays a crucial role in acute myeloid leukemia and is a potential marker for diagnosis and prognosis,but there are few systematic in-depth studies on the metabolic gene expression differences in non-M3 acute myeloid leukemia.The aim of this study is to explore the relationship between metabolic genes and the prognosis of nonM3 acute myeloid leukemia,construct a prognostic model of metabolic genes related to prognosis,and improve the existing prognostic stratification system of acute myeloid leukemia.Methods:1.Data download and sorting: all metabolic genes were downloaded based on C2 gene set of MSig DB database,and clinical data and RNA-seq data sets of AML patients were downloaded based on TCGA database.All metabolic genes in non-M3 acute myeloid leukemia were sorted out and divided into test group and validation group.2.Prognostic model construction: prognostic metabolic genes were screened in the test group and a prognostic model was constructed by univariate COX regression and LASSO regression.3.Prognostic model validation: Kaplan-Meier method,log-rank test,and ROC curve were used to verify the prognostic prediction model from various aspects.4.Based on the prognostic model,survival analysis and independent prognostic analysis were performed,and the survival nomogram was drawn.5.GSEA enrichment analysis was performed on the modeling genes to explore the possible metabolic pathways.Results: 1.The gene expression data of 117 patients with non-M3 acute myeloid leukemia screened in the TCGA database were randomly divided into groups and univariate COX regression analysis and LASSO regression analysis.Finally,a prognostic model of non-M3 acute myeloid leukemia with 11 metabolic genes(SLC4A7,C1GALT1,MXRA8,SOCS1,ZDHHC11,CAMK2 A,CBR1,ME1,SFXN3,KPTN,LCT)was constructed.2.Based on the median risk score of the prognostic model in the test group(0.621),the two groups of patients were divided into high and low risk groups,respectively.In the test group and validation group,the overall survival(OS)of Non-M3 AML patients in the high risk group was lower than that in the low risk group(p<0.05).3.The results showed that the overall survival(OS)of patients with intermediate was significantly different between the high and low risk groups(p < 0.05).The median survival time of the high risk group was 0.666(years),and the median survival time of the high risk group was 0.666(years).The median survival time of the low-risk group was 1.876 years.In patients with different age stratification and different mutation status of FLT3,IDH,RAS,and NPMc,the OS of the high-risk group was significantly lower than that of the low-risk group(p<0.05).The ROC curves of the test group and validation group based on the prognostic model suggested that the AUC values of the two groups were 0.922 and 0.744,respectively,which had good predictive ability.4.By univariate and multivariate independent prognostic analysis,the prognostic model was found to be an independent prognostic risk factor(multivariate independent prognostic analysis,HR=1.689,p<0.05;Univariate independent prognostic analysis,HR=1.545,p<0.05),which could independently predict the survival of patients.5.According to the enrichment analysis of metabolic genes in the high-risk group,P<0.05 and FDR<25% were used as screening conditions,and a total of 9 metabolic related signaling pathways were found.Conclusion: 1.A prognostic model of Non-M3 AML based on 11 metabolic genes(MXRA8,SOCS1,ZDHHC11,CAMK2 A,CBR1,ME1,SFXN3,KPTN,LCT,SLC4A7,C1GALT1)was constructed using TCGA database.2.Verified by a variety of bioinformatics analysis,this prognostic model is an independent prognostic factor,which can well predict the survival of Non-M3 AML patients.This prognostic model may play a complementary and perfect role in the risk classification of AML cell/molecular genetic indicators commonly used at present,and better guide the prognosis of Non-M3 AML.3.Based on the prognosis related metabolic model,9 metabolic pathways were found to be enriched in the high-risk group,which may provide new ideas and directions for the diagnosis and treatment of AML in the future.