Abstract
Acute myeloid leukaemia (AML) is a heterogeneous disease with a difficult to predict prognosis. Ferroptosis, an iron-induced programmed cell death, is a promising target for cancer therapy. Nevertheless, not much is known about the relationship between ferroptosis-related genes and AML prognosis. Herein, we retrieved RNA profile and corresponding clinical data of AML patients from the Gene Expression Omnibus (GEO) and The Cancer Genome Atlas (TCGA) databases. Univariate Cox analysis was employed to identify ferroptosis-related genes significantly associated with AML prognosis. Next, the least absolute shrinkage and selection operator (LASSO) regression was employed to establish a prognostic ferroptosis-related gene profile. 12 ferroptosis-related genes were screened to generate a prognostic model, which stratified patients into a low- (LR) or high-risk (HR) group. Using Kaplan-Meier analysis, we demonstrated that the LR patients exhibited better prognosis than HR patients. Moreover, receiver operating characteristic (ROC) curve analysis confirmed that the prognostic model showed good predictability. Functional enrichment analysis indicated that the infiltration of regulatory T cells (Treg) differed vastly between the LR and HR groups. Our prognostic model can offer guidance into the accurate prediction of AML prognosis and selection of personalized therapy in clinical practice.