Abstract
BACKGROUND: Neovascularization mechanisms are hyperactivated in tumors, leading to vascular dysfunction and contributing to tumor metastasis and growth. This study aims to comprehensively analyze angiogenesis-associated genes in relation to the prognosis of elderly patients with acute myeloid leukemia (AML). METHODS: Leukemia gene expression data were obtained from the GSE37642 (training set) and TCGA_LAML (validation set) datasets. Angiogenesis-associated genes were identified using the GeneCards database. Univariate Cox regression and LASSO analyses were employed to identify angiogenesis-associated genes linked to AML prognosis. A prognostic signature was constructed based on the selected genes, and its biological functions were analyzed. Finally, we predicted AML drug sensitivity and evaluated differences in drug activity based on the prognostic signature. RESULTS: Five angiogenesis-related genes associated with AML prognosis were identified: ECM1, EGLN1, FKBP5, FOXP1, and SIRT2. Kaplan-Meier analyses confirmed their prognostic value. A prognostic signature based on these genes demonstrated commendable efficacy in predicting patient outcomes. This signature was found to be an independent risk factor for AML and revealed distinct immune profiles. Furthermore, the signature was implicated in the tumor immune microenvironment, with high-risk patients exhibiting elevated levels of immune cell infiltration. Drug sensitivity analysis revealed negative correlations between FOXP1 and Daporinad, ABT737, and BI.2536, while SIRT2 showed positive correlations with ABT737, BI.2536, and ULK1_4989. CONCLUSION: We have constructed an angiogenesis-related gene prognostic signature that enriches the prognostic assessment system for AML and provides novel therapeutic directions for this disease.