Key points are not available for this paper at this time.
The existing risk stratification for acute myeloid leukemia (AML) reveals considerable heterogeneity in patient prognosis, underscoring the necessity for innovative risk stratification methodologies to optimize treatment responses. In this multicohort study, we explored the potential of carbohydrate metabolism and autophagy-related genes (CARGs) to enhance prognostic classification in AML patients. Employing univariate regression and least absolute shrinkage and selection operator (LASSO)-Cox stepwise regression analysis, we constructed a prognostic signature involving four genes related to CARGs in AML patients. By leveraging data from the TCGA cohort with 117 patients, the Gene Expression Omnibus (GEO) public data cohort with 1,431 patients, and our internal cohort of 117 patients, we showcased the robustness and accuracy of the CARG signature in forecasting survival outcomes among a collective sample of 1,665 non-Acute Promyelocytic Leukemia (APL) patients. Patients were categorized into high-risk and low-risk groups based on median risk score. The overall survival (OS) was significantly shorter in the high-risk group compared to the low-risk group. Differentially expressed genes (DEGs) were identified. Gene Ontology (GO) and Gene Set Enrichment Analysis (GSEA) analysis revealed that the DEGs were primarily associated with immune response signaling pathways. Immune-related analysis indicated that patients classified in the high-risk group exhibited a suppressive immune microenvironment. The results of the potential drugs for the risk groups demonstrated that inhibitors of PI3K/AKT/mTOR signaling pathway were effective. The novel risk model based on CARGs proposed in our study shows promise in prognostic classifications in AML, potentially providing new insights for the development of precise targeted cancer therapies.
Shen et al. (Fri,) studied this question.