Abstract
BACKGROUND: Acute myeloid leukemia (AML) is a complex hematological malignancy with high mortality, particularly in the elderly. Current treatments, including chemotherapy, targeted therapies, and emerging immunotherapies such as T cell engager (TCE) and CAR-T, face challenges such as drug resistance. Lactylation, a novel post-translational modification, has emerged as a potential regulator of cellular activities and may play a role in AML pathogenesis. METHODS: This study integrated data from three GEO datasets (GSE9476, GSE37642, GSE114868) and single-cell RNA sequencing (scRNAseq) data from GSE235857. Data preprocessing involved normalization and batch effect correction using the Seurat package in R. Cell clustering and subpopulation definition were performed using UMAP and Louvain clustering. Lactylation scores were calculated using the AddModuleScore method. Differential gene expression analysis was conducted using the limma package, and Weighted Gene Co-expression Network Analysis (WGCNA) was applied to identify gene modules related to lactylation scores. Machine learning techniques, including Least Absolute Shrinkage and Selection Operator (LASSO) regression, Support Vector Machine (SVM), and random forest, were used to screen hub genes. SHAP analysis was employed to evaluate the importance of these genes in AML diagnosis. RESULTS: Cell clustering identified 8 clusters and 7 subpopulations. Lactylation scores were significantly lower in AML patients compared to healthy controls (P < 0.0001). Differential expression analysis revealed significant gene expression differences between AML and controls, with key genes such as KHDRBS1 and RBM17 showing high importance in AML diagnosis. Machine learning identified five hub-genes (KHDRBS1, U2AF2, RBM17, RPL14, NCL) with high predictive value. SHAP analysis confirmed the importance of these genes, with KHDRBS1 having the highest average absolute SHAP value (0.0669). Immune cell infiltration analysis showed significant differences in immune cell levels between AML patients and controls, with key genes correlating with immune cell infiltration. CONCLUSION: Lactylation is associated with AML pathogenesis and may serve as a potential biomarker for diagnosis and treatment. The identified key genes and their correlation with immune cell infiltration provide new insights into AML’s immune microenvironment and potential therapeutic targets.