Identification and validation of immune-associated gene signatures for prognostic prediction in acute myeloid leukemia.

Acute myeloid leukemia (AML) is a heterogeneous hematologic malignancy with poor prognosis, and reliable prognostic biomarkers are essential for improving risk stratification and personalized treatment strategies. In the present study, our objective was to identify immune-associated differentially expressed genes (DEGs) that were correlated with overall survival of AML patients. Transcriptome data from multiple cohorts, including the cancer genome atlas, genotype-tissue expression, and gene expression omnibus were integrated to identify AML-specific biomarkers. DEGs were identified between AML and healthy samples, which were mainly involved in immune-associated processes. Based on the immune-associated DEGs, a prognostic model was constructed using least absolute shrinkage and selection operator regression. A 9-gene signature (CAPZB, TFEB, CAP1, ITGAX, ATP6V0D1, NCR1, LILRB3, LST1, and PAK1) model was constructed and validated in multiple independent datasets, showing robust predictive accuracy for overall survival, with high area under the curve values corresponding to 1-, 3-, and 5-year survival. Additionally, qRT-PCR experiment verified the differential expression of crucial genes in clinical AML samples. This study provides a promising immune-based prognostic model for AML, contributing to better patient stratification and personalized treatment approaches.