Background: Acute Myeloid Leukemia (AML) is a hematological cancer characterized by heterogeneous hematopoietic cells. Through the use of multidimensional sequencing technologies, we previously identified a distinct myeloblast population, CD34+CD117dim, the proportion of which was strongly associated with the clinical outcome in t (8;21) AML. In this study, we explored the potential value of the CD34+CD117dim population signature (117DPS) in AML stratification.
Methods: Based on the CD34+CD117dim gene signature, the least absolute shrinkage and selection operator (LASSO) Cox regression analysis was performed to construct the 117DPS model using the gene expression data from Gene Expression Omnibus (GEO) database (GSE37642-GPL96 was used as training cohort; GSE37642-GPL570, GSE12417-GPL96, GSE12417-GPL570 and GSE106291 were used as validation cohorts). In addition, the RNA-seq data from The Cancer Genome Atlas (TCGA)-LAML and Beat AML projects of de-novo AML patients were also analyzed as validation cohorts. The differences of clinical features and tumor-infiltrating lymphocytes were further explored between the high-risk score group and low-risk score group.
Results: The high-risk group of the 117DPS model exhibited worse overall survival than the low-risk group in both training and validation cohorts. Immune signaling pathways were significantly activated in the high-risk group. Patients with high-risk score had a distinct pattern of infiltrating immune cells, which were closely related to clinical outcome.
Conclusion: The 117DPS model established in our study may serve as a potentially valuable tool for predicting clinical outcome of patients with AML.
Keywords: Acute myeloid leukemia; ELN 2017; Gene expression profile; Prediction model.
© 2022. The Author(s).