JMIR Publications

ABSTRACT

Background:

Acute myeloid leukemia (AML) is a malignancy of the blood system. The commonly altered regions in the genome of AML encompass a multitude of gene modifications associated with epigenetic regulation. However, the prognostic significance of chromatin remodeling-related genes (CRRGs) as an overall indicator has yet to be assessed in AML. 

Objective:

The purpose of our analysis results was to establish a solid theoretical foundation for the clinical treatment of AML.

Methods:

Uunivariate Cox regression analysis was performed for CRRGs. The Differential expressed genes (DEGs) between chromatin remodeling related AML subtypes in TCGA-AML were measured. CIBERSORT and ssGSEA algorithms were employed to compare disparities in immune responses. GSVA analysis was performed and tumor mutation burden (TMB) analysis was conducted. Using TIDE to assess patients' differences in sensitivity to immunotherapy. The construction and verification of the nomogram were carried out. Expression of biomarkers in healthy and AML patients was analyzed by RT-qPCR.

Results:

In TCGA-AML, a total of 3995 differential genes were identified. There were differences between two AML clusters in T cells CD4 memory activated and Dendritic cells activated. A total of 6 prognostic biomarkers were identified, namely ARF6, ASF1B, CHD5, FLNA, KDM5B, and SPI1. The high-risk group exhibited higher TIDE scores. A total of 29 drugs had lower IC50 values in high-risk group. We found that risk score was an independent prognostic factor for AML.RT-qPCR results showed significant differences in expression of ARF6, KDM5B and CHD5 between healthy and AML patients.

Conclusions:

We identified six biomarkers, namely ARF6, ASF1B, CHD5, FLNA, KDM5B, and SPI1, thereby establishing a theoretical foundation for clinical diagnosis of AML.