An inflammatory response-related gene signature associated with immune status and prognosis of acute myeloid leukemia

Abstract

Objective: To determine the prognostic significance of inflammatory response-associated genes in acute myeloid leukemia (AML).

Methods: Transcriptomic profiles and related clinical information of AML patients were acquired from a public database. To establish a multi-gene prognosis signature, we performed least absolute shrinkage and selection operator Cox analysis for the TCGA cohort and evaluated the ICGC cohort for verification. Subsequently, Kaplan-Meier analysis was carried out to compare the overall survival (OS) rates between high- and low-risk groups. Biological function and single-sample gene set enrichment (ssGSEA) analyses were employed to investigate the association of risk score with immune status and the tumor microenvironment. Prognostic gene expression levels in AML samples and normal controls were confirmed by qRT-PCR and immunofluorescence.

Results: We identified a potential inflammatory response-related signature comprising 11 differentially expressed genes, including ACVR2A, CCL22, EBI3, EDN1, FFAR2, HRH1, ICOSLG, IL-10, INHBA, ITGB3, and LAMP3, and found that AML patients with high expression levels in the high-risk group had poor OS rates. Biological function analyses revealed that prognostic genes mainly participated in inflammation and immunity signaling pathways. Analyses of cancer-infiltrating immunocytes indicated that in high-risk patients, the immune suppressive microenvironment was significantly affected. The expression of the inflammation reaction-associated signature was found to be associated with susceptibility to chemotherapy. There was a significant difference in prognostic gene expression between AML and control tissues.

Conclusion: A novel inflammatory response-related signature was developed with 11 candidate genes to predict prognosis and immune status in AML patients.

Keywords: Acute myeloid leukemia; drug sensitivity; immune status; inflammatory response; overall survival; prognostic gene signature; tumor microenvironment.

Figure 1

The flow chart including data collection, analyses and representative experimental results in this study.

Figure 2

Determination of the candidate inflammatory response-related genes in the TCGA cohort. A. Venn diagram to determine DEGs between AML samples and normal controls. B. Heat map showing the expression levels of 21 overlapping genes between AML samples and normal controls. C. Forest plots revealing the results of the correlation between expression of 21 overlapping gene and OS. D. The correlation network of 21 overlapping genes.

Figure 3

Prognostic analysis of the eleven genes’ signature model in the TCGA cohort and ICGC cohort. TCGA cohort (A-E), ICGC cohort (F-J). (A, F) The median value and distribution of the risk scores. (B, G) The distribution of OS with risk scores. (C, H) PCA plot analysis of patients in the high- and low-risk groups. (D, I) Kaplan-Meier curves for OS of patients in the high- and low-risk groups. (E, J) AUC time-dependent ROC curves for OS at 1, 2, 3 years.

Figure 4

Gene set enrichment analysis of biological functions and pathways. A. Analysis of Gene Ontology (GO). B. Analysis of Kyoto Encyclopedia of Genes and Genomes (KEGG).

Figure 5

Analyses of immune status and the relationship between risk score and TME in the high- and low-risk groups. TCGA cohort (A, B), ICGC cohort (C, D). (A, C) The boxplots of showing the score of 16 immune cells and (B, D) 13 immune-related functions. (E) The relationship of risk score with RNAss, DNAss, Stromal Score and Immune Score. P values were presented as: ns, no significance; *P < 0.05; **P < 0.01; ***P < 0.001.

Figure 6

Analyses of detecting expression levels of PD-L1, PD-L2, MRP2 and MRP3 between different groups and the correlation between risk score and the expression of PD-L1, PD-L2, MRP2 and MRP3. A, C. PD-L1. B, D. PD-L2. E, G. MRP2. F, H. MRP3.

Figure 7

Representative top 9 scatter diagram of association between prognostic gene expression and drug susceptibility. A-I. Scatter plots sorted by Cor value. A, C, I. INHBA. B. CCL22. D, F. HRH1. E, H. IL 10. G. ACVR2A.

Figure 8

The mRNA expression analysis of the prognostic genes between AML tissues and normal controls by qRT-RCR. A. ACVR2A. B. CCL22. C. EBI3. D. EDN1. E. FFAR2. F. HRH1. G. ICOSLG. H. IL-10. I. INHBA. J. ITGB3. K. LAMP3. ***P < 0.001.

Figure 9

The protein expression analysis of the prognostic genes between AML tissues and normal controls by IF. The nucleus was stained with DAPI. A. ACVR2A. B. CCL22. C. EBI3. D. EDN1. E. FFAR2. F. HRH1. G. ICOSLG. H. IL-10. I. INHBA. J. ITGB3. K. LAMP3. Scale bars are 50 µm. Magnifications are 20X.