Identifying prognostic biomarker related to immune infiltration in acute myeloid leukemia

The immune cells of tumor microenvironment (TME) constitute a vital element of the tumor tissue. There is increasing evidence for their clinical significance in predicting prognosis and therapeutic outcomes. However, the TME immune cell infiltrating pattern of the bone marrow in acute myeloid leukemia (AML) patients remains unclear. Here, RNA-sequencing results of AML patients from TCGA database were used to quantify the abundance of 28 types of immune cells in the TME using the single-sample gene set enrichment analysis algorithm. We comprehensively evaluated the immune infiltration status in the TCGA-LAML cohort and defined two immunophenotypes: the immune hot and immune cold subtypes. Additionally, we constructed a TME score reflecting the immune infiltrating pattern of the patients using Cox regression algorithm. Subtypes with high TME score were characterized by over-activation of immune inflammation-related pathways, release of inflammatory factors, T-cell dysfunction, and poor prognosis. Subtypes with a low TME score were characterized by relatively low immune infiltration and immune exclusion. Our analysis indicated that patients in the low TME score group were more sensitive to chemotherapeutic drugs, and those in high TME score were more likely to respond to immunotherapy. Our study provides a new direction to evaluate anti-tumor therapy from immune infiltration of the TME, and the individualized scoring system in this study has important clinical significance in identifying patients who respond to immunotherapy.