Background: Acute myeloid leukemia (AML) continues to be a challenging leukemia to treat in the modern era. As with other malignancies, the role of the tumor immune microenvironment is helping elucidate how different patients respond to AML and what it could mean for prognosis. Better understanding of the tumor microenvironment, specifically how resting and activated macrophages, as well as resting and activated dendritic cells, could help determine prognostic tools as well as therapeutic targets. Regarding macrophages, they can either be in a M0, M1, or M2 state; corresponding to an undifferentiated, pro-inflammatory, or anti-inflammatory state. M1 macrophages prevent cell proliferation, produce cytokines, and reactive oxygen species. M2 macrophages are involved in cell proliferation and tissue repair. Likewise resting dendritic cells and activated dendritic cells respectively play a role in gathering antigens and activating T-cells. Other reports have shown a relationship between higher a percentage of activated macrophages and activated dendritic cells and longer overall survival in several different malignancies.

Methods: The Cancer Genome Atlas (TCGA) makes publicly available RNA sequencing data for several different tumor types. This project utilizes a sample of 147 patients from the AML TCGA dataset. The sequencing data from all 147 patients were inputted through CIBERSORT which is a Stanford application that helps break down each patient's leukemia microenvironment. Each patient sample is broken down into percentages of various immune cells: B-cells naïve, B-cells memory, plasma cells, CD8 T-cells, CD4 T-cells, T-cells follicular, T-cells regulatory, natural killer activated cells, natural killer resting cells, monocytes, macrophages (M0, M1, M2), activated dendritic cells, resting dendritic cells, activated mast cells, resting mast cells, eosinophils, and neutrophils. Once the immune cell fractions are obtained, a statistical tool named Radiant by Shiny is used to perform a linear regression to evaluate the relationship between overall survival and the different immune cells.

Results: M1 macrophages were independently prognostic of overall survival on multivariable analysis (p = 0.003). The coefficient of correlation was +3704 showing a positive association between the number of M1 macrophages and overall survival. Likewise activated dendritic cells were independently prognostic of overall survival on multivariable analysis (p = 0.038). The coefficient of correlation was +1925 showing a positive association between the number of activated dendritic cells and overall survival.

Conclusions: Patients with a higher number of M1 macrophages in their AML microenvironment had a longer overall survival. Likewise, patients with a higher number of activated dendritic cells in their AML microenvironment had a longer overall survival. M0 macrophages are in an undifferentiated state and can polarize into a M1 or M2 phenotype. Other publications have hinted at tumor microenvironment variability in different leukemias, but data is scarce on relationships between M1 macrophages and AML. Higher M1 percentages have been linked to better prognosis in colorectal cancer, lung cancer, and breast cancer. This is likely due, in part, to the M1 defense response to foreign invaders, namely the production of reactive oxygen species, nitric oxide, and interleukin-6. In this report, we show that higher amounts of M1 macrophages can also be linked to better prognosis in AML, however whereas higher M2 macrophages correlate to worse prognosis in other cancers, in our analysis there was no association between M2 macrophages and overall survival. This lack of relationship between M2 macrophages and overall survival could be due to the limited sample size of 147 patients. Regarding activated dendritic cells, newer reports are showing the promising utilization of dendritic cell based vaccines in AML therapy. This concept could be related to our findings of higher activated dendritic cells being linked to longer overall survival. The role of the leukemia microenvironment continues to evolve and can lead to substantial therapeutic advancements.

Disclosures

No relevant conflicts of interest to declare.

This content is only available as a PDF.
Sign in via your Institution