Characterization of myeloid signature genes for predicting prognosis and immune landscape in Ewing sarcoma.

Abstract

Myeloid cells as a highly heterogeneous subpopulation of the tumor microenvironment (TME) are intimately associated with tumor development. Ewing sarcoma (EWS) is characterized by abundant myeloid cell infiltration in the TME. However, the correlation between myeloid signature genes (MSGs) and the prognosis of EWS patients was unclear. In this research, we synthetically characterized the expression of MSGs in a training cohort and classified EWS patients into two subtypes. Immune cell infiltration analysis revealed that MSGs subtypes correlated closely with different immune statuses. Furthermore, a three-gene prognostic model (CTSD, SIRPA, and FN1) was constructed by univariate, LASSO, and multivariate Cox analysis, and it showed excellent prognostic accuracy in EWS patients. We also developed a nomogram for better predicting the long-term survival of EWS. Functional enrichment analysis showed immune-related pathways were distinctly different in the high- and low-risk groups. Further analysis revealed that patients in the high-risk group were tightly associated with an immunosuppressive microenvironment. Finally, we validated the expression of these candidate genes by Western blot (WB), qPCR, and immunohistochemistry (IHC) analysis. To sum up, our study identified that the MSGs model was strongly linked to prognostic prediction and immune infiltration in EWS patients, providing novel insights into the clinical treatment and management of EWS patients.