Abstract
With highly homologous epidermal growth factor (EGF)-like (EGFL) domains, the members of the EGFL family play crucial roles in growth, invasion, and metastasis of tumors and are closely associated with the apoptosis of tumor cells and tumor angiogenesis. Furthermore, their contribution to immunoreaction and tumor microenvironment is highly known. In this study, a comprehensive analysis of EGFL6, -7, and -8 was performed on the basis of their expression profiles and their relationship with the rate of patient survival. Through a pan-cancer study, their effects were correlated with immune subtypes, tumor microenvironment, and drug resistance. Using The Cancer Genome Atlas pan-cancer data, expression profiles of EGFL6, -7, and -8, and their association with the patient survival rate and tumor microenvironment were analyzed in 33 types of cancers. The expression of the EGFL family was different in different cancer types, revealing the heterogeneity among cancers. The results showed that the expression of EGFL8 was lower than EGFL6 and EGFL7 among all cancer types, wherein EGFL7 had the highest expression. The univariate Cox proportional hazard regression model showed that EGFL6 and EGFL7 were the risk factors to predict poor prognosis of cancers. Survival analysis was then used to verify the relationship between gene expression and patient survival. Furthermore, EGFL6, EGFL7, and EGFL8 genes revealed a clear association with immune infiltrate subtypes; they were also related to the infiltration level of stromal cells and immune cells with different degrees. Moreover, they were negatively correlated with the characteristics of cancer stem cells measured by DNAs and RNAs. In addition, EGFL6, -7, and -8 were more likely to contribute to the resistance of cancer cells. Our systematic analysis of EGFL gene expression and their correlation with immune infiltration, tumor microenvironment, and prognosis of cancer patients emphasized the necessity of studying each EGFL member as a separate entity within each particular type of cancer. Simultaneously, EGFL6, -7, and -8 signals were verified as promising targets for cancer therapies, although further laboratory validation is still required.