Screening of coexpression genes of immune cells in breast cancer tissues

This study aimed to investigate immune cell infiltration (ICI) in breast cancer tissues and its impact on the prognosis of patients. The whole transcriptome sequencing data sets of breast tissue (GSE126125, GSE190275 and GSE45498) were downloaded from Gene Expression Omnibus database. Data sets, including 281 breast cancer tissue samples and 59 normal breast tissue samples. In this study, the CIBERSORT algorithm was used to calculate the infiltration content of 22 immune cells subtypes in breast cancer tissues and normal breast tissues. The ICI between normal and breast cancer tissue samples was examined through the Rank-sum test. Furthermore, Kaplan-Meier and the log-rank test were used for survival analysis. Univariate and multivariate COX analysis was used to screen the prognostic risk factors of breast cancer based on ICI. The correlation between 22 kinds of immune cells was analyzed by the Pearson test. The results of univariate COX analysis indicated that resting dendritic cells, eosinophils, resting mast cells, monocytes, and memory CD4 T cells resting were protective factors for the prognosis of breast cancer patients (hazard ratio [HR] < 1, P < .05). The activation of macrophage M0 and mast cells were also prognostic risk factors for breast cancer patients (HR > 1, P < .05). Besides, multivariate COX analysis showed that resting mast cells were independent protective factors for the prognosis of breast cancer patients (HR < 1, P < .05). Macrophage M0 and mast cell activation were independent risk factors for the prognosis of breast cancer patients (HR > 1, P < .05). High infiltration of macrophage M0 and activated mast cells is associated with poor prognosis. Meanwhile, macrophage M0 and activated mast cells promote breast cancer progression. Low infiltration of resting mast cells is associated with poor prognosis, which inhibits breast cancer progression.