To analyze the immune implication of NR3C2 gene in breast cancer using bioinformatics methods and construct a prognostic model.

(1) The breast cancer cohort in the Cancer Genome Atlas (TCGA) database and the GSE42568 cohort in the Gene Expression Omnibus (GEO) database were used as training set (113 paracancerous samples and 1 019 breast cancer samples) and testing set (17 paracancerous samples and 104 breast cancer samples), respectively. The mRNA level of NR3C2 was compared between adjacent tissue and tumor tissue samples in the above 2 cohorts. The effect of NR3C2 expression on recurrence-free survival (RFS) was analyzed in cohorts of TCGA and Kaplan-Meier plotter(4 929 breast cancer samples), respectively. (2) The Gene Set Enrichment Analysis (GSEA) was employed to explore the potential biological functions of NR3C2. The 24 immune cells were quantitively accessed by single sample gene set enrichment analysis (ssGSEA) and the correlation of NR3C2 with 24 immune cells and 70 immunomodulator genes were conducted by Pearson coefficient. (3) A prognostic model was constructed by NR3C2-related immunomodulator genes in the TCGA cohort through multivariate stepwise Cox regression. The TCGA cohort was divided into two groups (high-risk group and low-risk group) by median risk value and RFS was compared between two groups. The sensitivity and specificity of the model was calculated using receiver operating characteristic (ROC) curves and validated in the GSE42568 cohort. Combined with other clinical parameters, the independent prognostic performance of this model was analyzed by multivariate Cox regression. (4) A nomogram was constructed based on pathological stage and risk value in TCGA cohort. The calibration curve was used to evaluate its accuracy and the predictive accuracy of different parameters was compared by time-dependent area under curve (tAUC). (5) In order to verify whether the NR3C2 mRNA expression is consistent with NR3C2 protein expression, we collected clinical specimens from three breast cancer patients who underwent surgical resection in the Department of Breast and Thyroid Surgery of the Army Medical Center in September 2021. The protein expression of NR3C2 in the paracancerous tissue and cancer tissue was detected by Western blot analysis.

(1) In TCGA cohort, the mRNA expression of NR3C2 in breast cancer tissue was significantly lower than that in paracancerous tissues (2.59±0.43 vs 0.98±0.62, t=35.990, P<0.001). In GSE42568 cohort, the mRNA expression of NR3C2 in breast cancer tissue was significantly lower than that in paracancerous tissues (5.35±1.47 vs 3.32±1.12, t=7.096, P<0.001). The results of survival analysis showed that in TCGA cohort and Kaplan-Meier plotter cohort, NR3C2 expression was positively correlated with RFS in breast cancer patients (HR=0.667, 0.725; 95%CI: 0.458-0.972, 0.653-0.804; both P<0.050). (2) GSEA results suggested that NR3C2 was mostly involved in JAK-STAT and TGF-β signaling pathways related to immunity. Correlation analysis found that the mRNA expression of NR3C2 was significantly correlated with 19 immune cells and 43 immunomodulator genes (all P<0.050). (3) The 43 NR3C2-related immunomodulator genes were included in Cox regression to construct a prognostic model which composed of 13 immunomodulator genes with a risk cutoff value equal to 0.988. Survival analysis showed that in TCGA cohort and GSE42568 cohort, the RFS in high-risk group was significantly lower than that in low-risk group (HR=2.682, 2.389; 95%CI: 1.839-3.910, 1.343-4.248; both P<0.010). ROC indicated that the AUC was 0.758 and 0.618 in TCGA cohort and GSE42568 cohort, respectively (95%CI: 0.662-0.857, 0.545-0.758; sensitivity: 0.833, 0.538; specificity: 0.614, 0.714; both P<0.010). Multivariate Cox regression showed that the risk score of this model could serve as an independent prognostic factor for breast cancer in TCGA cohort and GSE42568 cohort (HR=1.259, 1.163; 95%CI: 1.187-1.336, 1.068-1.266; both P<0.001). (4) The nomogram constructed based on the pathological stage and risk value could predict the 3-year, 5-year and 8-year RFS of breast cancer patients. The calibration curve suggested that it has good predictive accuracy and tAUC indicated that it is superior to pathological stage and a prognostic model. (5) Western blot analysis showed that the protein expression of NR3C2 was significantly decreased in breast cancer tissues.

NR3C2 is a potential immunotherapeutic target and prognostic biomarker in breast cancer patients.