A novel PD-1/PD-L1 pathway molecular typing-related signature for predicting prognosis and the tumor microenvironment in breast cancer

Abstract

Background: Currently, the development of breast cancer immunotherapy based on the PD-1/PD-L1 pathway is relatively slow, and the specific mechanism affecting the immunotherapy efficacy in breast cancer is still unclear.

Methods: Weighted correlation network analysis (WGCNA) and the negative matrix factorization (NMF) were used to distinguish subtypes related to the PD-1/PD-L1 pathway in breast cancer. Then univariate Cox, least absolute shrinkage and selection operator (LASSO), and multivariate Cox regression were used to construct the prognostic signature. A nomogram was established based on the signature. The relationship between the signature gene IFNG and breast cancer tumor microenvironment was analyzed.

Results: Four PD-1/PD-L1 pathway-related subtypes were distinguished. A prognostic signature related to PD-1/PD-L1 pathway typing was constructed to evaluate breast cancer's clinical characteristics and tumor microenvironment. The nomogram based on the RiskScore could be used to accurately predict breast cancer patients' 1-year, 3-year, and 5-year survival probability. The expression of IFNG was positively correlated with CD8+ T cell infiltration in the breast cancer tumor microenvironment.

Conclusion: A prognostic signature is constructed based on the PD-1/PD-L1 pathway typing in breast cancer, which can guide the precise treatment of breast cancer. The signature gene IFNG is positively related to CD8+ T cell infiltration in breast cancer.

Keywords: Breast cancer; IFNG; PD-1; PD-L1; Signature; Tumor microenvironment.

Fig. 1

The study flowchart. Firstly, WGCNA was performed on breast cancer samples in the TCGA database, and a total of 25 key genes of the PD-1/PD-L1 pathway in breast cancer were identified. TCGA breast cancer samples were classified accordingly. Next, univariate Cox analysis, LASSO, and multivariate Cox analysis were used to screen prognostic genes. A seven-gene signature was constructed, and the nomogram was further built. Subsequently, immune correlation analysis, somatic mutation, drug sensitivity analysis, GO, KEGG, and GSEA were applied to determine the clinical applicability of this signature. Finally, immunohistochemistry was used to verify the expression and function of the key prognostic gene

Fig. 2

The PD-1/PD-L1 pathway molecular subtype analysis. A Hierarchical clustering dendrogram of genes; B Module-phenotypic correlation analysis. Each cell contained the corresponding correlation and P value; C GS and MM correlation scatter diagram in the black module. GS was highly significantly correlated with MM. Results suggested that genes highly associated with the PD-1/PD-L1 pathway were also important in the black module; D The box diagram of expression of 25 key genes in TCGA breast cancer and normal tissues; E The consensus heatmap of NMF; F Survival analysis of different breast cancer subtypes in TCGA; G The violin plot of four subtypes of PD-1/PD-L1 pathway scores; H The heatmap of the relationship between characteristic gene expression and clinical parameters of different molecular subtypes of breast cancer in the TCGA database. ****P < 0.0001; **P < 0.01

Fig. 3

Tumor microenvironment analysis of different molecular subtypes. A ESTIMATE analysis of different molecular subtypes of breast cancer in TCGA database; B CIBERSORTx analysis of different molecular subtypes of breast cancer in TCGA database; C ImmuCellAI analysis of different molecular subtypes of breast cancer in TCGA database. ****P < 0.0001; ***P < 0.001; **P < 0.01; *P < 0.05

Fig. 4

Construction of the prognostic signature. A Ten-fold cross-validation of λ selection in LASSO analysis; B LASSO coefficient spectrum; C–E TCGA breast cancer samples were randomly divided into the training set and the internal test set at a 7:3 ratio. Survival analysis showing the difference in the prognosis of the training set, the internal test set, and the entire set in high- and low-RiskScore group, respectively; F–H 1-, 3- and 5-year ROC curves of the training set, the internal test set, and the entire set, respectively; I Survival analysis in the prognosis of GSE20711 in high- and low-RiskScore group; J Survival analysis in the prognosis of GSE42568 in high- and low-RiskScore group; K Survival analysis in the prognosis of GSE103091 in high- and low-RiskScore group

Fig. 5

Analysis of the clinical application of the prognostic signature. A The bubble plot of correlation between RiskScores and IC₅₀ values of antineoplastic agents; B Differences in PD-1/PD-L1 pathway scores of patients with different RiskScores; C Expression of PDCD1, CD274, CTLA-4, HAVCR2, and LAG3 in different RiskScore groups; D CIBERSORTx predicted tumor microenvironment in different RiskScore patients; E The heatmap of correlation between signature genes and the PD-1/PD-L1 pathway; F Differences in RiskScores for response to the immunotherapy; G The bubble map of correlation between signature genes and immunotherapy. ****P < 0.0001; ***P < 0.001; **P < 0.01; *P < 0.05; ns no significance

Fig. 6

Construction of the nomogram. A The nomogram was constructed by combining different clinical parameters. B Calibration curves, which could be used to show the actual survival probability and the predicted probability; C DCA, which could be used to evaluate the clinical utility of the nomogram. ***P < 0.001

Fig. 7

Exploration of molecular mechanisms between different RiskScore groups in the TCGA database: A The volcano plot: RiskScore-related DEGs in TCGA-BRCA samples; B GO analysis of RiskScore-related DEGs; C KEGG analysis of RiskScore-related DEGs; D GSEA of RiskScore-related DEGs

Fig. 8

Analysis of IFNG in breast cancer. A The bubble diagram of gene expression changes; B Survival analysis: effect of IFNG expression on the survival of TCGA-BRCA; C The box plot: IFNG expression in different breast cancer datasets; D Single-cell sequencing of breast cancer: expression of IFNG among different cells in the tumor microenvironment (Single Cell Portal); E Relationship between IFNG and different T cell infiltration in the breast cancer tumor microenvironment (TIMER 2.0); F IFNG staining of paired samples in the tissue microarray (×200); G IFNG staining score of paired samples in TMA: the left side showing the nuclear staining score and the right side showing the cytoplasm staining score. ****P < 0.0001; **P < 0.01; ns no significance