PD-1 and PD-L1 correlated gene expression profiles and their association with clinical outcomes of breast cancer

BACKGROUND: Immunotherapies that targeting programmed cell death 1 (PD-1) and programmed death-ligand 1 (PD-L1) have obtained prominent success in breast cancer (BC). However, not all the patients benefit from the antibody therapy. This study aimed to identify PD-1/PD-L1 correlated genes and pathways as well as investigate their potential as prognostic marker in BC. MATERIALS AND METHODS: By analysing transcriptional data of BC from TCGA, we identified PD-1 and PD-L1 correlated genes by WGCNA analysis and explored the biological process as well as pathways they enriched. Co-expression analysis were performed for PD-1/PD-L1 with immune infiltration and checkpoints. The prognostic value of PD-1 and PD-L1 were also investigated. RESULTS: PD-1 and PD-L1 expression showed significant difference in different molecular subtypes and stages. PD-1 correlated genes enriched in T cell activation, lymphocyte activation, leukocyte migration while PD-L1 correlated genes demonstrated enrichment including T cell apoptotic process, tolerance induction and cytolysis. Immune infiltration analysis suggested that PD-1 and PD-L1 were related with Neutrophils (r = 0.65, r = 0.48) and Fibroblasts (r = 0.59, r = 0.47). For immune checkpoints analysis, PD-1 was associated with HLA-A (r = 0.804) and INPP5D (r = 0.782) while PD-L1 correlated with CTLA4 (r = 0.843) and CD27 (r = 0.823). PD-1 was associated favorable survival of BC (HR = 0.67, P = 0.012) while PD-L1 did not demonstrate significant association with BC prognosis (HR = 0.85, P = 0.313). CONCLUSION: PD-1 and PD-L1 correlated genes participated in biological process including T cell activation, lymphocyte activation, leukocyte migration, T cell apoptotic process, tolerance induction and cytolysis. PD-1/PD-L1 expression also demonstrated relation with immune infiltration and immune checkpoints. High PD-1 expression predicted better survival of breast cancer patients.