The link between aberrant splicing and B cell infiltration in breast cancer hormone therapy

Abstract

Increasing evidence has suggested alternative splicing (AS) is an essential mechanism for tumor initiation and progression. However, the functions of aberrant splicing have not been well understood in hormone therapy-treated breast cancer (HTBC) patients yet. By mining the SpliceSeq data of TCGA, we characterized 1388 survival-associated AS events (SASEs) that are markedly related to the overall survival (OS) of HTBC patients. Subsequently, a SASE prognostic signature was deduced via the least absolute shrinkage and selection operator (LASSO) and multivariate Cox analysis, and the HTBC patients were stratified into high- and low-risk groups based on the signature. The differentially expressed genes between the high- and low-risk patients were enriched with immune response genes. Furthermore, we revealed that B cell infiltration was significantly low in the high-risk patients with single-sample gene set enrichment analysis (ssGSEA), suggesting a link between AS abnormality and B cell infiltration in tumor immune microenvironment (TIME). Next, a similar prognostic signature based on splicing factor expression was deduced using Molecular Taxonomy of Breast Cancer (METABRIC) HTBC patients. When this prognostic signature was applied to stratify HTBC patients in the METABRIC and TCGA cohorts, the high-risk subgroups both had significantly low B cell infiltration in TIME. CIBERSORT analysis further demonstrated plasma cells infiltrated less in TIME. Also, both the AS (P < 0.0001) and SF (P = 0.009) prognostic signatures performed well in predicting OS of HTBC patients. In summary, our study has demonstrated that AS abnormality is a potential cause of B cell infiltration alteration, which is a critical factor for cancer death risks in HTBC patients and might be exploited therapeutically in the future.