Multi-omics analysis revealing a senescence-relevant lncRNAs signature for the assessment of response to immunotherapy for breast cancer

Breast cancer (BRCA) is the most fatal malignancy of women. Immunotherapy has greatly improved the prognosis of advanced BRCA. Cellular senescence contributes to tumorigenesis and suppresses anti-cancer immunity. Identification of senescence-relevant long noncoding RNAs (SRlncRNAs) signature may benefit the predictions of prognosis and response to immunotherapy of BRCA. RNA-seq, mutation, and clinical data of BRCA were acquired from public databases. SRlncRNAs were screened using univariate Cox regression analysis. Consensus clustering classified BRCA patients into 2 clusters, and the differences of overall survival (OS) and immune status between the 2 clusters were analyzed by survival analysis, CIBERSORT, and ESITIMATE. The SRlncRNAs signature was constructed by least absolute shrinkage and selection operator (LASSO) regression analysis, and BRCA patients were divided into 2 risk groups. Enrichment analyses were performed to explore the cancer- and immunotherapy-relevant pathways. Transcriptome analysis was performed to investigate the differences of OS, immune infiltration, and ESITIMATE score of the 2 groups. Genome analysis was applied to investigate the differences of somatic mutation, tumor mutation burden (TMB) and microsatellite instability (MSI) between the 2 risk groups. A nomogram combined with calibration curves and decision curve analysis (DCA) was established for better clinical decision. Tumor Immune Dysfunction and Exclusion (TIDE) score and IMvigor-210 were applied for the predicting of response to immunotherapy. Profiling Relative Inhibition Simultaneously in Mixtures (PRISM) and the Cancer Therapeutics Response Portal resource (CTRP) databases were used for drug susceptibility analysis. Ten prognostic SRlncRNAs were identified and BRCA patients were divided into 2 clusters. Cluster 1 had better OS with anti-tumor immune microenvironment. The high-risk BRCA had poorer OS in the Cancer Genome Atlas (TCGA) training cohort, which was also verified by TCGA validation cohort and GSE20685 validation cohort. Low-risk patients also had anti-tumor immune microenvironment. Genome analysis demonstrated that the high-risk group had significant higher TMB. High-risk BRCA were more susceptive to immunotherapy according to the TIDE score and IMvigor-210. Finally, drug susceptibility analysis showed that 6 compounds were sensitive to high-risk BRCA patients. We developed and verified an original SRlncRNAs signature by multi-omics analysis, which could serve as a prognosis and immunotherapy predictor for BRCA.