A Minimal lncRNA-mRNA Signature Predicts Sensitivity to Neoadjuvant Chemotherapy in Triple-Negative Breast Cancer

Abstract

Background/Aims: Triple-negative breast cancer (TNBC) is a highly aggressive malignancy that responds in a diverse manner to neoadjuvant chemotherapy (NAC). This study was aimed to uncover an RNA signature in TNBC patients which predicts pathological complete responses (pCR) to NAC by analyzing long noncoding RNA (lncRNA) and coding gene expression. Methods: Microarray datasets from 26 TNBC patients receiving NAC including ten patients showing pCR were obtained from the Gene Expression Omnibus database. Results: A total of 172 coding genes and 84 lncRNAs were differentially expressed between patients achieving pCR and those who did not. Filtering based on the predictive efficacy of response to NAC using receiver operator characteristic curve (ROC) and area under the curve (AUC) shortlisted 23 lncRNAs and 15 coding genes from consideration. Finally, a response score consisting of 1 lncRNA and 2 coding genes was developed: response score = 2.595∗BPESC1-1.09∗WDR72-1.428∗GADD45A-0.731. The response score had good predictive performance (AUC=0.931, p< 0.01) and at the cut-off of 0.545, the response score had sensitivity and specificity of 0.8 and 0.9, respectively. Conclusion: We propose a simple gene expression signature of only three RNA species could be employed clinically to predict pCR in TNBC patients receiving NAC.