Recurrence prediction using microRNA expression in hormone receptor positive breast cancer during tamoxifen treatment

Abstract

PURPOSE: To identify miRNAs associated with distant recurrence during tamoxifen treatment and build a recurrence prediction model. MATERIALS AND METHODS: We measured the expression of five miRNAs (miR-134, miR-125b-5P, miRNA-30a, miR-10a-5p and miR-222). A total of 176 tumour tissues from 176 patients who had hormone receptor positive breast cancer with tamoxifen treatment were used to measure miRNA expression using quantitative real-time PCR (qRT-PCR). RESULTS: The five miRNAs were all up-regulated in distant recurrence cases within 5 years after surgery and during tamoxifen treatment. Kaplan-Meier survival analyses based on expression cut-offs determined by receiver characteristics curves (ROC) showed that high expression of miR-134, miR-125b-5P, miRNA-30a, miR-10a-5p and miR-222 were significantly (log-rank p-value =0.006, p-value <0.0001, p-value <0.0001, p-value <0.0001 and p-value <0.0001, respectively) associated with short relapse-free time. Our results were used to build a combined 3 miRNAs expression model. It could be used to categorize high-risk subset of patients with short relapse-free survival (AUC =0.891, p-value <0.0001). CONCLUSIONS: Distant recurrence during tamoxifen treatment of hormone positive breast cancer might be affected by tamoxifen resistance related miRNAs. Such distant recurrence can be predicted using miRNA measurement.