Usefulness of miRNA profiles for predicting pathological responses to neoadjuvant chemotherapy in patients with human epidermal growth factor receptor 2-positive breast cancer.

Pathological complete response (pCR) is considered to be a useful prognostic marker for neoadjuvant chemotherapy to improve the survival rate of patients with operable breast cancer. In the present study, we identified differentially expressed microRNAs (miRNAs) between pCR and non-pCR groups of patients with human epidermal growth factor receptor 2 (HER2)-positive breast cancer who received neoadjuvant chemotherapy with trastuzumab. Expression profiles were examined by miRNA microarrays using total RNA extracted from formalin-fixed, paraffin-embedded tissues from pretreatment biopsy specimens. Significant differences were observed in miRNAs associated with pCR between the luminal B-like (HER2-positive) and HER2-positive (nonluminal) subtypes, which were further classified according to their estrogen receptor (ER) status. Prediction models constructed with differentially expressed miRNAs performed well. In conclusion, the combination of miRNA profiles and ER status may improve the accuracy of pCR prediction in patients with HER2-positive breast cancer and enable the development of personalized treatment regimens.