Predictive and prognostic significance of cytoplasmic expression of ELAV-like protein HuR in invasive breast cancer treated with neoadjuvant chemotherapy

Cytoplasmic HuR is associated with reduced survival in invasive breast cancer. We designed this study to determine the predictive and prognostic value of HuR expression in women with breast cancer who underwent neoadjuvant chemotherapy followed by surgical resection. We immunohistochemically analyzed cytoplasmic HuR expression in tumor biopsy cores obtained from 139 patients with invasive breast cancers who received paclitaxel and anthracycline-based neoadjuvant chemotherapy. We evaluated the relationship of HuR expression level with pathologic complete response (pCR), local recurrence-free survival (LRFS), distant recurrence-free survival (DRFS), recurrence-free survival (RFS), and overall survival (OS). Cytoplasmic HuR expression was present in 60 cases (43.2 %). The expression of cytoplasmic HuR was significantly associated with high nuclear grade (P < 0.0001) and ER (P = 0.001) and PR (P = 0.005) status. Multivariate regression analysis further revealed that high nuclear grade (P = 0.023), negative ER status (P = 0.043), and human epidermal growth factor receptor 2 (HER2) overexpression (P < 0.0001), but not cytoplasmic HuR expression, were significant independent predictors of pCR. Interestingly, multivariate Cox analysis revealed that cytoplasmic HuR expression was a strong independent predictor of reduced LRFS (P = 0.014), DRFS (P = 0.001), RFS (P < 0.0001), and OS (P = 0.019) irrespective of pCR. Furthermore, the patient group with tumors showing both expression of cytoplasmic HuR and non-pCR had a worse prognosis in LRFS (P = 0.048), DRFS (P < 0.0001), RFS (P < 0.0001), and OS (P = 0.001) than did other patient groups; patients with tumors showing negative cytoplasmic expression of HuR and pCR had the best prognosis in all RFS and OS. Cytoplasmic expression of HuR is an independent prognostic marker in breast cancer patients undergoing chemotherapy. Combination analyses of HuR expression and pCR, compared with pCR alone, can better predict clinical outcome in patients with primary breast cancer.