Background: Recently, HER3-expression was postulated as independent risk factor for metastatic spread. Therefore, we investigated the role of HER3 expression as prognostic marker in metastatic breast cancer patients.
Methods: Patients of different breast cancer subtypes diagnosed with metastatic disease (visceral and/or brain metastases) were identified from a breast cancer database. Tissue samples of the respective primary tumors were retrieved, and immunohistochemical staining for estrogen-receptor, progesterone-receptor, HER2, and HER3 was performed. In HER2 equivocal and selected HER3 positive cases, subsequent fluorescent in situ hybridization (FISH) analysis was performed.
Results: Tissue specimens of 110 patients were available for this analysis. 21% had strong, complete, membranous HER3 staining of at least 10% of all tumor cells; HER3 protein expression was not associated with HER3 gene amplification. HER2/HER3 co-overexpression was observed in 12/110 (11%) specimens and HER3-overexpression showed a statistically significant association with HER2-overexpression (p = 0.02). No correlation was observed for HER3-overexpression and overall survival (OS), time to diagnosis of brain metastases, and incidence of brain metastases. Still, in patients with HER3 overexpression, a higher rate of 'brain only' metastatic behavior was observed (p = 0.042). In the HER2-positive subgroup, HER3-overexpression was significantly associated with shorter OS from diagnosis of metastatic disease (median 17 vs. 35 months; p = 0.04; log rank test).
Conclusions: HER2/HER3 co-overexpression is significantly associated with impaired OS from diagnosis of metastatic disease in patients with HER2-positive metastatic breast cancer. Co-inhibition of HER2 and HER3 or the inhibition of HER2/HER3 hetero-dimerization may improve clinical outcome in this subgroup.
Keywords: Breast cancer; HER2-positive; HER3; Trastuzumab.
Copyright © 2014 Elsevier Ltd. All rights reserved.