Fibroblast-secreted hepatocyte growth factor mediates epidermal growth factor receptor tyrosine kinase inhibitor resistance in triple-negative breast cancers through paracrine activation of Met.

INTRODUCTION: Epidermal growth factor receptor (EGFR) tyrosine kinase inhibitors (TKIs) have shown clinical efficacy in lung, colon, and pancreatic cancers. In lung cancer, resistance to EGFR TKIs correlates with amplification of the hepatocyte growth factor (HGF) receptor tyrosine kinase Met. Breast cancers do not respond to EGFR TKIs, even though EGFR is overexpressed. This intrinsic resistance to EGFR TKIs in breast cancer does not correlate with Met amplification. In several tissue monoculture models of human breast cancer, Met, although expressed, is not phosphorylated, suggesting a requirement for a paracrine-produced ligand. In fact, HGF, the ligand for Met, is not expressed in epithelial cells but is secreted by fibroblasts in the tumor stroma. We have identified a number of breast cancer cell lines that are sensitive to EGFR TKIs. This sensitivity is in conflict with the observed clinical resistance to EGFR TKIs in breast cancers. Here we demonstrate that fibroblast secretion of HGF activates Met and leads to EGFR/Met crosstalk and resistance to EGFR TKIs in triple-negative breast cancer (TNBC). METHODS: The SUM102 and SUM149 TNBC cell lines were used in this study. Recombinant HGF as well as conditioned media from fibroblasts expressing HGF were used as sources for Met activation. Furthermore, we co-cultured HGF-secreting fibroblasts with Met-expressing cancer cells to mimic the paracrine HGF/Met pathway, which is active in the tumor microenvironment. Cell growth, survival, and transformation were measured by cell counting, clonogenic and MTS assays, and soft agar colony formation, respectively. Student's t test was used for all statistical analysis. RESULTS: Here we demonstrate that treatment of breast cancer cells sensitive to EGFR TKIs with recombinant HGF confers a resistance to EGFR TKIs. Interestingly, knocking down EGFR abrogated HGF-mediated cell survival, suggesting a crosstalk between EGFR and Met. HGF is secreted as a single-chain pro-form, which has to be proteolytically cleaved in order to activate Met. To determine whether the proteases required to activate pro-HGF were present in the breast cancer cells, we utilized a fibroblast cell line expressing pro-HGF (RMF-HGF). Addition of pro-HGF-secreting conditioned fibroblast media to TNBC cells as well as co-culturing of TNBC cells with RMF-HGF fibroblasts resulted in robust phosphorylation of Met and stimulated proliferation in the presence of an EGFR TKI. CONCLUSIONS: Taken together, these data suggest a role for Met in clinical resistance to EGFR TKIs in breast cancer through EGFR/Met crosstalk mediated by tumor-stromal interactions.