Allele-specific regulation of FGFR2 expression is cell type-dependent and may increase breast cancer risk through a paracrine stimulus involving FGF10.

INTRODUCTION: SNPs rs2981582 and rs2981578, located in a linkage disequilibrium block (LD block) within intron 2 of the fibroblast growth factor receptor 2 gene (FGFR2), are associated with a mildly increased breast cancer risk. Allele-specific regulation of FGFR2 mRNA expression has been reported previously, but the molecular basis for the association of these variants with breast cancer has remained elusive to date. METHODS: mRNA levels of FGFR2 and three fibroblast growth factor genes (FGFs) were measured in primary fibroblast and epithelial cell cultures from 98 breast cancer patients and correlated to their rs2981578 genotype. The phosphorylation levels of downstream FGFR2 targets, FGF receptor substrate 2α (FRS2α) and extracellular signal-regulated kinases 1 and 2 (ERK1/2), were quantified in skin fibroblasts exposed to FGF2. Immunohistochemical markers for angiogenesis and lymphocytic infiltrate were semiquantitatively assessed in 25 breast tumors. RESULTS: The risk allele of rs2981578 was associated with increased FGFR2 mRNA levels in skin fibroblasts, but not in skin epithelial cell cultures. FGFR2 mRNA levels in skin fibroblasts and breast fibroblasts correlated strongly in the patients from whom both cultures were available. Tumor-derived fibroblasts expressed, on average, eight times more FGFR2 mRNA than the corresponding fibroblasts from normal breast tissue. Fibroblasts with higher FGFR2 mRNA expression showed more FRS2α and ERK1/2 phosphorylation after exposure to FGF2. In fibroblasts, higher FGFR2 expression correlated with higher FGF10 expression. In 25 breast tumors, no associations between breast tumor characteristics and fibroblast FGFR2 mRNA levels were found. CONCLUSIONS: The influence of rs2981578 genotypes on FGFR2 mRNA expression levels is cell type-dependent. Expression differences correlated well with signaling levels of the FGFR2 pathway. Our results suggest that the increased breast cancer risk associated with SNP rs2981578 is due to increased FGFR2 signaling activity in stromal fibroblasts, possibly also involving paracrine FGF10 signaling.

Anti-inflammatory M2 type macrophages characterize metastasized and tyrosine kinase inhibitor-treated gastrointestinal stromal tumors.

We have made a detailed inventory of the immune infiltrate of gastrointestinal stromal tumors (GISTs), which originate from mesenchymal cells in the intestinal tract. These sarcomas are heavily infiltrated with macrophages and T cells, while immune cells of other lineages were much less abundant. Dissecting the functional subtypes of T cells with multicolor fluorescent microscopy revealed substantial populations of cytotoxic T cells, helper T cells and FoxP3(+) regulatory T cells. The balance of cytotoxic T cells and FoxP3(+) T cells was toward immune suppression. Analysis of the macrophage population also showed a dominance of anti-inflammatory cells, as the M2 type scavenger receptor CD163 was abundantly present. Other subsets of macrophages (CD14(+)CD163(-)) were occasionally detected. M2 type CD163(+) macrophages were associated with the number of infiltrating FoxP3(+) regulatory T cells and twice as many macrophages were found in metastatic GIST compared to primary lesions. Most metastatic GISTs had been treated with the tyrosine kinase inhibitors imatinib and sunitinib, but the high macrophage infiltrate was not related to this treatment. However, imatinib and sunitinib did induce secretion of anti-inflammatory IL-10 in macrophage cultures, indicating that treatment with these inhibitors might contribute to an immune suppressive microenvironment in GIST. Overall, our data reveal a picture of GIST as an active site of tumor-immune interaction in which suppressive mechanisms overrule potential antitumor responses. Tyrosine kinase inhibitors might promote this negative balance.