Targeting the IL-6 Dependent Phenotype Can Identify Novel Therapies for Cholangiocarcinoma

Background: The need for new therapies for cholangiocarcinoma is highlighted by their poor prognosis and refractoriness to chemotherapy. Increased production of Interleukin-6 promotes cholangiocarcinoma growth and contributes to chemoresistance by activating cell survival mechanisms. We sought to identify biologically active compounds capable of ameliorating the phenotypic effects of IL-6 expression and to explore their potential therapeutic use for cholangiocarcinoma. Methodology: A genomic signature associated with Interleukin-6 expression in Mz-ChA-1 human malignant cholangiocytes was derived. Computational bioinformatics analysis was performed to identify compounds that induced inverse gene changes to the signature. The effect of these compounds on cholangiocarcinoma growth was then experimentally verified in vitro and in vivo. Interactions with other therapeutic agents were evaluated using median effects analysis. Principal Findings: A group of structurally related compounds, nitrendipine, nifedipine and felodipine was identified. All three compounds were cytotoxic to Mz-ChA-1 cells with an IC50 for felodipine of 26 mu M, nitrendipine, 44 mu M and nifedipine, 15 mu M. Similar results were observed in KMCH-1, CC-LP-1 and TFK-1 cholangiocarcinoma cell lines. At a fractional effect of 0.5, all three agents were synergistic with either camptothecin or gemcitabine in Mz-ChA-1 cells in vitro. Co-administration of felodipine and gemcitabine decreased the growth of Mz-ChA-1 cell xenografts in nude athymic mice. Conclusions: Computational bioinformatics analysis of phenotype-based genomic expression can be used to identify therapeutic agents. Using this drug discovery approach based on targeting a defined tumor associated phenotype, we identified compounds with the potential for therapeutic use in cholangiocarcinoma.