006, Figure 3C). At the end of the study, metastases were present in 15 out of Regorafenib msds 18 (83.3%) of the control group and 9 out of 18 (50.0%) of the vandetanib group (P=0.075). Lung and brain metastases were found in 20 and 6 mice, respectively. Most tumours were histologically identified as micronodules or assembled tumour cells (Figure 3C). Correlations between expression and gene amplification of EGFR in clinical samples Out of 90 cholangiocarcinoma samples, 19 (21.1%) were tested EGFR-positive by IHC. Among these cases, EGFR gene amplification was examined by FISH in the 19 EGFR-positive and 15 EGFR-negative samples. Of these 34 samples 8 had EGFR gene amplification, of which all 8 were confirmed as EGFR-positive by IHC. In contrast, none of the EGFR-negative samples were found to have gene amplification (P=0.

0045). Discussion Vandetanib is a tyrosine kinase inhibitor of both VEGFR-2 and EGFR, and preclinical studies have confirmed its anti-tumour effects in a range of cancer types (Wedge et al, 2002; Ciardiello et al, 2004; Taguchi et al, 2004; Williams et al, 2004; Yano et al, 2005). Phase III clinical studies are now underway with vandetanib in non-small-cell lung cancer following promising results in phase I and II studies (Holden et al, 2005; Natale et al, 2006; Tamura et al, 2006; Heymach et al, 2007a, 2007b). We have reported earlier that both EGFR and VEGF overexpressions are associated with progression of cholangiocarcinoma (Yoshikawa et al, 2008), and hypothesised that simultaneously blocking the EGFR and VEGF pathways might have synergistic therapeutic effects against cholangiocarcinoma.

In this study, we investigated the efficacy of vandetanib in cholangiocarcinoma cell lines and in xenograft models, and report here that vandetanib strongly inhibits tumour progression in vivo. Anti-proliferative effects of vandetanib in vitro As VEGFR-2 was not expressed in any of cholangiocarcinoma cell lines, we assumed that the anti-proliferative effects of vandetanib observed in this in vitro study were mainly because of the inhibition of EGFR signalling. All cholangiocarcinoma cell lines examined in this study expressed EGFR and VEGF, but the degree of the anti-proliferative effect of vandetanib in vitro varied between the cell lines. TKKK cells were sensitive to vandetanib, TGBC24TKB cells were moderately resistant to vandetanib, whereas OZ and HuCCT1 cells were refractory to vandetanib.

This finding is partly consistent with an earlier report that HuCCT1 cell line was resistant to the EGFR inhibitor, erlotinib (Jimeno et al, 2005). It is interesting that KRAS mutations were found in both cell lines (HuCCT1 and OZ) considered refractory to vandetanib in this study, and KRAS mutation has been reported as a mechanism of resistance to Batimastat EGFR inhibitors in lung and colorectal cancers (Pao et al, 2005; Lievre et al, 2006).