These in vitro results suggest that in cholangiocarcinoma cells, upregulation of the RAS/RAF/MAPK pathway by mutant KRAS might counteract the anti-growth effect of XL184 vandetanib by EGFR inhibition. The incidence of KRAS mutation in cholangiocarcinoma is estimated to be 54�C67% (Tada et al, 1990; Tannapfel et al, 2000), and therefore it may be important to examine the KRAS status when evaluating the activity of EGFR inhibitors in cholangiocarcinoma. In non-small-cell lung cancer, EGFR mutation and/or amplification have been reported as possible predictive factors of sensitivity to EGFR tyrosine kinase inhibitors (Lynch et al, 2004; Paez et al, 2004; Pao et al, 2004; Cappuzzo et al, 2005). Of the cell lines without KRAS mutation, TKKK, which has EGFR amplification, was most sensitive to vandetanib.

The incidence of EGFR mutations in cholangiocarcinoma is reported as 13.6�C15.0% (Gwak et al, 2005; Leone et al, 2006). However, we did not detect mutation in the kinase domain of the EGFR gene in the cell lines used in this study. We have reported earlier that EGFR overexpression occurs in ~20% of primary cholangiocarcinomas and is associated with tumour progression and poor outcome (Yoshikawa et al, 2008). In this study, our FISH analysis of clinical samples revealed that EGFR gene amplification was present in 42% (8 out of 19) of samples with EGFR overexpression, but absent in samples lacking EGFR overexpression. This result is consistent with an earlier report that EGFR amplification was found in 6.8% of cholangiocarcinomas (Nakazawa et al, 2005).

Collectively, the EGFR and KRAS gene status may be a potential biomarker for predicting the response to inhibitors of EGFR including vandetanib in cholangiocarcinoma. Anti-tumour effects of vandetanib in vivo On the basis of the in vitro data, we tested TKKK (the most sensitive) and OZ (the most resistant) cells in an in vivo therapeutic Dacomitinib model. As VEGFR-2 was not expressed in both cells, direct anti-tumour effect of vandetanib in this study was anti-EGFR inhibition, and anti-VEGFR-2 inhibition of vandetanib was exerted in the tumour stroma. Vandetanib greatly suppressed the tumour growth of the TKKK xenograft through anti-EGFR and VEGFR-2 inhibition, consistent with the in vitro study. However, vandetanib also inhibited tumour growth even in the OZ (refractory to EGFR inhibition) xenograft, when given at higher dose. Vascular endothelial growth factor receptor-2 expression in OZ xenograft with high-dose vandetanib treatment was also reduced, and histologically, both TKKK and OZ tumours treated with vandetanib showed necrosis, reduced microvessels, reduced proliferation, and increased apoptosis.