Abstract

The second-generation tyrosine kinase inhibitor and anticancer drug axitinib is a potent, orally active inhibitor of the vascular endothelial growth factor receptors 1, 2, and 3. Axitinib has clinical activity against solid tumors such as metastatic renal cell carcinoma and advanced pancreatic cancer. We studied axitinib transport using Madin-Darby canine kidney II cells overexpressing human ABCB1 or ABCG2 or murine Abcg2. Axitinib was a good substrate of ABCB1 and Abcg2, whereas transport activity by ABCG2 was moderate. These transporters may therefore contribute to axitinib resistance in tumor cells. Upon oral administration of axitinib, Abcg2(−/−) and Abcb1a/1b;Abcg2(−/−) mice displayed 1.7- and 1.8-fold increased axitinib areas under the plasma concentration-time curve from 0 to 4 compared with those of wild-type mice. Plasma concentrations in Abcb1a/1b(−/−) mice were not significantly increased. In contrast, relative brain accumulation of axitinib in Abcb1a/1b(−/−) and Abcb1a/1b;Abcg2(−/−) mice was, respectively, 6.8- and 13.9-fold higher than that in wild-type mice at 1 h and 4.9- and 20.7-fold at 4 h after axitinib administration. In Abcg2(−/−) mice, we found no significant differences in brain accumulation compared with those in wild-type mice. Thus, Abcb1 strongly restricts axitinib brain accumulation and completely compensates for the loss of Abcg2 at the blood-brain barrier, whereas Abcg2 can only partially take over Abcb1-mediated axitinib efflux. Hence, Abcg2 has a stronger impact on axitinib oral plasma pharmacokinetics, whereas Abcb1 is the more important transporter at the blood-brain barrier. These findings illustrate that in vitro transport data for ABCB1 and ABCG2 cannot always be simply extrapolated to the prediction of the relative impact of these transporters on oral availability versus brain penetration.