Stephanie van Hoppe

33 ABCG2 & ABCB1 transport afatinib and restrict its oral availability and brain accumulation D I S C U S S I ON A number of ABC transporters, including ABCB1 and ABCG2, cause resistance to a wide range of drugs. Expression of some ABC transporters, especially ABCB1 and ABCG2, was found to be associated with resistance to chemotherapy in several cancers, including NSCLC [27, 28]. This would suggest the possible importance of inhibiting these transporters when treating patients with chemotherapy to reverse such tumor resistance. Furthermore, NSCLC readily metastasizes to the brain [29], where ABCB1 and ABCG2 are expressed at the blood-brain barrier. Inhibiting both these transporters during chemotherapy could therefore possibly also improve treatment of metastases positioned in part or in whole behind the blood-brain barrier. Afatinib has been approved for the first-line treatment of patients with metastatic NSCLC whose tumors have epidermal growth factor receptor (EGFR) mutations [5, 17]. Because the ABC efflux transporters could affect the biological availability of afatinib at several levels, we wanted to investigate the possible effects these transporter proteins might have on afatinib disposition. We here show that both the ABC efflux transporters ABCG2 and ABCB1 are important factors that can limit uptake of afatinib. We demonstrated that afatinib is efficiently transported by hABCB1, hABCG2 and mAbcg2 in vitro (Fig. 1). Similar in vitro data were acquired in the past with sunitinib, ceritinib, topotecan and sorafenib [30- 32] The plasma afatinib concentrations in the Abcb1a/1b -/- ;Abcg2 -/- mice were highly increased relative to the WT levels. Single Abcg2 or Abcb1a/1b knockout mice also showed a marked increase in afatinib plasma concentrations, albeit not as high as seen in the combination knockout (Fig. 1). This effect was seen for all the experimental studies at the various end-points of 2, 8 and 24 hours after oral administration. These results indicate that the oral availability of afatinib is markedly restricted by both Abcg2 and Abcb1a/1b, which might result from decreased intestinal uptake and/or increased hepatic excretion mediated by these transporters. The results for the single and combination knockout strains suggested roughly additive effects of each transporter in limiting afatinib oral availability. Additionally we found that the plasma clearance was slower in the Abcb1a/1b -/- ;Abcg2 -/- mice as well as in the Abcg2 -/- mice, but perhaps less so for Abcb1a/1b -/- mice, compared to WT mice (Supplementary Fig. 4). Their mean elimination rate constants were 0.09 h -1 (t 1/2 = 7.6 h), 0.18 h -1 (t 1/2 = 3.8 h), 0.26 h -1 (t 1/2 = 2.7 h) and 0.36 h -1 (t 1/2 = 2.0 h) respectively. However, note that the elimination rate constant for the WT mice was calculated only until 8 h because the 24 h time point was below the quantification limit. Our data indicate that especially Abcg2, but also Abcb1a/1b, directly contribute to the plasma clearance of afatinib in mice. The accumulation of afatinib was highly increased in the brains of the Abcb1a/1b -/-