Stephanie van Hoppe

86 Chapter 4 The oral bioavailability of ibrutinib is clearly not restricted by mAbcb1 and mAbcg2 activity. This contrasts with the clear impact of mAbcb1 on the brain accumulation of ibrutinib. We (and others) have observed this for a range of different drugs: generally, when a drug is only moderately transported in vitro by ABCB1 and/or ABCG2, such as ponatinib and regorafenib, we see a more pronounced effect of these transporters in restricting the brain accumulation of these drugs than in reducing their oral availability [38, 39]. Only when drugs are very efficiently transported substrates, such as afatinib, we tend to see a clear role in oral availability [4]. Whereas the exact reason of this apparent discrepancy is unknown, it may be due to the presence of more broad-specificity uptake systems as well as a higher influx capacity in the intestine as compared to the much more restrictive BBB. An obvious translational implication of these findings is that, when inhibiting ABCB1 and/or ABCG2 with a pharmacological ABCB1/ABCG2 inhibitor, the oral availability of a drug will generally be (much) less enhanced than its brain penetration. Especially when there are potential toxic side effects of higher systemic exposure of the drug(s), this might be an advantage. Our finding that mAbcb1 deficiency unexpectedly decreased ibrutinib oral bioavailability in our new mouse facility, but not in our old facility, may be explained by the concomitant introduction of an entirely new intestinal microflora. For instance, bacterial inducing compounds present in our new microflora, but not the old, might cause upregulation of ibrutinib-metabolizing enzymes. If these inducers are normally kept out of the system by mAbcb1, this would result in higher metabolic clearance, and hence lower plasma levels, of ibrutinib in Abcb1-deficient mice in the new facility. In this context, it is interesting to note that the ibrutinib-DiOH-to-ibrutinib plasma ratios in the oldmouse facilitywere similar betweenAbcb1a/1b:Abcg2 -/- andWTmice (Supplemental Figure 3), whereas in the new mouse facility they were about 2-fold higher in both Abcb1a/1b-deficient strains compared to the WT mice (Supplemental Figure 4). This would be in line with a relatively more extensive metabolic conversion of ibrutinib to ibrutinib-DiOH in the newmouse facility in the Abcb1a/1b-deficient strains. Needless to say, given the (mostly uncharted) complexity of intestinal microflora composition, the identity of any intestinal inducing compounds would be difficult to resolve. Our CYP3A studies show that, like in humans, ibrutinib oral bioavailability is strongly restricted by CYP3A-mediated conversion to ibrutinib-DiOH. This is true for both the mouse Cyp3a family (encompassing some 8 functional Cyp3a genes) and the transgenic human CYP3A4. The observation that the endogenous mouse Cyp3a proteins had an even stronger effect on ibrutinib oral bioavailability than the transgenic human CYP3A4 (Figure 4) may reflect effectively higher expression of one or more of the mouse Cyp3a proteins, and/orhigher intrinsicefficacy inmetabolizing ibrutinib.Thisfinding is therefore not in itself surprising. In the CYP3A4 transgenic mice, liver and intestinal expression of