Stephanie van Hoppe

151 The impact of OATPs on disposition and toxicity of antitumor drugs; insights from KO and humanized mice 2 . 6 . 1 . S o ra fe n i b i n v i t ro and i n v i vo s t ud i e s The TKI sorafenib was further investigated in vivo in mouse models because it was readily transported by both OATP1B1 and OATP1B3 in vitro . Unlike its conjugate, sorafenib-glucuronide, which will be discussed later, the plasma pharmacokinetics of parent sorafenib, or its relative accumulation in the liver, was not substantially altered in either Oatp1b2-/- mice (DBA background strain), or in Oatp1a/1b(-/-) mice (FVB background strain) after administration of 10 mg/kg oral sorafenib (Zimmerman et al., 2013). A partial explanation might be that in vitro sorafenib was poorly transported by mouse Oatp1b2. However, sorafenib transport by the two other prominent sinusoidal uptake transporters in the mouse liver, Oatp1a1 and Oatp1a4, was not tested in vitro , so it remains uncertain whether there is an in vitro / in vivo discrepancy here (Zimmerman et al., 2013). Some TKIs have been further tested for their ability to inhibit human OATP1B1 in vitro and in vivo (Hu et al., 2014). These studies were triggered by clinical observations that coadministration of several TKIs can increase the systemic exposure to docetaxel in cancer patients. The underlying mechanism was poorly understood, but, as docetaxel clearance may in part depend on OATP function (see section 2.1.2), it might involve inhibition of OATPs by these TKIs (Hu et al., 2014). Nearly all 16 tested TKIs inhibited E2G uptake by human OATP1B1 in vitro when applied at 10 microM, and 4 of these (axitinib, nilotinib, pazopanib, and sorafenib) by more than 10-fold. Interestingly, three of these (axitinib, pazopanib, and sorafenib) are known to increase the docetaxel AUC by up to 50% or more when coadministered in patients (clinical data for nilotinib/ docetaxel coadministration are not available). Further analysis of sorafenib showed that it inhibited docetaxel uptake by OATP1B1 in vitro with an IC50 below 100 nM, and very extensively when applied at 10 microM. Mouse Oatp1b2 showed similar inhibition of E2G and docetaxel transport by sorafenib in vitro . In vivo , however, no significant effect of high-dose (60 mg/kg) oral sorafenib coadministration could be demonstrated on 10 mg/kg i.v. docetaxel plasma C max or AUC in wild-type, Oatp1b2(-/-), Oatp1a/1b(-/-), or OATP1B1-humanized mice. Also multiple sorafenib administrations did not elicit significant effects (Hu et al., 2014). Of note, in these specific experiments (see also section 2.1.2 above), the effects of Oatp1b2 knockout or OATP1B1 transgenics on docetaxel AUC were relatively modest (about 2-fold). This may have rendered these experiments less sensitive to picking up small pharmacokinetic effects of sorafenib- mediated inhibition. The absence of an obvious pharmacokinetic interaction between sorafenib and docetaxel in the various mouse strains suggests that there may be other factors in mice, perhaps alternative or compensatory mechanisms, that can sufficiently offset any changes in OATP1B-like activity towards docetaxel clearance. Also here we have to consider the possibility that interactions of OATPs with certain drugs may be