Stephanie van Hoppe

59 Brain accumulation of osimertinib and its active metabolite is restricted by ABCB1 & ABCG2 availability (53). We suspect that this could be due to a much more abundant presence of various other drug uptake systems as well as an overall higher influx capacity in the intestine as compared to the BBB. Thus, when removing or inhibiting ABCB1 and/or ABCG2, the oral availability of a substrate drug will generally be less enhanced than its brain penetration. Inhumans, osimertinib is thought tobepredominantlymetabolizedby theCYP3A4/5 enzymes, while in mice this appears to be primarily mediated by mouse Cyp2d proteins [44]. We observed no significant impact of Cyp3a deficiency on the osimertinib systemic availability and its tissue exposure. This is consistent with a study showing that codosing mice with osimertinib and the CYP450 inhibitor benzotriazole-1-amine did not have a significant effect on osimertinib metabolism [55]. Also, we observed only a borderline significant difference in AUC between the Cyp3aXAV and Cyp3a −/− mice. These findings can probably be explained by a dominant function of the murine Cyp2d proteins in thesemice [44]. Various clinical trials have assessed the therapeutic efficacy of osimertinib for NSCLC and metastatic NSCLC patients. A recent study demonstrated that patients treated with osimertinib developed CNS metastases to a lesser extent compared to the standard EGFR-TKIs treatment. In addition, another study showed that osimertinib had a reasonably high ORR in CNS metastases of 64% [18, 56]. These findings clearly suggest the promising therapeutic efficiency osimertinib could offer for NSCLC patients both with and without CNS involvement. One factor as to why osimertinib seems to be partially effective against brain metastases could be its intrinsically high brain penetration. In fact, the brain-to-plasma ratio in WT mice (14.5, Figure 3, Table 2) was only slightly lower than the liver-to-plasma ratio (29), and in Abcb1a/1b;Abcg2 -/- mice it was even considerably higher (93.0 in brain vs 28.4 in liver). This already favorable behavior of osimertinib with respect to brain metastases might therefore possibly be even further boosted by coadministration of efficient ABCB1 and ABCG2 inhibitors. Based on our findings, it is further likely that tumors substantially expressing ABCB1 and/or ABCG2 will also display some resistance to osimertinib-based chemotherapy. Thus, inhibiting these transporters with effective dual ABCB1 and ABCG2 inhibitors such as elacridar during osimertinib therapy could potentially further improve the tumor response. However, caution should always be exercised to prevent unexpected toxicities, and these possible approaches will first need to be carefully examined in clinical trials, as would also apply to efforts to increase osimertinib levels in the brain of patients with CNS tumors or metastases.