Category: Preclinical Development
Purpose: Entrectinib is a potent and orally bioavailable inhibitor against TRKA, TRKB, TRKC, ROS1, and ALK currently in clinical development for the treatment of tumors harboring NTRK or ROS1 fusions. Entrectinib was specifically selected for enhanced CNS distribution . Promising antitumor efficacy has been demonstrated with entrectinib in both preclinical and clinical studies across multiple tumor types, including brain tumors, despite the fact that entrectinib was found to be a P-gp substrate in vitro as defined by a conventional P-gp assay with a high efflux ratio (ER). Interestingly, entrectinib showed high steady-state brain-to-plasma concentration ratios of entrectinib (0.6 – 2.2) in vivo across several animal species following repeated daily oral administration, inconsistent with the P-gp interaction and apparent high ER observed in vitro. The purpose of this study was to determine an unequivocal assessment of the brain penetration properties and P-gp interaction of entrectinib, and compare these properties to other relevant drugs in this therapeutic space, including crizotinib (standard care for tumors harboring ROS1 fusion) and larotrectinib (first in class treatment for tumors harboring NTRK fusions).
Methods: The in vitro P-gp transport properties of entrectinib, crizotinib, larotrectinib, and digoxin (a positive control for a substrate of P-gp) were measured in the LLC-PK1 cell monolayers over-expressing human or mouse P-gp in the absence and presence of a known P-gp inhibitor. A novel “apical efflux ratio model” is defined by the apical efflux ratio (AP-ER) of the Papp (A-to-B) in the presence of a P-gp inhibitor to the Papp (A-to-B) in the absence of a P-gp inhibitor. In vivo brain penetration properties of entrectinib, crizotinib and larotrectinib were investigated in a rat in vivo study by plasma, brain and CSF sampling after an IV infusion study to achieve close to steady-state. Non-specific brain tissue binding was determined ex vivo by a novel kinetic Lipid Membrane Binding Assay (kinetic-LIMBA) particularly designed for compounds with a high tissue binding. The antitumor effects were examined in the TPM3-NTRK1 KM12-Luciferase (KM12-Luc) tumor cells inoculated intracranially to mice following daily oral administration of entrectinib.
Results: Entrectinib, crizotinib, larotrectinib, and digoxin all displayed significantly high ERs when assessed using the conventional bidirectional in vitro P-gp assay; all were ≥10 in human and mouse assays. However, when considering P-gp efflux after dosing in the apical-to-basolateral direction (i.e. AP-ER), entrectinib appeared to be remarkably distinct with a lower AP-ER than the other drugs (with values of 1.1, 3.5, 2.8 and 3.9 for the human P-gp, respectively; and 1.5, 4.9, 6.3, and 3.4 for the mouse P-gp, respectively). In vivo, following an IV infusion of the agent up to 6 hr to rats (maximum feasible time), entrectinib showed CSF-to-free plasma concentration ratio of ~0.2 under conditions where steady-state was not reached, suggesting that entrectinib slowly reaches equilibrium and thus this ratio may be under-estimated. Comparatively, crizotinib and larotrectinib reached steady state much faster (≤~6 hr), but the CSF-to-free plasma concentrations were ~10-fold lower than for entrectinib, indicating that crizotinib and larotrectinib are stronger P-gp substrates in vivo. The free brain concentrations were estimated to be within ~2-fold of the directly measured CSF concentrations, supporting that using CSF concentrations as a reasonable surrogate of active exposures in the brain for these molecules tested. Consistent with its brain exposure, entrectinib demonstrated tumor growth inhibition and full survival benefit of the intracranial tumor model at clinical relevant systemic exposures.
Conclusion: The present studies demonstrated that entrectinib is a poor substrate of P-gp which favors more brain penetration compared to typical P-gp substrates in this class, namely crizotinib and larotrectinib. The low AP-ER of entrectinib is consistent with its ability to penetrate to and maintain in the CNS at steady-state with strong antitumor efficacy in intracranial tumor models.
 Menichincheri et al., J. Med Chem, 2016, 59, 3392-3408
Mohhamed Ullah– Basel, Basel-Stadt, Switzerland
Claudia Senn– Basel, Basel-Stadt, Switzerland
Thomas Wirz– Basel, Basel-Stadt, Switzerland
Bjoern Wagner– Basel, Basel-Stadt, Switzerland
Dragomir Draganov– Basel, Basel-Stadt, Switzerland
Faye Vazvaei– New York, New York
Cecile De La Cruz– South San Francisco, California
Mark Merchant– South San Francisco, California
Axel Paehler– Basel, Basel-Stadt, Switzerland
Li Yu– DMPK/PD Project Lead, F Hoffman-La Roche Ltd, New York, New York