Abstract A143: A preclinical in vivo model to assess pharmacodynamic inhibition of ACK1.

Abstract

Introduction: Activated Cdc42-associated kinase (ACK1, TNK2) has been reported to be over-expressed in a range of tumor types including prostate and lung cancer, and has been suggested as a potential novel anticancer target. A number of kinase inhibitors currently in clinical development show inhibition of ACK1 biochemical activity in enzyme assays, yet their ability to inhibit ACK1 in vivo is unknown. This study describes the development of a pharmacodynamic model to evaluate ACK1 target coverage in vivo. Methods: MDA-MB-231 cells were engineered to over-express ACK1 under the control of a doxycycline-inducible promoter (MDA-MB-231-ACK1). MDA-MB-231-ACK1 cells were grown as subcutaneous xenografts in MF-1 nude mice and ACK1 expression induced by dosing doxycycline by oral gavage. Saracatinib/AZD-0530 (100mg/kg qd) or ASP-3026 (50mg/kg bid) were dosed for four consecutive days to reach steady state exposures at approximately MTD. Pharmacodynamic inhibition of ACK1 autophosphorylation was assessed using an in-house MSD assay and confirmed by Western blotting, in order to evaluate target inhibition in vivo. Results: ASP-3026 and saracatinib showed inhibition of cellular pACK1 of MDA-MB-231-ACK1 cells in vitro, with IC50s of 0.12 μM and 1.1 μM, respectively. Using the same cell line in vivo, expression of total and phosphorylated ACK1 was induced by daily oral dosing of doxycycline. We used this model to evaluate PK-PD correlations of ASP-3026 and saracatinib using pACK1 as a marker of target inhibition. Good plasma and tumor exposure of ASP-3026 was achieved at 50mg/kg bid and no drug-drug interactions were observed in combination with doxycycline. ASP-3026 showed 50-80% inhibition of pACK1 in vivo consistent with unbound plasma concentrations in excess of in vitro IC50 and in good agreement with predictions. Good exposure of saracatinib was observed at 100 mg/kg qd with total plasma Cmax of >4μM in excess of maximally tolerated clinical exposures (0.82μM). Tumor exposures of saracatinib were up to 10 x higher than plasma exposures (Cmax >40 μM), and in considerable excess of the cellular IC50, yet no ACK1 inhibition was observed in vivo. Conclusions: A robust PD model was established to evaluate ACK1 inhibition in vivo. Up to 80% inhibition of ACK1 in vivo was observed with ASP-3026, falling to approximately 50% at Cmin, suggesting that this agent does not completely inhibit ACK1 in vivo at steady state. In contrast, saracatinib did not significantly inhibit ACK1 at any of the exposures achieved, suggesting that it has limited in vivo activity against this target. Unbound plasma exposures predicted in vivo PD responses very well, whereas tumor exposures of these agents were not predictive of target inhibition. This model can be incorporated into a screening cascade for prioritising in vivo active ACK1 inhibitors. Citation Information: Mol Cancer Ther 2013;12(11 Suppl):A143. Citation Format: Mark Albertella, Shilina Roman, Mike Briggs, Aaron Cranston, Ralph Graeser, George Hynd, David Jones, Peter Lockey, Akbor Malik, Stephen Price, Susanna Stimpson, Patrizia Tisselli, Janine Arts, Simon Green. A preclinical in vivo model to assess pharmacodynamic inhibition of ACK1. [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference: Molecular Targets and Cancer Therapeutics; 2013 Oct 19-23; Boston, MA. Philadelphia (PA): AACR; Mol Cancer Ther 2013;12(11 Suppl):Abstract nr A143.