Abstract 4629: Physiologically based pharmacokinetic model and correlative in vitro assessment of metabolism-based interaction between Everolimus and PI3K/mTOR inhibitor BEZ235

Abstract

Abstract The PI3/Akt/mTOR kinase is a key intracellular signaling pathway and deregulation of this signaling pathway has been implicated in tumorigenesis and in numerous malignancies. Everolimus (RAD) is a potent allosteric inhibitor of mTORC1 and BEZ235 is a novel dual PI3K/ mTOR inhibitor (mTORCl and mTORC2). In preclinical studies, the combination of BEZ235 and RAD has shown synergistic effect and a phase 1b clinical trial to determine the safety, pharmacokinetics (PK) and pharmacodynamics in subjects with advanced solid tumors was conducted at University of Cincinnati. Here, we explored the potential mechanism and clinical impact of the PK interaction of this combination using a physiologically based PK (PBPK) model. The dose escalation consisted of five dose levels with a starting oral daily doses of RAD (2.5 mg) and BEZ235 (200 mg) with 28-day study treatment cycles. Serial blood samples were collected on Days 1 and 28 and plasma BEZ235 and blood RAD levels were analyzed using validated HPLC (with fluorescene detection) and LC/MS methods, respectively. The effect of BEZ235 on the expression and activity of cytochrome P450 CYP3A4, the principal RAD metabolizing enzyme, was studied employing pooled human liver microsomes (HLMs) and primary human hepatocytes. HLMs were employed to investigate potential competitive and/or time-dependent inhibition (TDI) or CYP3A4 by BEZ235. Potential CYP3A4 inductive effects were also examined in the hepatocyte model. A PBPK model was then developed along with a mechanistic static model to predict the change in the clinical PK of RAD. Population PK (NONMEM 7.2) was employed for analyzing data from our clinical trials and that of published studies. BEZ235 was observed to be a time-dependent CYP3A4 inhibitor (Kinact and Ki were, 0.0175 min-1 and 5.21 μM, respectively). It was also observed to be a mild CYP3A4 inducer (Emax, 2.78 µM and EC50 ,1.96 µM). Qualification of PBPK model for BEZ235 was performed using data from our clinical trial; everolimus was qualified using published data. PK analyses revealed that in our study the CL/F of RAD in the presence of BEZ235 was 10.5 ± 1.74 L/hr which was significantly lower than the reported RAD CL/F of 18.8 ± 0.98 L/hr when dosed alone. In-vitro studies and PBPK model suggests that TDI of CYP3A by BEZ235 may play a role in the observed DDI. In summary, the mechanistic model developed predicted the time and dose dependent effect of BEZ235 on RAD clearance and exposure. Our findings will aid efforts to optimize RAD and BEZ235 combination for future clinical trials. Citation Format: Ganesh P. Moorthy, Bilal Abuasal, Larry Sallans, John C. Morris, George Thomas, Pankaj B. Desai. Physiologically based pharmacokinetic model and correlative in vitro assessment of metabolism-based interaction between Everolimus and PI3K/mTOR inhibitor BEZ235. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr 4629. doi:10.1158/1538-7445.AM2014-4629