Tanaya R. Vaidya (1), Hardik Mody (1), Yesenia L. Franco (1), Ashley N. Brown (2), Sihem Ait-Oudhia (3)
(1) Center for Pharmacometrics and Systems Pharmacology, Department of Pharmaceutics, College of Pharmacy, University of Florida, USA, (2) Institute for Therapeutic Innovation, Department of Medicine, University of Florida, USA, (3) Merck & Co., Inc, Kenilworth, New Jersey, USA.
Introduction: Doxorubicin (DOX) belongs to the anthracycline class of chemotherapeutic agents and is known to cause dose-dependent and life-threatening cardiotoxicity (1). Several in vitro and in vivo studies have demonstrated activation of apoptotic pathways in response to anthracycline-induced stress as the major mechanism of cardiomyocytes death (2-4). In addition, release of cardiotoxicity biomarkers in the serum, such as B-type natriuretic peptide (BNP), has been demonstrated to be indicative of DOX-induced cardiac stress (5). Clinically, modification of the dosing schedule of DOX through dose-fractionation or slow continuous infusions has been associated with diminished cardiotoxicity response (6). Thus, a quantitative understanding of the pharmacokinetic (PK) drivers of toxicity might be useful at optimizing DOX dosing regimens to minimize cardiotoxicity.
Objectives: The objectives of this work were: 1.) To assess the utility of a multi-scale and translational quantitative systems toxicology (QST)-PK/toxicodynamics (TD) approach at optimizing DOX dosing regimens for early cardiotoxicity monitoring and minimization through the use of an in vitro three-dimensional and dynamic (3DD) cell culture system; 2.) To translate findings from our in vitro 3DD proof-of-concept studies to clinically relevant DOX PK and TD response profiles, and identify PK drivers of acute cardiotoxicity through model simulations.
Methods: A cellular-level QST model was established by exposing the human cardiomyocyte cell line, AC16, to DOX (100 and 500 nM), and measuring the dynamics of key cellular signaling proteins, cell viability and released biomarkers of cardiomyocyte injury, over a time course. Experiments were scaled up to a 3DD cell culture system to evaluate DOX-induced cardiotoxicity under various dosing regimen scenarios. PK determinants of DOX mediated cardiotoxicity were identified by assessing PK metrics that correlated with cardiotoxicity measures. Subsequently, in vitro findings were translated to the in vivo setting by scaling up PK and TD responses from the 3DD system to the clinical level through the use of a hybrid physiologically-based PK (PBPK)-TD approach. A hybrid PBPK model was established in pre-clinical species (mouse, rat, rabbit, dog) by utilizing plasma concentrations to drive perfusion input to the cardiac tissue that was further divided into interstitial fluid (ISF) space and intracellular space. System-related parameters were fixed from the literature and unknown drug-related parameters were estimated in the model. Finally, the model was scaled up to characterize cardiac tissue-level concentrations in humans based on system parameters for a typical 70 kg subject. Predicted cardiac tissue ISF concentrations were utilized to drive TD responses in humans, under various dosing regimen scenarios, and the PK driver of toxicity was determined. All QST-PK/TD modeling and simulations were performed using Monolix suite versions 2016R1 or higher [7].
Results: The developed cellular-level QST model captured well the observed protein dynamics, AC16 cell viability, and BNP release for the tested concentration levels. The first-order drug degradation rate constant in cell-culture medium was determined at (0.0222±0.0004 h-1), and incorporated in the model to describe well in vitro TD responses. In the 3DD setting, DOX PK displayed similar trends as that of in vivo PK, with a terminal elimination half-life of 23.2±1.03 h. DOX dose-fractionation displayed a significant reduction in cardiotoxicity with a once daily regimen as compared to larger doses administered every two or four days (p<0.01) over the span of the study, indicating peak concentrations as the determinant of DOX induced cardiotoxicity. The PK/TD models scaled up from the 3DD cell culture system were able to capture well clinical DOX PK profiles in human cardiac tissue ISF space, as well as clinical cardiotoxicity responses. Alleviation in cardiotoxicity was observed with dose-fractionated regimens administered once every week (Q1W), as compared to the standard once every three weeks (Q3W) regimen, with a 2.5-fold reduction in BNP release predicted for a DOX dose of 16.67 mg/m2 administered Q1W for three cycles, as against 50 mg/m2 administered Q3W, as short-term 15 minutes infusions.
Conclusion: The developed multi-scale and translational QST-PK/TD platform may serve as a tool for assessment of early toxicity and/or efficacy of developmental drugs in vitro.
References:
[1] Swain S.M. et al. Cancer (2003), 97(11): 2869-2879
[2] Liu J. et al. Am J Physiol. Heart Circ. Physiol. (2008), 295(5): 1956-1965
[3] Arola O.J. et al. Cancer Res. (2000), 60(7): 1789-1792
[4] Goorin A.M. et al. J Pediatr. (1990), 116(1): 144-147
[5] Tian S. et al. Front Oncol. (2014), 4:277
[6] Weiss A.J. et al Cancer Treat Rep. (1976), 60(7): 813-822
[7] Monolix version 2016R1. Antony, France: Lixoft SAS, 2016.
Reference: PAGE () Abstr 9586 [www.page-meeting.org/?abstract=9586]
Poster: Drug/Disease Modelling - Safety