In Vitro to In Vivo Translation of Complex Drug-Drug Interactions Involving Retinoic Acid

Thesis (Ph.D.)--University of Washington, 2018all-trans-retinoic acid (atRA), the active metabolite of vitamin A, is a ligand for several nuclear receptors and acts as a critical regulator of many physiological processes. The cytochrome P450 26 (CYP26) enzymes are responsible for atRA clearance and are potential drug targets to increase concentrations of endogenous atRA in a tissue-specific manner. The first project of this thesis aimed to establish the relationship between CYP26 inhibition, by the potent CYP26A1 and B1 inhibitor talarozole, and altered atRA concentrations in tissues, and to quantify the increase in endogenous atRA concentrations necessary to alter atRA signaling in target organs. The second part of this thesis focused on evaluating exogenous atRA and retinoid dosing on gene regulation. Several in vitro and preclinical studies have suggested that atRA down-regulates CYP2D6 expression and activity. Both atRA and its stereoisomer 13-cis retinoic acid (13cisRA) are used clinically, and steady state exposures of atRA are similar after dosing with atRA or 13cisRA. The second aim of this thesis was to determine whether 13-cis retinoic acid (13cisRA) or its active metabolites, atRA and 4-oxo-13cisRA, cause a drug-drug interaction (DDI) with CYP2D6, decreasing the clearance of the CYP2D6 probe dextromethorphan. Further, the effects of 13cisRA and its metabolites (i.e., atRA and 4-oxo-13cisRA) on CYP2D6 and CYP3A4 in human hepatocytes were determined and used to predict whether the in vivo results could be correctly predicted from in vitro. Following a single 2.5-mg/kg dose of talarozole to mice, atRA concentrations increased up to 5.7-, 2.7-, and 2.5-fold in serum, liver, and testis, respectively, resulting in induction of Cyp26a1 in the liver and testis and Rar and Pgc1 in liver. The increase in atRA concentrations was well-predicted from talarozole pharmacokinetics and in vitro data of CYP26 inhibition. After multiple doses of talarozole, a significant increase in atRA concentrations was observed in serum but not in liver or testis. This lack of increase in tissue atRA concentrations correlated with increase in CYP26A1 expression in liver and testis. The increased atRA concentrations in serum without a change in liver suggest that CYP26B1 in extrahepatic sites plays a key role in regulating systemic atRA exposure. The putative DDI between 13cisRA and dextromethorphan was studied in clinical study in eight healthy volunteers. The geometric mean ratio (GMR; 90% CI) of dextromethorphan area under the curve from time zero to infinity (AUC0-∞) prior to and after 13cisRA treatment was 0.822 (0.677 – 0.998) indicating that dextromethorphan clearance and CYP2D6 activity was increased following 13cisRA treatment. The dextrorphan-to-dextromethorphan AUC0-∞ ratio and dextrorphan formation clearance (Clf) were also increased consistent with CYP2D6 induction. In addition, the Clf of 3-methoxymorphinan, a CYP3A4-specific metabolite of dextromethorphan, and the Clf of 6β-hydroxycortisol, an endogenous marker of CYP3A4 activity, were also increased suggesting that CYP3A4 expression was also induced by 13cisRA. Quantitative DDI predictions were made with data for CYP2D6 mRNA down-regulation in human hepatocytes assuming that 13cisRA and its metabolites competitively bind to the same receptor to elicit CYP2D6 down-regulation. An ~50% decrease in CYP2D6 activity and 2-fold increase in dextromethorphan AUC were predicted to be observed in vivo after dosing with 13cisRA. These data demonstrate a clear disconnect between in vitro and in vivo CYP2D6 down-regulation. In contrast to data with CYP2D6, data from hepatocytes support the induction of CYP3A4 observed in the clinical study. This is the first study to fully characterize the concentration-response effect of retinoids on CYP2D6 activity and highlights the difficulty in translating in vitro observations of CYP down-regulation to the clinic