rat hepatocyte model (HepatoPac TM ) Diane Ramsden, Donald J. Tweedie, Roger St. George, Lin-Zhi Chen, and Yongmei Li

Transcription

1 DMD Fast This Forward. article has not Published been copyedited on and December formatted. The 23, final 2013 version as may doi: /dmd differ from this version. Generating an IVIVC for metabolism and liver enrichment of an HCV drug, faldaprevir, using a rat hepatocyte model (HepatoPac TM ) Diane Ramsden, Donald J. Tweedie, Roger St. George, Lin-Zhi Chen, and Yongmei Li Drug Metabolism & Pharmacokinetics, Boehringer Ingelheim Pharmaceuticals, Inc., Ridgefield, Connecticut, USA (D.R., D.J.T., R.S.G., L.Z.C., Y.L.) 1 Copyright 2013 by the American Society for Pharmacology and Experimental Therapeutics.

2 RUIG TITLE: Metabolism and liver enrichment of FDV in rat hepatocytes Please address correspondence to: Yongmei Li Drug Metabolism & Pharmacokinetics Boehringer Ingelheim Pharmaceuticals, Inc. 900 Ridgebury Rd., Ridgefield, CT 06877, USA Phone: Fax: umber of Text pages: 21 Tables: 4 Figures: 2 References: 36 umber of Words Abstract: 249 Introduction: 537 Discussion:

3 onstandard Abbreviations: ADME, absorption, distribution, metabolism, and excretion; DDI, drug-drug interaction; DMEs, drug metabolizing enzymes; QWBA, quantitative whole body autoradiography; HIM, hepatocyte incubation medium; HWM, hepatocyte wash medium; m/z, mass to charge ratio; CE, new chemical entities; rhcm, rat HepatoPac culture medium, rhim, rat HepatoPac incubation medium 3

4 ABSTRACT Hepatocytes provide an integrated model to study drug metabolism and disposition. Suspended and sandwich cultured hepatocytes have limitations in determining hepatocellular drug concentrations due to loss of polarity or significant decrease in expression of enzymes and transporters. Under-prediction of the extent of glucuronidation is also a concern for these hepatocyte models. Faldaprevir is an HCV protease inhibitor in late stage development that has demonstrated significant liver enrichment in in vivo rat models based on QWBA and liver to plasma AUC ratio. In bile duct cannulated rats the primary biliary metabolite was a glucuronide. It is difficult to assess liver enrichment in humans for ethical concerns and a lack of in vitro and in vivo correlation of glucuronidation has been reported. The current study was conducted to verify whether an hepatocyte model, rat HepatoPac, could overcome some of these limitations and provide validity for follow up studies with human HepatoPac. With rat HepatoPac, liver enrichment values averaged 34-fold and were consistent with rat QWBA (26.8-fold) and in vivo data (42-fold). In contrast, liver enrichment in suspended hepatocytes was only 2.8-fold. Furthermore, the extent of faldaprevir glucuronidation in HepatoPac studies was in agreement with in vivo, with glucuronidation as the major pathway (96%). Suspended rat hepatocytes did not generate the glucuronide or two key hydroxylated metabolites which were observed in vivo. verall, our studies suggest that HepatoPac is a promising in vitro model to predict in vivo liver enrichment and metabolism, especially for glucuronidation, and has demonstrated superiority over suspended hepatocytes. 4

5 ITRDUCTI Faldaprevir (BI , FDV) is an inhibitor of the HCV S3/4 protease and is currently in development for the treatment of hepatitis C virus infection (Sulkowski et al., 2013a; Sulkowski et al., 2013b). Based on an in vivo rat study, faldaprevir was highly enriched in the target organ, liver, with a liver/plasma AUC ratio of 42 (White et al., 2010). In addition, faldaprevir was extensively glucuronidated in the rat as evidenced by significant levels (37.6% of faldaprevirderived radioactivity) excreted in bile in a rat [ 14 C] absorption, distribution, metabolism, and excretion (ADME) study described herein. Recent regulatory drug-drug interaction (DDI) guidances have focused on providing a better understanding of the mechanisms of disposition and metabolism of drug candidates and relative clearance pathways (EMA, 2012; FDA, 2012). The EMA guidance on DDI has specifically pointed out that liver enrichment should be taken into account in DDI estimations, if available data indicate that the drug may accumulate in hepatocytes (EMA, 2012). The importance of understanding intracellular drug concentration in the liver, when evaluating drug efficacy, toxicity and DDI potential, is gaining increased attention (Korzekwa et al., 2012; Chu et al., 2013). Chu et al. (2013) reviewed a number of in vitro and in vivo model systems (such as membrane vesicles, recombinant proteins, suspended hepatocytes, sandwich cultured hepatocytes, perfused liver, and in vivo animal/human studies) used to assess hepatocellular concentrations of drugs in a quantitative manner. Methods for directly measuring intracellular concentrations, as well as modeling approaches to estimate the unbound fraction, were also presented. As highlighted by the authors, each currently available experimental model has its limitations in determining hepatocellular drug concentrations. For example, loss of polarity and 5

6 significant decrease in expression of enzymes and transporters are concerns for suspended and sandwich cultured hepatocytes, respectively. HepatoPac is an hepatocyte co-culture consisting of islands of hepatocytes surrounded by murine stromal cells (3T3-J2 fibroblasts), which has demonstrated long-term stable expression of drug metabolizing enzymes (DMEs) (both phase I and II) as well as canalicular efflux transport over an extended period (Khetani et al., 2008; Wang et al., 2010; Chan et al., 2013; Ukairo et al., 2013). In addition, human HepatoPac has been validated for uptake transporter function of ATP and TCP (Ramsden et al., 2014). Thus, HepatoPac provides an integrated system which enables simultaneous evaluation of the contributions of uptake, metabolism and efflux to the overall clearance of drugs. The expression of active uptake and efflux transporters makes HepatoPac an attractive model to evaluate intracellular drug concentrations and estimate the extent of hepatocellular enrichment. The purpose of the current study was to determine whether rat HepatoPac could overcome some of the limitations of other in vitro methodologies in assessing hepatocyte enrichment (Chu et al., 2013) and whether the extent to which metabolic pathways observed in vivo, particularly glucuronidation, would be replicated in vitro in rat HepatoPac. In addition, since HepatoPac has been demonstrated to accurately reproduce metabolite profiles (Wang et al., 2010), there was a specific intent to validate the rat HepatoPac model by showing an in vitro to in vivo concordance for faldaprevir metabolism and disposition with a goal of using the human HepatoPac system to help delineate the mechanisms of metabolite formation of faldaprevir in humans (Ramsden et al., 2014). 6

7 MATERIALS AD METHDS Chemicals, reagents and other materials. Rat HepatoPac cultures and proprietary maintenance and probing media were acquired from Hepregen Corporation (Medford, MA). Cultures were prepared from male cryoplateable Sprague Dawley rat hepatocytes purchased from Invitrogen Life Technologies (Grand Island, Y). Suspended male Spraque Dawley rat hepatocytes used to estimate liver partitioning were purchased from CellzDirect (Research Triangle Park, C). Suspended rat hepatocytes used for preliminary metabolite identification were isolated from male Sprague Dawley rats ( g) purchased from Charles River (Wilmington, MA). Hepatocytes were isolated using a two-step in situ collagenase perfusion method previously described (Lecluyse et al., 1994) and cryopreserved using a controlled freezing program (Houle et al., 2003). All protocols were approved by the Institutional Animal Care and Use Committee (IACUC). Hepatocyte incubation medium (HIM) for suspended hepatocytes was prepared with the following Ultrapure Bioreagent grade materials from J.T. Baker (Phillipsburg, J): magnesium sulphate 0.14 g/l, potassium phosphate, monobasic 0.16 g/l, potassium chloride 0.35 g/l, sodium choride 6.9 g/l, calcium chloride 0.15 g/l, sodium bicarbonate 2.4 g/l, HEPES 4.8 g/l and D-glucose, 2.0 g/l, ph adjusted to 7.4 with sodium hydroxide. Trypan Blue and William s E medium were purchased from Sigma Aldrich Inc. (St. Louis, M). A solution of Penicillin (10,000 units/ml), streptomycin (10 mg/ml), and GlutaMax (100X) were purchased from Gibco/Life Technologies (Grand Island, Y). Enhanced recovery plates and ITS premix were acquired from Corning Lifesciences (Bedford, MA). Faldaprevir, [ 14 C] faldaprevir, d7-faldaprevir, I79158 (M2a), I79157 (M2b) and BI (faldaprevir glucuronide) were synthesized at Boehringer Ingelheim Pharmaceuticals, Inc. (Ridgefield, CT). 7

8 All other reagents and solvents were of analytical grade or higher purity and were obtained from commercial suppliers. Evaluation of liver enrichment and metabolic profiling using rat HepatoPac. The rat HepatoPac model was prepared using cryoplateable rat hepatocytes as described previously (Khetani et al., 2008). HepatoPac cultures (5,000 hepatocytes in each well) were maintained in rat HepatoPac culture medium (rhcm). Culture medium was replaced every 2 days (50 µl per well) prior to incubation with test compounds. n day 9 post seeding faldaprevir was prepared at 0.15 and 2.86 µm in rat HepatoPac incubation medium (rhim; mixture of equal volume of rhcm containing 10% bovine serum and protein-free probing medium from Hepregen) and added to wells of the HepatoPac or fibroblast only plates. At various time points (up to 96 hr), the entire aliquot of medium (50 µl) was removed and quenched with 100 µl of quench solution consisting of acetonitrile/water (60:40, v:v, containing 0.1% acetic acid and 0.1 µm of an internal standard, d 7 -faldaprevir). These samples were designated as medium samples. The same volume (50 µl) of ice-cold blank medium was added to each well and rapidly aspirated and discarded in order to wash residual faldaprevir off the cell surface. The washing step was repeated. After the second wash, the same volume of blank medium (50 µl) was added to each well and the well surface was scraped with the end of a pipette tip and the mixture of cell debris and solution was triturated 3 times before depositing into corresponding wells in a separate 96- well plate. These samples were designated as lysate samples and were quenched with 100 µl of quench solution (as described for the medium samples). Samples were filtered through a membrane via centrifugation at 4 C for 15 min and the supernatant was analyzed by LC-MS/MS. All treatments were also repeated on plates containing only fibroblast feeder cells. 8

9 Uptake into suspended hepatocytes. Uptake studies using suspended hepatocytes were performed as previously described (Duan et al., 2012a). In short, reconstituted suspended rat hepatocytes (viability >85%) were incubated in HIM in an atmosphere of 5% C 2 for 10 min at 37 C. Faldaprevir and rat plasma were added to initiate the incubation. The final concentration of faldaprevir was 2 µm in 50% plasma. The final hepatocyte concentration was 1.6 x 10 6 cells/ml. Samples were incubated at 37 C for 16 min. After incubation samples were centrifuged at 11,000 g for 15 sec. A 400 µl aliquot of supernatant was retained for extraction. Remaining supernatant was aspirated and the cell pellet was resuspended in 0.8 ml of the hepatocyte wash medium (HWM=HIM without D-glucose). This step was repeated once more and the hepatocyte pellet was resuspended in 400 µl of ice cold HWM. The cell pellet and medium samples were extracted with 2x volume of acetonitrile containing 2% formic acid and centrifuged at 16,000 g for 5 min. Levels of faldaprevir in the medium and cell pellet were determined by LC-MS as described previously (Duan et al., 2012b). Metabolite profiling using suspended rat hepatocytes. Faldaprevir (final concentration 10 µm) was incubated with 2.5 x10 5 viable suspended rat hepatocytes in a volume of 1 ml up to 4 hr in William s E complete medium containing 100 nm dexamethasone, 100 units of penicillin, 0.1 mg streptomycin and 1% ITS premix solution in an atmosphere of 5% C 2 at 37 C. Viability of hepatocytes following reconstitution was >70%. At the end of the incubation, the reaction was stopped with 2 ml of quench containing 40% acetonitrile, 1 µm 1-naphthylglucuronide, and 0.1% acetic acid in water. Quantitative whole body autoradiography (QWBA). A QWBA study was conducted in accordance with Department of Health and Family Services, Radiation Protection Section (License o ). Male Long Evans rats were given [ 14 C]faldaprevir via gavage at a 9

10 dosage of 10 mg/kg in a solution of 80% PEG400/18% water/1% TRIS/1% meglumine. Animals were sacrificed at 1, 6, 12, 24, 48, 72, 336, and 672 hr postdose. Blood (approximately 2 to 10 ml) was collected into tubes containing sodium heparin from all animals at sacrifice and centrifuged to obtain plasma, which was analyzed by liquid scintillation counting. Concentrations of total radioactivity in the liver were obtained from the QWBA study using a validated image analysis system. Tissue concentrations were interpolated as nanocuries/g and then converted to ng equivalents/g on the basis of the faldaprevir specific activity (10.2 µci/mg in oral dose). Metabolite profiling of rat bile samples. Metabolic profiling of bile collected from bile duct-cannulated (BDC) Sprague Dawley rats dosed with [ 14 C]faldaprevir was performed as part of the rat [ 14 C]ADME studies. The animal protocol was reviewed and approved by Institutional Animal Care and Use Committee (IACUC). Briefly, each BDC rat received a single dose of faldaprevir of 10 mg/kg by oral gavage. Bile was collected into vials over dry ice pre-dose and at intervals of 0-2, 2-4, 4-6, 6-8, 8-24 and hr post dose. The radioactivity of the bile samples were determined by liquid scintillation counting. The 14 C profiles in rat bile samples were determined by LC radio-chromatography. Metabolite identification was performed by LC- MS/MS. LC/MS Conditions. The samples generated from HepatoPac incubations were analyzed on a 4000Qtrap (AB Sciex, Thornhill, ntario, Canada) attached to a Waters Acquity UPLC system (Milford, MA). The aqueous mobile phase (A) and organic mobile phase (B) consisted of 95:5 (v/v) water/acetonitrile and 95:5 (v/v) acetonitrile/water, respectively. Both mobile phases contained 0.1% acetic acid. Samples were eluted through an Acquity UPLC column BEH C µm (2.1 mm x 50 mm) following a 10 min gradient with 5% B to 38% B over 1 min at a flow 10

11 rate of 0.5 ml/min, then to 51.5% B over 9 min at a 0.7 ml/min flow rate. Multiple reaction monitoring (MRM) analysis was performed in positive ionization mode to quantify levels of faldaprevir (m/z , retention time 9.7 min), M2a and M2b (m/z , retention time 5.2 min for M2a and 5.5 min for M2b), faldaprevir glucuronide (m/z , retention time 5.8 min) and internal standard d 7 -faldaprevir (m/z , retention time 9.7 min) using authentic standards. The linear range of the metabolite curves was established to be between µm and 1 µm. The linear range of faldaprevir was determined to be between µm and 4 µm. Metabolite identification for suspended rat hepatocyte incubations was performed in positive ionization mode using a Sciex API 3000 (AB Sciex, Thornhill, ntario, Canada) coupled with Agilent 1100 pumps and autosampler (Agilent technologies, Santa Clara, CA). The aqueous and organic mobile phase consisted of 95:5 (v/v) water/acetonitrile and 95:5 (v/v) acetonitrile/water, respectively. The mobile phases contained 0.5% acetic acid. Samples were analyzed by elution through a phenomenex Synergi MAX-RP column (2.0 x 150 mm; 4 µm) using a 30-min gradient. Rat bile samples were analyzed in positive ionization mode using a 4000 Qtrap system coupled with LC-10AD pumps and an SIL-HTC autosampler (Shimadzu, orwell, MA) to obtain structural information of faldaprevir and metabolites. Samples were eluted through a phenomenex Gemini C 18 column (4.6 x 150 mm, 3 µm) over an 87-min gradient. 11

12 Data analysis Liver enrichment determined from rat HepatoPac. Faldaprevir concentrations were determined by LC-MS/MS in both medium and lysate samples. In order to calculate the intracellular concentration, a correction to exclude nonspecific binding of faldaprevir to fibroblasts was made. Fibroblast only plates were run during each experiment in the same manner as the HepatoPac plates. The faldaprevir concentrations in lysate samples from fibroblast only plates were measured as described for the HepatoPac plates and were multiplied by 0.75 to obtain the nonspecific binding of faldaprevir to fibroblasts in HepatoPac plates. This correction was made since each well in HepatoPac plates has 75% surface area as fibroblasts and 25% surface area as hepatocytes (Khetani et al., 2013). The hepatocyte specific concentration (pmol/ml) was calculated with the following equation. pmol Hepatocyte pmol HepatoPac lysate sample pmol Fibroblast lysate sample 0.75 ml ml ml This concentration was corrected by the hepatocyte intracellular volume ( ml for 5,000 hepatocytes) in order to calculate the intracellular concentration (pmol/ml). A hepatocyte volume of 6.48 pl/hepatocyte was used (personal communication with Dr. Kenneth Brouwer from Qualyst). This value is in-line with other literature values (Swift et al., 2010). pmol pmol hepatocyte 0.05 ml incubation volume ml Intracellular ml ml volume of 5,000 hepatocytes The enrichment value was calculated as below: pmol intracellular ml Enrichment factor pmol medium ml 12

13 Liver enrichment generated from suspended hepatocytes. The apparent concentration of faldaprevir in the liver was calculated using the following equation assuming that 1 g of liver occupies 1 ml of volume and using scaling factors of 1.2 x 10 8 hepatocytes per gram of liver (Houston et al.,1997). The enrichment value was then calculated by dividing the faldaprevir concentration in the liver by the concentration measured in the medium. Apparent conc. in liver nmole faldaprevir in pellet per.. 10 hepatoytes... In vitro formation rates of metabolites scaled to total liver based on rat HepatoPac data. In vitro formation rates of the faldaprevir oxidative metabolites M2a and M2b as well as faldaprevir glucuronide were scaled to total liver using the following equation and scaling factors of 1.2 x 10 8 hepatocytes per gram of liver (Houston et al.,1997) and 10 g of liver per rat (Davies et al.,1993). º»º»»º formation rate Total pmol formed ºº»º»»º per well Time point min hepatocytes in 10 g liver 5000 hepatocytes per well In vivo formation rates of M2a, M2b and faldaprevir glucuronide in the rat. In vivo formation rates of M2a, M2b and faldaprevir glucuronide in the rat were scaled from bile duct cannulated rat data using 0-24 hr samples. The equation below was used to calculate the rate of formation (pmol/min/total liver). The average dose given to rats in the study was 2.47 mg (2840 pmol). º»º»º formation rate Dose 2840 pmol % dose recovered in bile for each metabolite 1440 min a % dose recovered in bile, displayed in Table 3 13

14 RESULTS Liver enrichment in rat HepatoPac. Concentrations of faldaprevir, 0.15 and 2.86 µm, were selected to cover estimated free C max and the total C max, respectively, of faldaprevir observed during the [ 14 C]ADME study in the rat (data on file, Boehringer Ingelheim Pharmaceuticals, Inc.). on-specific binding of faldaprevir to the fibroblast only plates represented less than 1% of the total concentration added to the incubation and was not time or concentration dependent. Enrichment values were calculated after correction for non-specific fibroblast binding. The level of enrichment appeared to reach equilibrium by 2 hr (Fig. 1). To calculate the average enrichment value, time-points prior to 2 hr were not included. In addition, time-points after 48 hr were not included, due to an apparent loss of hepatocyte function after 48 hr (reduction in extents of enrichment and increased variability between replicates). The average extent of enrichment in rat hepatocytes was 34-fold based on the data from 2 hr to 48 hr (Table 1). Liver enrichment in suspended rat hepatocytes. A single concentration of faldaprevir (2 µm) was added to suspended rat hepatocytes in hepatocyte medium containing 50% v/v of rat plasma in order to estimate liver enrichment. Both medium and hepatocyte samples were collected after 16 min and analyzed for faldaprevir levels. The 16 min incubation time was selected based on an initial time course evaluation. The apparent concentration of faldaprevir in liver was estimated to be 5.6 nmol/g after correction of hepatocyte volume and assuming that 1 g of liver is equivalent to 1 ml of volume. The liver enrichment calculated from the ratio of the intracellular concentration over the measured concentration of faldaprevir in the medium (2.0 µm) was determined to be 2.8 (Table 1). 14

15 Liver enrichment determined from the QWBA study in the rat. Liver tissue and plasma concentrations were determined by whole-body autoradiography at specified times after a single oral administration of [ 14 C]faldaprevir (10 mg/kg) to male Long Evans rats. Liver enrichment was lower at earlier time-points and appeared to reach equilibrium by 6 hr (Table 1). The highest liver: plasma ratio observed was 26.8-fold at 6 hr. Metabolite profiling in in vitro incubations with suspended rat hepatocytes and in vivo rat bile. Two metabolites were detected in in vitro incubations of faldaprevir with suspended rat hepatocytes and six metabolites were identified in rat bile after oral dosing. The structures of the metabolites were proposed based on their molecular ions and product ion spectra (Table 2). The fragmentation patterns and characteristic ions of each metabolite are also included in Table 2. The levels of metabolites in the bile were determined by LC-radiochromatogram and are listed in Table 3. nly the data from the 0-24 hr period are shown because 91% of the total absorbed radioactivity was excreted into bile over the first 24 hr (equivalent to 31.2% of dose) and only 9% of the total absorbed radioactivity was excreted into bile from 24 hr to 48 hr (equivalent to 3.1% of dose). Faldaprevir glucuronide was the major metabolite in rat bile (approximately 38% of drug related materials in bile) while M2a and M2b were minor components (2.2% combined of drug related materials in bile). The levels of all other metabolites were also low (14% combined between 10 metabolites, of which only 3 were able to be positively identified and each was below 5% of drug related materials in bile). In suspended hepatocytes, M2a, M2b or faldaprevir glucuronide were not detected (Table 2). nly two amide hydrolysis metabolites were found (<3% of the parent based on peak area estimation after 4 hr-incubation). 15

16 Formation of M2a, M2b and faldaprevir glucuronide by rat HepatoPac. Authentic standards of faldaprevir glucuronide and two abundant human metabolites, M2a and M2b (Chen et al., 2014) were synthesized. The levels of M2a, M2b and faldaprevir glucuronide formed were quantified in medium and lysate samples from rat HepatoPac incubated with faldaprevir at 0.15 µm and 2.86 µm. The lower concentration of faldaprevir studied (0.15 µm) was too low for discernible formation of M2a and M2b. However, there was linear formation of both metabolites at 2.86 µm faldaprevir up to 48 hr (Fig. 2A). Linear formation occurred for faldaprevir glucuronide at both concentrations up to 24 hr (Fig. 2B) with a 21-fold difference in formation rates reflecting the difference in incubation concentrations of faldaprevir (19-fold) suggesting that these incubation concentrations are in the linear portion of the Michaelis-Menten curve. Comparison of in vitro and in vivo formation rates of faldaprevir metabolites. The formation rates of M2a, M2b and faldaprevir glucuronide with rat HepatoPac, at a substrate concentration of 2.86 µm, were scaled up to the whole liver (Table 4). Formation rates were also calculated based on metabolite levels collected in bile for 24 hr post-dose (Table 4). Scaling the bile data assumed that the formation and excretion of the metabolites into the bile were linear processes over the collection period. Both in vitro and in vivo, glucuronidation was the predominant clearance pathway and formation of M2a and M2b was minor. The ratios of the formation of faldaprevir glucuronide compared to the total formation of M2a and M2b were calculated and were comparable for HepatoPac (26-fold) and in vivo (17-fold) (Table 4). 16

17 DISCUSSI In vitro systems such as hepatocytes are a valuable tool to prospectively offer insights into the disposition of a drug prior to administering to humans. These systems can also be used retrospectively to provide a mechanistic understanding of in vivo phenomena. As an integrated metabolic system providing a full complement of phase I and II drug metabolizing enzyme capabilities, hepatocytes are becoming a primary tool in the pharmaceutical industry. This increasing reliance on hepatocytes, in comparison to liver microsomes, has also been promoted by a shift from a predominant role of CYP450 as a clearance mechanism for drugs, to an increasing importance of other DMEs such as aldehyde oxidase (Pryde et al., 2010; Hutzler et al., 2013) and UGT (Miners et al., 2010). This has partly come about as medicinal chemists develop structure-activity relationships (SAR) of new chemical entities (CE) to minimize metabolism (Hutzler et al., 2013). As models are developed to better predict and/or explain the disposition of a drug and the potential for DDIs, there has been an increasing awareness of the integrated role of DMEs and drug transporters. Thus, when using in vitro hepatocyte models, it is important to adopt an hepatocyte model that has fully functional DMEs as well as transporters, for accurate reproduction of in vivo parameters. Suspension cultures provide realistic levels and hence activities of DMEs for a limited incubation time, typically less than 4 hr, and are useful for determining metabolic clearance (Gomez-Lechon et al., 2008). While uptake transporters are functional in suspended hepatocytes, efflux transporters are internalized (Bow et al., 2008) and therefore this model is not capable of capturing the interplay between uptake and efflux. Sandwich cultured hepatocytes offer a viable system to assess biliary clearance including the 17

18 activity of efflux transporters but have diminished DMEs (Swift et al., 2010) and uptake transporters (Kotani et al., 2011; oel et al., 2013). An important aspect for modeling in vitro to in vivo metabolism and clearance is using the appropriate concentration of substrate. The recent regulatory DDI guidances summarize some of these considerations, namely, C max vs portal vein vs tissue concentrations (EMA, 2012; FDA, 2012). The role of drug transporters, particularly uptake transporters, in altering intracellular free drug concentrations, has been extensively considered (agar and Korzekwa, 2012; Chu et al., 2013) along with free concentrations corrected for binding to the test system (bach, 1999). Since the liver is the target organ for HCV therapy, there was an initial interest in considering whether liver concentrations of faldaprevir would provide a more accurate assessment of efficacy. In an in vivo study in rats, liver enrichment was approximately 42-fold (Table 1), a value higher than that observed in a rat QWBA study which reached a maximum liver enrichment value of 27-fold. Results from the QWBA study reflect total faldaprevir-related radioactivity and do not distinguish between parent and metabolites. A substantially lower value for liver enrichment (2.8-fold) was obtained with suspended hepatocytes (Table 1). It is likely that there were other factors contributing to liver enrichment that could not be captured with suspended hepatocytes (Duan et al., 2012a). In contrast, liver enrichment values achieved with rat HepatoPac were consistent with QWBA and in vivo data, ranging from approximately 13- to 51-fold. A steady state level of approximately 34-fold appeared to be reached by 2 hr of incubation (Fig. 1). It is expected that distribution of faldaprevir into rat liver is also a dynamic process, therefore the average enrichment value in rat HepatoPac may be reflective of average enrichment expected to occur in the rat at steady state. This appears to be consistent with observations from the QWBA study where maximal liver concentrations were reached at later 18

19 time points. Since there is an agreement in liver enrichment between in vitro rat HepatoPac data and in vivo rat data, liver enrichment generated from the human HepatoPac model has been included in the modeling of faldaprevir metabolism in humans (Li et al., 2014; Ramsden et al., 2014). A potential limitation of the current study may be that free concentrations in the liver were not calculated. Due to technical challenges in directly measuring unbound intracellular concentrations, it is uncertain whether the liver enrichment represents only changes in unbound concentrations or could also include differences in tissue binding. Further modeling may be able to address this issue. In general, understanding the relative contribution of glucuronidation to the overall metabolism has become increasingly important due to an increased role for UGT in the clearance of CEs (Miners et al., 2010) as well as requests from regulatory agencies to provide these data to help in assessing the potential for DDIs (FDA, 2012). Hepatocytes are an integrated model useful for the evaluation of Phase I and Phase II metabolism and generally provide better prediction for the extent of glucuronidation compared to microsomes (Soars et al., 2002; Engtrakul et al., 2005). However, in some cases, significant under-prediction has still been observed (Miners et al., 2006; Miners et al., 2010; Foster et al., 2009). A full profile of metabolites generated from bile samples (biliary excretion represented almost 100% of total absorbed radioactivity) as part of the rat [ 14 C]ADME studies showed that metabolism occurs primarily through glucuronidation in the rat (approximately 70% of total metabolites) with other metabolic pathways, including hydroxylation, amide hydrolysis, and conjugation with glucose, being minor (<5% each) (Table 2 and 3). The two predominant human metabolites, M2a and M2b, which are monohydroxylated products (Chen et al., 2014), constituted only 2.3 and 1.8% of the metabolites, respectively (Table 3). Initial metabolite 19

20 profiling was performed in vitro in suspended hepatocytes. Due to very low in vitro turnover within the 4-hr incubation period, the formation of M2a, M2b, and faldaprevir glucuronide was not detected (Table 2) and only two other minor amide hydrolysis metabolites were observed. These studies indicated that suspended hepatocytes were not an appropriate cell system for characterizing metabolite formation for faldaprevir. Metabolite formation studies using rat HepatoPac mainly focused on the three metabolites potentially relevant in human: faldaprevir glucuronide, M2a and M2b. The results showed that glucuronidation was 26-fold faster than the total formation of M2a and M2b in the rat HepatoPac model, a result similar to that observed in vivo (17-fold; Table 4). In addition, glucuronidation is the predominant metabolic pathway (94-96%) compared to the formation of M2a and M2b ( %) in both the in vitro rat HepatoPac study and the in vivo rat study. Thus, rat HepatoPac provided a metabolite profile similar to the in vivo results from the bile duct cannulated rats. verall, our studies suggest that HepatoPac is a promising in vitro model, able to predict in vivo liver enrichment and metabolism, especially for glucuronidation, and has demonstrated superiority over suspended hepatocytes. Assessing the relative contribution of glucuronidation can be particularly challenging in humans due to the challenges of bile sampling and the fact that quantitative analysis of individual metabolic pathways can be complicated by cleavage of glucuronide conjugates in feces by gut bacteria (Sousa et al., 2008). Human HepatoPac may be a promising model to predict the fraction of glucuronidation to the overall metabolism in human. The accurate representation of in vivo metabolic profiles (especially glucuronidation) and liver enrichment by the HepatoPac model, together with the work of others (Wang et al., 2010; Chan et al., 2013), has helped to validate an hepatocyte model through which the metabolic pathways of faldaprevir in humans can be proposed (Ramsden et al., 2014). This has been particularly 20

21 important considering the observation that significant metabolism of faldaprevir to M2a and M2b had occurred in humans despite the fact that these two metabolites were not detected in the systemic circulation (Chen et al., 2014). 21

22 ACKWLEDGMETS The authors would like to thank Gordon Bolger and Jianmin Duan for performing the in vitro Kp assessment. We would also like to thank Dr. Timothy S. Tracy for scientific advice. 22

23 AUTHRSHIP CTRIBUTIS Participated in research design: Ramsden, Tweedie, Chen, St. George, and Li Conducted experiments: Ramsden Performed data analysis: Ramsden and Li Wrote or contributed to the writing of the manuscript: Ramsden, Tweedie, and Li 23

24 References Bow DA, Perry JL, Miller DS, Pritchard JB, and Brouwer KL (2008) Localization of P-gp (Abcb1) and Mrp2 (Abcc2) in freshly isolated rat hepatocytes. Drug Metab Dispos 36: Chan TS, Yu H, Moore A, Khetani S, and Tweedie DJ (2013) Meeting the Challenge of Predicting Hepatic Clearance of Compounds Slowly Metabolized by Cytochrome P450 Using a ovel Hepatocyte Model, HepatoPacTM. Drug Metab Dispos 41: Chen LZ, Rose P, Mao Y, Yong CL, St. George R, Huang F, Latli B, Mandarino D, and Li Y (2014) Steady-State Disposition of the Hepatitis C S3/S4 Protease Inhibitor Faldaprevir in Healthy Male Subjects. Antimicrob Agents Chemother, in press. Chu X, Korzekwa K, Elsby R, Fenner K, Galetin A, Lai Y, Matsson P, Moss A, agar S, Rosania GR, Bai JP, Polli JW, Sugiyama Y, and Brouwer KL (2013) Intracellular drug concentrations and transporters: measurement, modeling, and implications for the liver. Clin Pharmacol Ther 94: Davies B and Morris T (1993) Physiological parameters in laboratory animals and humans. Pharm Res 10: Duan J, Bolger G, Garneau M, Amad M, Batonga J, Montpetit H, tis F, Jutras M, Lapeyre, Rheaume M, Kukolj G, White PW, Bethell RC, and Cordingley MG (2012a) The liver partition coefficient-corrected inhibitory quotient and the pharmacokineticpharmacodynamic relationship of directly acting anti-hepatitis C virus agents in humans. Antimicrob Agents Chemother 56: Duan J, Yong CL, Garneau M, Amad M, Bolger G, De MJ, Montpetit H, tis F, Jutras M, Rheaume M, White PW, Llinas-Brunet M, Bethell RC, and Cordingley MG (2012b) 24

25 Cross-species absorption, metabolism, distribution and pharmacokinetics of BI , a potent HCV genotype 1 S3/4A protease inhibitor. Xenobiotica 42: EMA (2012) Committee for Human Medicinal Products (CHMP): guideline on the investigation of drug interactions. [ WC pdf (Accessed on Dec. 16, 2013)] Engtrakul JJ, Foti RS, Strelevitz TJ, and Fisher MB (2005) Altered AZT (3'-azido-3'- deoxythymidine) glucuronidation kinetics in liver microsomes as an explanation for underprediction of in vivo clearance: comparison to hepatocytes and effect of incubation environment. Drug Metab Dispos 33: FDA (2012) Guidance for industry: drug interaction studies - study design, data analysis, implications for dosing, and labeling recommendations (Draft). [ ances/ucm pdf (Accessed on Dec. 16, 2013)] Foster JA, Hallifax D, and Houston JB (2009) How robust is the prediction of in vivo metabolic clearance from established in vitro systems? Drug Metab Rev 41(Suppl 1):26. Gomez-Lechon MJ, Castell JV, and Donato MT (2008) An update on metabolism studies using human hepatocytes in primary culture. Expert pin Drug Metab Toxicol 4: Houle R, Raoul J, Levesque JF, Pang KS, icoll-griffith DA, and Silva JM (2003) Retention of transporter activities in cryopreserved, isolated rat hepatocytes. Drug Metab Dispos 31: Houston JB and Carlile DJ (1997) Prediction of hepatic clearance from microsomes, hepatocytes, and liver slices. Drug Metab Rev 29:

26 Hutzler JM, bach RS, Dalvie D, and Zientek MA (2013) Strategies for a comprehensive understanding of metabolism by aldehyde oxidase. Expert pin Drug Metab Toxicol 9: Khetani SR and Bhatia S (2008) Microscale culture of human liver cells for drug development. at Biotechnol 26: Khetani SR, Kanchagar C, Ukairo, Krzyzewski S, Moore A, Shi J, Aoyama S, Aleo M, and Will Y (2013) Use of micropatterned cocultures to detect compounds that cause druginduced liver injury in humans. Toxicol Sci 132: Korzekwa KR, agar S, Tucker J, Weiskircher EA, Bhoopathy S, and Hidalgo IJ (2012) Models to predict unbound intracellular drug concentrations in the presence of transporters. Drug Metab Dispos 40: Kotani, Maeda K, Watanabe T, Hiramatsu M, Gong LK, Bi YA, Takezawa T, Kusuhara H, and Sugiyama Y (2011) Culture period-dependent changes in the uptake of transporter substrates in sandwich-cultured rat and human hepatocytes. Drug Metab Dispos 39: Lecluyse EL, Audus KL, and Hochman JH (1994) Formation of extensive canalicular networks by rat hepatocytes cultured in collagen-sandwich configuration. Am J Physiol 266:C1764-C1774. Li Y, Zhou J, Ramsden D, Taub ME, 'Brien D, Busacca C, Gonella, and Tweedie DJ (2014) Enzyme-Transporter Interplay: Effect on the Formation and Clearance of Abundant Metabolites of Faldaprevir Found in Excreta but not in Circulation. Drug Metab Dispos, doi: /dmd

27 Miners J, Knights KM, Houston JB, and Mackenzie PI (2006) In vitro-in vivo correlation for drugs and other compounds eliminated by glucuronidation in humans: pitfalls and promises. Biochem Pharmacol 71: Miners J, Mackenzie PI, and Knights KM (2010) The prediction of drug-glucuronidation parameters in humans: UDP-glucuronosyltransferase enzyme-selective substrate and inhibitor probes for reaction phenotyping and in vitro-in vivo extrapolation of drug clearance and drug-drug interaction potential. Drug Metab Rev 42: agar S and Korzekwa K (2012) Commentary: nonspecific protein binding versus membrane partitioning: it is not just semantics. Drug Metab Dispos 40: oel G, Le VM, Moreau A, Stieger B, Parmentier Y, and Fardel (2013) Functional expression and regulation of drug transporters in monolayer- and sandwich-cultured mouse hepatocytes. Eur J Pharm Sci 49: bach RS (1999) Prediction of human clearance of twenty-nine drugs from hepatic microsomal intrinsic clearance data: An examination of in vitro half-life approach and nonspecific binding to microsomes. Drug Metab Dispos 27: Pryde DC, Dalvie D, Hu Q, Jones P, bach RS, and Tran TD (2010) Aldehyde oxidase: an enzyme of emerging importance in drug discovery. J Med Chem 53: Ramsden D, Tweedie DJ, Chan T, Taub M, and Li Y (2014) Bridging in vitro and in vivo metabolism and transport of faldaprevir in human using a novel co-cultured human hepatocyte system, HepatoPac TM. Drug Metab Dispos, submitted. Soars MG, Burchell B, and Riley RJ (2002) In vitro analysis of human drug glucuronidation and prediction of in vivo metabolic clearance. J Pharmacol Exp Ther 301:

28 Sousa T, Paterson R, Moore V, Carlsson A, Abrahamsson B, and Basit AW (2008) The gastrointestinal microbiota as a site for the biotransformation of drugs. Int J Pharm 363:1-25. Sulkowski MS, Asselah T, Lalezari J, Ferenci P, Fainboim H, Leggett B, Bessone F, Mauss S, Heo J, Datsenko Y, Stern J, Kukolj G, Scherer J, ehmiz G, Steinmann GG, and Bocher W (2013a) Faldaprevir combined with pegylated interferon alfa-2a and ribavirin in treatment-naive patients with chronic genotype 1 HCV: SILE-C1 trial. Hepatology 57: Sulkowski MS, Bourliere M, Bronowicki JP, Asselah T, Pawlotsky JM, Shafran SD, Pol S, Mauss S, Larrey D, Datsenko Y, Stern J, Kukolj G, Scherer J, ehmiz G, Steinmann GG, and Bocher W (2013b) Faldaprevir combined with peginterferon alfa-2a and ribavirin in chronic hepatitis C virus genotype-1 patients with prior nonresponse: SILE- C2 trial. Hepatology 57: Swift B, Pfeifer D and Brouwer KL (2010) Sandwich-cultured hepatocytes: an in vitro model to evaluate hepatobiliary transporter-based drug interactions and hepatotoxicity. Drug Metab Rev 42: Ukairo, Kanchagar C, Moore A, Shi J, Gaffney J, Aoyama S, Rose K, Krzyzewski S, McGeehan J, Andersen ME, Khetani SR, and Lecluyse EL (2013) Long-term stability of primary rat hepatocytes in micropatterned cocultures. J Biochem Mol Toxicol 27: Wang WW, Khetani SR, Krzyzewski S, Duignan DB, and bach RS (2010) Assessment of a micropatterned hepatocyte coculture system to generate major human excretory and circulating drug metabolites. Drug Metab Dispos 38:

29 White PW, Llinas-Brunet M, Amad M, Bethell RC, Bolger G, Cordingley MG, Duan J, Garneau M, Lagace L, Thibeault D, and Kukolj G (2010) Preclinical characterization of BI , a C-terminal carboxylic acid inhibitor of the hepatitis C virus S3-S4A protease. Antimicrob Agents Chemother 54:

30 Footnote: This research was funded by Boehringer Ingelheim Pharmaceuticals, Inc. 30

31 FIGURE LEGEDS Figure 1: Faldaprevir enrichment into rat hepatocytes determined as ratios of intracellular concentrations to medium concentrations (n=3) or (n=6 at 8, 12 and 24 hrs). Figure 2: Formation of M2a, M2b at 2.86 µm faldaprevir (Panel A) and formation of faldaprevir glucuronide at two substrate concentrations (0.15 µm or 2.86 µm faldaprevir) (Panel B) (n=6) from rat HepatoPac over time. The fold differences between substrate concentrations and formation amounts are highlighted in Panel B. 31

32 Table 1: Comparison of liver enrichment values generated from in vivo rat models and in vitro test systems»ºº»»»¹ ºº ºº»¹» º» º¹º¹»º¹»»¹»»ºº»»»¹ ºº ºº¹»º¹ º» º¹¹º»º a (White et al., 2010) Enrichment values of faldaprevir In vivo rat study a 42.2 QWBA study 7.55 (1hr) 26.8 (6 hr) 22.8 (12 hr) 20.1 (24 hr) Suspended rat hepatocytes 2.8 Rat HepatoPac 13.3 ± 4.8 (1 hr) 34.0 ± 12.0 (after equilibrium was reached) 32

33 Table 2: Metabolite profile of faldaprevir using suspended rat hepatocytes compared with rat bile Parent and metabolites a (molecular ion) Faldaprevir ([M+H] + : 869) Chemical Structure Br H S H Elemental composition Characteristic fragment ions Suspended rat hepatocytes C 40 H BrS 448, 422, 380, Recovered in bile from BDC rats H H Glucuronide of amide hydrolysis product ([M+H] + :975) Hydroxylated metabolites (M2a and M2b) b ([M+H] + :885) , 448(624-glu) Br Br H H H 2 S H S H H Glu 422 H H C 42 H BrS 799, 624, 448, 380, 352, 223 C 40 H BrS 464, 422, 380, 223 ot observed ot observed bserved bserved 223 H

34 Table 2 cont: Metabolite profile of faldaprevir using suspended rat hepatocytes compared with rat bile Parent and metabolites a (molecular ion) Dealkylation product ([M+H] + : 759) Chemical Structure Br H S H Elemental composition Characteristic ions Suspended rat hepatocytes Recovered in bile from BDC rats C 34 H BrS 422, 338 bserved ot observed H 2 Amide hydrolysis product ([M+H] + : 799) Faldaprevir glucuronide ([M+H] + :1045) Acyl glycoside ([M+H] + :1031) 338 Br 448 Br Br H H H S H S H H H S H Glu H H C 36 H BrS 448, 352, 223 bserved bserved + H 624, 448=(624-glu) + H 610, 448=(610-glc) C 46 H BrS 869, 624, 448, 422, 380, 223 C 46 H BrS 869, 610, 448, 422, 380, 223 ot observed ot observed bserved bserved H Glc 223 H a. b 380 The metabolites are listed according to their respective order of elution from the HPLC column Correspond to M2a and M2b identified in the human [ 14 C] ADME study (Chen et al., 2014) 34

35 Table 3: Metabolite levels in bile from bile duct cannulated rats Analyte % of total radioactivity in bile % of dose recovered in bile a (0-24 hr) (0-24 hr) M2a M2b Faldaprevir glucuronide ther metabolites b [ 14 C]faldaprevir a. 31.2% of dose recovered in bile within 24 hr. Percentage of dose recovered in bile = % of total radioactivity in bile x b. 31.2% of dose The levels of metabolites other than M2a, M2b, and faldaprevir glucuronide were summed up. 35

36 Table 4: Comparison of the extent of the formation of M2a, M2b, faldaprevir glucuronide from in vitro rat HepatoPac study and in vivo rat study Metabolite Formation rates (pmol/min/total liver) In vitro rat HepatoPac data In vivo rat data % of the formation of M2a, M2b and faldaprevir glucuronide In vitro rat HepatoPac data In vivo rat data M2a M2b Faldaprevir glucuronide Fold of formation rates (glucuronide vs. M2a+M2b)

37 Enrichment value (lysate/medium) M M hr 1 hr 2 hr 4 hr 8 hr 12 hr 24 hr 48 hr Incubation time (hrs) Figure 1

38 Formation of faldaprevir glucuronide (pmol/ml) Formation of M2a or M2b (pmol/ml) A Time (hrs) B M2a M2b M M Time (hrs) Figure 2