ABSTRACT
Tenofovir alafenamide (TAF) is a prodrug of tenofovir (TFV) currently in clinical evaluation for treatment for HIV and hepatitis B virus (HBV) infections. Since the target tissue for HBV is the liver, the hepatic delivery and metabolism of TAF in primary human hepatocytes in vitro and in dogs in vivo were evaluated here. Incubation of primary human hepatocytes with TAF resulted in high levels of the pharmacologically active metabolite tenofovir diphosphate (TFV-DP), which persisted in the cell with a half-life of >24 h. In addition to passive permeability, studies of transfected cell lines suggest that the hepatic uptake of TAF is also facilitated by the organic anion-transporting polypeptides 1B1 and 1B3 (OATP1B1 and OATP1B3, respectively). In order to inhibit HBV reverse transcriptase, TAF must be converted to the pharmacologically active form, TFV-DP. While cathepsin A is known to be the major enzyme hydrolyzing TAF in cells targeted by HIV, including lymphocytes and macrophages, TAF was primarily hydrolyzed by carboxylesterase 1 (CES1) in primary human hepatocytes, with cathepsin A making a small contribution. Following oral administration of TAF to dogs for 7 days, TAF was rapidly absorbed. The appearance of the major metabolite TFV in plasma was accompanied by a rapid decline in circulating TAF. Consistent with the in vitro data, high and persistent levels of TFV-DP were observed in dog livers. Notably, higher liver TFV-DP levels were observed after administration of TAF than those given TDF. These results support the clinical testing of once-daily low-dose TAF for the treatment of HBV infection.
INTRODUCTION
Chronic hepatitis B (CHB) is a major global health problem, and the World Health Organization (WHO) estimates that ∼240 million people worldwide are chronically infected by hepatitis B virus (HBV) (1) (http://www.who.int/mediacentre/factsheets/fs204/en/). The small-molecule anti-HBV agents currently approved by U.S. Food and Drug Administration (FDA) are all nucleoside/nucleotide analogs targeting HBV reverse transcriptase. Tenofovir disoproxil fumarate (TDF), a prodrug of tenofovir (TFV), is a current first-line treatment for CHB (2) and has demonstrated efficacy in both hepatitis B e antigen (HBeAg)-positive and HBeAg-negative patients (3–5). TDF is also widely used as the backbone of current anti-HIV combination regimens (6, 7). For both HIV and HBV, the mechanism of action of TDF is intracellular metabolism to its pharmacologically active form, tenofovir diphosphate (TFV-DP), followed by competition with endogenous 2′-dATP for incorporation by the viral reverse transcriptase and subsequent chain termination of viral DNA replication (8, 9). TFV-DP is an effective inhibitor of HBV reverse transcriptase, with an inhibitor constant (Ki) of 0.18 μM in vitro (9).
A new prodrug of tenofovir, tenofovir alafenamide (TAF or GS-7340), was selected to more efficiently load HIV target cells while lowering circulating levels of TFV, resulting in reduced off-target exposure (10–13). The administration of 25 mg of TAF in HIV-infected patients resulted in significantly lower plasma TFV levels (∼90%) and higher TFV-DP levels in peripheral blood mononuclear cells (PBMCs) (>5-fold) than in those patients administered 300 mg of TDF (14–16). Consistent with the reduction in off-target TFV exposures with TAF, TAF has demonstrated improvement relative to TDF in clinical studies of bone mineral density at the spine and hip, as measured by dual-energy X-ray absorptiometry and in multiple renal function parameters, including estimated glomerular filtration, fractional excretion of phosphate, proteinuria, and glycosuria (17). TAF is in phase 3 clinical studies for HBV, and in the phase 1b HBV study, similar viral declines and reduced TFV exposures of >90% were observed with doses of ≤25 mg of TAF relative to 300 mg of TDF (16).
A previous pharmacokinetic study in dogs demonstrated that TAF was highly extracted in the liver, with a hepatic extraction ratio of ∼65%, suggesting that TAF was efficiently taken up by the liver (18). In the present study, we sought to better understand the potential of TAF as a therapy for HBV by evaluating TAF metabolism in primary human hepatocytes, interaction with hepatic uptake transporters, and in vivo plasma and liver pharmacokinetics in dogs.
MATERIALS AND METHODS
Materials.TAF, TDF, TFV, TFV-DP, telaprevir, boceprevir, and cobicistat were synthesized at Gilead Sciences, Inc. (Foster City, CA). Bis(4-nitrophenyl)phosphate (BNPP) was purchased from Sigma (St. Louis, MO). All other chemicals were of the highest grade available and purchased from Sigma.
Primary human hepatocytes.Primary human hepatocytes were purchased from Life Technologies (Grand Island, NY) in 12-well tissue culture plates seeded at confluence (0.88 × 106 cells/well). The cells were maintained in fresh Williams' medium E overnight in a 37°C incubator under a humid atmosphere of 95% air–5% CO2 (vol/vol) to a confluence of about 1.0 × 106 cells/well. TFV, TDF, or TAF was continuously incubated at 5 μM for 24 h. Primary human hepatocytes were also treated with 5 μM TAF for 2 h, and TAF was removed from the medium by replacing the TAF-containing medium with fresh medium. At selected time points, the cells were washed twice with 2.0 ml of an ice-cold saline (0.9% sodium chloride) solution. The cells were then scraped into 0.5 ml of 70% methanol containing 100 nM 2-chloro-adenosine-5′-triphosphate (Sigma-Aldrich, St. Louis, MO) as an internal standard. For the drug-drug interaction studies, the duplicate wells were incubated with 0.5 μM TAF and various concentrations (0, 0.08, 0.4, 2, 10, and 50 μM) of the inhibitor(s) dissolved in Williams' medium E. After 24 h of continuous incubation, the cells were harvested and extracted as described above. All the samples were stored overnight at −20°C to facilitate nucleotide extraction and centrifuged at 15,000 × g for 15 min, and the supernatant was transferred to clean tubes for drying in a MiVac Duo concentrator (Genevac, Gardiner, NY). The dried samples were then reconstituted in mobile phase A for analysis by liquid chromatography-tandem mass spectrometry (LC-MS/MS). Intracellular concentrations of TFV and its phosphorylated metabolites were determined using a seven-point standard curve prepared in blank matrices spanning 3 orders of magnitude, with quality control samples analyzed at the beginning and end of each sample set to ensure accuracy and precision.
OATP1B1 and OATP1B3 substrate uptake assay.Chinese hamster ovary (CHO) cells, either wild type or transfected with the genes encoding human organic anion-transporting polypeptide 1B1 or 1B3 (OATP1B1 or OATP1B3, respectively), were maintained in Dulbecco's modified Eagle's medium (DMEM) containing 1,000 mg/liter d-glucose, l-glutamine, 25 mM HEPES buffer, and 110 mg/liter sodium pyruvate, 1% penicillin-streptomycin (pen-strep), 10% fetal bovine serum, 0.05 mg/ml l-proline, and 0.5 mg/ml Geneticin G-418. The cells were maintained in incubators set at 37°C, 90% humidity, and 5% CO2. On the day before the assay, 10 mM sodium butyrate was added to the transfected cells to increase the protein expression level, and cells were grown to confluence overnight. The assay buffer contained 142 mM NaCl, 5 mM KCl, 1 mM KH2PO4, 1.2 mM MgSO4, 1.5 mM CaCl2, 5 mM glucose, and 12.5 mM HEPES (pH 7.4). After removal of the medium, the cells were washed with 1× phosphate-buffered saline (PBS) and treated with 5 ml of 0.05% trypsin. The cells were then resuspended with assay buffer at 37°C and 3 × 106 cells/ml, and 100 μl of the cell suspension was aliquoted to each well on a 48-well plate for a 30-min preincubation at 37°C. TAF, atorvastatin (positive control), and antipyrine (passive permeability control) with or without the transport inhibitor rifampin were diluted to 3-fold of the final target concentration in assay buffer and equilibrated at 37°C for 30 min. The final compound concentrations for TAF, atorvastatin, antipyrine, and rifampin were 10, 0.1, 10, and 40 μM, respectively. The assay was started by adding 50 μl of test compound solution to the 48-well plate containing the cell suspension, mixing, and incubating at 37°C for 1 min. The respective dilution factors took the cell density to 2 × 106 cells/ml and the compound test concentration to the targeted level. The entire reaction mixture was overlaid onto preprepared microcentrifuge tubes containing 100 μl of aqueous solution (bottom layer) and 100 μl of filtration oil (middle layer; 74.5%/25.5% silicon oil-to-mineral oil mix) and then centrifuged immediately at 13,000 × g for 30 s. The samples were placed in a freezer overnight at −80°C. Using a tubing cutter, the microcentrifuge tubes were cut in the middle, and the bottom layer was collected in an Eppendorf tube. Samples were extracted with organic solvents for analysis by LC-MS/MS.
Dog pharmacokinetics.Male beagle dogs (Canis lupus familiaris) were used for the in vivo portion of this study. The animals were housed in accordance with the standards of the American Association for Accreditation of Laboratory Animal Care and were receiving a standard commercial diet. The animals were handled in strict accordance with the Guide for the Care and Use of Laboratory Animals (19). The protocol was reviewed by the Institutional Animal Care and Use Committee at MPI Research Laboratories (Mattawan, MI).
The beagle dogs weighed approximately 10 to 12 kg at the time of dosing. The oral dosing vehicle for TAF was 0.1% (wt/wt) hydroxypropyl methylcellulose K100LV and 0.1% polysorbate 20 in water, and that for TDF was 5% ethyl alcohol in water. Each compound was dosed in 4 animals. Following repeat oral dosing for 7 days, blood samples were collected via the jugular vein at 0.083, 0.25, 0.5, 1, 2, 4, 8, 12, and 24 h for TAF and 0.25, 0.5, 1, 2, 4, 6, 8, 12, and 24 h for TDF on days 1 and 7 postdose. Liver samples were collected from 2 animals per time point at 4 and 24 h on day 7 postdose.
Blood samples (approximately 1 ml) were collected into prechilled collection tubes containing K2-EDTA and processed to plasma. At each terminal collection, each animal was anesthetized under isoflurane and the livers were harvested. The collected livers were wrapped in aluminum foil and flash-frozen in liquid nitrogen immediately following removal.
For plasma analysis, an aliquot was prepared following protein precipitation. Protein was precipitated from the samples by adding acetonitrile to a final concentration of 70% in the presence of internal standard (100 nM 5-iodotubercidin). Samples were dried completely for approximately 30 min under a stream of nitrogen at 40°C and reconstituted with 0.2% formic acid in water to 3 times the original volume of plasma.
Liver samples were prepared by sectioning into smaller pieces and distributing into preweighed 15-ml conical tubes, which were kept on dry ice. The ice-cold extraction buffer (0.1% KOH and 67 mM EDTA in 70% methanol containing 0.5 μM chloro-ATP as the internal standard; ∼2 ml) was added to ∼0.5 g of each liver sample. The mixtures were promptly homogenized using an Omni Tip tissue homogenizing (TH) kit with disposable hard-tissue homogenizer probes (Omni International). Aliquots of the homogenate were filtered by using a 0.2-μm 96-well polypropylene filter plate (Varian Captiva). The filtrates were evaporated to dryness and reconstituted with an equal volume of 1 mM ammonium phosphate buffer (pH 7) prior to LC-MS/MS analysis. In order to determine liver intracellular concentrations, tissue weights were converted to intracellular volume using the ratio of liver weight to an intracellular liver volume of 1.4 g/ml (20).
In order to assess the sample integrity, intracellular liver concentrations of endogenous adenosine nucleotides were measured. For the analysis of AMP, ADP, and ATP, a calibration curve was constructed by adding 13C, 15N-labeled stable isotopes of AMP, ADP, and ATP (Cambridge Isotope Laboratories, Inc.) into the blank dog liver homogenates. The dog liver homogenates were diluted an additional 1,000-fold in ammonium phosphate buffer for analysis on a Sciex API-4000 LC-MS/MS instrument. Liver concentrations of TFV, TFV monophosphate (TFV-MP), TFV-DP, and adenosine nucleotides were determined using seven-point standard curves prepared in blank liver spanning 3 orders of magnitude with quality control samples to ensure accuracy and precision. The observed endogenous ATP levels were in the low to normal range of the values reported in the literature, suggesting possible low-level dephosphorylation during the sample collections (data not shown) (21).
RESULTS
In vitro metabolism of TAF in primary human hepatocytes.TAF requires metabolic activation to become its pharmacologically active metabolite, TFV-DP, which is an alternate substrate of the HBV polymerase. Therefore, the metabolism of TAF was assessed in primary human hepatocytes in comparison with that of TDF and the nucleotide analog TFV. Primary human hepatocytes were continuously incubated with 5 μM TFV, TDF, or TAF for 24 h, and the levels of TFV-DP were determined (Fig. 1A). Intracellular TFV-DP levels persistently increased over the 24-h period. Incubation with TFV, TDF, and TAF resulted in TFV-DP levels of 12.1, 302, and 1,470 pmol/one million cells, respectively, at 24 h. Based on these values, TAF produced approximately 120- and 5-fold more TFV-DP than TFV and TDF, respectively. In order to assess the persistence of TFV-DP in primary human hepatocytes, the cells were incubated with 5 μM TAF for 2 h, TAF was removed by exchanging the TAF-containing cell culture medium with fresh medium lacking compound, and intracellular metabolite concentrations were monitored for 24 h. As shown in Fig. 1B, the TFV-DP levels reached approximately 700 pmol/one million cells at 3 h and decreased only slightly over the 24-h time course. These results illustrate the apparent half-life (t1/2) of TFV-DP to be >24 h in primary human hepatocytes.
In vitro metabolism in primary human hepatocytes. (A) Formation of TFV-DP in primary human hepatocytes after continuous incubation with 5 μM TAF, TDF, or TFV. (B) Intracellular TFV-DP levels following 2-h incubation with 5 μM TAF, followed by removal of TAF from the cell culture medium by replacing with fresh medium. All the data from the duplicate determinations at each time point in primary human hepatocytes from a single donor were plotted, and the lines were drawn based on the mean values.
Effects of esterase, peptidase, and cytochrome P450 (CYP) inhibitors on TAF metabolism.To determine the enzymes involved in the activation of TAF in primary human hepatocytes, cells were incubated with TAF together with known cathepsin A inhibitors (approved hepatitis C virus [HCV] nonstructural 3 [NS3] protease inhibitors telaprevir and boceprevir), carboxylesterase 1 inhibitor (bis-p-nitrophenyl phosphate [BNPP]), CYP3A4, and P glycoprotein (P-gp) inhibitor (cobicistat), or the combination of telaprevir and BNPP (8, 22–24). Consistent with the result described above, incubation with 0.5 μM TAF resulted in approximately 350 pmol/one million cells of TFV-DP in the absence of any inhibitors (data not shown). BNPP inhibited the metabolism of TAF in a dose-dependent manner, with approximately 37, 30, and 66% inhibition observed at 2, 10, and 50 μM BNPP, respectively (Fig. 2). Little or no effect on TFV-DP formation was observed with telaprevir, boceprevir, or cobicistat. While telaprevir alone caused little effect, greater inhibition was seen when telaprevir was added to BNPP than for BNPP alone at higher concentrations. For example, at concentrations of 10 μM and 50 μM each inhibitor, the inhibition of TFV-DP formation of approximately 84 and 95%, respectively, was observed. These results suggest that CES1 is the predominant enzyme activating TAF in hepatocytes and that cathepsin A also makes a minor contribution.
Effects of esterase and CYP inhibitors on TAF metabolism in primary human hepatocytes. Primary human hepatocytes were continuously incubated with 0.5 μM TAF for 24 h in the presence of increasing concentrations of CatA inhibitors (telaprevir or boceprevir), CES1 inhibitor (BNPP), CYP3A inhibitor (cobicistat), or both CatA and CES1 inhibitors (telaprevir plus BNPP). The values are means ± standard deviations (SD) from 3 independent experiments in primary human hepatocytes from different donors in duplicate.
OATP1B1- and OATP1B3-dependent uptake.As described above, TAF was efficiently taken up and metabolized in primary human hepatocytes. We then studied if hepatic transporters are involved in TAF uptake. The substrate potential for hepatic uptake transporters was tested using untransfected (wild-type) Chinese hamster ovary (CHO-WT) cells or CHO cells transfected with OATP1B1 (SLCO1B1 [CHO-OATP1B1]) or OATP1B3 (SLCO1B3 [CHO-OATP1B3]). TAF was taken up by CHO-WT at a rate of 9.0 pmol/min/106 cells, indicating that TAF has high passive permeability (Fig. 3). TAF uptake was higher in transfected cells, with rates of 12.0 and 24.1 pmol/min/106 cells in CHO-OATP1B1 and CHO-OAT1B3, respectively (Fig. 3). The transporter-dependent uptake was confirmed by assessing the transport activity in the presence of 40 μM rifampin, a known inhibitor of OATP1B1 and OATP1B3. Rifampin completely inhibited the transporter-dependent uptake, and the uptake rates in CHO-WT, CHO-OATP1B1, and CHO-OATP1B3 were 6.0, 6.2, and 5.8 pmol/min/106 cells, respectively (Fig. 3). The positive control atorvastatin showed uptake rates of 4.6 and 5.5 pmol/min/106 cells in CHO-OATP1B1 and CHO-OATP1B3, respectively, while minimal uptake was observed in CHO-WT. As expected, 40 μM rifampin inhibited the transporter-dependent uptake rates of atorvastatin by 5.0- and 4.8-fold in CHO-OATP1B1 and CHO-OATP1B3, respectively. Antipyrine, a passive permeability control, had uptake rates in CHO-OATP1B1 and CHO-OATP1B3 (23 and 24 pmol/min/106 cells, respectively) similar to that in CHO-WT (23 pmol/min/106 cells) and was not affected by rifampin.
OATP1B1- and OATP1B3-mediated uptake of TAF. The uptake rates of TAF, atorvastatin (positive control), and antipyrine (passive transport control) were measured in untransfected CHO cells (CHO-WT) or CHO cells expressing OATP1B1 (CHO-OATP1B1) or OATP1B3 (CHO-OATP1B3) in the absence or presence of the OATP inhibitor rifampin. The values are means ± SD of triplicate determinations.
Seven-day repeat-dose plasma and liver pharmacokinetics.TAF and TDF were dosed separately and given orally at 8.3 and 9.0 mg/kg of body weight (5 mg TFV equivalents/kg), respectively, as an aqueous solution to male beagle dogs. The plasma levels of TAF, TDF, and the major metabolite TFV were determined on days 1 and 7, and the results are summarized in Fig. 4 and Table 1. Following oral administration of TAF, the drug was rapidly absorbed and achieved maximum concentration of drug in plasma (Cmax) values of 4.51 and 2.36 μM on days 1 and 7, respectively, at the first sample collected at 0.17 h postdose. The apparent terminal elimination half-lives of TAF were 0.30 h and 0.28 h on days 1 and 7, respectively. The formation of TFV accompanied the rapid decline in plasma TAF. TFV had Cmax values of 1.47 and 2.12 μM at 0.75 h postdose on days 1 and 7, respectively, and persisted with a t1/2 of >12 h. The area under the concentration-time curve from time zero to time t (AUC0–t) values for TAF and TFV were 1.85 and 6.81 μM · h on day 1 and 0.77 and 11.8 μM · h on day 7, respectively (Fig. 4A and Table 1). TDF was below the limit of detection at all time points on both days 1 and 7 in dog plasma. The AUC0–t values for TFV following administration of TDF were 2.24 and 4.84 μM · h on days 1 and 7, respectively (Fig. 4B and Table 1).
Plasma pharmacokinetics following oral administration of 5 mg of TFV equivalents/kg of TDF or TAF for 7 days to male beagle dogs. (A) Plasma levels of TAF and TFV were determined on days 1 and 7 following 8.3 mg/kg of TAF (5 mg TFV equivalents [eq]/kg). (B) Plasma levels of TFV were determined on days 1 and 7 following 9.0 mg/kg of TDF (5 mg TFV eq/kg). TDF was below limit of quantification at all times. The pharmacokinetic parameters are summarized in Table 1.
Mean plasma pharmacokinetic parameters following oral administration of 8.3 and 9 mg/kg of TAF and TDF (5 mg of TFV equivalent/kg), respectively, to male beagle dogs
The liver levels of TFV, TFV-MP, and TFV-DP were determined at 4 and 24 h postdose on day 7 (Fig. 5). Neither TAF nor TDF was observed in liver tissue. The pharmacologically active metabolite TFV-DP was the predominant metabolite in the liver following oral administration of either TAF or TDF. When given at equivalent TFV doses, the TFV-DP levels were approximately 2-fold higher following administration of TAF than that with TDF, at 242 and 133 μM at 4 h postdose and 153 and 75.5 μM at 24 h postdose following TAF or TDF administration, respectively (Fig. 5). These results indicate more efficient hepatic delivery by TAF relative to that by TDF and suggest that the apparent t1/2 of hepatic TFV-DP is >20 h.
Liver levels of TFV, TFV-MP, and TFV-DP following oral administration of 5 mg TFV equivalents/kg of TDF or TAF for 7 days to male beagle dogs. The liver levels of TFV and its phosphorylated metabolites were determined at 4 and 24 h postdose on day 7 (n = 2). The gray and black bars represent metabolite levels following oral administration of TDF and TAF, respectively, to dogs.
DISCUSSION
TAF is a novel phosphonamidate prodrug of TFV that delivers higher levels of TFV and its phosphorylated metabolites to target tissues and lower circulating levels of TFV relative to those of TDF (14, 15, 17, 25). A previous study by Babusis et al. (18) demonstrated that TAF was highly extracted by the liver, with an extraction ratio of 65% in dogs, suggesting that TAF is efficiently taken up by liver cells (18). In this study, we assessed the cellular uptake, intracellular metabolism, and plasma and liver pharmacokinetics of TAF. In vitro incubation of primary human hepatocytes with TAF resulted in high intracellular concentrations of TFV-DP, approximately 5- and 120-fold higher than those observed when incubated with TDF and TFV, respectively. Consistently, the anti-HBV activity of TDF in HepG2 2.2.15 cells was approximately 50-fold more potent than that of TFV (9). TDF and TAF should have a higher permeability than that of TFV because of the masking of the negative charges on the phosphonate and increased lipophilicity imparted by the prodrug moieties, resulting in higher intracellular TFV-DP levels.
Our in vitro transporter experiments demonstrated that TAF is a substrate for the hepatic transporters OATP1B1 and OATP1B3 but also has high passive permeability. In our combination experiments in primary hepatocytes, telaprevir and boceprevir were used as cathepsin A (CatA) inhibitors, but they are also known to inhibit OATP1B1 (50% inhibitory concentrations [IC50s], 2.2 μM for telaprevir and 18 μM for boceprevir) and OATP1B3 (IC50s, 6.8 μM for telaprevir and 4.9 μM for boceprevir) (26, 27). Neither telaprevir nor boceprevir affected the intracellular TFV-DP levels at concentrations up to 50 μM (Fig. 3). TAF has also been shown to be a substrate for the efflux transporters P-gp and breast cancer resistance protein (BCRP) (22). Since telaprevir is also an inhibitor of the efflux transporter P-gp, inhibition of both the uptake and the efflux transporter may offset the effects. However, boceprevir was not an inhibitor of P-gp at up to 300 μM and was a weak inhibitor of BCRP (IC50, 81 μM) in vitro; therefore, only uptake transport should be inhibited by boceprevir under our experimental conditions (27). Since TAF also has high passive permeability, the contribution of OATP1B1- and OATP1B3-dependent uptake may be minimal during TAF loading of hepatocytes.
The intracellular metabolism of TAF has been studied intensively in PBMCs and macrophages, and the ester bond of TAF is known to be cleaved by lysosomal CatA (8, 28, 29). In primary human hepatocytes, CES1 is highly expressed, and in vitro combination results indicated that CES1 plays a major role in hydrolyzing TAF. Since a greater inhibition of TFV-DP formation was observed in the presence of both BNPP and telaprevir relative to that with BNPP alone, CatA may also contribute to TAF activation to a lesser extent. The inhibition of CYP3A or P-gp by cobicistat, a pharmacokinetic enhancer that inhibits P-gp and CYP3A, did not affect the activation of TAF in vitro. Based on the results, we proposed a metabolic activation pathway of TAF in primary human hepatocytes (Fig. 6).
Proposed mechanism of metabolic activation of TAF in hepatocytes. TAF enters hepatocytes by OATP1B1- and OATP1B3-mediated transport and passive permeability. The carboxyl ester promoiety of the phosphonamidate prodrug is primarily hydrolyzed by CES1. The alaninyl-TFV intermediate is then chemically and/or enzymatically converted to TFV. Histidine triad nucleotide binding protein 1 (HINT1) may be responsible for the reaction, as the P–N bonds in other phosphoramidate prodrugs have been shown to be cleaved by this enzyme (24, 30, 31). The host nucleotide kinases phosphorylate TFV to pharmacologically active TFV-DP, which inhibits HBV reverse transcriptase. TFV also is inefficiently effluxed from the cell and eliminated renally.
Consistent with in vitro results in hepatocytes, in which incubation with TAF resulted in approximately 5-fold higher TFV-DP levels relative to those with TDF, oral administration of TAF for 7 days resulted in approximately 2-fold higher liver levels of TFV-DP relative to those with TDF. These liver levels corresponded with a plasma TAF AUC0–t and Cmax that are higher than those achieved at the selected clinical dose of 25 mg. The clinical pharmacokinetic data have been reported, with a steady-state plasma TAF Cmax and AUC0–t of 0.47 μM and 0.24 μM · h, respectively, following 25 mg of TAF in 10-day monotherapy in HIV-infected patients (14). Assuming a similar plasma TAF and hepatic TFV-DP relationship between dogs and humans, this suggests that TFV-DP levels should be approximately 3- to 5-fold lower in human livers at the clinical dose than those observed in dogs in this study, concentrations that are still >100-fold higher than the Ki for HBV polymerase (9). The long intracellular half-life observed in liver cells in vitro and in vivo in this study support once-daily administration. Consistent with the results from these nonclinical studies, TAF showed anti-HBV activity consistent with 300 mg of TDF in a dose-ranging 28-day study even in the low-dose group at doses of ≤25 mg (16).
In summary, we describe the metabolism in primary human hepatocytes, effects on hepatic uptake transporters, and plasma and liver pharmacokinetics of TAF in dogs. TAF is efficiently taken up and activated by hepatocytes by a multistep process resulting in persistent intracellular levels of the potent HBV reverse transcriptase (RT) inhibitor TFV-DP. In contrast to lymphocytes, in which CatA activity is critical for ester hydrolysis, the activation of TAF in the liver is mediated primarily by CES1. These results establish the mechanistic understanding of the potent inhibition of HBV observed with low doses of TAF in the clinic and support the further development of TAF for CHB infection.
ACKNOWLEDGMENTS
We are all employees of Gilead Sciences.
FOOTNOTES
- Received 16 January 2015.
- Returned for modification 18 March 2015.
- Accepted 3 April 2015.
- Accepted manuscript posted online 13 April 2015.
- Copyright © 2015, American Society for Microbiology. All Rights Reserved.
