Clinical disposition, metabolism and in vitro drug–drug interaction properties of omadacycline

Abstract

1. Absorption, distribution, metabolism, transport and elimination properties of omadacycline, an aminomethylcycline antibiotic, were investigated in vitro and in a study in healthy male subjects.

2. Omadacycline was metabolically stable in human liver microsomes and hepatocytes and did not inhibit or induce any of the nine cytochrome P450 or five transporters tested. Omadacycline was a substrate of P-glycoprotein, but not of the other transporters.

3. Omadacycline metabolic stability was confirmed in six healthy male subjects who received a single 300 mg oral dose of [¹⁴C]-omadacycline (36.6 μCi). Absorption was rapid with peak radioactivity (∼610 ngEq/mL) between 1–4 h in plasma or blood. The AUC_last of plasma radioactivity (only quantifiable to 8 h due to low radioactivity) was 3096 ngEq h/mL and apparent terminal half-life was 11.1 h. Unchanged omadacycline reached peak plasma concentrations (∼563 ng/mL) between 1–4 h. Apparent plasma half-life was 17.6 h with biphasic elimination. Plasma exposure (AUC_inf) averaged 9418 ng h/mL, with high clearance (CL/F, 32.8 L/h) and volume of distribution (Vz/F 828 L). No plasma metabolites were observed.

4. Radioactivity recovery of the administered dose in excreta was complete (>95%); renal and fecal elimination were 14.4% and 81.1%, respectively. No metabolites were observed in urine or feces, only the omadacycline C4-epimer.

Keywords:

• Aminomethylcycline
• cytochrome P450
• human
• transporters

Introduction

Omadacycline is a first-in-class aminomethylcycline antibiotic and the first member of this class to enter clinical development (Draper et al., 2014). Omadacycline demonstrates broad-spectrum microbiological activity against Gram-positive and Gram-negative aerobes, some anaerobes and atypical bacteria that are common pathogens associated with serious infections (Macone et al., 2014). Results from nonclinical, Phase 1, and Phase 2 studies demonstrated the microbiological activity, characterized the pharmacokinetic profile and highlighted the potential clinical efficacy and tolerability in chronic skin and skin structure infections (Macone et al., 2014; Noel et al., 2012). Omadacycline is administered as once daily oral and intravenous (IV) monotherapy and is in Phase 3 clinical studies for acute bacterial infection skin and skin structure infections and community-acquired bacterial pneumonia. The present report describes the results of in vitro and in vivo investigations into the disposition, metabolism and elimination of omadacycline to inform appropriate use and potential risks (e.g. drug–drug interactions) of omadacycline use in patients.

Materials and methods

In vitro metabolism

The following in vitro preparations were from BD Biosciences (Franklin Lakes, NJ), Gentest (Woburn, MA): pooled human liver microsomes, S9 and cytosol from the same donor pool (n = 50 donors, mixed gender); human recombinant flavin monooxygenase enzymes, FMO1, FMO3 and FMO5. Pooled cryopreserved human hepatocytes (n = 20 donors, mixed gender) were from Celsis In Vitro Technologies (Baltimore, MD). The following chemicals were obtained from Sigma-Aldrich (St. Louis, MO): β-nicotinamide adenine dinucleotide 2′-phosphate (β-NADPH), uridine phosphate-glucuronic acid (UDPGA), alamethicin, dimethyl sulfoxide (DMSO), ammonium acetate, potassium phosphate (mono- and di-basic), MgCl₂, Krebs-Henseliet maintenance medium (KHB). Cell culture plates were from Corning (Corning, NY). Fetal bovine serum (FBS) was from Invitrogen (Walkersville, MD). [¹⁴C]-omadacycline (Figure 1) for in vitro studies was synthesized by the Isotope Laboratory of Novartis (East Hanover, NJ) with a chemical and radiochemical purity of  ≥90%.

Figure 1. [¹⁴C]-Omadacycline (asterisk indicates position of ¹⁴C label).

In vitro incubations of in human liver microsomes were performed in the presence of NADPH and/or UDPGA. Incubations contained (final concentrations): human liver microsomes (1 mg protein/mL) in 100 mM potassium phosphate buffer [(pH 7.4, previously preincubated with alamethicin (60 μg/mg protein, final concentration) for 15 min on ice)], MgCl₂ (5 mM), 4 mM UDPGA and/or 1 mM β-NADPH. Reactions were initiated with [¹⁴C]-omadacycline (12 or 48 μM, final concentrations) and incubated for 30 min at 37 °C (Eppendorf Thermomixer at ∼800 rpm). The reactions were quenched by adding an equal volume of cold acetonitrile and the precipitated protein was removed by centrifugation (39,000×g, 10 min, ∼4 °C). The supernatant was collected, evaporated and an aliquot analyzed by liquid chromatography (LC) with offline radioactivity detection as described later. Similarly, in vitro incubations of [¹⁴C]-omadacycline in human liver S9, cytosol or recombinant human FMO enzymes were performed as follows (final concentrations): [¹⁴C]-omadacycline (12 and 48 μM), protein (1 mg/mL, human liver S9, cytosol or recombinant human FMO1, FMO3, or FMO5 or control microsomes), 100 mM potassium phosphate buffer (pH 7.4), 5 mM MgCl₂, with or without β-NADPH (1 mM). The reactions were preincubated (3 min, 37 °C, Eppendorf Thermomixer at ∼800 rpm) and initiated by the addition of omadacycline. The samples were incubated (30 min, 37 °C), quenched with an equal volume of cold acetonitrile and precipitated protein was removed by centrifugation.

Hepatocyte incubations were performed in KHB media supplemented with sodium bicarbonate, fructose and glycine (final concentrations: 25, 10 and 3 mM, respectively). Frozen hepatocytes were thawed (37 °C, ∼90 s) and transferred to a flask containing the above medium. The hepatocytes were centrifuged (75 × g) and washed by resuspending in media. Cells were sedimented by centrifugation and the final cell pellets suspended in media supplemented with 2% (v/v) FBS (final concentration: ∼2 × 10⁶ cells·mL⁻¹). Cell viability was confirmed to be ∼81% prior to incubation (Supplemental Table S1). An aliquot (∼1 × 10⁶ viable cells) of the final cell suspension was added to each well of a 12-well plate predispensed with 0.5 mL of media and containing [¹⁴C]-omadacycline (2.5 μM or 12.5 μM) and incubated (37 °C, 5% CO₂, 95% air). At 2, 4, 8 and 24 h, the contents of individual wells were collected and quenched with acetonitrile. Positive metabolism control incubations using terfenadine and 7-ethoxycoumarin confirmed the metabolic activity of the hepatocyte incubations (data not shown).

In vitro assessment of omadacycline as an inhibitor of cytochrome (CYP) metabolism

The potential of omadacycline to reversibly inhibit human CYP enzyme activity was assessed using pooled human liver microsomes. This potential was determined by testing the effect of increasing concentrations of omadacycline (up to 100 μM) on the metabolism of probe substrates whose metabolism is known to be CYP enzyme selective. In the incubations, probe substrate concentrations were less than or equal to their reported K_m values and probe substrate metabolism was assessed by liquid chromatography-tandem mass spectrometry (LC-MS/MS) analysis of specific metabolite formation. Reversible enzyme inhibition potential was assessed for CYP1A2 (phenacetin O-deethylation; phenacetin is also a substrate of CYP1A1), CYP2A6 (coumarin 7-hydroxylation), CYP2B6 (bupropion hydroxylation), CYP2C8 (amodiaquine N-demethylation), CYP2C9 (diclofenac 4′-hydroxylation), CYP2C19 (S-mephenytoin 4′-hydroxylation), CYP2D6 (bufuralol 1-hydroxylation), CYP2E1 (chlorzoxazone 6-hydroxylation) and CYP3A4/5 (midazolam 1′-hydroxylation, testosterone 6β-hydroxylation) activities. The potential of omadacycline to act as a time-dependent inhibitor of human cytochrome P450 (CYP) enzyme activity was also determined with human liver microsomes. This was determined by measuring the CYP activity remaining after preincubation with omadacycline (concentrations up to 50 μM) for various time periods. The CYP activities investigated were CYP1A2, CYP2C9, CYP2D6 and CYP3A4/5. Positive inhibitor controls, furafylline (CYP1A2), tienilic acid (CYP2C9), paroxetine (CYP2D6) and troleandomycin (CYP3A4/5) also were tested. To assess the occurrence of nonspecific protein binding in the incubations, [¹⁴C]-omadacycline (4.8 or 48 μM) was incubated with varying concentrations of human liver microsomes (HLM) and the fraction of unbound omadacycline was determined by ultracentrifugation and liquid scintillation radioactivity analysis. The extent of binding was assessed over a protein concentration range of 0.01 to 2 mg protein·mL⁻¹.

In vitro assessment of omadacycline use and effect on drug transporters

Omadacycline as transporter substrate

Human embryonic kidney parental host strain (HEK293 Flp-In) cells or those stably expressing human organic anion transporters (hOAT1, hOAT3), human organic cation transporter 2 (hOCT2) or human organic anion transport polypeptide transporters (OATP1B1) (Novartis, NJ) or OATP1B3 (University of Heidelberg, Germany) were plated into 24-well poly-d-lysine-coated plates. The positive control substrates and model inhibitor used in the transport experiments are listed (Supplemental Table S2). On the day of the experiment, the cell culture medium was aspirated from the wells and replaced with 0.6 mL Hanks’s balanced salt solution (HBSS) pre-equilibrated to 37 °C. Transport was initiated by aspirating the HBSS and replacing it with 0.2 mL of HBSS containing [¹⁴C]-omadacycline. All subsequent incubations to examine interactions (i.e. transport and transport inhibition) with additional transporters were conducted for 5 min at 37 °C. At the designated time, uptake was stopped by removal of the radiolabeled substrate, washing the cells twice with ice-cold phosphate-buffered saline (PBS), and lysing the cells with 1% sodium dodecyl sulfate (SDS) and a freeze-thaw cycle. Protein content was determined using the bicinchoninic acid protein assay (Pierce Biotechnology, Rockford, IL), and quantification of [¹⁴C]-omadacycline was determined by liquid-scintillation counting (Packard Instruments, Meriden, CT).

Transport of [¹⁴C]omadacycline was evaluated across confluent Caco-2 cell monolayers in the presence or absence of the P-glycoprotein (P-gp) inhibitors LY335979 (1.0 μM; Dantzig et al., 1999) and GF120918 (4 μM, Matsson et al., 2009), the multidrug resistance-associated protein 2 (MRP2) inhibitor MK571 (10 μM, Matsson et al., 2009) or the human breast cancer resistance protein (BRCP) inhibitor Ko143 (1.0 μM, Allen et al., 2002). Caco-2 cells were grown on permeable polyethylene terephthalate membranes (BD Falcon HTS Multiwell, 1 um pore size, 0.31 cm² growth area) in Dulbecco's modified Eagle's medium (DMEM) containing 10% fetal calf serum (FCS), 1% nonessential amino acids and 100 μg/mL streptomycin at 37 °C until confluent (approximately 21 days). Transport experiments were conducted using 0.2 mL of prewarmed (37 °C) transport media in the apical and 1.0 mL in the basolateral chamber. Cell monolayers integrity was verified using an automated electrical volt-ohm meter (World Precision Instruments, Sarasota, FL). The donor solutions containing [¹⁴C]-omadacycline were added to either the apical or basolateral compartment, whereas the transport protein inhibitors (vide supra) were added to both compartments to final concentrations. To assess the permeability in either direction, aliquots (100 μL) were removed from the acceptor chamber at 120 min. The amount of drug transported in a given time interval was determined from the radioactivity content in the samples. The apparent permeability (Papp, cm·min⁻¹) across cell monolayers was determined.

Transport assays across Caco-2 cells in the Bl→Ap direction were performed using several [¹⁴C]-omadacycline concentrations (∼0.60–61 μM) in the basolateral compartment for 2 h at 37 °C without plate shaking. The apparent K_m associated with the observed P-glycoprotein (P-gp)-mediated omadacycline efflux was derived from the apparent net secretory flux measurement (i.e. P-gp-mediated transport activity). Net secretory flux (Gao et al., 2001) was estimated by subtracting the basolateral-to-apical (Bl→Ap) flux at each [¹⁴C]-omadacycline concentrations in the presence of the P-gp inhibitor GF120918 (Bl→Ap)i from that in the absence of GF120918 (Bl→Ap)0. A nonlinear regression analysis of the net secretory flux versus [¹⁴C]-omadacycline concentration was determined by fitting to the standard Michaelis–Menten equation (J = J_max× [S]/(K_m + [S]) (GraphPad Prism v. 4.02, GraphPad Software, Inc, La Jolla, CA).

Omadacycline as transporter inhibitor

Cell lines and reagents for inhibition studies are shown in Supplemental Table S2. On the day of the experiment, the cell culture medium aspirated from the wells and replaced with 0.6 mL HBSS pre-equilibrated to 37 °C. Transport was initiated by aspirating the HBSS and replacing it with 0.2 mL of HBSS containing labeled substrate with increasing concentrations of omadacycline up to 100 μM. Assays were terminated and processed essentially as described previously when examining omadacycline as a transporter substrate.

The potential for omadacycline 0.10–50 μM to inhibit transport-mediated efflux by the human orthologs of BCRP T8 cells, P-gp (MDA435 T0.3 cells), and the MRP2 Madin-Darby Canine Kidney II (MDCKII) cells overexpressed in mammalian cells was examined. Flow cytometry assays were used to assess the potential for omadacycline to inhibit the efflux of fluorescent substrates Bodipy FL prazosin (BDP) and rhodamine 123 (Rho123) by BCRP and P-gp, respectively. The potential for omadacycline to inhibit MRP2 was examined by testing the effect of omadacycline on [¹⁴C]-valsartan efflux from MDCKII–MRP2 cells.

In vitro assessment of omadacycline as an inducer of CYP metabolism and drug transporters

Omadacycline as an enzyme inducer

Omadacycline at concentrations up to 100 μM was examined in the hepatocytes from three donor livers. The in vitro enzyme induction potential was assessed based on messenger RNA (mRNA) and enzyme activity levels for CYP1A1, CYP1A2, CYP1B1, CYP2B6, CYP2C8, CYP2C9, CYP2C19, CYP2J2, CYP3A4, CYP3A5 and/or UGT1A1. The MTT (3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyl tetrazolium bromide) assay (In vitro toxicology assay kit, Sigma-Aldrich) was used to assess the viability of treated and untreated hepatocytes at the end of the treatment period (48 h). Messenger RNA was measured by real-time polymerase chain reactions (RT-PCR) following ∼48 h of treatment (mRNA from one donor was found to be degraded; therefore, only results from the remaining two donors are presented). Evaluation of changes in activity of CYP1A2, CYP2B6, CYP2C8, CYP2C9, CYP2C19, CYP3A, and UGT1A1 was determined by measuring selective enzyme probe substrate metabolism in the cells after the induction period using quantitative LC-MS/MS analysis. Rifampicin (RIF) was used as a positive control for pregnane X receptor (PXR) and/or constitutive androstane receptor (CAR) activation (CYP3A/2B/2C and ABCB1 induction) (Björnsson et al., 2003; Geick et al., 2001; Gerbal-Chaloin et al., 2001; Madan et al., 2003). Phenobarbital (PB) at a concentration of 1000 μM was also included as a well-known positive control inducer of CYP3A/CYP2B/2C and UGT1A (Gerbal-Chaloin et al., 2001; Madan et al., 2003). β-Napthoflavone (BNF) was included at a concentration of 10 μM as a positive control for CYP1A and UGT1A1 induction (Madan et al., 2003; Meunier et al., 2000; Yueh et al., 2005).

Omadacycline as transporter inducer

Omadacycline at concentrations up to 100 μM was examined for its potential to induce mRNAs of P-gp, ATP-binding cassette, subfamily B, protein 1 (ABCB1) and the MRP2, ATP-binding cassette, subfamily C, protein 2 (ABCC2) transporters in primary human hepatocytes of two individual donors after 48 h of treatment. Induction of mRNA, relative to the vehicle control, was determined by RT-PCR.

Human mass balance and disposition

Radiolabeled drug substance

[¹⁴C]-omadacycline (Figure 1) was synthesized by the Isotope Laboratory of Novartis (East Hanover, NJ). The chemical and radiochemical purity was  ≥90% (with no unspecified impurity  >1.0%) as verified by high-pressure liquid chromatography (HPLC). The major impurity was the omadacycline C4-epimer (a common impurity for this class of molecule (Nelson, 2001), with several trace impurities as described later.

Clinical study design and study medication

The study was approved by the local ethics and radiation safety committees and conducted in accordance with Good Clinical Practice guidelines and the Declaration of Helsinki. All subjects gave written informed consent before participation. The study population comprised six healthy male subjects aged 18–45 years, body weight at least 50 kg, body mass index (BMI) 18–30 kg/m², and in good health as determined by past medical history, physical examination, vital signs, electrocardiogram (ECG) and laboratory tests at screening. Subjects were excluded from the study if they were exposed to diagnostic radiation for occupational reasons or during participation in a clinical trial in the previous year (except dental X-rays and plain X-rays of thorax and bony skeleton [excluding spinal column]) or had a history of hypersensitivity or allergic reaction (e.g. anaphylaxis, urticaria, other significant reaction) to any tetracycline (e.g. minocycline, doxycycline or tetracycline) or to drugs of similar classes or to any of the excipients of these drugs. Subjects were required to fast overnight prior to the dose and until 4 h after dose administration.

Each subject received the single 300 mg oral dose of [¹⁴C]-omadacycline on the morning of Day 1. The dose of 300 mg was the proposed once-daily therapeutic oral dose for the treatment of skin and skin structure infection, based on safety data from prior clinical studies and expected exposure (plasma AUC) relative to minimum inhibitory concentrations of target bacteria. No dairy products, antacids, or other aluminum- or calcium-containing products on the day of dosing were allowed. Radiolabeled omadacycline was provided as individually packaged doses of 2 × 150 mg capsules with a total dose of 1.35 MBq or 36.6 μCi of [¹⁴C]-omadacycline. Subjects were confined to the study center for at least 20 h before administration of the study drug until 168 h (7 days) postdose. If subjects did not reach at least 85% recovery of the administered radioactive dose, then they remained at the clinical site until they met the discharge criteria for 85% recovery or up to Day 10 (216 h).

Sample collection

Blood was collected at predose, 0.5, 1, 1.5, 2, 2.5, 3, 4, 8, 12, 24, 36, 48, 72, 96, 120 and 144-h postdose by either direct venipuncture or an indwelling cannula inserted in a forearm vein. A total of 18 mL venous blood was collected at each time point in heparinized tubes. Of this, separate aliquots were used for the various study assessments (e.g. total radioactivity in blood and plasma, metabolite profiling). Analysis of total radioactivity was performed at the study site. The remaining whole blood was transferred into polypropylene screw-cap vials and immediately frozen and stored at −60 °C or colder until shipment for analysis. Urine was collected from each subject and pooled by time period postdose (0–4, 4–8, 8–12, 12–24, 24–48, 48–72, 72–96, 96–120, 120–144 and 144–168 h). Feces were collected as produced from time of dosing until at least 168-h postdose. All samples were stored at  < −20 °C until analysis.

Determination of total radioactivity and sample preparation for characterization of omadacycline and related components

Radioactivity measurements in plasma, blood, urine and feces were carried out using liquid scintillation counting. Processing for metabolite profiling is briefly described in the following paragraphs. The very low concentrations of radioactivity in plasma precluded radiochromatography profiling.

For each subject, urine samples were pooled to represent  >90% of the urinary total radioactivity. The pooled samples were centrifuged, and the supernatants analyzed by liquid chromatography (LC and LC/MS).

For each subject, a pooled 0–72 h fecal sample was prepared (0.5% by weight) from the homogenized fecal samples. The 0–72 h pools represented an average of 92% (85–97%) of excreted fecal radioactivity. From these fecal-pooled samples, an accurately weighed portion (∼9–17 g) of each was extracted with 20 mL acetonitrile containing 0.1% formic acid. The remaining pellet was extracted twice with water (5 mL each), followed by a methanol extraction (3 mL). The combined extracts were concentrated under N₂ at 30 °C. The resulting residue was reconstituted with water/acetonitrile (90/10, v/v). The radioactivity extraction recovery averaged 47% (range 34–64%). An aliquot of each was analyzed by LC and LC/MS.

Analysis of omadacycline

Plasma omadacycline concentrations were determined using a validated LC/MS method. Omadacycline and its internal standard (omadacyline-d₆) were isolated from sodium heparin human plasma samples (200 μL) using a solid-phase extraction procedure (Waters Oasis HLB, 10 mg, Milford, MA). The extracted analyte was analyzed by TurboIonSpray liquid chromatography/tandem mass spectrometry (Sciex API3000, Foster City, CA) in the positive ion mode [LC column: Phenomenex Chromolith RP 18 e, 4.6 × 100, Torrance, CA; mobile phase A: 10 mM ammonium acetate, pH 3.0, B: acetonitrile/methanol, 50:50 (v:v)]. The LC elution gradient was as follows: (1.0 mL/min flow rate): 0–0.5 min, 5% B, 0.5–6 min, 5–26% B, 6–6.01 min, 27–100% B, 6.01–7 min, 100% B, 7–7.01 min, 100–5% B. The MS/MS ion transitions monitored for omadacycline and internal standard were m/z 557.3 → 470.2 and m/z 563.3 → 476.4, respectively. The lowest limit of quantification (LLOQ) for the omadacycline measurement was 2 ng/mL. Calibration standard responses were linear over the range of 20–2000 ng/mL. A weighted (l/concentration2) under a linear calibration model was used. In plasma for omadacycline, the interday assay accuracy, expressed as percent relative error for quality control (QC) concentrations, ranged from −5.3% to 2.0% bias in QC samples. Assay precision, expressed as the interday percent coefficients of variation (% CV) of the mean estimated concentrations of QC samples ranged from 3.1% to 4.5%.

Metabolite analysis by HPLC

[¹⁴C]-Omadacycline and related components in urine and feces were analyzed by HPLC with offline radioactivity detection. The chromatographic separations were performed on a Waters Xbridge column (Milford, MA) (150 × 4.6 mm, 5 μm) maintained at a temperature of 25 °C. Components were resolved with gradient elution consisting of solvent A (10 mM ammonium formate with 0.1% formic acid, pH ∼3) and B (acetonitrile with 0.1% formic acid), at a flow rate of 1.0 mL/min. The gradient elution program was as follows (all steps were linear): 0 min, 2% B, 0–30 min, 2–15% B, 30–35 min, 15–25% B, 35–36 min, 25–90% B, 36–44 min, 90% B, and 44–45 min, 2% solvent B. Fractions were collected into 96-Deepwell LumaPlates (PerkinElmer, Waltham, MA) at a rate of 8 sec per well. Plates were evaporated to dryness, sealed and radioactivity counted [Packard TopCount NXT microplate scintillation counter (Downers Grove, IL)].

Structural elucidation by LC-MS/MS

Structural characterization and mass analysis of [¹⁴C]-omadacycline and related components were carried using the separation method described previously. The LC eluent was split (ratio 4:1) to deliver 800 μL min⁻¹ for collection into the 96-well plates and ∼200 mL/min to the MS system [Thermo Fisher LTQ Orbitrap Velos (Waltham MA) or Waters Synapt Q-TOF (Milford, MA)]. The MS systems were operated in positive ionization mode. MS parameter settings were as follows: Velos: Resolution 30000, source 5000 V, capillary 275 °C, sheath gas 40; aux gas 5; ion sweep gas 5; source type HESI; source temp 60 °C; Synapt Q-TOF: Resolution 9000, spray 3100 V, sampling cone 40 V, trap collision energy 5–40 eV, transfer collision energy 10 eV, argon collision gas.

Safety assessments

Safety assessments consisted of collecting all adverse events (AEs) and serious AEs, including their severity and relationship to study drug. Physical examination, vital signs, ECG and laboratory assessments (hematology, blood chemistry and urinalysis) were performed at baseline and at the end of study (Day 8). In addition, blood pressure and heart rate were assessed multiple times on Day 1 and again on Day 2, and hematology and blood chemistry also were performed on Day 2.

Pharmacokinetic analyses and statistics

Pharmacokinetic parameters maximum plasma concentration (C_max), time to maximum plasma concentration (T_max), half-life (T_1/2), area under the concentration–time curve from time 0 to the last measurable concentration (AUC_last), AUC from time 0 to infinity (AUC_inf), apparent total clearance (CL/F), apparent volume of distribution (Vz/F) of the parent omadacycline were derived from plasma concentration–time data for the parent drug in each subject by noncompartmental analysis using WinNonlin Pro Version 5.2 (Pharsight, Princeton, NJ). Data were presented in a descriptive way as means ± standard deviation. No formal statistical analysis was done.

Results

In vitro metabolism of omadacycline

Representative metabolism profiles obtained after incubation of [¹⁴C]-omadacycline with human liver microsomes and hepatocytes are shown in Figures 2 and 3, respectively. No metabolites were detected in any of these incubations, regardless of added cofactors. Subsequent experiments using other hepatic subcellular fractions (S9, cytosol) and various recombinant enzymes (FMO1/3/5) also indicated no metabolic turnover (Supplemental Figures S1 and S2). These in vitro results showed the metabolic stability of omadacycline. The C4-epimer of omadacycline (representing a change in stereochemistry at carbon-4) was the predominant non-parent omadacycline component, accounting for ∼8–9% of the radioactivity.

Figure 2. Incubation of [¹⁴C]-omadacycline in human liver microsomes. [¹⁴C]-omadacycline (12 or 48 μM) incubations were conducted in the absence or presence of metabolic co-factors (48 μM incubation results shown). Results shown include: (A) the absence of co-factors, (B) the presence of NADPH, (C) the presence of UDP-glucuronic acid (UPPGA) and (D) presence of NADPH and UDPGA. Incubations using 12 μM [¹⁴C]-omadacycline also showed no evidence for metabolic turnover (not shown).

Figure 3. Incubation of [¹⁴C]-omadacycline in human hepatocytes. Pooled cryopreserved human hepatocytes (1 × 10⁶ cells mL⁻¹) were incubated with [¹⁴C]-omadacycline (2.5 μM and 12.5 μM). The reactions were quenched after 2, 4, 8 or 24 h incubations, by the addition of an equal volume of cold acetonitrile. Incubation mixtures (A, 4 h; B, 24 h) were analyzed by LC with radiochemical detection (12.5 μM results shown).

In vitro assessment of omadacycline as an inhibitor of CYP metabolism

Omadacycline showed very little or no reversible inhibition of CYP1A2, CYP2A6, CYP2B6, CYP2C8, CYP2C9, CYP2C19, CYP2D6, CYP2E1 and CYP3A4/5 at omadacycline concentrations of up to 100 μM (all IC₅₀ values  >100 μM, Supplemental Table S3). Similarly, an assessment of potential time-dependent inhibition of selected CYP enzyme activities (CYP3A4/5, CYP2C9 and CYP2D6) showed no inhibition up to an omadacycline concentration of 50 μM (Supplemental Table S4). An assessment of nonspecific microsomal protein binding using [¹⁴C]-omadacycline and ultracentrifugation indicated that under the in vitro microsomal incubation conditions minimal nonspecific binding was seen, with  >92% of the omadacycline being unbound (Supplemental Table S5).

In vitro assessment of omadacycline as a transporter substrate

The uptake of omadacycline into control and transporter expressing HEK cells was determined at 8 μM over 15 min of incubation (Supplemental Table S6). Concentrations increased rapidly and reached 248 ± 9 pmol/mg protein by 15 min. The rate of uptake was not appreciably influenced in HEK cells expressing hOAT1 or hOAT3 with uptake of 312 ± 29 and 301 ± 35 pmol/mg protein, respectively. The rapid cellular uptake is consistent with the large volume of distribution and rapid distribution from plasma.

Additional experiments in HEK cells at a concentration of 25 μM [¹⁴C]-omadacycline showed no difference in intracellular concentrations with or without hOAT1 (395 ± 23 versus 403 ± 5.2 pmol/mg protein) or hOAT3 (442 ± 42 versus 403 ± 5.2 pmol/mg protein), whereas positive controls para-aminohippurate (PAH) uptake exhibited dependence on hOAT1 and estrone-3-sulfate (E3S) on hOAT3 (Supplemental Table S7). The inhibitor probenecid (100 uM) also had no effect on the accumulation of omadacycline but a pronounced inhibition of E3S and PAH uptake. The accumulation of [¹⁴C]-omadacycline was comparable in cells that expressed the OATP transporters (OATP1B1, 408 ± 40 pmol/mg protein; OATP1B3 137 ± 18 pmol/mg protein) to control cells (403 ± 5.2 pmol/mg protein and 103 ± 11 pmol/mg protein, respectively) indicating that omadacycline was not a substrate for OATP1B1 or OATP1B3. In contrast, the accumulation of the model substrates [³H]-estradiol-17ß-glucuronide (E₂17ßG) (OATP1B1) and [³H]-cholecystokinin 8 (CCK8) (OATP1B3) was 12.7- and 3.4-fold higher than that in control cells, respectively, indicating the suitability of the cellular test systems. In hOCT2 expressing cells versus control cells, the update of [¹⁴C]-omadacycline was similar (24.6 ± 2.1 and 24.0 ± 3.0 pmol/mg protein, respectively). These uptake values were unaffected by addition of inhibitor decynium 22 (20  μM) (23.8 ± 1.7 and 23.4 ± 2.2 pmol/mg protein, respectively). Positive control hOCT2 substrate metformin showed enhanced uptake in these cells and pronounced inhibition by decynium 22 as expected.

The average apparent permeability (Papp) values for [¹⁴C]-omadacycline across Caco-2 cells in the apical-to-basolateral (Ap→Bl) direction are shown in Table 1. LY335979 (P-gp inhibitor) reduced the omadacycline efflux ratio to near unity. In contrast, MK571 (MRP inhibitor), and Ko143 (BCRP inhibitor) had little effect on the Papp in either direction. These data suggest a primary involvement of P-gp in omadacycline transport and not BCRP or MRP2. Although saturation was not complete, the K_m and J_max values for P-gp-mediated efflux of omadacycline were estimated to be 81.5 μM and 1140 pmolh ⁻¹·cm ⁻², respectively, suggesting that omadacycline is a low affinity substrate. Based on the average Papp (Ap→B) of omadacycline in the presence of LY335979 (6.73 × 10⁻⁵cm·min⁻¹), it was classified as a poorly passive permeability compound [(compared to low permeability marker mannitol (4.0 × 10 ⁻⁵cm·min⁻¹) and high permeability marker propranolol (79.2 × 10⁻⁵cm·min⁻¹)].

Table 1. [¹⁴C]-omadacycline permeability across Caco-2 cell monolayers and the effect of transport protein inhibitors^a.

Omadacycline concentration (μM)	Inhibitor concentration (μM)	Transporter	Papp × 10^−5 Ap to BI (cm/min)	Papp × 10^−5 BI to Ap (cm/min)
2.0	None	–	2.73 ± 0.43	35.7 ± 3.7
	LY335979, 1.0	P-gp	7.66 ± 0.56	10.8 ± 1.2
	MK571, 10	MRP2	3.97 ± 0.76	35.1 ± 1.7
	Ko143, 1.0	BCRP	3.71 ± 0.99	36.9 ± 2.3
12	None	–	2.10 ± 0.14	26.6 ± 2.4
	LY335979, 1.0	P-gp	5.80 ± 0.22	7.0 ± 0.3

^aUptake experiments conducted over 120 min at 37 °C.

In vitro assessment of omadacycline as an inhibitor of drug transporters

The uptake of [³H]-PAH by hOAT1 cells was reduced by 32.1% at the highest concentration of omadacycline (25 μM) tested (Table 2) but showed no effect on basal PAH accumulation in control cells. Uptake of [³H]-E3S by hOAT3 cells was not altered at any omadacycline concentration. Moreover, omadacycline had no effect on the basal E3S accumulation in control cells. E3S accumulation in hOAT3-expressing cells was ∼25.7-fold higher than control cells, indicating adequate expression of the hOAT3 transporter. Additionally, probenecid dramatically reduced E3S accumulation in the hOAT3 cells, but not in control cells.

Table 2. Inhibition of human transporter proteins by omadacycline.

			Mean ± SD substrate accumulation^a (pmol/mg protein)
Transporter	Substrate (μM)	Inhibitor (μM)	HEK-Flp-In	Transfected Cells	% Inhibition
hOAT1	[^3H]-PAH (1.0)	None	6.4 ± 0.3	32.5 ± 1.2	–
		Probenecid (100)	6.4 ± 0.3	7.0 ± 0.5	98.0
		OMC (5.0)	5.9 ± 0.2	27.6 ± 2.6	16.7
		OMC (25.0)	6.0 ± 0.2	23.7 ± 1.9	32.1
hOAT3	[^3H]-E3S (1.0)	None	3.0 ± 0.5	68.2 ± 3.1	–
		Probenecid (100)	3.2 ± 0.4	5.5 ± 0.3	96.5
		OMC (25.0)	3.1 ± 0.3	68.4 ± 0.6	0
OATP1B1	[^3H]-E217βG (1.0)	None	0.78 ± 0.09	25.52 ± 0.67
		Rifamycin SV (25)	0.82 ± 0.03	0.94 ± 0.09	99
		OMC (100)	0.72 ± 0.15	23.12 ± 1.40	9.7
OATP1B3	[^3H]-CCK8 (0.033)	None	0.20 ± 0.01	0.67 ± 0.04	–
		Rifamycin SV (25)	0.18 ± 0.01	0.17 ± 0.01	105
		OMC (100)	0.23 ± 0.01	0.63 ± 0.07	8.5
hOCT2	[^14C]-Metformin (16)	None	2.5 ± 0.2	126.0 ± 2.2	–
		Decynium (22, 20)	1.4 ± 0.5	2.4 ± 0.5	99
		OMC (100)	1.9 ± 0.5	114.0 ± 2.8	9.2

^aCells incubated with substrate ± inhibitor for 5 min (hOAT) or 4 min (hOATP and hOCT) at 370C; OMC, omadacycline; Model substrates, para-aminohippurate (hOAT1), estrone-3-sulfate (hOAT3), estradiol-17β-D-glucuronide (hOATP1B1), cholecystokinin (hOATP1B3) and metformin (hOCT2) performed as expected with and without inhibitors.

Omadacycline (20 or 100 μM) minimally reduced [³H]-E₂17βG (1.0 μM) or [³H]-CCK8 (0.033 μM) accumulation in HEK-OATP1B1 and HEK-OATP1B3 cells, respectively (Table 2). In contrast, the accumulation of E₂17βG or CCK8 was reduced to the same levels as in control cells by the positive control inhibitor (rifamycin SV). These data indicate that omadacycline does not inhibit the transport activity of OATP1B1 or OATP1B3.

Omadacycline did not inhibit [¹⁴C]-metformin uptake into hOCT2-expressing cells (Table 2). In contrast, [¹⁴C]-metformin uptake in the presence of the known hOCT2 inhibitor, decynium-22 (20 μM), was reduced to background levels. The results of these studies demonstrate that omadacycline does not inhibit hOCT2 in vitro and suggest OCT2 would not be involved in the active renal secretion of omadacycline in vivo. Moreover, omadacycline would not be expected to alter the renal clearance of comedications that are actively secreted via renal hOCT2.

Omadacycline (50 μM) was not an in vitro inhibitor of BCRP, P-gp or MRP2. BCRP, P-gp and MRP2 efflux activity (based on prototypic substrates bodipy FL prazosin (BDP), rhodamine 123 (Rho123) and [¹⁴C]-valsartan (VAL), respectively) were largely unaffected by the concentrations of omadacycline up to 50 μM (50 μM omadacycline results, efflux without inhibitor, with inhibitor: BRCP in BCRP T8 cells, BDP efflux 23.3, 24.9 fluorescent units; P-gp in MDA435 T0.3 cells, Rho123 efflux 9.6, 7.0 fluorescent units; MRP2-MDCKII cells, VAL efflux, 404 ± 84, 455 ± 34 pmol/mg protein).

In vitro assessment of omadacycline as an inducer of CYP metabolism and drug transporters

Transporter induction

The induction potential of omadacycline was determined by the ability of omadacycline (100 μM) to induce the expression of P-gp and MRP2 mRNAs. All mRNA levels of activity with respect to the vehicle control were  <40% of the maximal positive control response and/or  ≤2-fold (Table 3). Omadacycline (up to 100 μM) did not induce mRNA levels of P-gp or MRP2 in the hepatocytes examined. Thus, it is unlikely that omadacycline would act as an inducer of the synthesis of these drug transporters in humans.

Table 3. Induction of mRNA of P-gp and MRP2 by omadacycline.

			Mean ± SD fold change
mRNA	Treatment	Inducer concentration (μM)	Liver 1	Liver 2
P-gp	None	–	1.0	1.0
	Omadacycline	100	1.16 ± 0.34	1.31 ± 0.25
	Rifampicin	10	2.53 ± 0.37	3.13 ± 0.99
MRP2	None	–	1.0	1.0
	Omadacycline	100	0.76 ± 0.23	1.01 ± 0.63
	Rifampicin	10	1.58 ± 0.40	1.48 ± 0.38

Enzyme induction

Omadacycline was examined for its potential to induce CYP450 enzymes and UDP-glucuronosyltransferase (UGT) UGT1A1 mRNA and activities in primary human hepatocytes of three individual donors after 48 h of treatment. Induction of mRNA, relative to the vehicle control, was determined by real-time PCR and evaluation of changes in enzyme activities were assessed after the induction period by quantitative LC-MS/MS analysis of enzyme-selective probe substrate metabolism. Cell viability of the hepatocytes was assessed by the MTT assay, a measurement of mitochondrial activity in living cells, after the treatment period. The cell viability, with respect to the vehicle control, was acceptable (∼ ≥75%) for all omadacycline and positive inducer (RIF, PB, or BNF) treatments. Omadacycline (up to 100 μM) did not induce activities of CYP1A2, CYP2B6, CYP2C8, CYP2C9, CYP2C19, CYP3A or UGT1A1 in the three hepatocyte donors (Table 4). All levels of activity with respect to the vehicle control were ∼≤2-fold and/or less than 40% of the maximal positive control response. In addition, omadacycline did not induce mRNA levels of CYP1B1, CYP2C9, CYP2C19, CYP2J2, CYP3A4, CYP3A5 or UGT1A1 in the hepatocytes examined. All mRNA levels with respect to the vehicle control were also ∼≤2-fold and/or less than 40% of the maximal positive control response. There was a slight mRNA induction response for CYP1A1, CYP1A2, CYP2B6 and CYP2C8 that was  >2-fold; however, this did not translate into induction of CYP activity. It can be concluded that it is not likely that omadacycline would act as an inducer of drug-metabolizing enzymes in humans.

Table 4. Induction of mRNA and enzyme activity of various drug-metabolizing enzymes.

			Mean±SD fold change
			mRNA		Activity
mRNA	Treatment^a	Inducer (μM)^b^,^c	Liver 1	Liver 2	Liver 1	Liver 2	Liver 3
CYP1A1	None	–	1.0	1.0	–^d	–	–
	OMC	100	2.97 ± 0.66	2.00 ± 0.93	–	–	–
	RIF	50	3.21 ± 0.90	3.84 ± 0.35	–	–	–
	PB	1000	3.86 ± 0.93	3.97 ± 0.41	–	–	–
	BNF	10	42.0 ± 11	66.6 ± 5.1	–	–	–
CYP1A2	None	–	1.0	1.0	1.0	1.0	1.0
	OMC	100	4.89 ± 1.2	3.25 ± 1.5	1.12 ± 0.14	1.41 ± 0.11	1.04 ± 0.11
	RIF	50	0.926 ± 0.12	0.847 ± 0.15	0.66 ± 0.09	1.36 ± 0.14	1.55 ± 0.16
	PB	1000	4.26 ± 3.3	2.09 ± 0.19	0.83 ± 0.09	1.39 ± 0.13	2.40 ± 0.21
	BNF	10	15.4 ± 4.4	22.4 ± 4.8	2.15 ± 0.24	27.0 ± 2.7	20.1 ± 1.6
CYP1B1	None	–	1.0	1.0	–	–	–
	OMC	100	0.933 ± 0.23	0.636 ± 0.36	–	–	–
	RIF	50	0.692 ± 0.15	0.878 ± 0.16	–	–	–
	PB	1000	1.02 ± 0.32	1.09 ± 0.078	–	–	–
	BNF	10	5.04 ± 1.0	5.04 ± 1.0	–	–	–
CYP2B6	None	–	1.0	1.0	1.0	1.0	1.0
	OMC	100	2.67 ± 0.98	1.43 ± 0.83	0.77 ± 0.05	0.80 ± 0.08	0.85 ± 0.11
	RIF	50	2.80 ± 0.36	4.78 ± 1.1	2.97 ± 0.17	6.65 ± 0.59	5.47 ± 0.72
	PB	1000	10.4 ± 2.3	17.0 ± 1.9	8.99 ± 0.60	7.16 ± 0.83	12.2 ± 1.6
	BNF	10	0.865 ± 0.13	1.72 ± 0.33	0.58 ± 0.048	2.51 ± 0.28	6.35 ± 0.86
CYP2C8	None	–	1.0	1.0	1.0	1.0	1.0
	OMC	100	2.99 ± 0.50	3.97 ± 0.97	1.24 ± 0.14	1.02 ± 0.09	0.75 ± 0.08
	RIF	50	4.42 ± 1.6	3.08 ± 0.31	1.78 ± 0.21	1.62 ± 0.17	2.02 ± 0.24
	PB	1000	4.74 ± 1.2	7.92 ± 1.1	2.62 ± 0.24	1.34 ± 0.15	3.20 ± 0.42
	BNF	10	0.37 ± 0.06	0.49 ± 0.06	0.95 ± 0.08	0.62 ± 0.07	1.40 ± 0.20
CYP2C9	None	–	1.0	1.0	1.0	1.0	1.0
	OMC	100	1.69 ± 0.97	1.11 ± 0.35	1.01 ± 0.13	1.09 ± 0.10	0.44 ± 0.05
	RIF	50	1.83 ± 1.3	2.04 ± 1.4	2.37 ± 0.25	3.36 ± 0.34	2.17 ± 0.29
	PB	1000	1.77 ± 0.35	2.29 ± 0.60	3.17 ± 0.35	3.43 ± 0.39	1.59 ± 0.18
	BNF	10	0.39 ± 0.12	0.45 ± 0.11	1.08 ± 0.12	1.35 ± 0.10	2.11 ± 0.22
CYP2C19	None	–	1.0	1.0	1.0	1.0	1.0
	OMC	100	1.98 ± 0.50	1.50 ± 0.15	1.17 ± 0.10	1.49 ± 0.14	1.04 ± 0.10
	RIF	50	2.97 ± 2.5	1.64 ± 0.29	5.24 ± 0.45	4.49 ± 0.52	5.45 ± 0.58
	PB	1000	3.37 ± 1.3	2.45 ± 0.61	7.68 ± 0.77	2.10 ± 0.21	2.45 ± 0.22
	BNF	10	0.67 ± 0.29	0.56 ± 0.065	1.71 ± 0.13	0.948 ± 0.11	1.96 ± 0.23
CYP2J2	None	–	1.0	1.0	–	–	–
	OMC		1.48 ± 0.32	1.62 ± 0.34	–	–	–
	RIF	50	0.66 ± 0.15	0.758 ± 0.12	–	–	–
	PB	1000	0.77 ± 0.20	0.869 ± 0.12	–	–	–
	BNF	10	0.55 ± 0.20	0.819 ± 0.30	–	–	–
CYP3A4	None	–	1.0	1.0	1.0	1.0	1.0
	OMC	100	1.60 ± 0.24	0.821 ± 0.29	0.765 ± 0.084	0.603 ± 0.061	0.719 ± 0.071
	RIF	50	18.0 ± 2.7	7.84 ± 2.7	9.99 ± 1.0	12.3 ± 1.6	14.7 ± 1.7
	PB	1000	13.6 ± 2.4	10.9 ± 4.4	21.5 ± 2.4	5.90 ± 0.69	10.1 ± 1.2
	BNF	10	0.19 ± 0.08	0.13 ± 0.04	0.947 ± 0.097	0.496 ± 0.059	2.04 ± 0.24
CYP3A5	None	–	1.0	1.0	–	–	–
	OMC	100	1.95 ± 0.85	0.523 ± 0.17	–	–	–
	RIF	50	1.39 ± 0.33	0.267 ± 0.23	–	–	–
	PB	1000	1.86 ± 0.16	0.408 ± 0.11	–	–	–
	BNF	10	0.28 ± 0.08	0.068 ± 0.02	–	–	–
UGT1A1	None	–	1.0	1.0	1.0	1.0	1.0
	OMC	100	1.77 ± 0.36	1.81 ± 0.55	1.21 ± 0.063	1.51 ± 0.17	0.706 ± 0.053
	RIF	50	1.23 ± 0.36	1.49 ± 0.20	1.11 ± 0.057	1.52 ± 0.15	0.500 ± 0.047
	PB	1000	2.21 ± 0.30	2.19 ± 0.47	1.44 ± 0.075	1.33 ± 0.17	0.721 ± 0.051
	BNF	10	1.31 ± 0.37	1.72 ± 0.20	2.43 ± 0.15	2.06 ± 0.27	1.54 ± 0.14

^aOMC, omadacycline; RIF, rifampicin; PB, phenobarbital; BNF, β-naphthoflavone.

^b Omadacycline concentrations were 1, 10, 50 and 100 μM, 100 μM results shown.

^c Rifampicin concentrations were 1, 5, 10, 25 and 50  μM, 50 μM results shown.

^d not determined. However, for CYP1A1 activity since there exists no CYP1A1-selective probe, the phenacetin probe results are reflective of both CYP1A1 and CYP1A2 activities.

Disposition of subjects and tolerability in human absorption, distribution, metabolism and excretion (ADME) study

A total of six healthy Caucasian male subjects mean (SD) age 25.7 (4.4) years, mean weight 75.5 (7.7) kg, and mean BMI 24.4 (2.6) kg/m² were enrolled in this open-label study and five completed the study. One subject withdrew consent due to a family emergency (Day 7) one day prior to the official end of study date. However, all six subjects were included in the PK and safety analyses.

There were no serious AEs and all AEs were mild in intensity. The most frequent AEs were gastrointestinal-related, including diarrhea (5 subjects), dyspepsia (2 subjects) and nausea (2 subjects), which resolved without intervention. There were no clinically relevant changes in biochemistry, hematology parameters, blood pressure or physical examination findings. Asymptomatic increases in heart rate following dose administration were observed in all subjects by examination and 12 lead ECG. The mean maximum increase from the predose heart rate was 22 beats/min and occurred at approximately 6 h after the dose. Heart rate returned to predose levels within 24 h in all subjects. No serious AEs or deaths were reported, and no subjects experienced AEs that led to study discontinuation.

Blood and plasma concentrations of radioactivity

Mean blood and plasma concentrations of radioactivity and key pharmacokinetic parameters are summarized in Table 5 (Supplemental Table S8 for individual data). The mean AUC_last of blood and plasma radioactivity was 1867 ngEq h/mL and 3096 ngEq h/mL, respectively. The mean C_max of radioactivity in blood and plasma was 608 μgEq/mL and 612 ngEq/ml, respectively, with median T_max of 2.75 and 2 h, respectively.

Table 5. Pharmacokinetic parameters (mean ± SD) after a single 300 mg oral dose of [¹⁴C]-omadacycline.

Pharmacokinetic parameters	Total blood radioactivity	Total plasma radioactivity	Plasma omadacycline
Tmax (h)^a	2.75 (range 1.5–4)	2 (range 1–3)	2.3 (range 1–4)
Cmax (ng/mL or ngEq/mL)	608 ± 73.6	612 ± 81.5	563 ± 79.5
AUC0-8^b (ng h/mL or ngEq h/mL)	NC^c	3143 ± 804	3019 ± 482
AUClast (ng h/mL or ngEq h/mL)	1867 ± 1322	3096 ± 1548	8573 ± 1941
AUCinf (ng h/mL or ngEq h/mL)	NC	NC	9418 ± 1857
T1/2 (h)	NC	NC	17.6 ± 1.45
CL/F (L/h)	NA	NA	32.8 ± 5.90
Vz/F (L)	NA	NA	828 ± 133

^aMedian (range).

^b Radioactivity only quantifiable up to 8 h due to very low levels.

^c NA/NC, not calculated or not applicable.

Parameters for total blood and plasma radioactivity could not be calculated due to insufficient data because of very low radioactivity (below quantification limit).

Omadacycline plasma concentrations

Similar to radioactivity, the omadacycline PK was associated with a median T_max of 2.3 h, with a mean C_max of 563 ng/mL. Mean CL/F for omadacycline was 32.8 L/h, mean AUC_inf was 9418 ng h/mL and mean Vz/F was 827.8 L (Figure 4, Table 5). The very low radioactivity concentrations in plasma precluded the generation of radiochromatography profiles. However, a comparison of unchanged omadacycline concentrations (as determined using the validated LC/MS bioanalytical method) to total radioactivity indicated omadacycline was the major drug-related circulating component in plasma (Figure 5), accounting for ∼96% of the drug-related radioactivity (based on AUC_0-8 h values).

Figure 4. Plasma omadacycline concentration-time profiles following a single oral dose of [¹⁴C]-omadacycline. Semilog plot is shown to illustrate biphasic kinetics.

Figure 5. Mean plasma radioactivity versus omadacycline concentration–time profiles following a single oral dose of [¹⁴C]-omadacycline (plasma radioactivity values were below quantification limit after 8 h).

Excretion and mass balance in urine and feces

Elimination of radioactivity after a single 300 mg oral dose of [¹⁴C]-omadacycline was primarily in feces (mean 81.1 ± 2.34% dose) representing primarily unabsorbed drug across the six subjects (Table 6 and Figure 6). Since only approximately 35% of the oral dose is systemically absorbed (Sun et al., 2012), it is not possible to determine the extent that hepatic excretion contributes to fecal recovery. The majority of fecal excretion (mean: 92% of fecal radioactivity, range 85–97%) occurred within 72-h postdose. Urinary excretion as a minor route (mean ± SD: 14.4 ± 2.33% of dose) represents approximately 40% of the absorbed dose. Overall dose recovery was judged to be complete (mean of 95.5%) within the 168-h collection period.

Figure 6. Cumulative urinary and fecal excretion of radioactivity following a single oral dose of [¹⁴C]-omadacycline.

Table 6. Excretion of total radioactivity (% of dose) in male subjects after a single 300 mg oral dose of [¹⁴C]-omadacycline.

	Subject
	5101	5102	5103	5104	5105	5106	Mean (SD)
Urine total	16.2	14.6	17.4	10.7	14.4	13.3	14.4 (2.33)
Omadacycline	8.28	6.95	8.48	5.10	7.26	7.45
C4-epimer	5.32	5.43	6.81	4.19	4.79	3.75
Feces total	82.1	80.0	77.5	80.1	82.8	84.0	81.1 (2.34)
Omadacycline	34.9	42.8	39.4	38.4	39.5	39.7
C4-epimer	28.2	28.1	27.9	34.5	33.1	33.4
Total in excreta	98.3	94.6	94.9	90.8	97.2	97.3	95.5 (2.73)

Analysis of omadacycline and related components in urine and feces

Pooled urine samples, representing  >90% (range: 90–95%) of the total urinary excretion of omadacycline-related radioactivity, were analyzed by LC and LC-MS/MS. Overall, all of the omadacycline-related components observed were present in the drug substance as either impurities or degradants, with no new structures indicative of metabolites. As with urine, fecal extracts (representing  >94%, range: 94–97%) of the total fecal radioactivity), also revealed no metabolites, only omadacycline and dose-related impurities and degradants. Representative radiochromatograms from urine and feces are shown in Figure 7. A summary of the omadacycline-related components observed in excreta is shown in Table 7, along with the corresponding MS results. In general, as shown in Figure 7, the C4-epimer was a prominent drug-related component in urine and feces.

Figure 7. Representative radio-profiles in human (A) urine and (B) feces, after a single oral 300 mg dose of [¹⁴C]-omadacycline. Results shown reflect pooled samples from time intervals representing ∼94% of urinary radioactivity (Subject 5103) and ∼96% of fecal radioactivity (Subject 5101).

Table 7. Structures of omadacycline and related components in male subjects following a single oral dose of 300 mg [¹⁴C]-omadacycline.

Peak ID	Exact mass MH^+	Measured mass MH^+	Mass Error (ppm)	Matrix	Proposed structures
Omadacycline	557.2970	557.2976	1.120	PlasmaUrineFeces (Dose)
C4-epimer	557.2970	557.2975	0.940	UrineFeces (Dose)
Minor or trace components^a
Impurity 2	571.3126	571.3127	0.129	UrineFeces (Dose)
4-keto	528.2340	528.2339	−0.179	UrineFeces (Dose)
M-2 impurity	555.2813	555.2799	−2.568	UrineFeces (Dose)
Omadacycline Degradation-1	557.2970	557.2969	−0.179	UrineFeces (Dose)	ND^b
Omadacycline Degradation-2	557.2970	557.2964	−1.077	UrineFeces (Dose)	ND
P26.8	573.2924	573.2932	1.395	Urine (Dose)
P28.3	573.2924	573.2914	−1.744	Urine (Dose)
P35.8	543.2819	543.2816	−0.552	Feces (Dose)

^aMinor or trace, each <1% of dose; ^bND, structure not defined

Discussion

Thorough evaluation of the drug disposition and drug–drug interaction potential of novel drugs is an important part of their development and essential to guide appropriate use in patients. Numerous examples are available of clinically significant drug–drug interactions that have led to serious complications and death (Russel, 2010). Drug interactions are attributed to a variety of factors, of which the best characterized are metabolism and/or transport-related. Metabolic interactions have been well studied and characterized, but the importance of membrane transporters has been increasingly recognized (Giacomini et al., 2010; Lai et al., 2010; Russel, 2010; Hutzler et al., 2011). Plasma and tissue concentrations of a broad variety of drugs are impacted by uptake and efflux transporters in small intestine, liver and kidney, which are critical for drug absorption, distribution, metabolism and elimination (König et al., 2013; Lu et al., 2010; Lin, 2007). Inhibition or induction of drug transporters can lead to increased toxicity or loss of efficacy (Lu et al., 2010). Tetracyclines as a drug class tend to have low drug–drug interaction potential, with the notable exception of interactions with divalent cations (i.e. Ca²⁺and Mg²⁺).

The lack of appreciable metabolism or transport of omadacycline in vitro at concentrations exceeding the expected therapeutic concentrations suggests that omadacycline is unlikely to be effected by other drugs that inhibit or induce enzymatic and/or transport pathways. Metabolic stability was found in multiple in vitro evaluations using both hepatocytes and hepatic microsomes. This metabolic stability was confirmed in the in vivo mass balance study in healthy subjects. The omadacycline C4-epimer (inversion of stereochemistry at carbon-4), which is pharmacologically inactive, was the only prominent additional chromatographic peak seen and was also present in the dose solution. Reversible epimerization at the 4-position is a well-known phenomenon for members of this antibiotic class and has been attributed to a variety of factors, such as pH, solvent and temperature (Nelson, 2001).

The absence of metabolites with omadacycline is in contrast to the major in vivo biotransformation reactions in humans for tigecycline, a structurally similar glycylcycline antibiotic, where both amide hydrolysis of the glycine-containing side chain and glucuronidation of the phenol nearest the side chain occur in vitro and in vivo (Hoffmann et al., 2007). Omadacycline lacks this amide bond and therefore is not susceptible to such a hydrolytic transformation. It also shows no evidence of glucuronidation of the phenolic hydroxyl nearby.

In addition to not being a substrate for any of the CYP isoforms tested, the results showed no relevant potential for omadacycline to inhibit or induce major CYP enzymes. This indicates a low likelihood of omadacycline as a perpetrator of CYP-mediated metabolism interactions, as would be expected for a tetracycline antibiotic.

Omadacycline was evaluated as both a potential transport substrate and inhibitor of several important transporters. The results demonstrated that only P-gp appeared capable of omadacycline transport, with a relatively high K_m (∼82 μM, compared to omadacycline C_max of ∼1 μM). Similarly, only weak inhibition potential for hOAT1 was seen with little or no inhibition observed with other transport systems. Finally, no evidence of induction of either P-gp or MRP2 was seen with omadacycline. The results from this series of studies revealed no significant interaction of omadacycline with a broad range of membrane transporters. These results suggest that drug–drug interactions based on inhibition or induction of human drug transporter activity are unlikely with omadacycline at therapeutic concentrations. Subsequent conduct of the ADME study in healthy male volunteers confirmed the lack of omadacycline metabolism. After administration of a single oral 300 mg (36.6 μCi) dose of [¹⁴C]-omadacycline to six subjects, recovery was judged to be complete (mean  >95% of dose). Elimination in feces was the predominant route of excretion accounting for ∼81% of the dose, the majority of which is likely to be unabsorbed drug (based on estimated ∼35% absorption, Sun et al., 2012). Urinary excretion of mostly drug and C4-epimer, accounted for ∼14% of dose which, accordingly, corresponds to ∼40% of the absorbed dose. In both urine and feces, no drug-related components suggesting metabolism had occurred were present, with only unchanged omadacycline and several impurities and/or degradant were detected. The most notable component in excreta other than parent omadacycline was its C4-epimer, as seen in the in vitro experiments. The lack of apparent metabolism in vivo indicates metabolic clearance is not a key factor in the clearance of omadacycline.

All six subjects experienced at least one gastrointestinal AE, but all such AEs were mild and resolved during the study. Asymptomatic increases in heart rate were observed for several hours following dosing; this finding has been seen in healthy volunteers in other Phase-1 studies of omadacycline. In contrast to this finding in healthy volunteers, no apparent effect on heart rate was observed in a Phase-2 study of omadacycline for the treatment of patients with complicated skin and skin structure infections, although monitoring was not as intense as in the healthy volunteer studies (Noel et al., 2012).

Conclusions

The in vitro and in vivo evaluation of omadacycline metabolism, disposition and drug–drug interaction properties in humans demonstrated metabolic stability (i.e. no significant biotransformation) and a lack of inhibitory or inductive effects on metabolizing enzymes or transporters. These results suggest that clinical use of omadacycline is unlikely to affect or be affected by concomitant pharmacotherapies that interfere with comedications via metabolism or transporter-related mechanisms. Similarly, it is not expected that omadacycline will be influenced by co-administration with medications that inhibit or induce metabolic enzymes.

Declaration of interest

The authors acknowledge the editorial assistance of Richard S. Perry, PharmD in the preparation of this manuscript, which was supported by Paratek Pharmaceuticals, King of Prussia, PA. Jimmy Flarakos, Yancy Du, Helen Gu, Lai Wang, Heidi J. Einolf, Dung Y. Chun, Bing Zhu, Natalia Alexander, Adrienne Natrillo, Imad Hanna, Lillian Ting, Wei Zhou, Kiran Dole, Haiying Sun, Steven J. Kovacs, Daniel S. Stein and James B. Mangold were employees of Novartis Institute for Biomedical Research at the time of these studies. SKT and SV are employees of Paratek Pharmaceuticals.

Supplementary material available online

Acknowledgements

The authors would like to thank the Bioanalytics group (Novartis, East Hanover, NJ) for providing the omadacycline concentration data used in the pharmacokinetic analysis, and the Isotope Laboratory (Novartis, East Hanover, NJ) for providing radiolabeled drug substance and study doses.