Abstract

The disposition and metabolism of bempedoic acid, a selective inhibitor of ATP citrate lyase, were examined in healthy male subjects. After a single administration of [14C] bempedoic acid (240 mg, 113 μCi) oral solution, mean concentrations of total radioactivity in plasma as a function of time indicated absorption was rapid with peak concentrations achieved at 1 hour after dose administration. Radioactivity was decreased in a multiexponential fashion with an estimated elimination half-life of 26.0 hours. Radiolabeled dose was predominantly recovered in urine (62.1% of dose) and a smaller amount in feces (25.4% of dose). Bempedoic acid was extensively metabolized with 1.6%–3.7% of dose excreted unchanged in urine and feces combined. Overall, the major clearance route of bempedoic acid is metabolism by uridine 5′-diphosphate glucuronosyltransferases. Metabolism in hepatocyte cultures of human and nonclinical species were generally in agreement with clinical metabolite profiles. Pooled plasma samples were characterized by the presence of bempedoic acid (ETC-1002), which accounted for 59.3% of total plasma radioactivity, ESP15228 (M7; a reversible keto metabolite of bempedoic acid), and their respective glucuronide conjugates. The acyl glucuronide of bempedoic acid (M6) represented 23%–36% of radioactivity in plasma and accounted for approximately 37% of dose excreted in urine. In feces, the majority of radioactivity was associated with a co-eluting mixture of a carboxylic acid metabolite of bempedoic acid (M2a), a taurine conjugate of bempedoic acid (M2c), and hydroxymethyl-ESP15228 (M2b), which collectively accounted for 3.1%–22.9% of bempedoic acid dose across subjects.

SIGNIFICANCE STATEMENT This study characterizes the disposition and metabolism of bempedoic acid, an inhibitor of ATP citrate lyase for hypercholesterolemia. This work provides further understanding of bempedoic acid clinical pharmacokinetics and clearance pathways in adult subjects.

Introduction

Alterations in lipid and lipoprotein metabolism may contribute to the pathogenesis of cardiovascular disease. Accumulated evidence has demonstrated that reductions in low-density lipoprotein cholesterol (LDL-C) achieved with statin and non-statin therapies are associated with lower cardiovascular risk (Baigent et al., 2010; Silverman et al., 2016). However, some patients with hypercholesterolemia are unable to achieve their LDL-C goals despite statin therapy at maximally tolerated doses (Boekholdt et al., 2014). Because of the limited effectiveness, tolerability, or adherence of available therapeutic options for some patients, additional lipid-modifying therapies are needed to complement statin and non-statin options to achieve LDL-C goals.

Bempedoic acid (8-hydroxy-2,2,14,14-tetramethyl-pentadecanedioic acid, ETC-1002, ESP55016, Nexletol, Nilemdo) is a first-in-class small molecule drug approved in the United States, European Union, United Kingdom, and Switzerland for the treatment of hypercholesterolemia. The activation of bempedoic acid, a dicarboxylic acid prodrug, to a pharmacologically active CoA thioester (bempedoic acid-CoA) is catalyzed by very-long-chain acyl-CoA synthetase 1 (ACSVL1; SLC27A2). Bempedoic acid-CoA is a potent and selective inhibitor of ATP citrate lyase, a key enzyme in the pathway of de novo cholesterol and fatty acid synthesis (Pinkosky et al., 2016). Inhibition of ATP citrate lyase by bempedoic acid results in decreased cholesterol synthesis and upregulation of hepatic low-density lipoprotein (LDL) receptor expression, leading to an increase in LDL particle clearance and decreased LDL-C levels in blood. In phase 3 studies, bempedoic acid 180 mg demonstrated LDL-C lowering across a range of populations, including patients with hypercholesterolemia who are intolerant to statin use and those with atherosclerotic cardiovascular disease or heterozygous familial hypercholesterolemia receiving maximally tolerated lipid-modifying therapy (Ballantyne et al., 2018; Goldberg et al., 2019; Laufs et al., 2019; Ray et al., 2019). Based on the tissue expression of ACSVL1, primarily localized in human liver, activation of bempedoic acid is not believed to occur in skeletal muscle or other tissues (Pinkosky et al., 2016).

Human pharmacokinetics (PK) of bempedoic acid are linear over the dose range of 120–240 mg, and oral administration of a 180 mg tablet was characterized by an average steady-state peak plasma concentration (mean ± SD) of 20.6 ± 6.1 µg/ml and elimination half-life of 21 ± 11 hours (data on file). The intrinsic physicochemical properties of bempedoic acid confer its low aqueous solubility and high permeability. However, bempedoic acid is a weak acid with high solubility at intestinal pH, resulting in rapid oral absorption from the gastrointestinal tract. The characteristics of pH-dependent solubility and high permeability are manifested clinically as rapid oral absorption of a 180 mg dose of bempedoic acid with a median time to peak concentration of 3.5 hours after dosing (Hanselman et al., 2020; Amore et al., 2022).

ESP15228, a keto metabolite of bempedoic acid, is activated to a CoA thioester conjugate similar to the mechanism of bempedoic acid activation (Oniciu et al., 2006). The metabolism of ESP15228 by ACSVL1 leads to the formation of ESP15228-CoA, a selective inhibitor of hepatic ATP citrate lyase activity with similar potency to bempedoic acid-CoA. Although the CoA thioesters of ESP15228 and bempedoic acid are equipotent, ESP15228 plasma exposure is approximately 82% lower than bempedoic acid levels with an average steady-state peak plasma concentration of 2.8 ± 0.9 ng/ml following bempedoic acid 180 mg tablet administration. Nonclinical studies characterized circulating metabolites of bempedoic acid in rat and cynomolgus monkey plasma that included ESP15228 and an acyl glucuronide conjugate of bempedoic acid. In rabbit, a sulfate conjugate of bempedoic acid was observed in plasma. A secondary metabolite, ESP15228-glucuronide, and minor oxidized metabolites were also observed in plasma of nonclinical species (data not included).

The studies presented here describe the disposition and metabolism of radiolabeled bempedoic acid in human subjects. The specific objectives were to 1) determine the absorption, excretion, and PK of [14C] bempedoic acid after a single 240 mg oral solution dose in healthy human subjects; 2) characterize bempedoic acid metabolites recovered in plasma, urine, and feces of human subjects; 3) compare bempedoic acid metabolite profiles of clinical samples to profiles of in vitro incubations with hepatocytes isolated from human donor tissue and nonclinical species; and 4) characterize the human enzymes involved in bempedoic acid biotransformation. The safety and tolerability of single administration bempedoic acid at the 240 mg dose level were confirmed previously, and the same dose was studied to maximize the recovery of metabolites formed by linear processes.

Materials and MethodsMaterials and Reagents

[14C] bempedoic acid, reference standards of non-labeled bempedoic acid, ESP15228, and bempedoic acid-glucuronide, and the deuterium-labeled internal standards (IS), [2H] bempedoic acid, and [2H] ESP15228, were synthesized by Ricerca Biosciences (Ann Arbor, MI). [14C] bempedoic acid (0.47 µCi/mg) was prepared with the radiolabel incorporated at each of two carbonyl carbon positions and radiochemical purity of 99.7%. Deuterium was incorporated at positions 3 and 13 of [2H] bempedoic acid and [2H] ESP15228 (Supplemental Fig. 1). The chemical purity of non-labeled bempedoic acid and ESP15228 were >98% and 94.4%, respectively. Other chemicals and solvents were obtained from commercial sources and were of analytical grade or higher.

Clinical Study

The PK and disposition of bempedoic acid following administration of a single oral dose of [14C] bempedoic acid were evaluated in a phase 1, open-label study. Six healthy male subjects 33–59 years old with body mass index of 19.7–28.5 kg/m2 and confirmed normal kidney and liver functions were enrolled. Recovery of total radioactivity in excreta was monitored up to 10 days after dosing and metabolites of bempedoic acid were characterized in plasma, urine, and feces. Safety was evaluated by monitoring for treatment-emergent adverse events (AEs). The study was conducted by Quotient Bioresearch, Ltd. (Northamptonshire, UK) at a single site in accordance with Good Clinical Practice Guidelines as defined by the International Conference on Harmonization, the Declaration of Helsinki, and/or all applicable federal and local regulations and institutional review boards, as appropriate. Approval for administration of radioactive substances was obtained from the Department of Health (UK) Administration of Radioactive Substances Advisory Committee. All subjects provided written informed consent before participation in the study.

Treatment and Sampling

Subjects received a 240.0 mg ± 1% dose (range, 237.6–242.4 mg) of [14C] bempedoic acid (specific activity, 0.47 μCi/mg) formulated as an oral solution in disodium hydrogen phosphate buffer at a final pH 7.8 with >98% chemical purity and >98% radiochemical purity (Lot 60309-13-001). Subjects were fasted with water ad libitum for approximately 8 hours overnight before dosing the next morning and continued to fast for at least 4 hours after dosing. Water was withheld for 2 hours prior to dosing and for 2 hours after dosing. At the time of dosing, subjects received water with the dose solution for a combined volume of 240 ml and received water (240 ml) again 2 hours after dose administration.

Venous blood was collected into tubes that contained potassium EDTA for analysis of whole blood and plasma. Samples of whole blood (2 ml) were collected within 1 hour prior to dose administration (pre-dose) and at 1, 2, 4, 6, 12, 24, and 48 hours after dosing and stored at −80°C until analysis for total radioactivity. Plasma samples (4–8 ml) were collected at pre-dose and at 1, 2, 3, 4, 5, 6, 8, 10, 12, 24, 48, 72, 96, 120, 144, and 168 hours after dosing. Plasma was promptly separated by centrifugation, transferred to clean tubes, and stored at −80°C until analysis. Urine was collected in polypropylene containers up to 12 hours before dosing (pre-dose) and at 0–6 hours, 6–12 hours, and 12–24 hours after dosing and at 24-hour intervals thereafter up to 240 hours post-dose. Samples were stored frozen at −80°C until analysis. Feces were collected in tarred polypropylene containers up to 1 day before dosing (pre-dose) and at 24-hour intervals up to 240 hours after dosing, weighed, and stored at −80°C until analysis. Chromatographic, radiochemical, and mass spectrometric methodologies employed to analyze clinical samples and in vitro samples are summarized in Supplemental Table 1.

Bempedoic Acid and ESP15228 Concentrations in Plasma

Bempedoic acid and ESP15228 PK were estimated from plasma concentrations determined by liquid chromatography (LC) tandem mass spectrometry (MS/MS) using a previously described method (Engel et al., 2020). Briefly, plasma was introduced to a 96-well sample plate with an ethanolic IS solution and subjected to solid phase extraction. Extracts were evaporated to dryness and reconstituted in methanol with 0.1% aqueous formic acid for LC-MS/MS analysis. Calibration ranges were 0.020–20.0 µg/ml for bempedoic acid and 0.010–10.0 µg/ml for ESP15228. Regression analyses and computation of sample concentrations were determined using the Thermo Scientific Watson Laboratory Information System (v.7.4 SP4).

Preparation of Clinical Samples for Total Radioactivity and Metabolite Profiling AnalysesUrine

Samples were thawed at 4°C, vortex mixed, and centrifuged (12,500g) to separate particulate matter. Excretion of total radioactivity in urine was determined from individual samples analyzed by liquid scintillation counting (LSC). Analyte recoveries and metabolite structure characterizations were determined from pooled urine samples, in which urine from each subject (0–120 hours) was pooled by combining equivalent aliquots based on total sample weight (1% w/w). Pooled samples were subsequently analyzed by radiochemical profiling and LC-MS/MS analyses.

Feces

Samples were thawed at 4°C and homogenized in water. Total radioactivity excreted in feces was determined by combustion of individual sample aliquots and analysis by LSC. Analyte recoveries and metabolite structure characterizations were determined from pooled samples, in which feces homogenates were pooled gravimetrically (1% w/w) for each subject (24–216 hours). Pooled homogenate samples (2 g) were extracted by adding 10 ml acetonitrile followed by vortex mixing, sonication, and centrifugation (3200g). If recovery of extracted radioactivity was low (<90%) relative to the cumulative recovery of total radioactivity from the same subject by LSC, then sample homogenates were extracted a second time and the organic layers were combined. Organic layers were transferred to clean tubes, dried under a stream of nitrogen gas, reconstituted in methanol (0.2 ml) with ammonium formate (5 mM, pH 3; 0.8 ml) and centrifuged (12,500g) to sediment particulate matter. Final extraction recoveries ranged from 89.5%–110.3%. Sample extracts were subjected to radiochemical profiling and LC-MS/MS analyses.

Plasma

Samples were thawed at 4°C and prepared by liquid–liquid extraction to determine total radioactivity, analyte recovery, and metabolite structure. Total radioactivity was determined in individual plasma samples by LSC. Analyte recovery and metabolite structure were determined from plasma samples that were pooled to estimate the area under the concentration–time curve (AUC) from time 0 to 48 hours using a strategy published previously (Hamilton et al., 1981). Pooled plasma (0.85 ml) was extracted by adding 10 ml acetonitrile followed by vortex mixing and sonication. After centrifugation, the organic layer was transferred to a clean tube and the aqueous phase was subjected to three additional extraction cycles: once with methanol and twice with methanol acidified with formic acid (0.1% v/v) using the same procedure. The combined extracts were dried under a steady stream of nitrogen gas, reconstituted in methanol (0.1 ml) and ammonium formate (5 mM, pH 3; 0.4 ml), and centrifuged (12,500g). Sample recoveries ranged from 76.3%–115.1%. Extracted plasma was analyzed by LSC to determine total radioactivity and subjected to radiochemical profiling and LC-MS/MS.

Determination of Total Radioactivity

Total radioactivity in sample aliquots was determined gravimetrically with LSC detection using a Packard 2300TR counter (Perkin Elmer, UK). Cumulative excretion and recovery in urine and feces were expressed as equivalents of radioactivity and as a percentage of administered dose. Total radioactivity in whole blood and feces samples were processed by combustion using a Perkin Elmer Automatic Sample Oxidizer (Model 307; Perkin Elmer, UK) and trapped [14C] CO2 was measured by LSC. The lower limits of detection by LSC were 17, 72, 30, and 95 ng-eq/ml for urine, feces, plasma, and whole blood, respectively.

Radiochemical Profiling and Metabolite Structure Characterization of Clinical Samples

Reconstituted plasma, urine, and feces were transferred to vials, placed in an autosampler and injected (0.1–0.2 ml) onto a Thermo Accela (Thermo Scientific, UK) high-performance liquid chromatography (HPLC) instrument equipped with a Zorbax Eclipse XBD-C8 column (150 × 4.6 mm, 3.5 µm; Agilent) arranged in series with a guard column (Security Guard C8, 4 × 3 mm; Phenomenex, UK). Analyte separation was achieved with a mobile phase mixture of 5 mM ammonium formate at pH 3 (solvent A) and 50 mM ammonium formate:methanol (10:90, v/v; solvent B) introduced at a constant flow rate of 1 ml/minute and gradient elution with solvent B starting from 50% for 3 minutes, increasing to 80% by 70 minutes, and to 100% by 70.1 minutes. The column was washed with 100% solvent B for 5 minutes and re-equilibrated at initial conditions for 5 minutes before the next injection.

Radiochromatographic analyses were conducted by in-line fraction collections every 13 seconds into Deepwell LumaPlate-96-well microplates (Perkin Elmer, UK) using a CTC HTX Pal fraction collector (Presearch, UK). Extracts were evaporated using a Savant Speedvac concentrator and analyzed using a TopCount NXT microplate scintillation counter (Perkin Elmer Life Sciences, UK). Characterization of metabolite structure was achieved by single stage and product ion spectra analyses using a Thermo Fisher Scientific LTQ Orbitrap XL mass spectrometer interfaced to the HPLC system. Mass spectral analyses were conducted using electrospray ionization in negative ion mode with nitrogen as the collision gas and capillary temperature at 320°C with an electrospray ionization voltage of 3.5 kV and tube lens voltage maintained at −56 V. Metabolite structures were derived from product ion mass spectra and elemental composition determined by exact mass measurements.

Pharmacokinetic Methods

PK parameters were estimated for bempedoic acid and ESP15228 in plasma and for total radioactivity in blood and plasma by standard noncompartmental methods of analysis using Phoenix WinNonlin version 6.3 (Certara USA, Inc., St. Louis, MO). Primary PK parameters included Cmax, time to maximum drug concentration (Tmax), AUC from time zero to last measured time point (AUC0-last), AUC from time zero extrapolated to infinity (AUCinf), and elimination half-life. Additional parameters of oral clearance (CL/F), apparent volume of distribution (Vz/F), and renal clearance (CLr) were determined for bempedoic acid. PK parameters were summarized using descriptive statistics. Blood to plasma partitioning was estimated as the ratio of blood and plasma radioactivity concentrations determined at each time point up to 48 hours after dosing.

In Vitro Plasma Protein Binding

Plasma protein binding of bempedoic acid, ESP15228, and bempedoic acid-glucuronide was determined in vitro by equilibrium dialysis. Blank human plasma collected on sodium heparin (mixed gender, Lot GLP530-5) was obtained from Bioreclamation (Westbury, NY). Stock solutions of each compound in DMSO were added to human plasma (0.300 ml) to achieve a final concentration of 5 µM (0.005% v/v final DMSO) and incubated at 37°C for 4 hours against PBS in the receiver well. Aliquots of each side of the dialysis chamber were removed to a 96-well plate where blank plasma was added to the receiver well sample, and PBS was added to the donor well sample to achieve final volumes of 0.250 ml. After addition of acetonitrile (0.500 ml) and centrifugation, supernatants were diluted with water (50% v/v) and analyzed by LC-MS/MS on a Perkin Elmer Sciex API 4000 instrument with turbo ionspray operated in positive ion mode. Detection of bempedoic acid ([MH]+, mass-to-charge [m/z] 345.4/263.5), ESP15228 ([MH]+, m/z 343.4/299.4) and bempedoic acid-glucuronide ([MH]+, m/z 521.5/345.5) was achieved using multiple reaction monitoring.

Human Liver Microsomes

The in vitro metabolism of bempedoic acid and ESP15228 was studied in human liver microsomes (HLMs) pooled from 50 donors (25 male and 25 female; Lot LBA) obtained from Celsis In Vitro (Baltimore, MD). [14C] bempedoic acid (1 and 60 µM) and [14C] ESP15228 (1 and 8 µM) were incubated with HLM (0.5 and 1.0 mg/ml protein), in which the upper bempedoic acid (60 µM, 20.7 µg/ml) and ESP15228 (8 µM, 2.74 µg/ml) substrate concentrations were selected to approximate their corresponding clinical steady-state Cmax of 20.6 ± 6.1 µg/ml and 2.8 ± 0.9 µg/ml, respectively. Stock solutions of bempedoic acid and ESP15228 were prepared in ethanol and all incubations were carried out in triplicate with a final ethanol concentration <1% (v/v).

Microsomal incubations were conducted in two sets of experiments. In a first set, incubations were conducted in the presence and absence of 1 mM NADPH to evaluate the potential formation of oxidized metabolites. Radiolabeled substrate was pre-incubated with microsomes in 100 mM phosphate buffer (pH 7.4) at 37°C for 5 minutes and reactions were initiated by the addition of NADPH cofactor and maintained at 37°C. In a second set of experiments, incubations in the presence and absence of 2 mM uridine diphosphoglucuronic acid (UDPGA) were conducted to evaluate the formation of glucuronide metabolites. HLM mixtures were pretreated with 0.025 mg/ml alamethicin, 20 mM saccharic acid 1,4-lactone, and 4 mM MgCl2 in phosphate buffer at 4°C for 10 minutes. Following pre-treatment, microsomal mixtures were preincubated with radiolabeled substrate at 37°C for 10 minutes and incubation reactions were initiated by UDPGA cofactor addition and maintained at 37°C. In both experiments, HLM incubations were terminated after 0, 15, 30, and 60 minutes with addition of ice-cold acetonitrile (1 ml) and samples were vortex mixed and centrifuged to remove precipitated proteins. Supernatants were transferred to clean tubes and stored at −20°C until analysis.

Recombinant UDP-Glucuronosyltransferases

The in vitro metabolism of [14C] bempedoic acid (60 µM) and [14C] ESP15228 (8 µM) was studied in cDNA-expressed recombinant uridine 5′-diphospho-glucuronosyltransferase (UGT; 1.0 mg/ml protein) enzyme preparations in the presence of 2 mM UDPGA at 37°C. Recombinant UGT enzymes (Supersomes) prepared from baculovirus-infected insect cells were procured from Corning Inc. (Woburn, MA): UGT1A1 (Lot 3273576), UGT1A3 (Lot 3296661), UGT1A4 (Lot 3303828), UGT1A6 (Lot 4036003), UGT1A9 (Lot 3287625), UGT2B7 (Lot 3287588), and UGT2B15 (Lot 3247829). Incubation reactions with each recombinant UGT enzyme were conducted in triplicate and terminated after 0 and 60 minutes with addition of ice-cold acetonitrile containing 6% acetic acid v/v (1 ml). Samples were vortex mixed and centrifuged with transfer of the resultant supernatants to a clean tube and stored at −20°C until analysis by HPLC with radiometric detection. Individual UGT enzyme (0.5 mg/ml protein) activities were confirmed using 4-methylumbelliferone (100 µM) as a positive control. The rates of 4-methylumbelliferone glucuronidation were ≥1.43 nmol/min/mg protein (range, 1.43–19.8 nmol/min/mg protein) across UGT enzyme preparations.

Hepatocyte Cultures

The in vitro metabolism of [14C] bempedoic acid at 29.1 µM (1.2 µCi) and 72.6 µM (3.0 µCi) was studied in freshly isolated hepatocytes from CD-1 mouse (male, n = 4), Wistar rat (male, n = 1 and female, n = 1), cynomolgus monkey (female, n = 1), and human (donor 1, female; donor 2, male; donor 3, female) livers at Quotient Bioresearch. Animals were obtained from Quotient Bioresearch stock or were sourced from commercial suppliers. Human hepatocytes were isolated from resected surgical liver tissue obtained from donor subjects 64–80 years old. The viability of cell preparations was determined using trypan blue and ranged from 74.3% to 87.9% in preclinical species and was 82.0%, 87.0%, and 91.0% viable in human hepatocytes from donor livers 1, 2, and 3, respectively. All hepatocyte incubations were conducted in Hepatocyte Maintenance Media supplemented with Hepatocyte Maintenance Media SingleQuot (insulin, gentamicin sulfate, and amphotericin) sourced from Lonza Bioscience (Slough, UK).

[14C] bempedoic acid was incubated in 24-well plates (0.5 ml/well) at final concentrations of 29.1 µM and 72.6 µM (10 and 25 µg/ml, respectively) with approximately 0.35 × 106 viable cells per incubation across species, except for mouse hepatocyte incubations, which contained approximately 0.2 × 106 viable cells per incubation. Reactions were terminated by addition of 0.25 ml acetonitrile at 0, 2, 6, 12, and 24 hours, and incubation mixtures were transferred to clean tubes and stored at −80°C until analysis. Samples were sonicated and supernatants isolated by centrifugation were analyzed by HPLC with radiometric detection.

HPLC radiochromatograms from each species were evaluated for metabolite formation and disappearance of the parent [14C] bempedoic acid. The metabolic rate, expressed as picomoles [14C] bempedoic acid consumed/h/106 viable cells, was assessed after 12 hours when hepatocytes from all species exhibited approximately linear reaction rates. The amount of each metabolite formed and [14C] bempedoic acid remaining were expressed as a percentage of total eluted sample radioactivity for each chromatogram. Bempedoic acid metabolites were characterized by LC-MS/MS from selected incubation samples.

HPLC Radiochromatography and Metabolite Structure Characterization of In Vitro Samples

Biotransformation samples from incubations with HLM, recombinant UGTs, and hepatocytes were analyzed by HPLC with radiometric detection using an Agilent 1200 Series instrument interfaced with a LabLogic β-Ram Model 4B radiodetector. Conditions for analyte separation and characterization of metabolite structure using an LTQ Orbitrap XL mass spectrometer were identical to those for clinical sample analyses, as described above.

Results[14C] Bempedoic Acid Disposition in Human Subjects

A single 240 mg (113 μCi) dose of [14C] bempedoic acid was administered as an oral solution to six healthy male subjects (race: White [n = 4/6, 66.7%], Black [n = 1/6, 16.7%], and Asian [n = 1/6, 16.7%]). Subjects (33–59 years of age) had a median (range) body weight of 79.3 kg (58.2–89.4 kg) and body mass index of 25.2 kg/m2 (19.7–28.5 kg/m2). One subject experienced mild headache, which was possibly related to the study drug. No serious AEs, deaths, or clinically important changes in laboratory assessments, vital signs, electrocardiograms, or physical examinations were reported during the study. Total recovery of radioactivity was monitored in excreta for 240 hours after dosing (Fig. 1). The cumulative recovery of radioactivity in urine and feces combined was 87.4% of dose, with less than 5% of dose being excreted unchanged. Excretion of radioactivity in urine was the dominant route of elimination, accounting for 62.1% of dose recovery on average. Approximately 25.4% of dose was recovered in feces. Approximately 50% of total recovery was achieved within 72 hours of radioactive dose administration.

Fig. 1.

Mean (± SD) cumulative percent of radioactivity excreted in urine and feces and in combined excreta after single oral dose administration of 240 mg [14C] bempedoic acid. Mean cumulative recovery of radioactivity was 87.4% of dose, with approximately 62.1% of dose excreted in urine and 25.4% in feces.

The distribution of radioactivity between whole blood and plasma was determined after [14C] bempedoic acid dosing. Geometric mean estimates of the ratio of blood to plasma radioactivity ranged from 0.518 to 0.572 over the 0- to 48-hour sampling time course, with corresponding geometric mean (CV%) blood:plasma ratios of 0.489 (6.7%) for Cmax and 0.498 (14.8%) for AUCinf parameters.

Mean (± SD) plasma concentration–time profiles of total radioactivity, bempedoic acid, and ESP15228 after administration of a single oral solution dose of 240 mg [14C] bempedoic acid are shown in Fig. 2. The time course of total radioactivity and bempedoic acid concentrations in plasma were similar, rapidly attaining peak levels at 1 hour after oral dosing and characterized by similar multiexponential profiles. Geometric mean (CV%) estimates of elimination half-life were 26.0 (15.9%) hours for total radioactivity and 22.4 (24.0%) hours for bempedoic acid (Table 1). The corresponding ESP15228 concentration–time profile indicated peak concentrations were attained at 11.0 hours with a similar elimination half-life of 31.1 (26.7%) hours as observed for bempedoic acid. A comparison of AUCinf estimates indicates bempedoic acid accounted for approximately 59.3% of total radioactivity in plasma and approximately 90.4% of peak radioactivity concentration. Based on AUCinf exposure, the metabolites ESP15228, bempedoic acid-glucuronide, and ESP15228-glucuronide accounted for approximately 12.1%, 28.9%, and 10.4% of total radioactivity in plasma, respectively (Table 1).

Fig. 2.

Mean (± SD) plasma concentrations of total radioactivity, bempedoic acid, and ESP15228 after single oral dose administration of 240 mg [14C] bempedoic acid. Inset: Linear concentration–time profiles plotted from time 0 to 48 hours.

TABLE 1

Pharmacokinetic parameters after a single oral 240 mg dose of [14C] bempedoic acid

Bempedoic acid and ESP15228 PK parameters were estimated from plasma concentrations determined by quantitative analysis using analytical standards and deuterium-labeled IS. LC-MS/MS analysis of bempedoic acid was achieved with an assay lower limit of quantitation of 20.0 ng/ml, inter-run precision of 1.3%–11.8% CV, and accuracy of −1.0% to 1.8% mean bias. ESP15228 was determined with a lower limit of quantitation of 10.0 ng/ml, inter-run precision of 3.1%–8.6% CV, and accuracy of −4.8% to 6.3% mean bias. Geometric mean estimates of bempedoic acid CL/F, Vz/F, and CLr were 808 ml/h, 26.1 L, and 8.55 ml/h, respectively. Circulating concentrations of ESP15228 were lower relative to parent, as mean estimates of ESP15228 Cmax and AUCinf were approximately 3.5% and 20.4% of bempedoic acid exposures, respectively (Table 1). The observed ratio of metabolite to parent AUCinf after bempedoic acid 240 mg dosing is consistent with AUCinf exposure ratio (18%) reported previously following single bempedoic acid 180 mg dose administration (Amore et al., 2021b).

Accurate Mass LC-MS/MS Analyses and Quantitative Radiometric HPLC Profiles of Plasma, Urine, and Feces

A representative chromatographic profile of plasma collected from subject 1 and pooled over 0–48 hours was characterized by four main peaks of radioactivity (Fig. 3A). Following a single 240 mg bempedoic acid dose, the most abundant analyte in circulation was bempedoic acid (ETC-1002) with a retention time (r.t.) of 65.41 minutes and accounted for approximately 37.3% of the radioactivity in circulation up to 48 hours post-dose. Radioactivity associated with bempedoic acid ranged from approximately 37%–58% of total radioactivity injected across pooled plasma samples (0–48 hours) from six subjects (Table 2). Bempedoic acid was characterized by accurate mass LC-MS/MS operated in negative ion mode with a molecular ion at m/z 343.2489 and sodium adduct with m/z 365.2309, consistent with a molecular weight of 344 Da. Accurate mass LC-MS/MS analyses and molecular formula assignments are shown in Table 3. A product ion spectrum obtained by collision-induced dissociation of m/z 343 (nominal mass-to-charge ratio) yielded a base peak ion at m/z 299 formed by a neutral loss of 44 Da that corresponds to the loss of CO2, and a fragment ion with m/z 297 representing subsequent dehydration. Product ions at m/z 325 and m/z 281 correspond to the loss of water from m/z 343 and m/z 299, respectively, while m/z 253 represents a loss of CO2 from m/z 297. Peak assignments and results of LC-MS/MS analyses for bempedoic acid and metabolites are summarized in Supplemental Table 2.

Fig. 3.

Representative HPLC radiochromatograms of pooled (A) plasma (0–48 hours), (B) urine (0–120 hours), and (C) feces (24–216 hours) from subject 1 following single 240 mg oral dose administration of [14C] bempedoic acid. CPM, counts per minute; ETC-1002, bempedoic acid; HPLC, high-performance liquid chromatography; M1, ESP15228-taurine; M2, mixture of bempedoic acid-carboxylic acid, hydroxymethyl-ESP15228 and bempedoic acid-taurine; M3, ESP15228-glucuronide; M4, mixture of hydroxymethyl-bempedoic acid and hydroxymethyl-bempedoic acid-taurine; M5, mixture of ETC-1002-sulfate and unassigned analyte; M6, bempedoic acid-glucuronide; M7, ESP15228.

TABLE 2

Distribution of bempedoic acid and metabolite-related radioactivity in plasma, urine, and feces following single 240 mg oral dose administration of [14C] bempedoic acid

TABLE 3

Accurate mass LC-MS/MS analyses

In addition to bempedoic acid, three analytes present at lower abundances in the pooled plasma sample from subject 1 (Fig. 3A) included bempedoic acid-glucuronide (M6; r.t. 48.13 minutes), ESP15228 (M7; r.t. 60.97 minutes), and ESP15228-glucuronide (M3; r.t. 41.96 minutes). The proportions of radioactivity ranged from approximately 23%–36% for bempedoic acid-glucuronide, 8%–12% for ESP15228, and 7%–18% for ESP15228-glucuronide across pooled plasma samples from six subjects. Bempedoic acid-glucuronide (M6) was characterized by [M-H]− of m/z 519.2803 and sodium adduct with m/z 541.2624, consistent with a molecular weight of 520 Da. The product ion spectrum of M6 revealed a base peak ion with m/z 343 that resulted from a neutral loss of 176 Da, representing the loss of glucuronic acid. Product ions observed at m/z 299 and m/z 297 were formed by neutral loss of 44 Da with subsequent dehydration of the aglycone bempedoic acid, respectively. ESP15228 (M7) was characterized by [M-H]− of m/z 341.2333 and sodium adduct with m/z 363.2154, consistent with a molecular weight of 342 Da. Similar to the fragmentation pattern of bempedoic acid, the product ion spectrum of ESP15228 included ions at m/z 297, corresponding to loss of CO2, m/z 295, and m/z 253 representing the loss of CO2 from m/z 297. A product ion at m/z 139, postulated to represent the loss of CO2 and 2,2-dimethyl-heptanoic acid with subsequent dehydration, was also observed. ESP15228-glucuronide (M3) was characterized by [M-H]− of m/z 517.2647 and sodium adduct with m/z 539.2468, consistent with a molecular weight of 518 Da. The product ion spectrum of M3 indicated a neutral loss of 176 Da to yield a product ion with m/z 343, corresponding to the aglycone ESP15228, plus ions at m/z 297, m/z 295, m/z 253, and m/z 139, consistent with the fragmentation pattern observed for ESP15228 (M7). The product ion spectra of bempedoic acid and M7 in plasma were confirmed by LC-MS/MS analyses of bempedoic acid and ESP15228 analytical standards, respectively.

The excretion of radioactivity in urine was the dominant route of [14C] bempedoic acid elimination. A representative chromatogram of a pooled urine sample (0–120 hours) from subject 1 showed three main peaks of radioactivity (Fig. 3B). The most abundant analyte in urine was bempedoic acid-glucuronide (M6), which accounted for approximately 37.3% of total radioactivity excreted over 120 hours. Radioactivity associated with bempedoic acid-glucuronide ranged from approximately 32%–42% of total radioactivity across the pooled urine samples from the six study subjects. ESP15228-glucuronide metabolite (M3) and bempedoic acid were also excreted in urine, accounting for 10%–14% and 0.9%–3.3% of excreted radioactivity, respectively (Table 2). The presence of bempedoic acid, M3, and M6 in urine was confirmed on the basis of comparisons of chromatographic retention times and product ion spectra of corresponding plasma analytes. A small peak with r.t. 52.33 minutes represented an average of 1.3% of radioactive dose and was not characterized due to its relatively minor contribution to overall [14C] bempedoic acid dose recovery.

The excretion of radioactivity in feces accounted for approximately 25.4% of [14C] bempedoic acid dose. A representative chromatographic profile of pooled feces extracts (24–216 hours) from subject 1 revealed six main peaks of radioactivity (Fig. 3C). The proportions of radioactive dose recovered in feces ranged from approximately 0.1%–1.9% for bempedoic acid and approximately 0.1%–1.5% for ESP15228 (M7) across six study subjects. Two small peaks with a r.t. of 34.80 minutes and 55.78 minutes represented an average of 0.52% and 0.15% of radioactive dose, respectively, but due to the minor contribution to overall [14C] bempedoic acid dose recovery, these smaller peaks were not characterized further. In addition to bempedoic acid and ESP15228, several primary and secondary metabolites were also present in human feces. The first peak (M1, r.t. 32.34 minutes) was characterized by [M-H]− at m/z 448.2374 and sodium adduct with m/z 470.2197, consistent with a molecular weight of 449 Da, and accounted for approximately 2.9% of dose. Across pooled samples from six subjects, peak M1 represented 0.1%–3% of [14C] bempedoic acid dose. LC-MS/MS analyses revealed a base peak at m/z 384, corresponding to the loss of SO2, with multiple smaller product ions resulting from combinations of CO2 and water loss from m/z 384. From the observed fragmentation of m/z 448.2374, taken together with a proposed molecular formula of C21H38O7NS, M1 was assigned as the taurine conjugate of ESP15228 (Table 2 and Supplemental Table 2).

HPLC radiochromatographic peaks M2 (r.t. 38.26 minutes), M4 (r.t. 43.20 minutes), and M5 (r.t. 47.64 minutes) represented an average of 11.0%, 2.85%, and 0.92% of the radioactive dose recovered in feces, respectively (Table 2). Each of these peaks was comprised of a mixture of analytes due to incomplete chromatographic resolution. Mass spectrometric analyses of peak M2 indicated a mixture of three co-eluting analytes that accounted for 3%–23% (11.0% mean) of dose excreted across samples from six subjects and included a carboxylic acid metabolite of bempedoic acid (M2a; [M-H]− = m/z 373.2240), hydroxymethyl-ESP15228 (M2b; [M-H]− = m/z 357.2292), and a taurine conjugate of bempedoic acid (M2c; [M-H]− = m/z 450.2530). Peak M4 was also a mixture of co-eluting analytes that accounted for 2%–6% (2.85% mean) of dose excreted. The molecular ion of metabolite M4a has an even m/z value that suggested the presence of a nitrogen atom (M4a; [M-H]− = m/z 466.2309), as observed with the metabolites M1 and M2c. In addition, the molecular ion of M4a differed from metabolites M1 and M2c by 18 Da and 16 Da, respectively. Subsequent interrogation of M4a indicated a fragmentation pattern similar to bempedoic acid-taurine (M2c), with losses of SO2 (64 Da) plus OH and CO2 to yield a base peak with m/z 342. The product ion at m/z 324 can be similarly rationalized after loss of HSO3 from the molecular ion. Taken together, M4a was assigned as the taurine conjugate of hydroxymethyl-bempedoic acid. The second analyte of peak M4 (M4b; [M-H]− = m/z 359.2441) has a molecular ion that differed from bempedoic acid by 16 Da and was assigned as hydroxymethyl-bempedoic acid. Peak M5 contained bempedoic acid-sulfate (M5a; [M-H]− = m/z 423.2067) plus an unidentified analyte (M5b; [M-H]− = m/z 448.2743). Product ion formation was insufficient to further characterize M5b. As a mixture, metabolites M5a and M5b accounted for 0.3%–2.7% (0.92% mean) of dose excreted across samples from six subjects.

Hepatocyte Incubations

Human hepatocyte incubations (three donor livers; one male [donor 2], two female) were conducted at 29.1 µmol/L (10 µg/ml; 1.2 µCi) and 72.6 µmol/L (25 µg/ml; 3.0 µCi) [14C] bempedoic acid concentrations. The extent of metabolism, estimated by the disappearance of bempedoic acid over a 24-hour incubation period, was variable across donor hepatocytes ranging from 62.1% (donor 1) to 96.0% (donor 3) at 10 µg/ml bempedoic acid and from 58.5% (donor 1) to 92.8% (donor 3) at 25 µg/ml bempedoic acid. The rates of bempedoic acid metabolism in hepatocyte cultures were linear up to 12 hours, with metabolism rates of 2150, 3710, and 5530 pmol/h/106 viable cells (donors 1, 2, and 3, respectively) determined at the 12-hour time point of incubations with 25 µg/ml initial substrate concentration.

HPLC radiochromatographic profiles of 10 µg/ml [14C] bempedoic acid incubations were characterized by parent drug eluting at 63.2 minutes and a large peak (m3) eluting at 45.2 minutes plus the presence of several minor peaks (Supplemental Fig. 2A). After 24 hours, the chromatographic peak assigned to bempedoic acid accounted for 3.91%–36.9% of total eluted radioactivity and peak m3 accounted for 50.0%–86.3% of eluted radioactivity. Due to incomplete chromatographic resolution, peak m3 represents a pair of co-eluting analytes that included bempedoic acid-glucuronide and hydroxymethyl-bempedoic acid. Using mass spectrometry to monitor the molecular ions eluting at 45.2 minutes in full scan mode, the ion intensity associated with bempedoic acid-glucuronide ([M-H]− = m/z 519) was approximately 66-fold greater than the intensity of hydroxymethyl-bempedoic acid ([M-H]− = m/z 359) after incubation of bempedoic acid (25 µg/ml) for 24 hours. Although quantitative assignment of the relative contributions of bempedoic acid-glucuronide and hydroxymethyl-bempedoic acid to the radiochromatographic m3 peak is not possible based on relative ion intensities, mass spectrometric analyses suggest peak m3 is primarily comprised of radioactivity associated with bempedoic acid-glucuronide. Higher relative formation of bempedoic acid-glucuronide in human hepatocytes would also be consistent with higher average levels of radioactivity assigned to bempedoic acid-glucuronide in human plasma (29.4% of total radioactivity on column) and urine (36.2% of dose).

The time courses of metabolite formation in hepatocytes at 10 µg/ml and 25 µg/ml [14C] bempedoic acid starting concentrations were characterized (Supplemental Figs. 2B and 2C). The profiles show decreases in bempedoic acid parent over time with linear increases of peak m3. In addition to bempedoic acid and peak m3, radiochromatographic peak m2 (39.1 minutes) accounted for 2.83%–5.26% of total eluted radioactivity and was characterized as the putative ESP15228-glucuronide by LC-MS/MS ([M-H]− = m/z 517). The remaining minor radiochromatographic peaks of the 24-hour incubation profiles shown in Supplemental Fig. 2A each accounted for less than 5% of total eluted radioactivity. Peaks eluting at 38.0 minutes (m1), 55.62 minutes (m4), and 58.8 minutes (m5) were tentatively assigned as a glycerol conjugate of bempedoic acid-glucuronide ([M-H]− = m/z 593), glycerol conjugate of bempedoic acid ([M-H]− = m/z 417), and ESP15228 ([M-H]− = m/z 341), respectively. There was general concordance between metabolites formed in human hepatocyte incubations and metabolites of bempedoic acid observed clinically, with predominant formation of bempedoic acid-glucuronide.

A comparison of hepatocytes isolated from human and nonclinical species revealed that bempedoic acid metabolites ESP15228-glucuronide (m2), bempedoic acid-glucuronide (m3), bempedoic acid-glycerol (m4), and ESP15228 (m5) were formed upon incubation of [14C] bempedoic acid in human hepatocytes and in cynomolgus monkey, rat, and mouse hepatocyte cultures. A glycerol conjugate of bempedoic acid-glucuronide (m1) was formed in hepatocytes from human, cynomolgus monkey, and rat, but not from mouse (data not shown). The rates of bempedoic acid metabolism in hepatocyte cultures were linear up to 12 hours across species at 10 µg/ml and 25 µg/ml [14C] bempedoic acid. Average rates of metabolism were similar in 25 µg/ml [14C] bempedoic acid incubations with human (3790 ± 1700 pmol/h/106 viable cells), cynomolgus monkey (3350 pmol/h/106 viable cells), and mouse (2740 pmol/h/106 viable cells) hepatocytes and slower with male and female rat (619 and 698 pmol/h/106 viable cells, respectively) hepatocytes.

Human Liver Microsomal Incubations with NADPH

[14C] bempedoic acid and [14C] ESP15228 were each incubated with human hepatic microsomes and the disappearance of substrate was monitored in the presence and absence of NADPH cofactor. Incubations of 60 µM [14C] bempedoic acid in 1 mg/ml microsomal protein and NADPH showed 98.2% ± 0.21% of initial substrate concentration remained after 60 minutes, with the formation of ESP15228 accounting for approximately 1% total radioactivity. Incubations starting with 8 µM [14C] ESP15228 in 1 mg/ml microsomal protein plus NADPH showed 94.3% ± 0.07% of initial substrate remained, with bempedoic acid accounting for approximately 5.3% total radioactivity after 60 minutes. Substrate turnover was not observed in the incubations conducted without NADPH. Human liver microsomal incubations of [14C] bempedoic acid and [14C] ESP15228 with NADPH provided supportive in vitro evidence for the reversible metabolism of bempedoic acid.

Human Liver Microsomal Incubations with UDPGA

[14C] bempedoic acid and [14C] ESP15228 were incubated with human hepatic microsomes pretreated with alamethicin and saccharic acid 1,4-lactone to monitor substrate disappearance in the absence and presence of UDPGA. Incubations of [14C] bempedoic acid (60 µM) in 1 mg/ml microsomal protein and UDPGA showed bempedoic acid-glucuronide accounted for approximately 2% total radioactivity after 60 minutes with 97.3% ± 0.27% of initial substrate concentration remaining. Incubations starting with [14C] ESP15228 (1 µM) in 1 mg/ml microsomal protein plus UDPGA showed the highest rate of ESP15228-glucuronide formed after 60 minutes, accounting for approximately 6.3% total radioactivity with 91.1% ± 0.78% of initial substrate remaining. Substrate turnover was not observed in the incubations conducted without UDPGA.

Recombinant UGT Incubations

[14C] bempedoic acid and [14C] ESP15228 were incubated with individual cDNA-expressed UGT enzymes and formation of glucuronide conjugates were observed after 60-minute incubations with UGT2B7, but not with UGT1A1, UGT1A3, UGT1A4, UGT1A6, UGT1A9, or UGT2B15. At concentrations that approximated their observed clinical Cmax, incubations of [14C] bempedoic acid (60 µM) or [14C] ESP15228 (8 µM) with UGT2B7 resulted in formation of bempedoic acid-glucuronide (2.02% ± 0.190% total radioactivity) and ESP15228-glucuronide (7.59% ± 0.490% total radioactivity) metabolites, respectively. Conditions of substrate depletion were ruled out by the high fraction of [14C] bempedoic acid (95.7% ± 0.380%) and [14C] ESP15228 (90.6% ± 0.370%) remaining at the end of a 60-minute incubation period.

Plasma Protein Binding

Bempedoic acid, bempedoic acid-glucuronide, and ESP15228 were exogenously added to blank human plasma and the extent of protein binding was determined by equilibrium dialysis. In human plasma, bempedoic acid (5 µM), bempedoic acid-glucuronide (5 µM), and ESP15228 (5 µM) were 99.3%, 98.8%, and 99.2% bound to plasma protein, respectively.

Discussion

The PK, metabolism, and disposition of bempedoic acid were characterized in six healthy male adult subjects following single bempedoic acid 240 mg oral solution administration. To augment the mass balance recovery of metabolites formed by linear kinetic processes, a 240-mg dose of [14C] bempedoic acid was selected. Although higher than the approved dose of 180 mg, no serious AEs were reported after single-dose administration, indicating that bempedoic acid 240 mg was safe and well tolerated among healthy subjects in the current study and consistent with a previous study that administered a single 240 mg dose (Amore et al., 2021a).

Total recovery of radiolabeled dose was 87.4% (range, 78.3% –94.1%) over a 10-day collection period, with an average of 62.1% recovered in urine and 25.4% in feces. Bempedoic acid was the primary analyte in plasma, accounting for 59.3% of total radioactivity AUCinf exposure. By contrast, 1.6%–3.7% of administered dose was recovered as unchanged drug in urine and feces combined, with CLr (8.55 ml/h) representing approximately 1% of bempedoic acid CL/F (808 ml/h). On average, less than 1% of dose was recovered as parent drug in feces, suggesting that incomplete oral absorption did not contribute to radioactivity recovered in feces. These mass balance results indicate that bempedoic acid was extensively absorbed (>95%) with low recovery of unchanged drug in feces and rapidly absorbed after oral solution administration, as evidenced by plasma Cmax of parent drug accounting for 90.4% of total radioactivity at Tmax. The rate and extent of bempedoic acid absorption is consistent with its high passive permeability and pH-dependent solubility. As a weak acid, bempedoic acid is relatively insoluble at gastric pH but extremely soluble at intestinal pH conditions. Based on low recoveries of unchanged drug in urine and feces, metabolism represents the main pathway of bempedoic acid elimination with a majority of the [14C] bempedoic acid dose being excreted in urine. Although elimination is predominantly by metabolic clearance, bempedoic acid 180 mg was shown to be well-tolerated in patients with mild and moderate hepatic impairment, and corresponding PK exposures (Cmax and AUC) were similar to those in subjects with normal liver function (Amore et al., 2021b).

Analyses of specimens from healthy subjects who received a single [14C] bempedoic acid dose revealed the formation of multiple oxidized and conjugated metabolites of bempedoic acid (Fig. 4). Pooled plasma samples were characterized by the presence of bempedoic acid, the primary metabolites ESP15228 (M7) and bempedoic acid-glucuronide (M6), and a secondary metabolite ESP15228-glucuronide (M3). These circulating metabolites represent the principal pathways of bempedoic acid metabolism.