Abstract
Lurbinectedin is a novel and potent selective inhibitor of active transcription of protein-coding genes, triggering apoptosis of cancerous cells. It has been approved for the treatment of patients with metastatic small-cell lung cancer with disease progression on or after platinum-based chemotherapy. Studies exploring the disposition and metabolism of lurbinectedin were performed in vitro and in vivo (by intravenous administration of lurbinectedin). Low blood cell partitioning for lurbinectedin in rats, nonhuman primates (NHP), and humans was determined as 23.4%, 29.8%, and 9.8%, respectively. Protein binding was very high (>95%) in total plasma (rat, NHP, and human), albumin, and α-1-acid glycoprotein (both human). In vitro, lurbinectedin underwent intense liver microsome–mediated metabolism—in 10 minutes, 80% of the compound is metabolized in human—with CYP3A4 being the isoform involved in that metabolism. Results also showed NHPs being the nonclinical species which, metabolically, most closely resembles humans. Mass balance studies performed in rats (both genders), NHPs (male only), and patients (both genders) demonstrated that the principal route of excretion of 14C-lurbinectedin–related radioactivity was through the feces (88.7% ± 10.1% in patients), with only a minor fraction recovered from the urine (5.6% ± 2.0% in patients). In plasma samples, the majority of lurbinectedin-related radioactivity was attributed to unchanged compound (95% ± 3.1% and 70.2% ± 10.9% in NHPs and humans, respectively). Plasma metabolic profiling demonstrated the major (% compared with unchanged compound) circulating metabolites were N-Desmethyl-lurbinectedin (0.4% ± 0.2% and 10.4% ± 2.2% in NHPs and patients, respectively) and 1',3′-Desmethylene-lurbinectedin (0.9% ± 0.7% and 14.3% ± 10.4% in NHP and patients, respectively).
SIGNIFICANCE STATEMENT Lurbinectedin is a novel and potent selective inhibitor of active transcription of protein-coding genes, triggering apoptosis of cancerous cells, and was recently approved for the treatment of patients with metastatic small-cell lung cancer with disease progression on or after platinum-based chemotherapy. The present study provides a complete set of information on the pharmacokinetics, biotransformation, and elimination of 14C-lurbinectedin and its metabolites, following a single intravenous administration to nonclinical species (rats and nonhuman primates) and patients.
Introduction
Lurbinectedin [(1R,6′'R,6a′R,7′R,13′S,14′S,16′R)-8′,14′-dihydroxy-6,9′-dimethoxy-4′,10′,23′-trimethyl-19′-oxo-2,3,4,6′,7′,9,12′,13′,14′,16′-decahydro-6a′H-spiro[β-carboline-1,20′-[7,13]epimino[6,16](epithiopropanooxymethano)[1,3]dioxolo[7,8]isoquino[3,2-b][3]benzazocin]-5′-yl acetate] is a synthetic tetrahydroisoquinoline alkaloid (Fig. 1). Lurbinectedin-induced antitumor activity is the result of a specific inhibition of transcription, this being produced by the selective binding to CG-rich sequences, mainly those located around promoters of protein-coding genes. This induces an irreversible stalling of elongating RNA Pol II on the DNA template, and its specific degradation by the ubiquitin/proteasome machinery that finally leads to the generation of DNA breaks and subsequent apoptosis in tumor cells (Santamaria Nunez et al., 2016). In transcriptionally addicted tumor cells, lurbinectedin causes a detachment of transcription factors from their target promoters, inhibiting the block of its transactivating activity (Di Giandomenico et al., 2014; Harlow et al., 2016), as well as reducing the tumor-associated macrophages and the inflammatory tumor microenvironment (Cespedes et al., 2016; Belgiovine et al., 2017). Lurbinectedin demonstrated a significant in vivo activity in a wide variety of human tumors xenografted in mice (Leal et al., 2010; Vidal et al., 2012; Romano et al., 2013; Cespedes et al., 2016; Harlow et al., 2016; Takahashi et al., 2016), which led to the initiation of the clinical development of lurbinectedin. The results gathered from the first in-human phase I study in patients with advanced solid tumors, following the administration of 1-hour intravenous infusion of lurbinectedin every 3 weeks (Elez et al., 2014), demonstrated that myelosuppression (mainly consisting of nonfebrile, brief, transient, grade 4 neutropenia) was the primary toxicity at the recommended dose (4 mg/m2). Other adverse events were fatigue (mild/moderate), nausea, and vomiting. Owing to the relevance of the results gathered from a recent clinical trial (Trigo et al., 2020), lurbinectedin has been approved in the United States (https://www.fda.gov/), the United Arab Emirates (https://www.mohap.gov.ae/), Australia (https://www.tga.gov.au/), Singapore (https://www.hsa.gov.sg/), and Canada (https://www.canada.ca/) for the treatment of patients with metastatic small-cell lung cancer (SCLC) with disease progression on or after platinum-based chemotherapy.
Chemical structure of lurbinectedin. The site for one 14C substitution was C4; the four 2H substitutions in the internal standard were located at C3 and C4.
This article summarizes the results obtained from a series of in vitro and in vivo experiments aimed at describing the disposition and metabolism of lurbinectedin in nonclinical species, namely rats and nonhuman primates (NHPs) (cynomolgus; Macaca fascicularis). The results gathered from an open-label, mass balance clinical trial (including metabolite identification) of lurbinectedin in patients with advanced cancer, are also presented.
Material and Methods
Test Articles
Lurbinectedin, as well as reference standards for primary metabolites, namely N-Desmethyl-6-Hydroxy-lurbinectedin (Metabolite 1; coded as PM030036), 6-Hydroxy-lurbinectedin (Metabolite 2; coded as PM01158), N-Desmethyl-lurbinectedin (Metabolite 6; coded as PM030047), and 14’-Dehydroxy-lurbinectedin (Metabolite 7; coded as PM030779), and the impurity 8’,11’-Dioxo-lurbinectedin (coded as PM080323) were synthetized by PharmaMar S.A. (Colmenar Viejo, Spain). The metabolite 1′,3′-Desmethylene-lurbinectedin (or Metabolite 4) was obtained by the method described elsewhere (Altares et al., 2019). 14C-lurbinectedin (specific activity, 54 μCi/mg; radiochemical purity, 92.2%, coded as PM100567) labeled at substituted C4 indole (Fig. 1), which was proven to be metabolically stable (both, in vitro and in vivo), was also synthetized by PharmaMar S.A., as well as the deuterated lurbinectedin (Fig. 1) (coded as PM040038) that was used as the bioanalytical internal standard.
Chemical and Reagents
Human and rat plasma were purchased from Sera Laboratories, Ltd. (Sussex, UK); NHP plasma was obtained from the animal science laboratory at Aptuit (Verona, Italy). Human albumin and α-1-acid glycoprotein were purchased from Sigma-Aldrich (St. Louis, MO). Rat, NHP, and human blood were sourced from Aptuit (Verona, Italy). For the in vitro metabolism experiments, pooled batches of rat, NHP, and human liver microsomes were obtained from Xenotech (Kansas City, KS). Pooled human liver microsomes (BD Biosciences; San Jose, CA), as well as CYP2B6 and CYP2C19 supersomes (Corning Inc.; Corning, NY) were used for the identification, by specific chemical inhibition, of the cytochrome P450 isoforms involved in the oxidative metabolism of lurbinectedin. The specific chemical inhibitors used in these experiments were obtained either from Merck KGaA (Darmstadt, Germany): furafylline, 8-methoxypsoralen, sulfaphenazole, quinidine, diethyldithiocarbamate, ketoconazole, montelukast; Cypex Ltd. (Dundee, Scotland, UK): 3-benzyl-phenobarbital; or TargetMol (Boston, MA): sertraline. The CYP-specific chemical substrates used were verapamil, phenacetin, coumarin, diclofenac, omeprazole, dextromethophan, chlorzoxazone, testosterone, paclitaxel, midazolam (all from Merck KGaA; Darmstadt, Germany), and efavirenz (TargetMol; Wellesley Hills, MA). Glucose-6-phosphate dehydrogenase, glucose-6-phosphate and NADP were provided by Corning Inc. (Corning, NY). All other chemicals were reagent grade unless otherwise stated.
Blood Cell Partitioning Determination
Five milliliters of rat, NHP, or human whole blood (all collected with K3EDTA as anticoagulant) were spiked with 14C-lurbinectedin to reach a final concentration of 200 and 500 nM (157 and 393 ng/mL, respectively; 1 nM = 0.79 ng/mL). Following 15 minutes of equilibration time at 37°C, samples were centrifuged at 2000g for 10 minutes at 5°C, after which 2 aliquots were taken (250 μL each). The first aliquot was centrifuged to obtain plasma; the second was solubilized by Solvable (Perkin Elmer, Waltham, MA), added with 30% hydrogen peroxide, and incubated for at least 2 hours at approximately 50°C. Then, both solubilized blood and plasma were assayed for radioactivity as described below. Before and after incubation, blood was collected into glass capillaries and hematocrit (Hto) determined by centrifuging at 15,000g for 10 minutes. Blood cell partitioning (BCP, %) was determined by applying the following equation:
Where Cp and Cb represent lurbinectedin concentration in plasma and blood, respectively. Each experiment was carried out in triplicate. Details on the determination of equilibration time, as well as compound stability in whole blood, are not described in this article.
Plasma Protein, Albumin, and α-1-Acid Glycoprotein Binding
Rat and NHP (male only) plasma, human (pooled genders) plasma, human albumin (50 and 25 mg/mL), and α-1-acid glycoprotein (1.5 and 0.5 mg/mL) were fortified with lurbinectedin to reach a final concentration of 500, 200, 50, and 10 nM (392, 157, 39.2, and 7.8 ng/mL, respectively). Solutions were gently mixed for 15 minutes at 25°C to equilibrate prior to ultrafiltration. Samples were transferred to Centrifree ultrafiltration devices with an Ultracel regenerated cellulose membrane (30 KDa) (Merck KGaA; Darmstadt, Germany), centrifuged (1500g at 25°C) for either 5 minutes (albumin and α-1-acid glycoprotein) or 15 minutes (plasma) and then retained and ultrafiltrated fluids were analyzed by LC-MS/MS for lurbinectedin concentration (see Method 1). The binding (B, %) of lurbinectedin to proteins was determined by using the equation:
Where C0 and Cu represent the concentration of lurbinectedin at t = 0 minutes and in the ultrafiltrate, respectively. Each experiment was done in triplicate.
In Vitro Microsome Incubations and Metabolic Profiling
Incubations of either 14C-lurbinectedin or lurbinectedin with liver microsomes from rats and NHPs (both genders), as well as from humans (pooled both genders), were carried out at a protein concentration of 0.5 mg/mL in a total volume of 1 mL. The cofactor mixture in each incubate contained 3.3 mM of glucose-6-phosphate, 0.4 U/mL units of glucose-6-phosphate dehydrogenase, 3.3 mM of MgCl2, and 1.3 mM of NADP+ in 1.0 mL of a 0.1 M potassium phosphate buffer pH 7.4 (Corning Inc.; New York, NY). Lurbinectedin or 14C-lurbinectedin solutions were prepared to give a final concentration of 400 nM (314 ng/mL) in 1 mL of enzymatic reaction. After 5 minutes of the preincubation phase, the reaction was started with the addition of the NADPH regenerating system, according to the supplier’s instructions (BD Biosciences; San Jose, CA). Reaction was incubated for 10 minutes at 37°C under continuous shaking. Following each incubation (replicated 10 times per species and gender) with activated microsomes, the enzymatic reaction was stopped by adding 1 mL of cold acetone with 0.1% formic acid (1:1, v/v). Then the mixture was centrifuged (15,500g at room temperature). Radioactivity was determined in the pellet and in 100 µL aliquot from the supernatant. Another 100 μL aliquot from the supernatant was injected into the HPLC system (see Method 2). Column eluting fractions were collected and the recovered radioactivity was determined in the column eluent that was collected into 100 different fractions (15 seconds each; cumulative collection in the same vial coded as F1 to F100), using a 1260 Infinity Fraction Collector AS (Agilent Technologies, Inc.; Santa Clara, CA). This procedure was repeated in each incubation, collecting each fraction in the same vial. After mixing with 5 mL of scintillation liquid (Ultima-FLO M; PerkinElmer; Waltham, MA), the radioactivity was measured by a scintillation liquid counter (Hidex 300 SL; Hidex; Turku, Finland). The value of radioactivity (dpm) obtained in each fraction was plotted versus time (minutes), and then a radiochemical chromatogram was constructed for each species evaluated. In these radiochemical chromatograms, any fraction with radioactivity values 2× higher than the corresponding fraction in the blank solvent was considered for further calculations.
Chemical Inhibition of Cytochrome P450 Activities
Lurbinectedin (200 nM; 157 ng/mL) was incubated (in triplicate) at 37°C in NADPH-activated human liver (protein concentration, 0.5 mg/mL) microsomes or supersomes with CYP-selective chemical inhibitors: furafylline 10 μM (CYP1A2); 8-methoxypsoralen 10 μM (CYP2A6); sertraline 100 μM (CYP2B6); montelukast 10 μM (CYP2C8); sulfaphenazole 10 μM (CYP2C9); phenobarbital 10 μM (CYP2C19); quinidine 3 μM (CYP2D6); diethyldithiocarbamate 100 μM (CYP2E1); and, ketoconazole 10 μM (CYP3A4). At different time points (0, 5, 10, 15, 20, and 25 minutes) an aliquot (100 µL) of the incubation mixture was added to cold acetone 0.1% with formic acid (100 µL), centrifuged (10000g at 5°C for 10 minutes), and the resulting supernatant subjected to LC-MS/MS analysis (see Method 3). Blank control (lurbinectedin incubation without NADPH) and blank reagent (incubation with NADPH, without lurbinectedin) were included as a control in each replicated incubation.
Mass Balance Experiments
All animal protocols were reviewed and approved according to regional Institutional Animal Care and Use Committees. Animals (rats and NHPs) received a single intravenous bolus administration of 14C-lurbinectedin at the maximum tolerated dose, which was determined in separate experiments that are not detailed in this article. Sprague-Dawley rats (approximately 7 to 9 weeks of age) were purchased from Charles River (Milano, Italy), whereas NHPs (approximately 4.5 years old) were obtained from LCL-Cynologics IBL House (Port Louis, Mauritius) and quarantined at SILABE ADUEIS (Strasburg, France).
The open-label, mass balance clinical trial of lurbinectedin in patients with advanced cancer was conducted in conformance with the International Conference on Harmonization guidelines for Good Clinical Practice and in accordance with the Declaration of Helsinki. The protocol was approved by the Netherlands Cancer Institute Independent Ethics Committee. All patients signed the written informed consent before they were enrolled in the clinical trial.
Mass Balance in Rats
Sprague-Dawley rats (N = 27/gender) were intravenously administered at 1.2 mg/m2 (11.2 μCi/kg; male) and 0.6 mg/m2 (5.6 μCi/kg; female) of 14C-lurbinectedin (the maximum tolerated dose in male and female, respectively) and placed in stainless steel metabolic cages. After 0.25, 0.5, 1, 2, 4, 8, 24, 48, and 72 hours of compound administration, blood samples were collected (N = 3/gender) in K3EDTA-containing tubes from the abdominal aorta of euthanized animals. In a separate experiment, additional animals (N = 3/gender) were dosed with the compound (at the same dose levels) and then refrigerated urine (intervals: 0–8, 8–24 hours and then daily) and feces (daily) were collected for a period of 7 days. Residual radioactivity in body carcasses was determined at the end of the experimental period. Also, the biliary excretion of 14C-lurbinectedin-related material was studied separately. Bile duct–cannulated Sprague-Dawley rats (N = 3/gender) were treated at the same dose levels, and their bile was then collected at 0–8, 8–24, 24–48, and 48–72 hours postadministration; feces and carcasses were also sampled at the end of the experiment (72 hours). Residual radioactivity was measured by daily rinsing with ethanol:water (1:1, v/v) of the metabolic cages and by washing with an aqueous solution of Radiacwash (Nukepills Inc.; Edgewater, MD) at the end of the collection period (168 hours). Mixtures of feces and ultrapure water (1:1.5, w:v) were homogenized (Precellys 24 Dual; Bertin Instruments; Montigny-le-Bretonneux, France), and then weighed homogenates were mixed with Solvable (PerkinElmer; Waltham, MA), 30% hydrogen peroxide, isopropyl alcohol and incubated for 2 hours at 50°C. Animal carcasses were solubilized with a solution of 40% KOH in methanol (w:v) and further mixed with 30% hydrogen peroxide. Radioactivity was measured in fecal homogenates (4/sampling time), solubilized carcasses (N = 4/sample) and in original liquid samples (namely, urine, bile, plasma, cage rinse; N = 2/sample).
Mass Balance in NHPs
Three non-naïve, male NHPs were dosed at 1.7 mg/m2 (9.4 μCi/kg) of 14C-lurbinectedin by intravenous route and then housed in metabolic cages. At selected time points (0.08, 0.25, 0.5, 1, 2, 4, 8, 24, 48, 72, 96, 120, and 144 hours) postdose, animals were sampled from the cephalic/femoral vein or the femoral artery, and blood collected in K3EDTA-containing tubes and divided into 2 aliquots: the first (blood) was kept at 5°C; the second aliquot was centrifuged and plasma obtained and divided into 2 aliquots, which were stored at −80°C until further analysis. In the first aliquot, 14C-lurbinectedin-related total radioactivity was determined by liquid scintillation counting; the second plasma aliquot was precipitated by adding (1:3, v/v) of cold acetone 0.1% formic acid, and then supernatants were analyzed by LC-MS/MS (see Method 3) to identify the parent compound and its main metabolites. Protein pellets and the three consecutive rinses were counted by liquid scintillation counting for residual 14C-lurbinectedin-related radioactivity retained in plasma proteins.
Urine was collected, refrigerated at intervals of 0–8, 8–24, 24–48, 48–72, 72–96, 96–120, 120–144, and 144–168 hours after compound administration; feces were collected daily up to 7 days postdose and weights recorded. Metabolic cages were rinsed daily with ethanol/water (1:1; v/v) and washed with an aqueous Radiacwash (Nukepills Inc.; Edgewater, MD) solution at the end of the collection period (168 hours).
Feces samples were homogenized with ultrapure water (1:1; w:v). Further, weighed aliquots (4/sampling time) of homogenates were solubilized by Solvable (PerkinElmer; Waltham, MA) and 30% hydrogen peroxide for 2 hours at 50°C; blood samples were treated likewise. All samples were then counted for radioactivity as described below.
Mass Balance in Patients
Six patients with advanced cancer (namely, oropharynx carcinoma, melanoma, pancreatic adenocarcinoma, breast cancer, neuroendocrine carcinoma, and cancer of unknown primary origin) were enrolled in a phase I, open-label, uncontrolled study to characterize the mass balance following a single administration of 14C-lurbinectedin as 1-hour intravenous infusion at 5 mg (flat dose). The median age (range) of patients was 59 (52–68) years. Infusion bags were prepared by PRA Health Sciences (Groningen, the Netherlands) by adding the appropriate amounts of 14C-lurbinectedin and nonradiolabeled lurbinectedin to give a final solution containing exactly 5.0 mg of lurbinectedin, with 100 μCi of radioactivity, in 124 mL of sodium chloride (0.9%). Infusion bags were weighed before and after drug infusion, and then the exact dose (14C-lurbinectedin and nonradiolabeled lurbinectedin) administered to each patient was calculated; subsequent doses of lurbinectedin were prepared with a nonradiolabeled compound at a dose of 3.2 mg/m2. Venous blood samples (from the contralateral arm to intravenous infusion) were collected in Vacutainer K3EDTA tubes (Becton, Dickinson and Company; Franklin Lakes, NJ) at the following time points: predose, just before the end of infusion (EOI), 0.08, 0.25, 1, 1.5, 2, 4, 6, 8, 10, and 12 hours, and on the morning of days 2–8. Feces were collected until less than 1% of the administered radioactive dose was excreted over 24 hours for two consecutive days (cessation criteria). Urine was collected at predose and during the following intervals: 0–12 hours, 12–24 hours, and then daily until the cessation criteria were met. Blood samples were aliquoted (2 × 200 μL each) and added with Solvable (PerkinElmer, Waltham, MA), 30% hydrogen peroxide, NaEDTA, and incubated at approximately 40°C until samples became translucent. The remaining blood, was centrifuged and the plasma obtained was used for total radioactivity counting determination. Feces samples were digested by incubating (until transparency reached) a mix with Solvable (PerkinElmer, Waltham, MA), 30% hydrogen peroxide and isopropyl alcohol at 40°C. Plasma and urine samples were counted for total radioactivity without performing any previous treatment. Metabolic profiling was evaluated in plasma, feces, and urine samples frozen at −70°C until LC-MS/MS (see Method 4) analysis was carried out.
Determination of Total Radioactivity
Aliquots of samples (weighed fecal homogenates, solubilized animal carcasses, and original liquid samples, such as urine, blood, plasma, or cage rinses) were added with an appropriate volume of scintillation cocktail (Ultima Gold; PerkinElmer; Waltham, MA) and then total radioactivity was measured by liquid scintillation counting using Tri-Carb 2900 TR liquid scintillation spectrometer (PerkinElmer; Waltham, MA).
LC-MS/MS Methods
Four LC-MS/MS methods were used throughout these experiments. The first method was used in the quantitation of lurbinectedin in samples from in vitro studies of binding to plasma proteins albumin and α-1-acid glycoprotein. The second method was used for the quantitation of lurbinectedin and metabolic profiling in samples from both in vitro and in vivo studies (NHPs). The third method was dedicated to the quantitation of lurbinectedin in samples obtained from the in vitro chemical inhibition of Cytochrome P450. Lurbinectedin and its metabolites in patient-derived samples were quantified by applying the fourth LC-MS/MS method. In all samples obtained from both NHP and patients, the quantitation of the compounds was accomplished by using the corresponding MS/MS transitions obtained for lurbinectedin and/or 14C-lurbinectedin and its related metabolites.
Method 1 (Used for Lurbinectedin Analysis in In Vitro Studies of Plasma Binding to Proteins Albumin and α-1-Acid Glycoprotein)
Lurbinectedin was analyzed in samples already prepared by adding cold acetone with 0.1% formic acid (1:1, v/v) and then centrifuged. Lurbinectedin was separated on a 2.1 × 50 mm column, packed with 1.7 µm BEH Phenyl (Acquity; Waters; Milford, MA) and kept at 45°C. Chromatography (Acquity UPLC; Waters; Milford, MA) was carried out (at a flow rate of 0.7 mL/min) with 0.1% formic acid in 10 mM ammonium formate (A) and acetonitrile (B) using isocratic conditions (A:B; 60:40, v/v). Detection was done by tandem mass spectrometry (Sciex API5000; AB SCIEX, Framingham, MA) operated in the positive mode. Multiple reaction monitoring was used for quantification.
Method 2 (Used for the Quantitation of Lurbinectedin and Metabolic Profiling in Microsomes Incubations and Plasma Samples from NHPs)
Samples were prepared by adding (1:1, v/v in microsomes; 1:3, v/v in plasma) cold acetone with 0.1% formic acid and then centrifuged. Supernatants were subjected to a gradient phase chromatography in which analytes (lurbinectedin and its metabolites) were separated on an ACE C18 PFP, 3 μm, and 2.1 × 150 mm column (Advanced Chromatography Technologies; Aberleen, UK) heated to 50°C. Liquid chromatography (Agilent 1290 infinity; Agilent Technologies, Inc.; Santa Clara, CA) was carried out at a flow rate of 0.4 mL/min using a gradient of water (A) in acetonitrile (B), both with 0.1% formic acid: 90 to 70% A from 0.0 to 12.0 minute; 70 to 30% A from 12.0 to 18.0 minute; 30 to 10% A from 18.0 to 20.0 minute; isocratic for 1.0 minute and then returning to initial conditions in 0.1 minute, kept isocratic up to the minute 25.0. Relevant information on MS/MS fragmentation of lurbinectedin is presented in Fig. 3.
Method 3 (Used for Quantitation of Lurbinectedin In Vitro Chemical Inhibition of Cytochrome P450)
Lurbinectedin was analyzed in samples already prepared by adding cold acetone with 0.1% formic acid (1:1, v/v) and then centrifuged. Lurbinectedin was separated on a 2.1 × 30 mm column, packed with 3 μm ACE C18 PFP (Advanced Chromatography Technologies; Aberleen, UK) and kept at 50°C. Gradient chromatography (Agilent 1290 infinity; Agilent Technologies, Inc.; Santa Clara, CA) was carried out (at a flow rate of 0.6 mL/min) using 0.1% formic acid in water (A) and 0.1% formic acid in acetonitrile (B) with the following profile: gradient to 90% to 10% A from 0.0 to 2.5 minute; isocratic 10% A from 2.5 to 3.5 minute; and then initial conditions during 1.5 minute. Detection was by tandem mass spectrometry (API4000 or API4500 QTrap; AB SCIEX; Framingham, MA) operated in the positive mode. Multiple reaction monitoring was used for quantification.
Method 4 (Used for the Quantitation of Lurbinectedin and Its Metabolites in Plasma and Urine Samples Obtained in the Clinical Trial in Patients)
A previously described method used for the quantification of lurbinectedin in human plasma and urine (van Andel et al., 2018) was modified to optimize the analysis of lurbinectedin-related metabolites (Method 2). In short, 100 µL of plasma samples and 10 µL of urine diluted with 90 µL control human K3EDTA plasma (to prevent adsorption of the drug to the tubes) were pretreated with tert-butyl methyl ether (TBME). Samples were dried by gentle N2 stream (at 25°C), and dried extracts were resuspended in 0.1% formic acid in water:acetonitrile (70:30; v/v). Liquid chromatography (UPLC Acquity I class; Waters; Milford, MA) was done on a 150 × 2.1 mm column, packed with Zorbax-SB C18, 5 μm, (Agilent Technologies, Inc.; Santa Clara, CA) and kept at a temperature of 50°C. The mobile phase was a mixture of (A) 0.1% formic acid in water:acetonitrile (90:10; v/v) and (B) 0.1% formic acid in acetonitrile:water (90:10; v/v). The flow was 0.4 mL/min. Following an isocratic period of 1 minute (A/B; 100/0), a gradient was used from (A/B; 100/0) to (A/B; 83/17) over 31 minutes, then to (A/B; 0/100) in 0.1 minute, kept until 34 minutes and back to initial conditions in 0.1 minute; column was then reconditioned for 3 minutes. Detection was performed by tandem mass spectrometry (Sciex API5500 Q-Trap; AB SCIEX; Framingham, MA) with turbo ion spray, operated in the positive ion mode.
Control human plasma and urine were spiked with lurbinectedin to prepare calibration standards (8 concentration levels from 0.1 to 100 ng/mL) and quality control (QC) samples (at 0.3, 3, and 75 ng/mL). Separate calibration standards and QC samples were spiked for the metabolites in plasma (ranging from 0.1 to 25 ng/mL) and in urine (ranging from 1 to 250 ng/mL). The linear regression of the peak area ratio versus the concentration was weighted by 1/χ2 (the reciprocal of the squared concentration). For every calibration standard, the concentrations were back-calculated from the responses. At least 75% of the nonzero standards were within ±15% of the actual value (or ±20% for the lower limit of quantification) and at least 50% of the QC samples at each concentration level and 2/3 of all QC samples were within ±15% of the nominal concentration for each analytical run, hence adequate assay performance in terms of intra- and interassay accuracy and precision was demonstrated.
To allow the direct comparison between the concentrations calculated from the total radioactivity counting, metabolite calculated concentrations were converted into the ng-equivalent (eq)/mL of lurbinectedin by multiplying the concentration of the given metabolite by the molecular weight ratio of the metabolite and lurbinectedin.
Structural Elucidation of Metabolites
The structure of lurbinectedin-related metabolites found in either in vitro or in vivo experiments was elucidated based on molecular mass, charged molecular ions, and collision-induced dissociation fragments (Oliveira and Watson, 2000). All the structures of the proposed lurbinectedin-related metabolites were fully confirmed by comparison of chromatography retention times and fragmentation patterns obtained with authentic synthetic standards, which were subjected to 1H NMR structural elucidation (with the exception of 1′,3′-Desmethylene-lurbinectedin; Metabolite 4).
Data Analysis
The total radioactivity excreted in urine, bile, and feces was expressed as a percentage of the given radioactivity. Individual plasma concentrations at nominal sampling times were used to assess pharmacokinetic parameters using standard noncompartmental methods via the WinNonlin 6.3 Phoenix64 software (Pharsight Co.; St. Louis, MO). Results were presented as mean ± S.D., unless otherwise specified. Statistical analyses (2-tailed Mann–Whitney U test) were done and graphs plotted using GraphPad Prism, version 9.2.0 (GraphPad Software; San Diego, CA).
Results
Blood Cell Partitioning
The blood cell partitioning in rats, NHPs, and humans was determined by spiking whole blood with 14C-lurbinectedin to reach a final concentration of 200 and 500 nM (157 and 393 ng/mL, respectively). Regardless of the concentration assayed, low values for blood cell partitioning were determined for rats (23.2%–23.6%), NHPs (29.7%–29.8%), and humans (9.5%–10.6%), these values being in agreement with the lack of compound distribution (evaluated as lurbinectedin-related radioactivity) into blood cells determined in the in vivo studies carried out either in nonclinical species (rats and NHPs) or patients.
Plasma Protein, Albumin, and α-1-Acid Glycoprotein Binding
The binding to plasma proteins was studied by LC-MS/MS analysis of lurbinectedin-fortified, ultrafiltrated plasma at 10, 50, 200, and 500 nM (7.8, 39.2, 157, and 392 ng/mL, respectively) from rats, NHPs, and humans. Regardless of either the species examined or the concentration of lurbinectedin, a very high percentage of binding to proteins of plasma from rats, NHPs, and humans was found (>95%). Lurbinectedin (at identical concentrations as above) displayed a very high percentage of binding to physiologic concentrations of specific human-obtained proteins, namely albumin (50 and 25 mg/mL) and α-1-acid glycoprotein (1.5 and 0.5 mg/mL).
In Vitro Microsome-Mediated Lurbinectedin Metabolism, Metabolism Profiling, and Cytochrome P450 Isoforms Involved
Incubations (at 37°C) of 14C-lurbinectedin (400 nM; 315 ng/mL; 0.02 μCi) with rat-, NHP-, or human-derived liver microsomes (0.5 mg/mL of protein) in the presence of NADPH regeneration were kept for 10 minutes, stopped by the addition of cold acetone (with 0.1% of formic acid), and then centrifuged. In supernatants, recovered radioactivity ranged from 82% (rat, female) to 88% (rat, male; human, pooled); resulting pellets retained between 13% (rat, male) and 23% (NHP, male) of the initially incubated 14C-lurbinectedin-related radioactivity. After the incubation period, 14C-lurbinectedin was nearly or even completely metabolized (% remaining unchanged) by liver microsomes from NHPs (5.9% ± 0.1% and 5.2% ± 1.2% male and female, respectively; N = 3/gender), humans (17.8% ± 5.8% pooled-both-genders), and, to a lesser extent, rats (65.1% ± 8.9% and 76.9% ± 11.6% male and female, respectively; N = 3/gender).
Thus, supernatants (100 μL) obtained from 10 independent incubations (triplicated per species and gender) of 14C-lurbinectedin with microsomes were subjected to HPLC separation by using chromatographic Method 3 (see above). Each of the 100 column-eluting fractions (15 seconds each) was accumulated in the same vial (coded as F1 to F100) and then counted for radioactivity content. Reconstructed radiochemical chromatograms (counts/fraction versus time) obtained for each microsomal incubation per gender and species are displayed in Fig. 2, which strongly suggests that NHPs and humans have the most qualitatively and quantitatively similar microsome-mediated metabolism of 14C-lurbinectedin. The corresponding integration of these reconstructed radiochemical chromatograms are summarized in Table 1. Results suggest a relationship between the total postcolumn recovery and the metabolic rate, which may be explained by an apparent lack of radioactivity within the small, nonintegrable peaks, likely belonging to redundant metabolites. Two main 14C-lurbinectedin-related metabolites were identified and coded as Metabolite 4 and Metabolite 6, respectively—the latter being found in a similar amount, regardless of the species or gender assayed. However, Metabolite 4 was far more abundant in NHPs and humans than in rats. Two other low-intensity peaks were found exclusively in NHPs and humans and coded as Metabolite 1 and Metabolite 2. Altogether, these results demonstrated the evident metabolic dissimilarities between rodent and nonrodent (NHP and human) species. The structures of these lurbinectedin-related metabolites were confirmed by the analysis of fragmentation patterns obtained with standards, as well as by the comparison of retention times following the chromatography of those standards.
Representative reconstructed radio-chromatograms following 10-minute incubations of 14C-lurbinectedin with NADPH-activated liver microsomes from (A) both genders pooled humans, (B) male NHPs, (C) female NHPs, (D) male Sprague-Dawley rats, and (E) female Sprague-Dawley rats. Dotted line: change of component B (acetonitrile, 0.1% formic acid) in the gradient. Metabolites coding: M1 or Metabolite 1, N-Desmethyl-6-Hydroxy-lurbinectedin; M2 or Metabolite 2, 6-Hydroxy-lurbinectedin; M4 or Metabolite 4, 1',3′-Desmethylene-lurbinectedin; M6 or Metabolite 6, N-Desmethyl-lurbinectedin.
Comparative quantitative in vitro metabolism of 14C-lurbinectedin in microsomes from nonclinical species (both genders) and humans (pooled) following 10 min of incubation.
Italics denote metabolites that were produced from primary lurbinectedin-related metabolites.
The mass spectrum of Metabolite 6 revealed a protonated molecular ion at m/z 771.5, corresponding to a 14-amu shift relative to lurbinectedin (m/z 785.3; Fig. 3); the same 14-amu shift was noticed in the specific protonated molecular ion MS/MS fragment, m/z 479.2 (Fig. 4). The chromatography of PM030047 confirmed that Metabolite 6 is the result of N-Desmethylation at 23′ position in the lurbinectedin structure (Fig. 5). The metabolite coded as Metabolite 4 (m/z 773.1) was shifted 12 amu as compared with lurbinectedin (Fig. 3), as well as the protonated molecular ion MS/MS specific fragments (m/z 481.1 versus 493.0, respectively; Fig. 4). The chromatography of the biologically obtained standard demonstrated that Metabolite 4 is 1′,3′-Desmethylene-lurbinectedin, as a result of the aliphatic ring opening at 1′ and 3′ positions of the parent compound (Fig. 5). There were two other minor peaks identified as lurbinectedin-related metabolites, namely Metabolite 2 and Metabolite 1 (Table 1). The protonated molecular ion MS/MS specific fragment of M2 (m/z 259.1; Fig. 4) corresponds to a 14-amu shift relative to the equivalent specific fragment of lurbinectedin (m/z 273.1; Fig. 4), suggesting the lack of a methyl group as the more likely metabolic change produced in lurbinectedin; HPLC analysis of PM01158 confirmed Metabolite 2 as the O-Desmethylated lurbinectedin (6-Hydroxy lurbinectedin; Fig. 5). The protonated molecular ion of Metabolite 1 had lost 28 amu (m/z 757.3; Fig. 4) compared with lurbinectedin, and the MS/MS specific fragments (m/z 259.1 and 479.2; Fig. 4), as well as the HPLC of the compound PM030036, strongly suggest that this metabolite is the result of a di-Desmethylation, leading to N-Desmethyl-6-Hydroxy-lurbinectedin (Fig. 5). The other two minor compounds, coded as Metabolite 3 and Metabolite 5, were identified as the aliphatic ring opening and the oxidation of Metabolite 6 (data not shown), respectively (Table 1). Therefore, these two compounds, rather than true lurbinectedin metabolites, were likely to be the result of an artifact, owing to the very intensive in vitro metabolism, which might produce a metabolite-of-metabolite (redundant in vitro metabolism).
Mass spectrometric fragmentation pattern of lurbinectedin.
MS/MS spectra of (A) lurbinectedin, (B) Metabolite 1 (PM030036), (C) Metabolite 2 (PM01158), (D) Metabolite 4, and (E) Metabolite 6 (PM030047).
In vitro metabolic pathway of lurbinectedin in liver microsomes of nonclinical species (rats and NHPs) and humans. Dotted arrows represent potential redundant metabolism. Bold arrows indicate major human and NHP metabolites.
As a summary, a proposal for the in vitro metabolic pathway of lurbinectedin in NADPH-activated liver microsomes is depicted in Fig. 5. Lurbinectedin was metabolized via several biotransformation pathways including N-Desmethylation, aliphatic ring opening, and/or O-Desmethylation.
To identify the specific cytochrome P450 isoforms involved in its metabolism, lurbinectedin (at 200 nM; 157 ng/mL) was incubated with pooled human liver microsomes in the presence of cytochrome P450–selective chemical inhibitors. The disappearance of lurbinectedin as a function of incubation time in the absence and presence of cytochrome P450 isoform–selective chemical inhibitors is displayed in Fig. 6. Following a 20-minute incubation period with pooled human liver microsomes, only 2.0% ± 0.1% of lurbinectedin remained unchanged. The selective chemical inhibition of the different cytochrome P450 isoforms resulted in residual (mean ± S.D.; N = 3) lurbinectedin percentages of 2.0% ± 0.4% (CYP1A2), 3.0% ± 1.0% (CYP2A6 and CYP2C9), 5.0% ± 1.0% (CYP2C8), 2.0% ± 1.0% (CYP2C19), 2.0% ± 0.3% (CYP2D6), 2.0% ± 0.5% (CYP2E1), and 99.0% ± 4.0% (CYP3A4/5). After a 20-minute incubation in Supersomes, results demonstrated the lack of CYP2B6- and CYP2C19-mediated metabolism. As such, the percentage (median) of unchanged lurbinectedin was 87% and 71% (P = 0.100; CYP2B6) and 96% and 86% (P = 0.300; CYP2C19) without and with chemical inhibition of each isoform, respectively. Results demonstrated that the metabolism of lurbinectedin was completely abolished by ketoconazole, a selective chemical CYP3A4 inhibitor, this isoform being considered, therefore, as the single responsible of lurbinectedin metabolism.
Effect of CYP (between brackets) isoform-selective chemical inhibitors on the transformation of lurbinectedin (200 nM; 157 ng/mL) in pooled human liver microsomes; (mean, N = 3).
Mass Balance in Rats
A mass balance study was performed in which Sprague-Dawley rats received a single intravenous bolus dose of 14C-lurbinectedin at 1.2 mg/m2 (male) and 0.6 mg/m2 (female). Blood samples were collected (N = 3/gender) at 0.25, 0.5, 1, 2, 4, 8, 24, 48, and 72 hours postdose. In another experiment, additional animals (N = 3/gender) were dosed with the compound, urine and feces were collected for a period of 7 days, and then the carcasses were analyzed for residual radioactivity. Also, the biliary excretion of lurbinectedin-related material was studied separately. Bile duct–cannulated Sprague-Dawley rats (N = 3/gender) were treated at the same dose levels and their bile was then collected up to 72 hours postadministration. The multicompartment distribution of the concentration-time curves in blood and plasma were similar in both genders, with Cmax values (64.2 and 33.0 ng-eq/g in blood of males and females, respectively) obtained at 0.25 hours postdose (Fig. 7). Although males were treated at a higher dose (2×) than females, similar systemic exposure to total lurbinectedin-related radioactivity (AUC0-t) was obtained in both genders (340 and 235 ng-eq·h/g in males and females, respectively). The combined recovery of radioactivity in carcasses and excreta was about 99%. The data indicated that in rats, the excretion of radioactivity was quick, with 3% (male) and 6% (female) of radioactivity recovered from the carcass at 7 days postdose. In this species, the principal route of excretion was through the feces (91.0% ± 2.1% and 91.2% ± 1.7% of total radioactivity in males and females, respectively; N = 3/gender), with only a minor fraction recovered from the urine (3.4% ± 1.4% in males and 2.6% ± 0.5% in females; N=3/gender). From these data, no gender differences were recorded (Table 2).
Representative concentration-time curves of total radioactivity in plasma (solid symbol) and blood (open symbol) following a single intravenous dose of 14C-lurbinectedin in (A) rats (mean, N = 3): at 1.2 mg/m2 (male) and 0.6 mg/m2 (female); (B) NHPs (mean ± S.D., N = 3; male): at 1.7 mg/m2; and, (C) patients (mean ± S.D.; N = 6; both genders: 5 males, 1 female): at 5 mg flat dose (∼2.3 mg/m2); mean concentration of lurbinectedin in plasma (as ng/mL) is also presented (gray symbol).
Cumulative excretion and mass balance (as % of the radioactivity dosed) after a single intravenous bolus dose of 14C-lurbinectedin at 1.2 and 0.6 mg/m2 in male and female Sprague-Dawley rats (mean ± S.D.; N = 3/gender), respectively.
Note: divide by 6 to determine the dose in mg/kg.
Mass Balance in NHPs
After administration, the maximum levels of 14C-lurbinectedin-related total radioactivity were observed at the first sampling time (0.08 hours), declining rapidly, followed by a more gradual decrease. This curve shape suggests multicompartmental kinetics of 14C-lurbinectedin-related radioactivity in male NHPs, with no meaningful differences observed between plasma and blood (Fig. 7). Regardless of the sampling time, results showed similar blood-to-plasma ratio (range, 0.9 to 1.3) of 14C-lurbinectedin-related radioactivity, this being consistent with the values obtained following the in vitro evaluation of the blood distribution of lurbinectedin. The main pharmacokinetic parameters (blood versus plasma; N = 3/matrix) of 14C-lurbinectedin-related radioactivity in male NHPs following a single intravenous bolus dose (1.7 mg/m2) were: Cmax (ng-eq/g), 156 ± 74 versus 159 ± 72; AUC0-∞ (ng-eq·h/mL), 956 ± 164 versus 888 ± 133; t1/2 (h), 77 ± 8 versus 94 ± 2; CL (L/h/kg), 0.15 ± 0.03 versus 0.16 ± 0.03; and, Vdss (L/kg), 12.5 ± 1.7 versus 16.3 ± 2.2.
The combined recovery of radioactivity in this study (feces, urine, and cage rinse) was 82.0% ± 3.1% (N = 3) (Fig. 8). These data indicated that in NHPs the principal route of lurbinectedin`s excretion was through the feces, with cumulative (0 to 168 hours postdose) radioactivity excretion determined as 75.9% ± 0.9% (N = 3) of the dose administered; only a minor fraction of the 14C-lurbinectedin-related radioactivity was recovered from the urine (4.0% ± 1.1%; N = 3).
Cumulative radioactivity recovered from feces and urine after a single i.v. administration of 14C-lurbinectedin: (A) bolus dose (1.7 mg/m2) to male NHP (N = 3); (B) 1-hour i.v. infusion dose (5 mg flat dose; ca. 2.3 mg/m2) to patients (N = 6).
Plasma samples obtained in the study were subjected to further analysis, in which lurbinectedin and its metabolites were quantified by LC-MS/MS. Regarding lurbinectedin metabolites, only Metabolite 4 and Metabolite 6 were found at levels high enough to allow an accurate quantitation, with a percentage relative to lurbinectedin (N = 3) of 0.9% ± 0.7% and 0.4% ± 0.2%, respectively. However, results showed that a sustained amount (2.1 ± 0.7 ng-eq/mL) of 14C-lurbinectedin-related radioactivity was retained in the pellets after plasma protein precipitation from 24 hours postdose onwards that might produce a slight increase in exposure and half-life of total radioactivity versus 14C-lurbinectedin (Fig. 9).
Representative concentration-time curves: (Top) of total radioactivity, lurbinectedin, Metabolite 4, and Metabolite 6, as well as pellet and rinse obtained in plasma of male NHPs (mean ± S.D.; N = 3) after a single i.v. bolus dose at 1.7 mg/m2 (0.142 mg/kg) of 14C-lurbinectedin. (Bottom): of total radioactivity, lurbinectedin, Metabolite 1, Metabolite 2, Metabolite 4, and Metabolite 6 in plasma of patients (mean ± S.D.; N = 6) after 1-hour i.v. infusion dose (5 mg flat dose; ca. 2.3 mg/m2) of 14C-lurbinectedin up to 10 hours after the infusion end.
With regards to excretion (N = 3), only 49.9% ± 4.0% of the radioactivity administered was recovered from the feces. The main lurbinectedin-related metabolite in feces was Metabolite 4 (2.9% ± 0.3%) and its N-Desmethylated derivative (1.1% ± 0.2%); other minor (≤0.6%) metabolites were identified as Metabolite 1, Metabolite 2, and Metabolite 6, as well as other oxidized derivatives, these being likely products of a chemical degradation rather than metabolism. Unchanged lurbinectedin only accounted for 0.4% ± 0.1% of the dose administered to animals.
Owing to the low amount of 14C-lurbinectedin-related radioactivity found in the urine, metabolic profiling was not attempted.
Mass Balance in Patients
Human mass balance studies with anticancer agents are performed in patients with advanced solid tumors who have exhausted therapeutic options (Beumer et al., 2006; Nijenhuis et al., 2016). The approach of performing these studies by recruiting patients with the same tumor type may offer some advantages (such as reduced variability or increased accuracy of estimated pharmacokinetic parameters), although the feasibility of the study may be hampered due to the difficulty of patient recruitment. In the human mass balance with lurbinectedin, no tumor type restrictions were applied because the pharmacokinetic variability of lurbinectedin cannot be explained by the tumor type (Fernandez-Teruel et al., 2019), as in most of the human mass balance studies (Beumer et al., 2006).
Near-complete recovery of administered radioactivity (94.3% ± 8.7%; N = 6) was achieved within 500 hours postdose, following a single mass dose of 4.29 ± 0.23 mg of 14C-lurbinectedin (radioactive dose of 74.28 ± 8.62 µCi) administered as a 1-hour intravenous infusion.
Whole blood-to-plasma ratio of total 14C-lurbinectedin-related radioactivity (as determined with mean AUC0-last values) was 0.7, being in agreement with the in vitro results that suggested a limited lurbinectedin distribution to blood (Fig. 6). The plasma tmax values for total 14C-lurbinectedin-related radioactivity and lurbinectedin were similar (1.22 ± 0.34 hours); the other main pharmacokinetic parameters were higher than the parent compound: Cmax (1.4-fold: 154 ± 88 ng-eq/mL versus 108 ± 46 ng/mL), AUC0-last (1.6-fold: 1438 ± 1241 ng-eq·h/mL versus 889 ± 883 ng·h/mL), AUC0-∞ (1.9-fold: 1789 ± 1555 ng-eq·h/mL versus 962 ± 961 ng·h/mL), t1/2 (1.2-fold: 55 ± 40 hour versus 47 ± 14 hour), CL (1.7-fold: 6.0 ± 4.9 L/h versus 10.2 ± 7.4 L/h) and Vdz (0.5-fold: 285 ± 105 L versus 598 ± 366 L). The value of AUC0-last calculated for lurbinectedin was 66.6% ± 27.5% of the corresponding value obtained for total 14C-lurbinectedin-related radioactivity, therefore demonstrating that lurbinectedin is the main plasma circulating chemical entity (Fig. 7C). Up to 10 hours after the end of infusion (Fig. 9B), 87.0% ± 10.3% of the total 14C-radioactivity in plasma can be attributed to lurbinectedin (70.2% ± 10.9%) plus four metabolites (17.1% ± 7.6%), among which the main circulating metabolites were Metabolite 4 (9.4% ± 6.2% of total radioactivity; 14.3% ± 10.4% of unchanged lurbinectedin) and Metabolite 6 (6.7% ± 1.4% of total radioactivity; 10.4% ± 2.2% of unchanged compound). Other minor circulating metabolites were Metabolite 1 (0.6% ± 0.3% of total radioactivity; 0.8% ± 0.5% of unchanged lurbinectedin), Metabolite 2 (0.5% ± 0.2% of total radioactivity; 0.8% ± 0.3% of unchanged lurbinectedin) and Metabolite 7 (PM030779), which was only detected at concentrations below the lower limit of quantitation (0.1 ng/mL) (Fig. 9B).
Regarding excretion, the majority of radioactivity administered was found in feces (88.7% ± 10.1%) (Fig. 8B), with minimal excretion in urine (5.6% ± 2.0%) (Fig. 8C). In feces, the profiling of secreted metabolites was hampered by a very high interpatient fecal excretion profile (likely owing to gastrointestinal drug-induced disturbances, e.g., constipation) and a very low recovery of radioactivity (28.5%). Only traces (<0.2% of the administered radioactivity) of unchanged lurbinectedin were detected, although many metabolites were found in feces, the more abundant (1.1% of the administered radioactivity) being Metabolite 4. In urine, only 5.6% ± 2.0% of the administered radioactivity was recovered, having been quantified lurbinectedin and Metabolite 6 (1% each).
Discussion
The results summarized in the present study, in which nonclinical species (rats and NHPs) and patients received 14C-lurbinectedin as a single intravenous administration, as well as results gathered from in vitro experiments also presented herein, provide a complete set of information about the pharmacokinetics, biotransformation, and elimination of lurbinectedin and its metabolites.
In both nonclinical species and patients treated with 14C-lurbinectedin, the highest concentration of total radioactivity was always recorded at the shortest sampling time, or immediately after the end of the infusion time (in patients). Thereafter, 14C-related lurbinectedin radioactivity quickly declined in a multiexponential manner, followed by more prolonged distribution and terminal phases. In patients, the pharmacokinetic behavior of 14C-related lurbinectedin was similar to that seen in nonclinical species, and was characterized by a low-to-moderate systemic clearance, an extensive volume of distribution, and a long terminal half-life.
In nonclinical species and patients, total radioactivity values found in blood and plasma were numerically close consistently, regardless of the sampling time. These results were in agreement with the in vitro blood-to-plasma partitioning ratio calculated in rat, NHP and human samples, in which preferential distribution of lurbinectedin into cellular blood components, was not detected.
In vitro studies indicated that the CYP3A4/5-mediated metabolism of lurbinectedin is very intense in NHPs and in humans, and to a lesser extent in rats. Following 10 minutes of incubation, 80% and 95% of compound was metabolized in human and NHP microsomes, respectively, although only 30% was transformed in rat-derived microsomes. The microsomal-mediated biotransformation N-Desmethylation and/or aliphatic ring opening and/or the O-Desmethylation are the pathways involved in the metabolism of lurbinectedin. These in vitro results demonstrated that the aliphatic ring opening is a change that occurs only in NHP and human lurbinectedin microsomal-mediated metabolism. In the three species, similar amounts of the other metabolites were found, except the O-Desmethylated (as well as the N-, O-Desmethylated) and the aliphatic ring opening of the N-Desmethylated metabolites that were absent in the in vitro incubations with rat microsomes. These results suggest that the CYP3A4/5 equivalent isoform in the rat is unable to act on the methylendioxy ring resulting, therefore, in the absence of this metabolic transformation or related, such as its N-demethylation, in rodents.
The in vitro results described above were reflected in the plasma metabolic profiling in vivo, in which the same systemic lurbinectedin-related metabolites are present in both NHPs and patients. In the analysis of plasma samples generated in the mass balance study conducted in NHPs, results demonstrated that the sum of the 5 components, lurbinectedin, pellets, rinses (quantified by liquid scintillation), and Metabolites 4 and 6 (quantified by LC-MS/MS) gave a total concentration very similar to the total radioactivity measured in plasma (by liquid scintillation counting), resulting in overlapped concentration-time curves. Results showed that a sustained amount (2.1 ± 0.1 ng-eq/mL) of 14C-lurbinectedin-related radioactivity was retained in the pellets after plasma protein precipitation from 24 hours postdose onwards. This consistent finding explained the increased exposure (expressed as AUC0-144h) and half-life of total radioactivity versus 14C-lurbinectedin; however, similar values were determined for Vdss and CL. Regarding lurbinectedin metabolites, only Metabolites 4 and 6 were found at levels high enough to allow an accurate quantitation, with a percentage relative to lurbinectedin of 0.9% ± 0.7% and 0.4% ± 0.2%, respectively. Similar results were obtained in patients. Systemic 14C-lurbinectedin–related radioactivity may be mainly attributed to the unchanged compound and 17.1% ± 7.6% to circulating metabolites, namely Metabolites 4 and 6 and, to a lesser extent, Metabolites 1 and 2.
Notwithstanding, some quantitative differences were seen between NHPs and patients. In NHPs, the systemic exposure of lurbinectedin was lower than the corresponding values in patients, which is accepted as a common finding in antineoplastic-treated patients (DeGeorge et al., 1998). In particular, the recommended dose of lurbinectedin in patients is ∼2× higher than the maximum tolerated dose in NHPs, which resulted in a higher systemic exposure to lurbinectedin, as well as to its main metabolite (Metabolite 4) in patients than in NHPs.
Regardless of the study, the mean total radioactivity recovery successfully achieved an adequate (Roffey et al., 2007) cutoff (99.3% ± 1.6% in rats, 82.0% ± 1.6% in NHPs, and 94.3% ± 8.7% in patients) for raising accurate conclusions. In rats, the radioactivity was principally recovered in the feces (cumulative 90.1% ± 0.1% at 168 hours postdose), with excretion via the urine occurring to a much smaller degree (3.0% ± 0.6%). In bile-cannulated animals, a mean biliary excretion of 49.9% ± 11.2% of the administered dose was found during the 0–72 hours period. Similar results were found in male NHPs, with the cumulative (up to 168 hours postdose) excreted radioactivity mainly found in feces; only 4.0% ± 1.1% was found in urine. In agreement with these nonclinical results, the excretion of radiolabeled material in patients occurred predominantly through the feces, with only small amounts of radioactivity recovered from the urine. The metabolic profiling in feces (from NHPs and patients) was hampered by the very low amount of radioactivity recovered during the extraction procedures that were tried (49.0% ± 4.0% in NHPs and 28.5% in patients), likely owing to the concomitant chemical interaction of 14C-lurbinectedin, its metabolites, and/or degradation products with matrix components, as well as gastrointestinal drug-induced or condition-related (in patients) disturbances.
As already mentioned, lurbinectedin is a new synthetic ecteinascidin, chemical analog of trabectedin. Both compounds only differ because of the replacement of a tetrahydro β-carboline ring (in lurbinectedin) by a tetrahydroisoquinoline ring (in trabectedin) (He et al., 2019). Although the nonclinical evaluations carried out with both compounds demonstrated that the resultant in vitro and in vivo antitumor activity was unaffected by these small structural changes (Romano et al., 2013), pharmacokinetic differences and decreased toxicity were recorded in patients, which allows higher dose intensities in lurbinectedin than in trabectedin. Thus, in phase I clinical studies, the recommended doses for the treatment of patients with solid tumors were determined to be 1.5 mg/m2 and 4 mg/m2 for trabectedin (van Kesteren et al., 2000; Taamma et al., 2001; Twelves et al., 2003; Chuk et al., 2012) and lurbinectedin (Elez et al., 2014), respectively. More importantly, at the recommended dose level, biochemical hepatic toxicity (increases in transaminases, alkaline phosphatase, and/or hyperbilirubinemia) was more severe in trabectedin (Twelves et al., 2003) than in lurbinectedin-treated patients, in whom only mild alkaline phosphatase and transaminase increases were recorded (Elez et al., 2014).
One possible explanation for the different safety profiles of trabectedin and lurbinectedin in patients could be that the metabolism of trabectedin involves the generation of an additional 1,2-cathecol from the 6-Hydroxy-7-Methoxy substitution in the dopamine ring (Vermeir et al., 2009). However, because of the dopamine ring replacement by a β-carboline in lurbinectedin’s structure, the substitution described above is not feasible. The presence of this additional catechol group in a ring that can interact with adjacent nuclear proteins and can be further oxidized to a reactive quinone intermediate that is capable of adducting protein (McDonald and Rettie, 2007) may lead to the possibility of trabectedin-induced liver injury, owing to the local generation of this specific metabolite. As previously detailed, because of the structural differences between both compounds, this mechanism of inducing toxicity is unfeasible in lurbinectedin.
Taken together, the in vitro and in vivo results demonstrated that lurbinectedin is mainly cleared by oxidative metabolism mediated by the CYP3A4/5 isoform. The higher metabolic similarities were found between NHPs and humans. In all the species analyzed, none of the metabolites produced were generated in a large amount as circulating metabolites. Rather, both unchanged lurbinectedin and related metabolites are directly eliminated, via biliary excretion, to feces, with only a minor contribution in the urine.
Acknowledgments
The authors gratefully acknowledge the members of PharmaMar S.A. Organic Chemistry Laboratories for providing lurbinectedin, 14C-lurbinectedin (coded as PM100567), deuterated lurbinectedin (coded as PM040038), as well as the reference standards for primary metabolites, namely, PM030036, PM01158, PM030047, and PM030779, and the impurity PM080323.
Authorship Contributions
Participated in research design: Altares, van Andel, Lubomirov, Fudio, Rosing, Tibben, Nan-Offeringa, Schellens, Beijnen, Aviles.
Conducted experiments: Altares, van Andel, Márquez del Pino, Benedit, Tibben, Nan-Offeringa, Estefan.
Contributed new reagents or analytic tools: Altares, van Andel.
Performed data analysis: Altares, van Andel, Lubomirov, Fudio, Rosing, Tibben, Nan-Offeringa, Francesch, Schellens, Beijnen, Aviles.
Wrote or contributed to the writing of the manuscript: Altares, van Andel, Lubomirov, Fudio, Rosing, Tibben, Nan-Offeringa, Francesch, Zeaiter, Cuevas, Schellens, Beijnen, Aviles.
Footnotes
- Received September 6, 2021.
- Accepted January 2, 2022.
The Netherlands Cancer Institute received financial support from PharmaMar to execute the study.
Aviles, Altares, Fudio, Lubomirov, Marquez del Pino, Benedit, Luepke, Francesch, Zeaiter and Cuevas are PharmaMar S.A. employees or PharmaMar S.A. shareholders or both.
ABBREVIATIONS
- HPLC
- high performance liquid chromatography
- LC-MS/MS
- liquid chromatography coupled in tandem with mass spectrometry/mass spectrometry detection
- NHP
- nonhuman primates
- QC
- quality control.
- Copyright © 2022 by The American Society for Pharmacology and Experimental Therapeutics
