Abstract
IDN 5390 (13-(N-Boc-3-i-butylisoserinoyl)-C-7,8-seco-10-deacetylbaccatin III) is a new taxane, derived from 7,8-C-seco-10-deacetylbaccatin, selected for its ability to inhibit angiogenesis, mainly by acting on endothelial cell motility, and for its selective activity on class III β-tubulin. In vivo, IDN 5390 shows activity against paclitaxel-sensitive and -resistant tumors when administered on a prolonged, continuous dosage schedule. We studied the pharmacokinetics and bioavailabilty of the drug in mice after single and repeated oral treatment. IDN 5390 was rapidly absorbed after oral administration, with good bioavailability (43%). After intravenous injection, it was extensively distributed in tissue, mainly the liver, kidney, and heart, with low but persistent levels in brain. The kinetics appear dose-dependent with a clearance of 2.6, 1.4, and 0.9 l/kg at, respectively, 60, 90, and 120 mg/kg, and a half-life 24, 36, and 54 min. After prolonged daily oral doses given for 2 weeks, we found that there was a decrease in drug availability; i.e., the area under the concentration-time curve value after p.o. daily administration on day 14 was 2-fold lower than that on day 1. Metabolism plays a major role in elimination of the drug, and at least 12 metabolites were identified in feces and urine. The percentage excreted as metabolites after an oral dose (42%) was higher than that after the i.v. dose (33%), suggesting a first-pass effect. Four metabolites were found in plasma at detectable levels; one of them, with restored taxane scaffold, is a species 3 times more potent than IDN 5390, possibly contributing to the observed anti-tumor activity.
Taxanes are one of the most active classes of antitumor agents developed in the last two decades (Verweij et al., 1994; Huizing et al., 1995). Paclitaxel is the main taxane used in clinical practice and is effective on a series of human tumors. It is currently used for breast and ovarian cancer and several other human malignancies such as non-small cell lung cancer, melanoma, and head and neck cancer (McGuire et al., 1989; Huizing et al., 1995; Rowinsky and Donehower, 1995; Aisner and Cortes-Funes, 1997). However, important limitations to its clinical use are 1) its poor solubility in the traditional aqueous medium of infusion, that has made it essential to develop an appropriate formulation, containing Cremophor Elp and ethanol, which is not altogether devoid of toxicity; 2) its limiting toxicities: neutropenia, mucositis, and neuropathies; and 3) the multiple mechanisms of resistance, including those related to overexpression of proteins involved in intracellular drug transport and retention.
Another inconvenience is the acquired drug resistance that develops in many patients after an initial response to treatment (Gupta, 1983; Horwitz et al., 1993; Kavallaris et al., 1997). These limitations have driven research for new, more soluble paclitaxel analogs with less toxicity and better activity, particularly in paclitaxel-resistant cell lines.
IDN 5390, 13-(N-Boc-3-i-butylisoserinoyl)-C-7,8-seco-10-deacetylbaccatin III, is a prototype of a new class of synthetic taxane derivatives (Appendino et al., 1997). This C-seco paclitaxel analog is characterized by C-ring opening through the cleavage of bond 7–8 (Fig. 1). Initially, IDN 5390 was selected for its ability to inhibit endothelial cell motility from a screening of paclitaxel analogs with antiangiogenic properties (Taraboletti et al., 2003). This antiangiogenic effect was confirmed in a paclitaxel-resistant tumor model (Petrangolini et al., 2004). Moreover, the compound has high antitumor activity in paclitaxel-resistant cells overexpressing class III β-tubulin. This implies that the mechanism of IDN 5390 is different from that of other taxanes (e.g., paclitaxel and docetaxel), and helps explain why the cytotoxicity of the combination of IDN 5390 with paclitaxel was synergistic (Ferlini et al., 2005).
IDN 5390 showed high activity against a variety of human tumor xenografts, both sensitive and resistant to paclitaxel, including ovarian and colon carcinoma and glioblastoma (Taraboletti et al., 2002; Petrangolini et al., 2004). The optimal schedule for antitumor activity requires protracted administration of subcutaneous or oral doses (Taraboletti et al., 2002; Pratesi et al., 2003). Repeated oral treatments seemed to be very effective at doses causing no apparent toxicity (Pratesi et al., 2003). Paclitaxel given orally is poorly bioavailable (<5%), and its limited absorption may be caused by the P-gp transport system, present in large amounts in gastrointestinal tissues (Sparreboom et al., 1997; Synold et al., 2001; Woo et al., 2003).
Because IDN 5390 is more active than paclitaxel in multidrug-resistant tumors (Pratesi et al., 2003), it is plausible that it is not a strong substrate for P-gp. Thus, it might achieve much better bioavailability after oral dosing, allowing prolonged treatment, which is especially relevant in light of the antiangiogenic properties.
These peculiarities of IDN 5390 prompted us to investigate its pharmacokinetic profile in mice, determining the bioavailability at different doses, the disposition after repeated administration, the distribution, fecal and urinary elimination, and the metabolism.
Materials and Methods
Reagents, Solutions, and Equipment. IDN 5390 (lot 553/16), its 9-methylether derivative (IDN 5517, lot 572/1/A), used as internal standard (IS) for the assay in plasma, and paclitaxel (lot 27759/K3), used as IS for the urine and feces assays, were obtained from Indena S.p.A. [Milan, (MI), Italy].
IDN 5390 stock solution was prepared in methanol at a concentration of 1 mg/ml. The stock solution was diluted with methanol to obtain working solutions at concentrations of 100, 10, and 1 μg/ml. IDN 5390 solution was prepared in methanol at the concentration of 4 mM to study the metabolism in vitro with liver microsomes. The IS stock solution was prepared in methanol at the concentration of 1 mg/ml. It was diluted with methanol to obtain a working solution at the concentration of 20 μg/ml.
Control mouse plasma was obtained from IFFA (Credo, France). Methanol, HPLC grade, was obtained from J. T. Baker B.V. (Deventer, Holland). Glacial acetic acid for analysis and acetonitrile of HPLC grade were obtained from Carlo Erba (Milan, Italy). Triethylamine was obtained from Fluka (Buchs, Switzerland). 7-Ethoxycoumarin (lot 054K3687) and 7-hydroxycoumarin (lot 1123370), NADPH generating system [β-NADP, uridine diphosphate glucuronic acid (UDPGA), glucose-6-phosphate dehydrogenase, glucose 6-phosphate, magnesium chloride, d-saccharic acid 1,4 lactone], RPMI 1640 culture medium, and sulforhodamine B sodium salt were obtained from Sigma-Aldrich (St. Louis, MO). 7-Ethoxycoumarin was dissolved in methanol and diluted with 0.1 M pH 7.4 KH2PO4 buffer to the final concentration of 0.1 mg/ml. Disposable borosilicate glass tubes, 16 × 100 mm, were obtained from Corning Inc. (Corning, NY). Water, HPLC grade Milli Ro 60 Water System, was obtained from Millipore (Billerica, MA). A Bench Mate Workstation system for solid-phase extraction was obtained from Zymark (Hopkinton, MA).
The HPLC system, consisting of a model 717 WISP autosampler, a model 510 pump, a detector, model W2487 at variable UV-visible wavelength, and an acquisition system, Empower Software Chromatography Manager, was obtained from Waters (Milford, MA). Sep-pak-CN cartridges (100 mg, 1 ml) for solid-phase extraction, Symmetry C18 HPLC column (3.5 μm, 4.6 × 150 mm) and precolumn (5 μm, 4.6 × 20 mm), vials, and limited volume inserts for the WISP 717 autosampler were obtained from Waters.
A liquid chromatography/tandem mass spectrometry (LC/MS/MS) system consisting of an Alliance series 2695 coupled with a Micromass Quattro Ultima Pt triple-quadrupole mass spectrometer was obtained from Waters. An HPLC system consisting of a Surveyor autosampler and Surveyor MS pump was obtained from Thermo Electron (Waltham, MA), combined with an LCQ Deca XP Plus (Thermo Electron) ion trap mass spectrometer. An Omnispher 3 C18 column, 100 × 2.00 mm × 1/4 inch, was obtained from Varian (Palo Alto, CA). An optical density plate-reader, Labsystem Multiskan MS, was obtained from Labsystems (Dasit, Italy).
Pharmacokinetic Study.Mice. Female CDF1 mice (body weight 20 ± 2g) were obtained from Charles River Italia (Calco, Italy). Procedures involving animals and their care are conducted in conformity with the institutional guidelines that are in compliance with national (D.L. n. 116, G.U., suppl. 40, 18 Febbraio 1992, Circolare N0. 8, G.U., 14 Luglio 1994) and international laws and policies (EEC Council Directive 86/609, OJ L 358,1, Dec. 12, 1987; Guide for the Care and Use of Laboratory Animals, U.S. National Research Council, 1996).
Drug. Vials of IDN 5390 formulated in polysorbate 80 (Montanox 80 VG DF) at the concentration of 80 mg/ml were provided by Indena S.p.A. The stock solution was diluted with 0.9% NaCl just before use and administered in a volume of 10 ml/kg body weight.
Drug treatment. To assess the pharmacokinetics and the bioavailability of IDN 5390, six groups of female mice were given single oral or intravenous doses of 60, 90, and 120 mg/kg. To assess the bioavailability after repeated oral treatment, two groups of animals were treated once or twice a day (at 9:00 AM and 5:00 PM) with 90 mg/kg for 7 or 14 consecutive days. The effect of vehicle on the bioavailability of IDN 5390 was evaluated on day 7 by giving the drug to a third group of mice pretreated for 1 week with polysorbate 80 twice a day. To determine the fecal and urinary excretion of IDN 5390, six groups of five mice were given 60, 90, and 120 mg/kg doses of the drug intravenously and orally.
Sampling. Pharmacokinetic studies were done on day 1 after the single i.v. and p.o. treatments and on days 7 and 14 after 1 and 2 weeks of daily oral treatment. In all studies, blood samples were collected at 5, 15, and 30 min and 1, 1.5, 2, and 4 h after p.o. administration, and at 5, 15, and 30 min and 1, 2, and 4 h after i.v. injection of IDN 5390. Four mice were used per time point. Blood was obtained from the retro-orbital plexus under diethyl ether anesthesia and collected in heparinized tubes. The animals were sacrificed by cervical dislocation. The plasma fraction was immediately separated by centrifugation (12,000 rpm, 10 min, 4°C) and stored at –20°C until analysis. The tissue distribution of IDN 5390 was studied only after the intravenous dose of 90 mg/kg; brain, liver, heart, lung, and kidney were removed and frozen at –20°C until assayed.
Feces and urine were collected from mice housed in metabolic cages, before dosing and after 24 and 48 h. The fractions were immediately measured, aliquoted, and frozen at –20°C until LC/MS/MS analysis.
IDN 5390 Analysis.Plasma samples. IDN 5390 was determined according to a previously published HPLC method after the addition of IDN 5517 as internal standard and automated solid-phase extraction with Sep-pak-CN cartridges (Zaffaroni et al., 2002). To assay the unknown plasma samples, six calibration standards of 0.05 to 2.00 μg/ml were prepared in duplicate, by combining 0.3 ml of control mouse plasma with different amounts of IDN 5390 working solution. The limit of quantitation was 50 ng/ml.
On plasma samples of mice given 90 mg/kg, identification of the circulating metabolites was attempted by LC/MS/MS analysis: 60 μl of pooled plasma were extracted with 240 μl of HCOOH/CH3CN (1:1000), centrifuged at 12,000 rpm for 10 min, and the supernatant was dried under nitrogen. The residue was dissolved in 30 μl of mobile phase (0.1% HCOOH/0.1% HCOOH acetonitrile, 90:10) and injected into the HPLC coupled with an LCQ Deca XP Plus ion trap mass spectrometer.
Chromatographic separation was carried out on an Omnispher 3 C18 column, 100 × 2.00 mm × ¼ inch, using a mobile phase system gradient of 0.1% HCOOH (solvent A) and 0.1% HCOOH acetonitrile (solvent B). The gradient conditions were from 10% of B to 100% of B in 27 min at the flow rate of 0.2 ml/min.
Tissues samples. Tissues were homogenized in 0.2 M acetate buffer pH 4.5 (1:3 w/v), and 1 ml of the homogenate was mixed with 1 μg of paclitaxel as IS and extracted at 30 min with 5 ml of acetonitrile. Then, samples were centrifuged at 4000 rpm, the supernatant was separated and dried under vacuum, and the residue was dissolved with 120 μl of the mobile phase and 30 μl of sulfosalicylic acid 10% (w/v); 80 μl of the solvent mixture were injected into the HPLC apparatus, separated on a Symmetry C18 column (3.5-μm, 4.6 × 150 mm) with an isocratic mobile phase of 0.01 M acetate buffer/acetonitrile/methanol (50:40:10) at the flow rate of 1.2 ml/min, and quantified at 227 nm.
On the day of the analysis, a standard calibration curve of seven IDN 5390 concentrations, made in duplicate in the range 0.2 to 10 μg/g, was prepared in control tissue to measure the unknown samples. The limit of quantitation of the method was 100 ng/g.
Urine and feces. IDN 5390 was assayed in feces and urine by LC/MS/MS. In brief, urine samples (100 μl) were mixed with 200 ng of paclitaxel as IS, diluted (1:1) with CH3CN/H2O 10:90 containing 0.1% HCOOH, and 5 μl were injected into the HPLC system.
Feces were weighed and homogenized 1:4 (w/v) with 4% bovine serum albumin solution prepared and diluted (1:4) in bidistilled water; then, 1 ml was mixed with 400 ng of paclitaxel as IS. Finally, for extraction, 3 ml of acetonitrile were added, and the supernatant was dried under nitrogen and rinsed with 200 μl of mobile phase. Five microliters were injected into the HPLC system consisting of an Alliance series 2695 coupled with a Micromass Quattro Ultima Pt triple quadrupole mass spectrometer operating in positive ion mode. This mass spectrometer was used to obtain both the mass spectra (MS1) and the product ion spectra (MS2). The instrument was equipped with an electrospray ionization interface. The biological samples were analyzed with the ionspray needle operating at +3500 V and the cone voltage at 60 V. The mass spectrometer was programmed to allow the [M + H]+ ions of IDN 5390 and paclitaxel at m/z 788.4 and 854.6 to pass through the first quadrupole into the collision cell. The characteristic product ions of these two compounds were monitored to the third quadrupole at m/z 309.0, 327.0 for IDN 5390, and 286.4, 509.5, and 569.5 for paclitaxel. The chromatographic separation was carried out on an Omnispher 3 C18 column, 100 × 2.00 mm × ¼ inch, using a mobile phase system gradient of 0.1% HCOOH (solvent A) and 0.1% HCOOH acetonitrile (solvent B). The gradient conditions were from 10% of B to 100% of B in 27 min at the flow rate of 0.2 ml/min. Extracts of urine and feces were also analyzed by HPLC coupled with an LCQ Deca XP Plus ion trap mass spectrometer for the search for IDN 5390 metabolites.
Pharmacokinetic analysis. The concentration data at each time point considered for pharmacokinetic elaborations were the means ± standard deviation from four animals. Plasma samples with a concentration above the highest standard point of the calibration curve were diluted and reassayed.
The experimental area under the concentration versus time curve (AUC) of IDN 5390 was calculated by the linear trapezoidal rule extrapolated to infinity by the Ke. Pharmacokinetic parameters, half-life (t1/2), clearance (CL), and volume of distribution (Vd) were calculated using WinNonlin Pro Node 4.1 pharmacokinetic software (Pharsight, Mountain View, CA), as follows: t1/2 = 0.693/Ke; CL = Dose/AUC; Vd = CL/Ke; where Ke is the constant of elimination of the drug obtained by the fitting. The bioavailability (F) was derived from: F = (AUCp.o./AUCi.v.) × (dosei.v./dosep.o.) × 100.
In Vitro Metabolism and Assessment of Cytotoxicity of IDN 5390 after Incubation with Mouse Liver Microsomes.Microsome test system. Metabolic profile of IDN 5390 was determined in hepatic microsomes (batch 0410046) of CD1 female mice obtained from Xenotech, LLC (KS). Six groups of samples were prepared and incubated at 37°C according to the reported scheme (Scheme 1).
The same scheme was used for incubation of 7-ethoxycoumarin to evaluate the metabolic activity of the microsomes: 7-hydroxycoumarin was identified on the basis of the retention time in comparison with the standard, and used as a marker of oxidative metabolism.
To prepare the NADPH regenerating solution (NRS), the following reagents were used in samples I to IV at the concentrations indicated: 4 mM MgCl2, 10 mM β-d-glucose 6-phosphate, 1 mM β-NADP, and 1.5 units/ml glucose-6-phosphate dehydrogenase. UDPGA (5 mM) and saccharic acid 1,4 lactone (5 mM) were added to a portion of NRS, obtaining the NRS for groups V and VI. Samples were prepared by adding 250 μl of 0.1 M pH 7.4 KH2PO4 buffer, 100 μl of NRS (with or without UDPGA and saccharic acid 1,4 lactone), 45 μl of bidistilled water, and 5 μl of IDN 5390 or 7-ethoxycoumarin to a final concentration of 40 μM or 1 μg/ml, respectively. The reaction was started by adding 100 μl of the microsomal suspension at the concentration of 0.5 mg/ml to the samples. The same volume of 0.1 M pH 7.4 KH2PO4 buffer was added to samples incubated without microsomes or drugs. The reaction was stopped at selected times by adding 500 μl of methanol. Samples were shaken, centrifuged at 12,000 rpm for 10 min (4°C), and the supernatants were used for LC/MS/MS analysis, as reported above, to determine the metabolites in feces and urine.
In vitro cytotoxicity of IDN 5390 incubated with mouse microsomes. The cytotoxic activity of IDN 5390 after incubation with mouse microsomes was evaluated on the A2780 cell line. Cells were grown in RPMI 1640 medium containing 10% fetal bovine serum and 1% penicillin/streptomycin; 48 h before treatment, cells were seeded in 96-well plates. IDN 5390 stock solution was prepared in CH3OH at the concentration of 100 μM. This solution was further diluted with methanol to obtain working solutions at the concentrations of 5, 10, 15, 25, 50, 65, and 80 μM. IDN 5390 was incubated at the concentrations of 0, 50, 100, 150, 250, 500, 650, and 800 nM with mouse liver microsomes, as described before. Control samples prepared in RPMI, boiled microsomes, and incubation buffer (0.1 M pH 7.4 NaH2PO4 + NRS) were added with the same concentrations of IDN 5390.
After 4 h at 37°C, 22 μl of the incubation mixtures were added to the cells, obtaining final theoretical drug concentrations of 0 (control groups), 5, 10, 15, 25, 50, 65, and 80 nM. After 72 h, the medium was removed, and the cells were washed with phosphate-buffered saline, then fixed with cold trichloroacetic acid for 1 h in ice, washed with distilled water, and stained for 30 min with 0.4% (w/v) sulforhodamine dissolved in 1% acetic acid. The optical density was determined after 15 min at 540 nm using a plate-reader.
Results
Pharmacokinetics of IDN 5390 after Single Treatment.Figure 2 shows the pharmacokinetic profiles of IDN 5390 in plasma of CDF1 mice after 60, 90, and 120 mg/kg i.v. in comparison with the profiles after the oral doses. The main pharmacokinetic parameters are reported in Table 1.
After the intravenous treatments, Cmax and AUC increased superproportionally and corresponded to decreased clearance with increasing dose. The elimination half-life also appeared to be shorter at the lower dose. After oral administration, IDN 5390 was rapidly absorbed, achieving the Cmax within 15 to 30 min; then the drug concentrations declined with a t1/2 in the range of 17 to 42 min, with measurable drug plasma levels up to 4 h. The oral route resulted in much lower concentrations and AUCs than intravenous treatment: bioavailability calculated on the basis of the AUCs after oral and intravenous treatment was 43% at the dose of 60 mg/kg but lower at 90 and 120 mg/kg (Table 1).
Tissue distribution was studied after the intravenous administration of 90 mg/kg (Fig. 3); the drug concentrations in liver, kidney, heart, and lung were approximately 3 to 4 times the concentration in plasma. In brain, the drug concentration was lower than in the other organs at every time point, but persisted longer, with concentrations at 4 h even higher than those in the other tissues.
Pharmacokinetics of IDN 5390 after Oral Repeated Treatment. The AUCs on days 7 and 14, after 90 mg/kg given once or twice a day, were approximately half those after the single dose on day 1 (Table 2). To assess the effect of the vehicle on the disposition of IDN 5390, mice received polysorbate 80 daily for 1 week, followed by a single dose of IDN 5390 at 90 mg/kg. The AUC in these mice (7.9 μg/ml h) was superimposable with that obtained after a single dose without polysorbate 80 pretreatment (9.1 μg/ml h).
To determine whether liver metabolic induction was responsible for the decrease in the IDN 5390 AUC, we studied the pharmacokinetics after an i.v. dose of 90 mg/kg IDN 5390 given after a week of daily treatment with 90 mg/kg oral IDN 5390. There was no reduction of plasma levels, and the AUC after chronic treatment was 37.9 μg/ml h, close to the 42.0 μg/ml h found after a single dose.
Fecal and Urinary Elimination of the Unchanged IDN 5390. The fecal and urinary excretion of IDN 5390 was investigated in CDF1 female mice treated intravenously and orally with 60, 90, and 120 mg/kg. Excretion of the parent drug per se was low (Table 3). The percentage of drug eliminated in feces after intravenous treatment was relatively low and did not appear to depend on the doses, being in the range of 2.9 to 3.2% of the dose. After oral treatment, the range of excretion was slightly higher, ranging between 3.1 and 4.7%. The elimination of IDN 5390 in urine was much lower than in feces, but was higher after i.v. (0.5–1.4%) than after oral doses (0.2–0.5%).
Metabolism.Figure 4 shows the typical fragmentation pattern of IDN 5390 obtained by LC/MS/MS analysis. Feces and urine of mice given 120 mg/kg revealed that IDN 5390 was extensively metabolized, and after both p.o. and i.v. single doses, we found several taxane-related structures. On the basis of the m/z ratio, it seemed that the metabolites were mainly formed from oxidation, mono- and dihydroxylation of IDN 5390, and conjugation with sulfate (Fig. 5).
The exact identity of the products of metabolic modifications of taxane structure could not be ascertained from the fragmentation pattern. In analogy with information on the murine metabolism of paclitaxel (Sparreboom et al., 1996; Bardelmeijer et al., 2003), we consider position 6 as a feasible target of hydroxylation. From Fig. 5, the hydroxyl function most likely to be conjugated with sulfate is in position 9 for the parent drug (see Fig. 5, series A) in view of its enolic nature and in position 7 for 7,8-cyclized IDN 5390 derivatives (see Fig. 5, series B).
Table 4 tentatively indicates the percentages of fecal and urinary excretion of the main hypothesized metabolites, assuming that the metabolites and IDN 5390 had similar responses in the mass spectrometer. After oral treatment, the most abundant metabolites in feces, as a percentage of the dose, were M4 (15.0%), M1 (9.1%) and M5 (7.9%). M4 and M1 are products of sulfate conjugation and hydroxylation of IDN 5390, respectively, whereas M5 is the sulfate of M1. Modification of the IDN 5390 structure involving formation of the typical taxane scaffold and dihydroxylation gave C4, amounting to 4.8% of the dose. These metabolites (M1, M4, M5, and C4) were also the most abundant after intravenous treatment. As reported in Table 4, several other metabolites were quantified in feces after both treatments. The total amount of the dose recovered, as IDN 5390 per se and metabolites, amounted to, respectively, 45.5 and 33.2% after oral and intravenous treatment.
In urine, 3.3% of drug and metabolites was recovered after intravenous treatment and 0.9% after oral administration. M2 and C3, two isomers of M1 and C2 with different retention times, were found in urine. M2 is probably an epimer of M1 and C3 could be an epimer or a structural isomer of C2. From the fragmentation pattern of these monohydroxylated derivatives (Fig. 5), it was not possible to establish which stereogenic center was involved in the epimerization and which part of the ring system in the isomerism.
LC/MS/MS analysis showed that among the metabolites found in feces and urine, the most abundant in plasma were C1, M1, C3, and M3. As shown in Fig. 6, which reports the plasma profile of formation and elimination of these metabolites in comparison with that of the parent drug, these metabolites were found at levels 4 to 5 times lower than IDN 5390.
In Vitro Metabolic Profiles. The formation of several metabolites was observed after 4 h of incubation with mouse liver microsomes. On the basis of the LC/MS/MS analysis, we assumed the metabolites mainly formed through oxidation, mono- and dihydroxylation. We found six peaks with molecular weight of 785, and one, namely C1 (7,8-cyclized IDN 5390, Fig. 5), was common to plasma and urine. C1 is characterized by the restored C-taxane ring through the formation of bond 7,8. Such a modification involves an oxidation and, therefore, loss of hydrogen. The other five metabolites are probably isomers of C1. According to their relative abundance, we found one monohydroxylated derivative of IDN 5390 (mol. wt. 803) that amounted to 20.1% of the parent drug and four monohydroxylated derivatives of C1 amounting to 1.2 to 7.1%. We also found one dihydroxylated metabolite of IDN 5390 and one of C1 amounting to 1.8 and 3.4%, respectively. No differences were detectable in the concentrations of IDN 5390 and metabolites from the samples incubated with and without UDPGA, indicating that UDP glucuronosyl-transferases do not play any role in the metabolism of IDN 5390.
Cytotoxicity Results. The effect of incubation with mouse liver microsomes on IDN 5390 cytotoxicity in A2780 cells was evaluated in two separate experiments with six replicates each (Fig. 7). The cytotoxic activity was enhanced approximately 10 times. There was a slight increase in cytotoxicity with boiled microsomes, too, probably due to the residual activity of the enzymes.
Discussion
It has been previously proposed that to exploit both the cytotoxic and the antiangiogenic effect of IDN 5390, continuous exposure for prolonged time is preferable (Pratesi et al., 2003; Taraboletti et al., 2003). Experimental evidence suggests that IDN 5390 given orally daily for a prolonged time has good antitumor activity with very limited toxicity (Pratesi et al., 2003). The present pharmacokinetic study shows that the drug has good oral bioavailability in mice. This finding could be explained by the fact that IDN 5390, unlike paclitaxel (Malingre et al., 2001) is a poor substrate for the P-gp system (Bernacki et al., 2002) that pumps the drug out of the cells and is abundant in the gastrointestinal mucosa. The plasma AUC after the three oral doses tested indicated a dose-dependent increase of plasma AUC, but not of bioavailability. This is because of the dose-dependent clearance after intravenous administration. The dose-dependent kinetics was probably not due to saturation of the elimination of IDN 5390, since it was excreted to a similar extent at all three doses, but metabolism appears to play an important role in its systemic clearance. The amount of metabolites excreted in the feces expressed as percentage of the administered dose was higher after oral (41%) than i.v. treatment (30%), suggesting a first-pass effect through the gastrointestinal tract and the liver.
We found at least 12 possible metabolites of IDN 5390 in feces and urine, mainly formed by oxidation, leading to mono- and dihydroxylated derivatives of IDN 5390, products of retrograde C cyclization and several sulfate derivatives. In contrast to the metabolism of paclitaxel (Bardelmeijer et al., 2003), metabolites related to the baccatin structure were not found in any of the biological specimens. Repeated oral administration led to a decrease of drug disposition after 1 or 2 weeks of single or double daily doses of 90 mg/kg IDN 5390, and drug elimination was halved. This can be ascribed to metabolic autoinduction, as already suggested for paclitaxel (Gustafson et al., 2005), at the level of the intestinal metabolism rather than the liver. In fact, there was no AUC reduction in the pharmacokinetic study using a single i.v. dose after 1 week of daily oral treatment. In addition, we verified that the decreased disposition of IDN 5390 was not due to the vehicle, polysorbate 80, that was reported to potentially affect the absorption of several drugs, reducing their bioavailability.
Since the drug showed greater activity with repeated oral administration (Pratesi et al., 2003), the lower plasma levels after this schedule presumably do not have a deleterious impact on antitumor activity. The formation of highly cytotoxic metabolites including compound C1, 7,8-cyclized IDN 5390, present in plasma at a concentration 4 times lower than that of the parent compound, probably contributed to the antitumor activity. Our in vitro data support this idea, since C1, known as compound FMA73A (synthesized and provided by Indena S.p.A.) and characterized by the presence of the closed taxane ring, was 3 times more cytotoxic than IDN 5390 in A2780 cells (data not shown). It is likely that C1, together with other metabolites, enhances the in vivo antitumor activity of the parent drug, a notion supported by the cytotoxicity in A2780 cells of the incubate of IDN 5390 with mouse liver microsomes. The metabolism of IDN 5390 seems to be a crucial feature, and characterization of the metabolites requires further studies, particularly to assess their antitumor properties.
In contrast to paclitaxel in mice (Eiseman et al., 1994; Sparreboom et al., 1996), IDN 5390 entered the central nervous system and persisted at active concentrations for a long time. This feature might be exploitable for the therapy of glioblastoma, considering the high sensitivity of these tumor cells to taxanes (Silbergeld et al., 1995; Terzis et al., 1997).
In summary, IDN 5390 is an active taxane that can be administered orally and seems effective even in tumors resistant to taxanes. However, its metabolic-pharmacokinetic profile has to be carefully evaluated in case of further clinical development, particularly with a prolonged schedule of daily doses.
Footnotes
-
The generous contribution of the Italian Association for Cancer Research is gratefully acknowledged.
-
Article, publication date, and citation information can be found at http://dmd.aspetjournals.org.
-
doi:10.1124/dmd.106.012153.
-
ABBREVIATIONS: IDN 5390, 13-(N-Boc-3-i-butylisoserinoyl)-C-7,8-seco-10-deacetylbaccatin III; IDN 5517, 13-(N-Boc-β-isobutyl-isoserinoyl)-9-O-methyl-C7,8-seco-10-deacetylbaccatin III; P-gp, P-glycoprotein; AUC, area under the concentration-time curve; Cmax, maximum plasma concentration; Tmax, time of Cmax; CL, clearance; Vd, volume of distribution; t1/2, half-life; F, bioavailability; IS, internal standard; HPLC, high-performance liquid chromatography; UDPGA, UDP-glucuronic acid; LC/MS/MS, liquid chromatography/tandem mass spectrometry; NRS, NADPH regenerating solution.
- Received July 20, 2006.
- Accepted September 7, 2006.
- The American Society for Pharmacology and Experimental Therapeutics