QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) All Versions of this Article: dmd.104.002592v1 33/6/739    most recent Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Bonate, P. L. Articles by Rose, J. Q. Search for Related Content PubMed PubMed Citation Articles by Bonate, P. L. Articles by Rose, J. Q.

THE DISTRIBUTION, METABOLISM, AND ELIMINATION OF CLOFARABINE IN RATS

Genzyme Oncology, San Antonio, Texas (P.L.B., L.A.); and Quintiles, Kansas City, Missouri (J.S., R.J.P., J.Q.R., P.Y.)

(Received October 4, 2004; Accepted February 23, 2005)

	Abstract

Top Abstract Materials and Methods Results Discussion References

 
The distribution, metabolism, and elimination of intravenous [¹⁴C]clofarabine was studied in Fischer 344 male rats under a once daily for 5 days dosing schedule of 25 or 50 mg/kg/day. Also, the in vitro metabolism in rat, dog, and human hepatocytes was studied. Plasma radioactivity (of which clofarabine accounted for 63% to 93%) exhibited three phases of exponential elimination, with half-lives of 0.3, 1.3, and 12.8 h after administration of the 25 mg/kg/day regimen. Unscheduled deaths occurred after one to three doses with the 50 mg/kg regimen, possibly due to nonlinear pharmacokinetics; therefore, mass balance and radiokinetic profiles could not be obtained. A total of 77.1% (of which 87.2% was clofarabine) and 10.8% (of which 6.9% was clofarabine) of the dose was recovered in urine and feces, respectively. 6-Ketoclofarabine, believed to be formed via adenosine deaminase, was the metabolite of greatest concentration found in urine and feces, but in each matrix, it accounted for only 7% of the daily recovery of radioactivity. 6-Ketoclofarabine was also found in myocardium and liver but accounted for less than 2% of the total radioactivity in those tissues. Clofarabine was the major analyte found in myocardium (>97% region of integration) and liver (>94% region of integration). Whole body autoradiography demonstrated that the highest postdistributive concentrations of radioactivity were in the excretory organs, kidney, bladder, and gastrointestinal tract, with no remarkable suborgan distribution. In rat, dog, and human hepatocytes, 95, 96, and 99.8% [¹⁴C]clofarabine remained, respectively, after 6 h of incubation. Eleven metabolites were observed, with the largest constituting 2.5% of the radioactivity.

View larger version (22K): [in this window] [in a new window]   FIG. 1. Postulated structures for clofarabine and its metabolites. denotes placement of 14C-radiolabel.

Previous studies in adults have shown clofarabine activity in both solid tumors and hematologic malignancies. Kantarjian et al. (2003b) administered clofarabine by 1-h intravenous infusion once daily for 5 days in a phase I dose-escalation study. The maximum tolerated dose (MTD) in adult patients with solid tumors was 2 mg/m² with the dose-limiting toxicity (DLT) being myelosuppression. In patients with acute leukemias, the MTD was 40 mg/m² with the DLT being hepatotoxicty. In 32 patients with acute leukemias, two achieved a complete response and three had a complete response without platelet recovery for an overall response rate of 16%. Later, in a phase II study in 62 adult patients with acute refractory or relapsed leukemia treated daily for 5 days at a dose of 40 mg/m², clofarabine had an overall response rate of 48%, with the major adverse events being transient liver dysfunction, skin rashes, palmoplantar erythrodyesthesia, and mucositis (Kantarjian et al., 2003a). Clofarabine was also active in the treatment of pediatric patients with advanced leukemias. Jeha et al. (2004) studied the safety and efficacy of clofarabine in 25 pediatric patients, 1 to 19 years old, with relapsed or refractory leukemia in a phase I dose-escalation study. Clofarabine was administered daily (for 5 days) for 1 h at 11.25 to 70 mg/m². The MTD was 52 mg/m², with the DLT being reversible hepatotoxicity and skin rash. The overall response rate was 32%.

The purpose of this study was to characterize the absorption, distribution, metabolism, and elimination of clofarabine in male Fischer 344 rats at two target doses, 25 and 50 mg/kg/day for 5 days, and to further examine the in vitro metabolism of clofarabine by rat, dog, and human hepatocytes. The strain and doses used in this study were consistent with those used in the good laboratory practices-toxicology studies to support clinical development.

	Materials and Methods

Top Abstract Materials and Methods Results Discussion References

 
Chemicals and Reagents. Clofarabine (2-chloro-9-(2'-deoxy-2'-fluoro-ß-D-arabinofuranosyl) 9H-purin-6-amine), [¹⁴C]clofarabine, 2-chloroadenine, 2-chloro-9-(2'-deoxy-2'-fluoro-ß-D-arabinofuranosyl)-9H-purin-6-one, 6-ketoclofarabine, and clofarabine phosphate were supplied by Genzyme Oncology (San Antonio, TX). The specific activity of [¹⁴C]clofarabine was 50.1 mCi/mmol with a radiochemical purity of >98% as determined by HPLC coupled to radioactivity detection and chemical purity of >99.3% as determined by HPLC and ultraviolet detection. Cryopreserved rat, dog, and human hepatocytes and hepatocyte isolation kits were purchased from XenoTech LLC (Lenexa, KS). Hepatocyte incubation medium was purchased from In Vitro Technologies (Baltimore, MD). All chemicals were of reagent grade or better.

Animals. Male Fisher 344 rats (F344/NHsd) approximately 9 to 11 weeks old were obtained from Charles River Laboratories, Inc. (Wilmington, MA) and quarantined for 1 week before dosing. Rats were kept in metabolism cages under a 12-h light/dark cycle maintained at 22 ± 4°C and 50 ± 15% humidity. Rats were allowed food and water ad libitum except overnight prior to the day 5 dose, when food was withheld until 4 h postdose.

Formulation. Clofarabine and [¹⁴C]clofarabine were weighed and transferred into a glass container. PEG 400 was added and the material was stirred until the test article was completely dissolved. The solution was diluted to the final volume using isotonic saline such that the final volume was 25% v/v PEG 400 in 0.9% sodium chloride. All solutions were stored at 4°C. A single stock solution was prepared for 5 days of dosing. On each day of dosing the required volume was removed from the stock solution and sterilized through a 0.2 µm syringe filter under aseptic conditions. Solutions were equilibrated to room temperature prior to administration. Two stock solutions were prepared, 2.5 and 5.0 mg/ml, to allow dose volumes of 2.5 ml to be pushed over the dosing interval.

Dose Administration. Each animal was cannulated via the femoral vein at least 1 week before the first dose. Each day, rats received a slow-push (3-min) intravenous dose of either 25 or 50 mg/kg [¹⁴C]clofarabine (19 µCi/day, 10 ml/kg). Mass balance doses were administered by weight, whereas doses to determine the plasma concentration-time profile were administered by volume. After dose administration, a small volume of heparinized saline was injected into the cannula to maintain patency.

Sample Collection. Pharmacokinetic Study. Rats (two per time period per dose group) were anesthetized with isoflurane at 0, 0.5, 1, 2, 3, 4, 6, 8, 12, and 24 h postdose on day 5 and whole blood was collected by cardiac puncture or from the abdominal aorta. Because of unscheduled deaths in the 50 mg/kg dose group, sample times deviated from the nominal times. Heparinized blood was centrifuged at 5°C and plasma was harvested for radioanalysis and radiochromatographic profiling.

Mass Balance. Urine was collected daily during the 5-day treatment period and for 3 days after the final dose. Urine and feces were collected frozen in glass receivers embedded in dry ice and then daily thawed and processed for radioactivity. The remainder was stored at -20°C for further analysis. After each collection, metabolism cages were rinsed with water and, at the end of the mass balance study, were washed with detergent and wiped with gauze pads. Rinse volumes were collected and stored frozen for analysis. Carcasses from the mass balance study were frozen and stored for analysis. Due to unexpected deaths, mass balance was not determined in the 50 mg/kg dose group.

Liver and Heart Analyses. The myocardium and liver were collected from rats at 2, 8, and 24 h after the last dose, flushed with saline, and stored frozen at -20°C until analyzed. Only those samples in the 50 mg/kg dose group were profiled for metabolites because samples in the 25 mg/kg dose group had too little radioactivity for analysis.

Quantitative Whole Body Autoradiography (QWBA). In the QWBA study, rats received only 25 mg/kg [¹⁴C]clofarabine; no 50 mg/kg dose group was studied. Rats (two per time point) were sacrificed predose, and at 0.5, 2, 6, and 30 h after the last dose on day 5. Rats were anesthetized with isoflurane and sacrificed by immediate immersion in a hexane dry ice bath. Carcasses were then placed on dry ice for 2 h, stored at -70°C, removed from cold storage, and embedded in carboxymethylcellulose together with ¹⁴C radioactivity standards used for section thickness quality control. Samples were stored as a carboxymethylcellulose block at -20°C until they were sectioned.

Quantitative Radiochemical Analysis. Plasma, Urine, Cage Wash, and Cage Rinse. Duplicate aliquots (0.2 ml for plasma and urine and 0.5 ml for cage rinse and cage wash) were transferred to tared scintillation vials and weighed. Ultima Gold (PerkinElmer Life and Analytical Sciences, Boston, MA) scintillant (7 ml) was added to each vial and counted for radioactivity. All samples were counted for radioactivity using a Beckman Coulter (Fullerton, CA) LS-6000LL liquid scintillation counter. Quench correction was made by the external standard method.

Feces, Tissue, and Carcass. Feces, myocardium, and liver were thawed and homogenized in 3 volumes of water (25% w/v). Whole carcasses (including skin) were digested for several days in 5 M potassium hydroxide and methanol (50:50 v/v). The digested carcasses were homogenized with a Tekmar homogenizer (Tekmar-Dohrmann, Mason, OH). Duplicate homogenate aliquots (0.5 g for feces and tissues and 0.2 g for carcass) were transferred to tared cellulose cups and weighed. Carcass cups were neutralized with concentrated hydrochloric acid (50 µl). Samples were air dried overnight in a hood and then combusted in a sample oxidizer. Approximately 10 ml of CarboSorb E CO₂ absorber (PerkinElmer Life and Analytical Sciences) and 10 ml of Permafluor E⁺ scintillant (PerkinElmer Life and Analytical Sciences) were added to each vial and counted for radioactivity. All samples were counted for radioactivity using a Beckman Coulter LS-6000LL liquid scintillation counter.

QWBA Sample Preparation. Samples were sectioned by a Leica CM 3600 cryomicrotome maintained at -18°C. The 40-µm sections were collected at five sagittal planes. The sections were dried in the cryomicrotome chamber, and a representative section from each plane, together with ¹⁴C autoradiographic standards for subsequent calibration and image analysis, was mounted on an aluminum support, wrapped tightly with Mylar film, and exposed to phosphorimaging screens (Amersham Biosciences Inc., Piscataway, NJ). The exposed screens were scanned with an Amersham Biosciences Inc. Storm 820 scanner. The autoradiographic standard image data, generated using American Radiolabeled Chemicals (St. Louis, MOS) ARC 146 standards, were captured with Imaging Research Inc. (St. Catharines, ON, Canada) AIS software to create a calibrated standard curve. Specific tissue concentrations were determined from interpolation of a standard curve.

Pharmacokinetic Analysis. Pharmacokinetic parameters were determined in the 25 mg/kg dose group by compartmental analysis using WinNonlin Pro (Version 4; Pharsight, Mountain View, CA). Two- and three-compartment models were fit to the mean clofarabine radiokinetic profile (expressed as µg-Eq/ml) using inverse concentration as the weights. Clofarabine concentration was calculated based on the total radiokinetic profile that was determined to be parent clofarabine based on metabolic profiling. Model selection was based on the smallest Akaike information criteria. Area under the curve was estimated using the linear trapezoidal rule.

Metabolite Profiling and ¹⁴C-LC-MS/MS. Rat urine was aliquoted, centrifuged to remove any particulates, and transferred to a clean tube. The supernatant was injected on-column. Plasma from two rats at the same time point were pooled, precipitated, and extracted into 9 ml of 2% acetic acid in acetonitrile, vortexed for 30 min, and centrifuged. The supernatant was transferred to a clean tube, dried under nitrogen at 40°C in a TurboVap, and reconstituted in 300 µl of 10% acetonitrile in mobile phase. The reconstitute was vortexed for 30 min, centrifuged, transferred to an HPLC vial, and the supernatant was injected on-column.

Rat fecal homogenates (2 ml) from the 25 mg/kg dose group were extracted into 3 ml of acetronitrile/water (50:50 v/v) and sonicated for 30 min followed by 30 min of vortexing. Samples were centrifuged and the supernatant wascollected into a test tube. The remaining precipitant was then extracted, sonicated, vortexed, and centrifuged, and the supernatant was collected and pooled with the first supernatant sample. The supernatant was dried in a SpeedVac (Thermo Electron, Waltham, MA) overnight and reconstituted with 10% acetonitrile in mobile phase. The reconstitute was sonicated for 20 min, centrifuged, and transferred to an HPLC vial, and the supernatant was injected on-column.

Rat myocardium homogenates collected at the same time point were pooled. Rat myocardium and liver homogenates (2 ml) were precipitated and extracted into 6 ml (9 ml for heart) of 2% acetic acid in acetonitrile. Samples were vortexed for 30 min, centrifuged, and transferred to a clean tube. Samples were dried overnight under nitrogen at 40°C using a TurboVap. The residue was reconstituted with 50 µl of acetonitrile and 50 µl of mobile phase. Samples were briefly vortexed, sonicated, transferred to a clean tube, centrifuged, and then transferred to an HPLC vial.

Metabolite profiling and characterization were performed using a radio-LC-MS/MS system consisting of a Sciex API 3000 (Applied Biosystems/MDS Sciex, Foster City, CA) triple quadrupole tandem mass spectrometer utilizing the ionspray interface in negative mode. The HPLC system interfaced with the mass spectrometer consisted of two PerkinElmer Series 200 micro LC pumps, a PerkinElmer Series 200 autosampler, and a ß-RAM radioactivity detector equipped with a 200-µl ultra high pressure lithium glass cell (IN/US Systems Inc., Tampa, FL). Separation was performed on a Primesphere C18-HC (5-µm, 4.6 x 250 mm) column with a C18 (7-µm, 3.2 x 15 mm) guard column, both of which were connected in-line with the radiodetector and mass spectrometer. The mobile phase, consisting of solvent A (water containing 0.1% formic acid and 0.1% ammonium hydroxide) and solvent B (90% acetonitrile, 10% water, 0.1% formic acid, and 0.1% acetonitrile) was used throughout. The flow rate was 1 ml/min (of which 0.2 ml was injected) and elution was accomplished with a 10-min isocratic flow at 5% B and a 40-min linear gradient from 5 to 12% B, followed by a 5-min gradient from 12 to 80% B. The effluent from the radioflow detector was split, approximately 0.2 ml/min to the mass spectrometer and the remainder to waste. A switch valve was used to direct highly aqueous eluate to waste for the first 3 min. The MS/MS experiments were performed by collision-induced dissociation with N₂ as the target gas, and full-scan data were acquired. Multiple components were monitored in one LC-MS/MS experiment by separating the total acquisition time in segments, during which MS/MS data were acquired for specific ions.

Radioactivity profiles in plasma and tissue samples were monitored using a radio-LC system consisting of liquid chromatography with accurate radioisotope counting (LC-ARC system). The column was a Primesphere C18-HC (5 µm, 4.6 x 250 mm) kept at 40°C with a C18 guard column (7 µm, 3.2 x 15 mm). The flow rate was 1.0 ml/min with a total run time of 55 min. A PerkinElmer Radiomatic 150 radioactivity detector was used postcolumn with a 2.5:1 Stop-Flo AQ scintillant to mobile phase ratio. Counting was done in 20-s fractions with a 2-min counting time for each fraction. The gradient system used was the same as the radio-LC-MS/MS system. This system was not in-line with the mass spectrometer.

Mass Characterization. Mass analyses by LC-MS in the negative ion mode were performed on urine, fecal, plasma, heart, and liver extracts. Characterization of metabolites was conducted using the following approach. Fullscan mass spectra were generated over the regions where radioactivity was present. The spectra were compared with control samples, and then collision-induced dissociation (CID) MS/MS spectra were generated from the unique ions. When CID mass spectra could not be obtained, selective mass transitions were monitored (MRM analysis) corresponding to unique fragment m/z 168 (2-chloroadenine). Note that the dosing solution of the in vivo study was predominantly parent clofarabine (47:1). LC-MS data were evaluated for the corresponding nonradiolabeled ions.

Metabolism of [¹⁴C]Clofarabine in Hepatocytes. The in vitro metabolism of [¹⁴C]clofarabine in rat, dog, and human hepatocytes was evaluated at 0, 2, and 6 h using 10 µM [¹⁴C]clofarabine incubated at 37°C. Cryopreserved cells were thawed and poured into a Percoll solution diluted with Dulbecco's modified Eagle's medium. The sample was mixed and centrifuged, and the dead cell debris was removed. The cells were resuspended in supplemented Dulbecco's modified Eagle's medium solution and centrifuged. The supernatant was discarded and the cells were resuspended in 6 ml of pregassed (95% O₂/5% CO₂) incubation media. Viability was determined using trypan blue exclusion. Incubations with [¹⁴C]clofarabine were performed in a six-well plate using an incubation volume of 2 ml per well. Each well received 1 ml of diluted cells (approximately 2 x 10⁶ live cells) and 1 ml of a 20 µM solution of [¹⁴C]clofarabine. Incubations were terminated at 2 and 6 h by the addition of 4 ml of 2% acetic acid in acetonitrile. Negative controls were prepared with heat-deactivated cells (80°C for 5 min). Positive controls were prepared by spiking the cells with 10 µM dextromethorphan, 5 µM midazolam, and 25 µM 7-hydroxycoumarin.

Preparation of Samples for Mass Spectral Analysis. Hepatocyte samples were centrifuged at 3000 rpm for 10 min. The supernatant was evaporated to dryness under N₂ and the residue was resuspended in 250 µl of 20% acetonitrile/H₂O for subsequent analysis by LC-MS/MS coupled to radioactivity detection. Urine samples (1 ml) were clarified by centrifugation at 12,000 rpm for 5 min. The resulting supernatants were analyzed by LC-MS/MS coupled to radioactivity detection. Equal volumes of plasma (1.5 ml) were combined for each selected time point. Preweighed myocardium and liver samples were homogenized in 3 volumes of deionized water. The proteins from each matrix were precipitated with 3 volumes of acetonitrile containing 2% acetic acid. After centrifugation at 3500 rpm for 10 min, the supernatant was evaporated to dryness under N₂ and the residue was resuspended in 300 µl of 10% acetonitrile/mobile phase A for subsequent analysis by LC-MS/MS coupled to radioactivity detection. Weighed fecal samples were homogenized with 3 volumes of deionized water. Each homogenate (2 g) was extracted twice with 3 ml of 50% acetonitrile/water. The two organic extracts were combined and evaporated to dryness in a vacuum evaporator overnight. The residue was resuspended with 200 µl of 10% acetonitrile/mobile phase A, sonicated, vortexed and centrifuged at 12,000 rpm for 5 min. The clear supernatant was analyzed by LC-MS/MS coupled to radioactivity detection.

View larger version (16K): [in this window] [in a new window]   FIG. 2. Plasma concentration-time profiles for clofarabine, total radioactivity, and 6-ketoclofarabine in the 25 mg/kg dose group at steady state on day 5. Solid line is the predicted concentration based on a three-compartment model. Open circles are parent clofarabine concentrations based on metabolite profile.

	Results

Top Abstract Materials and Methods Results Discussion References

Pharmacokinetic Results in the 25 mg/kg Dose Group. Mean trough concentrations of radioactivity in plasma at steady state were low (<0.1 µg-Eq/g). Total radioactivity expressed as a parent concentration based on metabolic profiling was best described by a three-compartment model with -, ß-, and -half-lives of 0.3, 1.3, and 12.8 h, respectively (Fig. 2). The final parameter estimates (±S.E.) were 1264 ± 57 ml/h/kg for clearance, 1.1 ± 0.18 l/kg for the central volume of distribution, 829 ± 85 ml/kg for the volume of the peripheral compartment, 934 ± 151 ml/h/kg for intercompartmental clearance to the peripheral compartment, 2.9 ± 2.1 l/kg for the volume of the deep compartment, 182 ± 40 ml/h/kg for intercompartmental clearance to the deep compartment, and 4.9 ± 2.2 l/kg for the volume of distribution at steady state. The volume of the deep compartment was not estimated with the same degree of precision as other model parameters, but inclusion of the deep compartment was necessary to adequately characterize the data.

Mass Balance and Profiling in the 25 mg/kg Dose Group. Mean trough concentrations of radioactivity in liver and myocardium at steady state were 8- to 9-fold higher than plasma concentrations (0.8 and 0.9 µg-Eq/g, respectively). Concentrations of radioactivity in liver averaged 16.1, 2.2, and 0.8 µg-Eq/g at 2, 8, and 24 h postdose, respectively. Tissue/plasma ratios for radioactivity in liver averaged 6.3, 10.9, and 11.3 at 2, 8, and 24 h postdose, respectively. Concentrations of radioactivity in myocardium averaged 12.3, 3.5, and 0.9 µg-Eq/g at 2, 8, and 24 h postdose, respectively. Tissue/plasma ratios for radioactivity in myocardium averaged 4.8, 16.7, and 12.8 at 2, 8, and 24 h postdose, respectively.

Radioactivity recoveries 0 to 24 h after the fifth dose averaged 77.1 ± 3.1% in urine and 10.8 ± 0.6% in feces. An additional 6.0 ± 2.7% of the radioactive dose was recovered in cage wash. Less than 1% of the dose was recovered in carcass. The total mass balance recovery, determined for 3 days after the final (day 5) dose, was 95.3 ± 0.5%. A total of 87.2% of the radioactivity in urine excreted 0 to 24 h after the last dose could be attributed to unchanged clofarabine. If cage wash radioactivity is assumed to represent primarily urinary radioactivity (which is a reasonable assumption because no significant fecal contamination was evident in cage wash), then 72.4% of the administered dose was excreted in urine as unchanged clofarabine and 10.6% of the dose was excreted in urine as metabolites. Similarly, 10.8% of the administered dose was recovered in feces on day 5.

Metabolic profiling results indicated that about 6.9% of the radioactivity in feces was unchanged clofarabine. Therefore, approximately 0.75% of the dose was eliminated in feces as unchanged clofarabine and 10.1% of the dose was eliminated in feces as metabolites. If these values (based on the observed mean recovery of 95.3% of the dose) are scaled to 100%, approximately 76.0% of the 25 mg/kg dose was excreted in urine as unchanged clofarabine, 21.7% was eliminated in urine and feces as metabolites, and 0.8% was eliminated as unchanged clofarabine by either biliary secretion, colonic secretion, or hydrolysis of glucuronide conjugates in the gut. Clofarabine metabolites were eliminated in approximately equal fractions by renal clearance (51%) and biliary clearance (49%). The total amount of administered dose recovered in urine as unchanged clofarabine and metabolites was 4.7 mg and the AUC_0–24 of radioactivity was 17.1 µg-Eq · h/g. Therefore, the total estimated renal clearance of radioactivity was 275 ml/h. Because 87.2% of the radioactivity in urine represented unchanged clofarabine, the estimated renal clearance of clofarabine was 240 ml/h.

Pharmacokinetic Results in the 50 mg/kg Dose Group. Because of the unscheduled deaths after one to three doses of this regimen, it was not possible to characterize the concentration versus time profile of radioactivity. Blood samples were collected as soon as possible following death for quantification of plasma radioactivity concentrations. Actual sample collection times from treated animals at the time of death ranged from midinfusion to 3 h postdose, with most samples collected at or near the end of the 3-min infusion. Because samples could not be obtained at the nominal times after the 50 mg/kg dose, it was not possible to directly compare radioactivity concentrations between the 25 and 50 mg/kg dose groups. Therefore, linearity was assessed by compartmental modeling. Based on a three-compartment model following intravenous infusion of a 25 mg/kg dose, the estimated maximum plasma concentration at the end of a slow bolus would be 20.4 µg-Eq/g. Assuming linear pharmacokinetics, after a slow bolus of 50 mg/kg, the estimated plasma radioactivity concentration would be 40.8 µg-Eq/g. The observed concentration at 0.05 h after the 50 mg/kg dose was 120 µg-Eq/g, which was 3-fold higher than expected. Similarly, the theoretical concentration for a 50 mg/kg dose was 28 µg-Eq/g at 0.25 h postdose, but the observed mean concentration was 50.8 µg-Eq/g, or 2-fold higher than expected. These differences exceed the anticipated range of variability from experimental error and suggest nonlinear pharmacokinetics between 25 and 50 mg/kg.

Metabolite Profiling in the 25 mg/kg Dose Group. The counting efficiency of the radio-LC-MS/MS system ranged from 30 to 43% with a counting efficiency of <10%. The mean recovery after injecting urine and fecal extracts on-column was 100%. The LC-ARC system had a counting efficiency of 60 to 96% with a counting precision of 6 to 21%. The extraction recoveries of fecal, plasma, and tissue homogenates ranged from 89% to 115%.

Based on region of integration (ROI), instead of daily recovery of radioactivity (DRR), six metabolites were observed in plasma (Table 1). Clofarabine accounted for 63% (24 h postdose) to 93% (0.5% postdose) of the ROI. Of the metabolites, P9a (which was later identified and confirmed as 6-ketoclofarabine) accounted for 4% of the ROI at 0.5 h postdose but was 37% of the ROI 24 h postdose (Table 2). The ratio of P9a to clofarabine increased over time.

The peak numbering system of the radiochromatograms incorporated all peaks observed in all the in vitro and in vivo matrices evaluated (Fig. 3). Overall, radiochromatograms were divided into 17 radioactivity regions, each of which may contain one or more metabolites. Regions P6 and P10 were observed exclusively in the in vitro hepatocyte study. Eight metabolite peaks were observed in urine, of which clofarabine (P15) accounted for 68 to 74% of the DRR. The next most abundant peaks were P9a (7% DRR), P8 (1.5% DRR), and P13 (1.5% DRR). The minor metabolites accounted for less than 0.7% of the DRR. Thirteen peaks were observed in feces, which accounted for 0.7% of the DRR. The major metabolite observed was P9a, which accounted for 7% of the DRR, followed by P14 (1.7% of the DRR), with the remainder accounting for less than 0.9% of the DRR.

View larger version (42K): [in this window] [in a new window]   FIG. 3. Sample radiochromatograms from clofarabine standard, urine, feces, plasma, liver, and myocardium. Urine and fecal samples were collected 24 h postdose. The plasma, liver, and myocardial samples were collected 0.5, 1, and 1 h after dosing.

Concentrations and Metabolite Profiling of Plasma, Myocardium, and Liver in the 50 mg/kg Dose Group. No excreta were analyzed in the 50 mg/kg dose group due to premature deaths. Because of unscheduled deaths, concentrations of radioactivity in liver and myocardium could not be determined at the nominal sampling times. Radioactivity concentrations in liver averaged 144 µg-Eq/g immediately after the third dose and 135 µg-Eq/g at 0.25 h after the third dose. Corresponding concentrations of radioactivity in myocardium averaged 125 and 91.0 µg-Eq/g. Tissue/plasma ratios were not calculated because the data were not from Day 1 and concentrations were not at steady-state. Therefore, tissue/plasma ratios could not be compared between the 25 and 50 mg/kg doses.

In those plasma samples collected immediately after death, clofarabine accounted for more than 91% of the ROI in most animals. One animal survived to 3 h postdose on the third day. In that animal, six plasma metabolites were observed, of which clofarabine accounted for 77% of the ROI, followed by P12 (10% ROI), and P9a (6.5% ROI). The remaining metabolites accounted for less than 2% of the ROI. In myocardium and liver radiochromatograms, parent clofarabine accounted for 94% to 99% of the ROI. Five metabolites were observed in myocardium, whereas seven were observed in the liver. The major metabolite in heart and liver was P9a, which accounted for <2% of the ROI.

Mass Characterization. Table 3 presents the parent ions and major fragment ions of clofarabine and authentic reference metabolites following LC-MS/MS analyses under negative mode of detection. Figure 1 presents the postulated metabolites of clofarabine across the in vivo and in vitro studies. In the mass spectrum of clofarabine (P15, mol. wt. 303, (M - H)^- = 302), the largest ion in the mass spectrum at m/z 348 corresponds to the formic acid adduct of the parent drug and an artifact of the ammonium formate buffer used in the LC mobile phase. The CID spectra of m/z 302 had characteristic fragments m/z 282, 264, 246, 234, 224, 168, and 132. CID analysis of m/z 348 showed fragment ions m/z 302, 282, 246, 224 and 168 (data not shown).

The molecular ion for P7 was m/z 168, which corresponds to the N-dealkylation of the furanosyl moiety to yield 2-chloroadenine. CID analysis of m/z 168 in P7 was congruent with that of the 2-chloroadenine standard (data not shown). P7 was only observed in fecal samples.

The molecular ion for P9a was m/z 303, which corresponds to the addition of 1 amu to clofarabine. The most likely change to account for this mass difference is the displacement of the 6-amino group (NH₂, 16 amu) of the adenosyl moiety by a hydroxyl group (OH, 17 amu). CID analysis of m/z 303 showed similar fragmentation pattern as clofarabine, with 1 unit mass shifts for each of the fragment ions. CID spectrum of P9a was congruent with the CID spectrum of 2-chloro-9-(2'-deoxy-2'-fluoro-ß-D-arabinofuranosyl)-9H-purin-6-one standard. The metabolite was subsequently renamed 6-ketoclofarabine. P9a was observed in every matrix.

The molecular ion for P11 was m/z 316, which corresponds to an addition of 14 Da indicating the presence of a double-bonded oxygen or methylation. The most likely site for the addition of the double-bonded oxygen is at the 5' carbon of the furanosyl moiety, changing the alcohol to an acid; methylation could occur at either of the furanosyl–OH groups. CID analysis of this peak showed primarily the characteristic fragment ion m/z 168, indicating that the adenine part of the molecule is unmodified. The exact location of the change and whether it corresponds to methylation or oxidation is unknown. P11 was observed in urine, feces, and plasma.

The molecular ion for P12 and P13 was m/z 480, resulting from the addition of 176 Da, and corresponds to the glucuronide of clofarabine. Due to low intensity, no fragmentation of the m/z 480 was observed in CID mass spectra. The presence of the characteristic fragment ions m/z 304, 226, and 170 in the MRM analysis verifies that m/z 480 is a metabolite of clofarabine. The possible site for the addition of glucuronide is on the oxygen at the 3' or 5' carbon. Although m/z 480 was not observed in the MS spectra of human hepatocyte samples, MRM analysis indicated its presence. Clofarabine glucuronides were not observed in rat samples, but only in the in vitro hepatocyte study.

The molecular ion for P13 (urine) and P14 (feces) was also m/z 382. The addition of 80 Da would correspond to either the phosphate or sulfate conjugate of clofarabine. However, CID analysis of the m/z 382 peak from urine or feces did not show the presence of characteristic phosphate fragment ion m/z 79. The chromatographic separation (different retention times) of P13 and P14 indicates the presence of two isomers of the sulfate conjugate. In addition to feces, P14 was also observed in rat plasma and tissues.

Clofarabine phosphate has been reported to be the major metabolite in human T-lymphoblast leukemia CCRF-CEM cells (Xie and Plunkett, 1996). Clofarabine phosphate has weak ionization response and a retention time similar to that of hydroxy-clofarabine (m/z 303, P9b). The CID spectrum of m/z 382 from the clofarabine phosphate standard indicated the presence of m/z 168, which is a characteristic fragment ion of the unmodified adenine moiety, and m/z 79 of the phosphate ion. Initial evaluation of the electrospray mass spectrum of the urine, feces, plasma, heart, and liver sample matrices did not show the presence of clofarabine phosphate (m/z 382). Reanalyses by selectively monitoring the transitions of m/z 382 to product ions m/z 168 and 79 indicated the possible presence of clofarabine phosphate in the region of P9b in urine, plasma, myocardium, and liver, although these were below the detection limit. Clofarabine phosphate was not observed in feces. Attempts to chromatographically separate 6-ketoclofarabine and the clofarabine phosphate peaks were unsuccessful and, therefore, the relative amount of clofarabine phosphate could not be estimated. Moreover, experimentation with spiking clofarabine phosphate to fecal extracts showed the disappearance of the m/z 382 response, indicating either ion suppression or instability of the compound. Extracted ion chromatogram analyses of the total ion spectra from the biological samples for the clofarabine diphosphate ion (m/z 462) or triphosphate ion (m/z 542) did not indicate the presence of these metabolites. MRM analyses for these metabolites were not performed.

Mass spectral analysis of the radioactive regions P1 to P5, P12, and P16 to P17 from the chromatograms of rat urine, feces, plasma, or tissues did not reveal distinguishable ions apart from those seen with control samples. Therefore, CID analysis of these peaks was not conducted. In total, these peaks represented <3% of the excreted radioactivity.

Quantitative Whole Body Autoradiography. The limit of quantification (BLQ) was set at 0.274 µg-Eq/g. Distribution of radioactivity to tissues was widespread following the final intravenous dose of 25 mg/kg/day (Fig. 4). The distribution of radioactivity in male rats was widespread at 0.5 h after the final dose, with most tissues reaching their maximal concentration at this point. The highest concentrations of radioactivity were found in the urine and the urinary bladder (440 and 254 µg-Eq/g, respectively). High concentrations of the administered dose were also found in the spleen, kidney, cecum, and thymus (123, 82, 65, and 62 µg-Eq/g, respectively). Most measured tissues had concentrations greater than plasma; i.e., tissue to plasma ratios were larger than 1. Distribution of radioactivity in the kidney was relatively homogeneous. Biliary excretion could not be confirmed, since radioactivity was not measurable in the bile (within hepatic ducts of the liver). Low concentrations (approximately 13- to 34-fold less than blood) of ¹⁴C-clofarabine-derived radioactivity were detected in the tissues of the central nervous system. Distribution of radioactivity in the tissues of the central nervous system was homogeneous and showed no evidence of differential localization. Elevated radioactivity concentrations in the lower gastrointestinal (GI) tract were probably the result of continued elimination of drug-derived radioactivity from the previous day's dose. The lowest measurable tissue concentrations of radioactivity were found in the seminal vesicle, cerebellum, cerebrum, spinal cord, and medulla (3, 0.8, 0.6, 0.5, and 0.3 µg-Eq/g, respectively).

View larger version (80K): [in this window] [in a new window]   FIG. 4. Representative whole body autoradioluminograph of radioactivity taken 0.5 h postdose on day 5 after a 5-day, once daily 25 mg/kg/day intravenous dosing regimen of [14C]clofarabine. All sections are from the same animal and represent a lateral sagittal (top figure) to midsagittal (bottom figure) view.

By 2 h after the final dose, the overall distribution of [¹⁴C]clofarabine-derived radioactivity in male rats remained widespread, although overall concentrations began to decline. The highest concentration of radioactivity was found in urine (241 µg-Eq/g) and urinary bladder (175 µg-Eq/g). High concentrations of radioactivity were found in the cecum contents, small intestinal contents, and large intestinal contents, indicating transit of the administered dose in the GI tract. The small intestinal wall was the only measured tissue that reached maximal concentrations at 2 h postdose. Radioactivity concentrations decreased by 2- to 4-fold in the major organs of excretion (liver and kidney). The spleen, thymus, harderian gland, and lymph nodes (44, 28, 23, and 23 µg-Eq/g, respectively) had concentrations of radioactivity that exceeded the concentrations found in the liver and kidney. Drug-derived radioactivity incorporated into the hair 2 h after the final dose was 39 µg-Eq/g. The myocardium concentration was 2-fold higher than that found in skeletal muscle and 4-fold greater than the blood concentration. The remainder of the non-GI tract tissues had radioactivity concentrations that were less than 16 µg-Eq/g. The lowest quantifiable concentrations were found in the preputial gland, spinal cord, and medulla of the brain (<0.4 µg-Eq/g, respectively) at this sampling time.

At 6 h postdose, further declines in [¹⁴C]clofarabine-derived radioactivity were found in the majority of tissues, although distribution of radioactivity remained widespread. Excluding the GI tract, the urine and urinary bladder had the highest concentrations of radioactivity (111 and 33 µg-Eq/g, respectively). The large intestinal contents and cecum contents reached the maximum observed concentration at 6 h after the final dose, indicating further elimination of drug-derived radioactivity. Hair (25 µg-Eq/g) was the only other measured matrix that contained more than 5 µg-Eq/g at this time point. The following tissues had radioactivity concentrations that were below the limit of quantification (BLQ): cerebellum, cerebrum, medulla, and spinal cord.

By 30 h after the final day 5 dose, overall radioactivity continued to decline as continued elimination of drug-derived radioactivity was evident. Radioactivity concentrations in the spleen, kidney, and liver fell 322-, 212-, and 73-fold, respectively. The majority of the remaining radioactivity was associated with the contents of the GI tract, urine, and hair. The highest concentration value, 23 µg-Eq/g, was found in the hair, followed by the large intestinal contents, urine, cecum contents, and urinary bladder (18, 13, 11, and 5 µg-Eq/g, respectively). All other measured matrices had concentrations lower than 1.5 µg-Eq/g. Twenty-three of 53 measured tissues were BLQ by 30 h after the final day 5 dose. Elimination of drug-derived radioactivity was not complete at the final sampling time.

In summary, high radioactivity was present in urine and kidney, suggesting rapid renal excretion of drug-derived radioactivity. The majority of tissues reached maximum concentrations at 0.5 h postdose and then declined steadily over time. Although distribution was fairly homogeneous, decreased radioactivity was observed in tissues with special barriers compared with other tissues at all time points. The distribution of radioactivity in the tissues of the central nervous system was uniform, with no evidence of differential distribution. Hence, the highest postdistributive concentrations of radioactivity were in the excretory organs, kidney, bladder, and GI tract, with no remarkable suborgan distribution.

In Vitro Hepatocyte Metabolism. The appearance of the appropriate phase I and II metabolites in the positive controls, and the absence of metabolites in the negative control, indicated that the system was acting appropriately. Metabolism of clofarabine in rat, dog, and human hepatocytes was minimal after a 6-h incubation of 10 µM [¹⁴C]clofarabine. The percentage of clofarabine remaining ranged from 95 to 98.8% (Table 4). In rat and dog hepatocytes, the major metabolite observed was P11, accounting for 1.2 and 2.5% of the radioactivity in the chromatogram, respectively. P11 is proposed to be a carboxy- or methoxy-clofarabine. The only metabolite observed in human hepatocytes was P14, which is proposed to be the sulfate conjugate of clofarabine and accounted for 0.2% of the radioactivity in the chromatogram.

	Discussion

Top Abstract Materials and Methods Results Discussion References

 
The results of this study indicate that clofarabine exhibited nonlinear pharmacokinetics at the doses examined: 25 and 50 mg/kg daily for 5 days. Qian et al. (1994) suggested that clofarabine has nonlinear pharmacokinetics after single-dose intravenous and oral administration of 10 and 25 mg/kg. The results from this study corroborated their observation. In the good laboratory practices-toxicology studies, clofarabine administration in rats induced adverse histopathological effects on the heart at 50 mg/kg/day after 5 days of treatment and at 25 mg/kg/day for 5 days after six cycles of treatment, but was not observed at lower doses (internal data). Nonlinear pharmacokinetics may be playing a role in the degree of toxicity, since it has been observed that prolonging the infusion length and changing the route of administration from intravenous to oral (which also blunts maximal concentrations) reduces the degree of cardiotoxicity. Furthermore, such cardiotoxicity has only been seen in rats and may be species-dependent. In this study, the maximal clofarabine concentration after intravenous bolus administration of 25 and 50 mg/kg was 20 and 120 µg/ml, respectively, the latter being a predicted value. In the phase II human studies in pediatric and adult patients with leukemia, the recommended doses are 52 mg/m² and 40 mg/m², respectively, given by 1- to 2-h infusion. At these doses, the observed maximal concentrations ranged from 0.14 to 1.05 µg/ml in pediatric patients and from 0.24 to 1.92 µg/ml in adults. Hence, these data suggest a more than 15-fold safety margin between animals and humans.

At steady state, tissue distribution of radioactivity based on QWBA was widespread and showed highest concentrations in the rapidly perfused tissues. Radioactivity distribution was widespread, with the highest concentrations observed in the excretory organs; i.e., gastrointestinal tract and kidneys. No tissue showed selective accumulation, and only tissues with special barriers showed restrictive distribution. Previous studies in mice have shown that the distribution of nucleoside analogs, clofarabine included, is uniform and rapid, with the highest concentrations found in the rapidly perfused tissues (Lindemalm et al., 1999). Low plasma protein binding has been reported for clofarabine in rats (13%; Qian et al., 1994) and humans (47%; Reichelova et al., 1995), indicating that a large fraction of the compound circulating in blood is present in the free form and available for cellular uptake. This difference in protein binding may also help explain the greater toxicity in the rat than in humans, since more unbound drug is available for tissue distribution in rats than in humans. Furthermore, clofarabine has low lipophilicity, with a mlog P of -0.17, indicating almost equal preference for water and lipid. With a large fraction of drug available for distribution and low lipophilicity, it was expected that distribution in the rat would not favor particular tissues, unless that tissue had a specific nucleoside transporter (which did not appear to be the case).

The major metabolite formed was 6-ketoclofarabine, which was found in all matrices examined but accounted for less than 10% of the radioactivity in all the tissues examined. Interestingly, clofarabine was metabolically stable in isolated hepatocytes, having more than 95% remaining intact after 6 h of incubation in all three species, indicating that clofarabine does not undergo cytochrome P450 metabolism to any significant extent and that the location of the enzyme responsible for its metabolism to 6-ketoclofarabine is extrahepatic. The identity of the enzyme responsible for metabolism of clofarabine to 6-ketoclofarabine is unknown, but given the close structural similarity of clofarabine to adenosine, a likely possible enzyme is adenosine deaminase, a ubiquitous enzyme of purine metabolism that catalyzes the irreversible deamination of adenosine and deoxyadenosine to inosine and deoxyinosine, respectively. Further studies are needed to confirm this hypothesis.

	Footnotes

 
Financial Disclosure: All work was sponsored by Genzyme Oncology. P.L.B. and L.A. are employees of Genzyme Oncology. All other authors have no financial affiliations with Genzyme Oncology.

Article, publication date, and citation information can be found at http://dmd.aspetjournals.org.

doi:10.1124/dmd.104.002592.

ABBREVIATIONS: MTD, maximum tolerated dose; DLT, dose-limiting toxicity; HPLC, high-performance liquid chromatography; DRR, daily recovery of radioactivity; ROI, region of integration; QWBA, quantitative whole body autoradiography; LC-MS/MS, liquid chromatography-tandem mass spectrometry; MRM, multiple reaction monitoring; BLQ, below the limit of quantitation; GI, gastrointestinal.

¹ Current address: XenoTech, 16825 W. 116th Street, Lenexa, KS 66219.

Address correspondence to: Peter L. Bonate, Genzyme Oncology, San Antonio, TX 78229. E-mail: peter.bonate{at}genzyme.com

	References

Top Abstract Materials and Methods Results Discussion References

 


Jeha S, Gandhi V, Chan KW, McDonald L, Ramirez I, Madden R, Rytting M, Brandt M, Keating M, Plunkett W, et al. (2004) Clofarabine, a novel nucleoside analog, is active in pediatric patients with advanced leukemia. Blood 103: 784-789.[Abstract/Free Full Text]

Kantarjian H, Gandhi V, Cortes J, Verstovsek S, Du M, Garcia-Manero G, Giles F, Faderl S, O'Brien S, Jeha S, et al. (2003a) Phase 2 clinical or pharmacologic study of clofarabine in patients with refractory or relapsed acute leukemia. Blood 102: 2379-2386.[Abstract/Free Full Text]

Kantarjian HM, Gandhi V, Kozuch P, Faderl S, Giles F, Cortes J, O'Brien S, Ibrahim N, Khuri F, Du M, et al. (2003b) Phase I clinical and pharmacology study of clofarabine in patients with solid and hematologic cancers. J Clin Oncol 21: 1167-1173.[Abstract/Free Full Text]

Lindemalm S, Liliemark J, Larsson BS, and Albertioni F (1999) Distribution of 2'-chloro-2'-deoxyadenosine, 2-chloro-2'-arabino-fluoro-2'-deoxyadenosine, fludarabine and cytarabine in mice: a whole body autoradiography study. Med Oncol 16: 239-244.[Medline]

Parker WB, Shaddix SC, Chang CH, White EL, Rose LM, Brockman RW, Shortnacy AT, Montgomery JA, Secrist JA 3rd, and Bennett LL Jr (1991) Effects of 2-chloro-9-(2-deoxy-2-fluoro-ß-D-arabinofuranosyl)adenine on K562 cellular metabolism and the inhibition of human ribonucleotide reductase and DNA polymerases by its 5'-triphosphate. Cancer Res 51: 2386-2394.[Abstract/Free Full Text]

Qian M, Wang X, Shanmuganathan K, Chu CK, and Gallo JM (1994) Pharmacokinetics of the anticancer agent 2-chloro-9-(2-deoxy-2-fluoro-ß-D-arabinofuranosyl)adenine in rats. Cancer Chemother Pharmacol 33: 484-488.[Medline]

Reichelova V, Liliemark J, and Albertioni F (1995) Structure-activity relationships of 2-chloro-2'-arabino-fluoro-2'-deoxy-adenosine and related analogues: protein binding, lipophilicity and retention in reversed phase HPLC. J Liq Chromatogr 18: 1123-1135.

Xie KC and Plunkett W (1996) Deoxynucleotide pool depletion and sustained inhibition of ribonucleotide reductase and DNA synthesis after treatment of human lymphoblastoid cells with 2-chloro-9-(2-deoxy-2-fluoro-ß-D-arabinofuranosyl)adenine. Cancer Res 56: 3030-3037.[Abstract/Free Full Text]

This article has been cited by other articles:

				K. M. King, V. L. Damaraju, M. F. Vickers, S. Y. Yao, T. Lang, T. E. Tackaberry, D. A. Mowles, A. M. L. Ng, J. D. Young, and C. E. Cass A Comparison of the Transportability, and Its Role in Cytotoxicity, of Clofarabine, Cladribine, and Fludarabine by Recombinant Human Nucleoside Transporters Produced in Three Model Expression Systems Mol. Pharmacol., January 1, 2006; 69(1): 346 - 353. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: dmd.104.002592v1 33/6/739    most recent Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Bonate, P. L. Articles by Rose, J. Q. Search for Related Content PubMed PubMed Citation Articles by Bonate, P. L. Articles by Rose, J. Q.