Abstract
Losoxantrone is an anthrapyrazole derivative in Phase III development in the U.S. for solid tumors, notably breast cancer. To obtain information on the routes of elimination of the drug, a study was conducted in four patients with advanced solid tumors, which involved intravenous administration of 100 μCi of [14C]losoxantrone for a total dose of 50 mg/m2 during the first course of losoxantrone therapy. Blood, urine, and feces were collected for up to 2 weeks and were analyzed for total radioactivity and parent drug. In addition, feces were profiled for the presence of metabolites. Plasma concentrations of total radioactivity exhibited a temporal pattern similar to the parent drug. Combined recovery of administered total radioactivity from urine and feces was 70% with the majority (87%) of this radioactivity excreted in the feces, presumably via biliary excretion. Feces extracts were profiled for metabolites using a high-performance liquid chromatography method developed to separate synthetic standards of previously identified human urinary metabolites. Only intact losoxantrone was found in the feces. About 9% of the dose was excreted in the urine, primarily during the first 24 h and mostly in the form of parent compound. Collectively, these data indicate that fecal excretion of unmetabolized drug via biliary and/or intestinal excretion is the primary pathway of intravenously administered losoxantrone elimination in cancer patients with refractory solid tumors.
Losoxantrone (7-hydroxy-2-[2-[(2-hydroxyethyl)amino]ethyl]- 5-[[2-[(2-hydroxyethyl)amino]ethyl]amino]anthra[1,9-cd]pyrazol-6(2H)-one, dihydrochloride) (also known as CI-941, DuP 941, or biantrazole, Fig.1) is a member of the anthrapyrazole class of DNA intercalating agents in Phase III development in the U.S. Although not completely elucidated, the mechanism of action of losoxantrone is related to its ability to intercalate into DNA and block cell cycle division (Showalter et al., 1986). The clinical pharmacokinetics of losoxantrone have been well characterized during Phase I studies in cancer patients (Foster et al., 1992; Diab et al., 1999); however, the nature and extent of losoxantrone's elimination pathway have not been elucidated. In humans, renal excretion accounts for less than 10% of the dose (Graham et al., 1992). Two identified urinary metabolites (Blanz et al., 1993; Proksch et al., 1994; Richards and Sun, 1995) represent less than 0.6% of the administered i.v. dose. There is no evidence to suggest the presence of glucuronide conjugates in human urine. Therefore, in humans, the primary elimination pathway is presumably through the feces.
Attempts to elucidate losoxantrone metabolism using in vitro metabolic systems have not been successful; therefore, this study was designed to determine the time course, as well as the metabolic profile, for the elimination of [14C]losoxantrone when given as a single i.v. dose to patients with refractory solid tumors during their first course of chemotherapy.
Materials and Methods
Study Medication.
Individual vials of radiolabeled losoxantrone contained 20 μCi/ml of [14C]losoxantrone and 10 mg/ml of losoxantrone in sterile normal saline. The [14C]losoxantrone (98.4% radiochemically pure, lot CFQ7074-1) was synthesized at Amersham Pharmacia Biotech (Buckinghamshire, England) under Good Manufacturing Practices conditions and was uniformly labeled on the phenolic ring. Nonradiolabeled losoxantrone was supplied in vials containing 25 mg of lyophilized losoxantrone per vial and was reconstituted with normal saline.
Study Design.
This was an open label, multiple dose study in which four patients with advanced solid tumors received 10 min losoxantrone infusions every 3 weeks. One hundred microcuries of radiolabeled drug was mixed with sufficient unlabeled drug to make up the 50 mg/m2infusion dose for the first course of therapy. The protocol and informed consent form were reviewed and approved by the Institutional Review Board, and each patient gave written informed consent before receiving the study medication. Radiation dosimetry calculations (Snyder et al., 1975) estimated that the radiolabeled dose used in this study would result in exposures (∼3000 milliroentgen-equivalent-man) well below allowable limits (Massé and Miller, 1985).
Sample Collection.
In the first patient, blood (plasma), urine, and feces were collected for up to 2 weeks after [14C]losoxantrone administration for the determination of total radioactivity, parent drug, and metabolites (feces only). Based on data from the first patient, blood, urine, and fecal samples were collected for 1 day, 1 week, and 2 weeks, respectively, for the remaining patients.
Analyses of Total Radioactivity in Plasma, Urine, and Feces.
Total radioactivity was determined in plasma and urine by direct liquid scintillation counting (LSC1) (model 1900CA or 2250CA, Packard Instruments, Downers Grove, IL) of 0.2- to 1.0-ml aliquots to which 15 ml of Ultima Gold Scintillation fluid (Packard) was added. Fecal samples were freeze-dried, reweighed, and pulverized to a fine homogeneous powder. Duplicate aliquots of ∼0.1 g were added to Combusto-Cones (Packard), combusted in a sample oxidizer (Tri-Carb model B306, Packard), and radioactivity was determined by LSC. A specific activity individualized for each patient (mg/m2 dose/100 μCi; range: 0.83–1.10 mg/μCi) was used to convert dpm concentrations to losoxantrone equivalents (ng/ml). The validated lower limit of quantification was 7 dpm (after background subtraction) based on precision and accuracy estimates of ≤15%.
Analyses of Parent Drug in Plasma and Urine.
The extraction procedure was a previously published method for losoxantrone (Graham et al., 1989). Chromatographic separation of extracts was accomplished with a Waters (Milford, MA) Symmetry C8 HPLC column (3.9 × 150 mm). The mobile phase consisted of acetonitrile/20 mM sodium heptane sulfonic acid/100 mM sodium acetate/40 mM tetrabutylammonium hydrogen sulfate (8:8:44:40, v/v/v/v) at a flow rate of 1 ml/min. A Waters fraction collector was used for collecting 2-ml fractions of the HPLC eluents. Fifteen milliliters of Ultima Gold was added to each vial, and radioactivity was measured by LSC as described for plasma total radioactivity.
HPLC Profiling of Radioactivity in Feces.
Fecal samples containing the highest concentration of radioactivity from a patient (collected 80.7 h after dosing) were used for determining the parent drug radioactivity. Fecal samples (3 × 200 mg) or [14C]losoxantrone spiked control feces were repeatedly extracted with methanol and concentrated HCl (95:5 v/v, 5–10-ml volumes), and extracts were concentrated under a nitrogen stream to approximately 2 ml. A diluted aliquot was filtered using 0.45-μm Ultrafree centrifugal filters (Millipore Corp., Bedford, MA) and taken for analysis by LSC for tracking the recovery of radioactivity. A portion of the remaining filtrate (150 μl) was analyzed by HPLC equipped with a YMC (Wilmington, NC) phenyl column (100 × 4.6 mm) run in the isocratic mode using a mobile phase of ammonium phosphate buffer (pH 3, 54 nM)/isopropanol (93:7) at a flow rate of 1 ml/min. The eluent was mixed with Flo-Scint (Packard) in a 1:3 ratio and was monitored using a Radiomatic Flo-one/Beta model A500 radioactivity detector system (Packard). The remaining feces pellets were burned completely in a sample oxidizer.
Synthetic standards of losoxantrone (1), its monocarboxylic acid metabolites (2 and 3), and its dicarboxylic acid metabolite (4) were dissolved in mobile phase and analyzed by HPLC using UV/Vis detection (at 492 nm). The HPLC column, method, and mobile phase were identical to the one used for profiling the radioactivity in feces, as described above. Metabolites were adequately separated from each other and from losoxantrone (Fig. 1).
Studies to determine whether the feces extraction methodology using methanolic HCl would cleave an ether glucuronide conjugate, if present, were conducted using phenolphthalein glucuronide as a model compound. Phenolphthalein glucuronide (5 mg, Sigma Chemical Co., St. Louis, MO) was added to 95:5 methanol/HCl (2 ml) or incubated with Helix Pomatia β-glucuronidase (Sigma) in sodium acetate buffer (0.1 M, pH 5) and shaken overnight at 37°C. Samples were treated with 100 μl of NaOH (11.6 N) to neutralize the HCl, and then an additional 3 ml of NaOH was added to be sure that the solution was alkaline. Only the samples incubated with β-glucuronidase turned pink, indicating that hydrolysis had occurred. This result indicates that if a glucuronide conjugate were present in the feces, the glucuronide bond would remain intact during methanolic HCl extraction.
Results and Discussion
The urinary and fecal recoveries (mean ± S.D.) were 9.0 ± 2.3 and 60.9 ± 18.3%, respectively. Overall recovery (for up to 2 weeks) of total radioactivity was 69.9 ± 16.8%. A mean cumulative excretion plot is presented as an inset in Fig.2. Radioactivity was excreted into urine within the first 4 h after dosing with [14C]losoxantrone and was essentially complete within the first 24 h. Fecal excretion was evident within the first 24 h and persisted for up to 9 days.
The concentration of [14C]losoxantrone closely paralleled the concentration of total radioactivity in plasma. Mean losoxantrone and total radioactivity concentrations in plasma are shown in Fig. 2. Intact losoxantrone plasma area under the curve represented 82% of the mean total radioactivity area under the curve (2198 ng·h/ml). Radioactivity in the urine accounted for 13% of the total radioactivity recovered. Intact losoxantrone accounted for the majority (88%) of the radioactivity recovered in urine (0–24 h).
Radioactivity in the feces accounted for the majority (87%) of the total radioactivity recovered, suggesting biliary or intestinal excretion. The major portion (77%) of this radioactivity was in the form of intact losoxantrone, based on radiochemical profiling and HPLC using UV/Vis retention time comparison to synthetic standards (Fig. 1). Extraction recoveries from control feces treated with [14C]losoxantrone as well as patient feces samples were more than 80% of the total radioactivity. Losoxantrone metabolites (2, 3, and 4), as previously identified in patient urine (Blanz et al., 1993; Proksch et al., 1994; Richards and Sun, 1995), were not evident in the feces sample from a patient given [14C]losoxantrone.
There are several potential mechanisms by which intravenously administered losoxantrone could be excreted intact in the feces. First, it could go directly into the bile from the liver, as has been described for the analogous anticancer agent mitoxantrone (Savaraj et al., 1982). Second, losoxantrone could be transported by intestinal P-glycoprotein and thereby secreted from the systemic circulation into the intestine. In vitro evidence that losoxantrone is a substrate for P-glycoprotein was demonstrated using a multidrug-resistant cell line (Chen et al., 1996). Resistance to losoxantrone by the human breast cancer cell line (MCF-7) was overcome by simultaneous exposure to thaliblastine, a known P-glycoprotein substrate. Separate studies have suggested that mitoxantrone and doxorubicin are substrates for intestinal P-glycoprotein in Caco-2 cells (Peters and Roelofs, 1992). Our study was not able to differentiate between these proposed mechanisms.
Acknowledgments
We thank Nancy Cooper, the study nurse at the University of Chicago; Dr. Ryan from the University of Chicago, for radioisotope approval and dosing with the radiolabeled compound; Mark Joslin of DuPont Pharmaceuticals, for analyses of part of the samples; Peter King of DuPont Pharmaceuticals, for technical discussions of the protocol design; Thomas Emm of DuPont Pharmaceuticals, for technical suggestions on HPLC methodology; and Mary Paler of DuPont Pharmaceuticals, for study monitoring.
Footnotes
-
Send reprint requests to: Amita S. Joshi, Ph.D., Genentech, Inc., 1 DNA Way, MS-70, South San Francisco, CA 94080. E-mail: amita{at}gene.com
-
This work was presented in abstract form at the Annual Conference of the American College of Clinical Pharmacology and published as an abstract in the College's Journal [J Clin Pharmacol37:868 (1997)].
- Abbreviations used are::
- LSC
- liquid scintillation counting
- HPLC
- high-performance liquid chromatography
- UV/Vis
- UV/visible
- Received August 25, 2000.
- Accepted October 30, 2000.
- The American Society for Pharmacology and Experimental Therapeutics