Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Cetrorelix (Fig. 1) has been characterized as a potent luteinizing hormone-releasing hormone (LH-RH)² receptor antagonist free of edematogenic side effects (Bajusz et al., 1988). With cetrorelix a suppression of ovarian and testicular functions is achieved by competitive inhibition of LH-RH at its receptor in the pituitary gland. Cetrorelix was recently granted in the European Union for the prevention of premature ovulation in patients undergoing a controlled ovarian stimulation, followed by oocyte pick-up and assisted reproductive technique. Clinical evaluation in several indications concerning benign and malignant diseases is currently running.

View larger version (14K): [in this window] [in a new window]   Fig. 1.   Structural formula and amino acid sequence of (D-4-chloro[ring-U-14C] Phe)cetrorelix. The asterisk denotes the site of the 14C label.

Several pharmacokinetic studies in rats and dogs were performed and showed that pharmacokinetics after single dose are in good correlation with data on pharmacokinetics of LH-RH and its analogs (Schwahn et al., 1997; M.S. and H. Derendorf, submitted). This paper describes absorption, disposition, metabolism, and excretion of [¹⁴C]cetrorelix in rats and dogs as a part of the preclinical development of this LH-RH antagonist.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Chemicals. ¹⁴C-labeled cetrorelix acetate [D-Nal¹;(D-4-chloro-[ring-U-¹⁴C]Phe)²;D-Pal³; D-Cit⁶; D-Ala¹⁰-NH₂]LH-RH (D-20761; SB-075; Fig. 1) was synthesized by Amersham International plc (Little Chalfont, UK) and provided in 0.01 mol/liter acetic acid solution. Specific radioactivity was determined with HPLC to be 10.09 MBq/mg peptide base and radiochemical purity was >95%. The 4-chloro-phenyl-ring of the second amino acid 4-Cl-D-phenylalanine in the decapeptide cetrorelix is uniformly ¹⁴C-labeled ([¹⁴C]cetrorelix).

Animals. Albino Wistar rats of both sexes weighing between 250 and 400 g were purchased from Charles River Wiga (Sulzfeld, Germany). The rats were allowed free access to standard diet and tap water.

Dosage Form, Dose, and Route of Administration. All doses were calculated on the basis of peptide base. The doses were administered as solutions in 5.2% aqueous mannitol. In rats the test article was administered s.c. into the lower abdominal area near the thigh of the left hindleg. In the excretion study each rat received a dose of 0.1 mg/kg b.wt. The bile duct-cannulated rats received a dose of 1 mg/kg. The administered volume was 1 ml/kg b.wt. Injection (s.c.) in dogs was performed under the skin of the dorsal trunk between the scapulae. Each dog in the study received a dose of 0.08 mg/kg. The bile duct-cannulated dogs received a dose of 0.75 mg/kg (labeled test article diluted with unlabeled by 1:4, v/v). The administered volume was 0.5 ml/kg b.wt. Before each injection the cannula was controlled to be free moving under the skin of the animal to ensure the s.c. application.

Protein Binding (PB). For the determination of ex vivo PB of [¹⁴C]cetrorelix acetate salt the ultracentrifugation method (Pacifici and Viani, 1992) was chosen. Samples of 4.5 ml of dog plasma (from individual dogs and pooled from the groups) were incubated for 15 min at 37°C and subsequently centrifuged at 250,000g for 16 h at 37°C (Beckman ultracentrifuge L5-65B; Beckman Instruments, Palo Alto, CA). For the determination of the PB, 5 × 0.2 ml of the protein-free supernatant were transferred into minivials and the radioactivity of these samples was measured. The first two to three samples from the top phase were used for the calculation of the PB. The PB (%) was determined from the ¹⁴C concentration in the supernatant (C_s; dpm) and the protein solution before ultracentrifugation (C_p; dpm) using the following formula:

Pharmacokinetic Experiments.

Rats In the tissue distribution study rats (n = 4 per sex and time point) were thoracotomized under deep ether anesthesia at the respective sampling time, and blood was collected by heart puncture. The sampling times were 0.5, 1, 2, 4, 8, 24, 48, 96, and 264 h postdose. Triplicates of 500-µl aliquots of blood were combusted in an oxidizer (OX 500; Zinsser, Frankfurt, Germany). The remaining blood was used to prepare plasma. Duplicates of 500-µl aliquots of plasma were used directly for liquid scintillation counting (LSC).

Dogs. Blood samples (6 ml) were taken at 0 (predose), 0.5, 1, 1.5, 2, 4, 8, 12, 24, 48, 72, 96, 120, 168, 216, and 336 h postdose. The sample volume at 1.5, 4, and 8 h was increased to 9 ml for additional determination of ex vivo PB. After plasma preparation, aliquots of 500 µl were used directly for LSC.

Study Procedures and Sample Collection.

Bile duct cannulation The rats (n = 4 per sex) were laparotomized under pentobarbital anesthesia (60 mg/ml, 1 ml/kg nembutal-sodium; Sanofi Ceva, Hannover, Germany). The bile duct was stripped with forceps if necessary to be cleaned from surrounding tissue. Then the bile duct was positioned under tension using artery forceps for easier handling during cannulation and ligature. A silk suture was loosely placed around the duct and a V-shaped cut using microchirurgic scissors was performed in the proximal (liver) part of the duct and the anterior cannula was inserted and was fixed by tightening the ligature. A successful cannulation is visible when bile immediately flows into the cannula. The same procedure was repeated for the distal (duodenum) cannula to insert the posterior tube. The cannulae inserted are rigid, 2-cm-long polyethylene tubes (inner/outer diameter 0.28/0.5 mm; Kronlab, Sinsheim, Germany) connected with silicon tubes (inner/outer diameter 0.5/0.75 mm; Kronlab). The polyethylene tubes were placed in the abdominal cavity in a wide slope to prevent obstruction or disconnection. The catheters were s.c. tunneled to the neck of the rat and exteriorized between the shoulder blades. Both catheters were connected to enable bile flow into duodenum during reconvalescence of the rat and hidden under a "jacket" (Harvard Apparatus, South Natick, MA) used to protect the rat from self-injuries. After surgery the rats had access to water containing an isotonic glucose solution.

Collection of urine and feces. Urine and feces from both species were collected at room temperature by use of metabolic cages. Feces samples were homogenized after adding 1.5-fold of the sample weight of water. Fractionated excretion of total radioactivity in urine and feces of rats was determined in the groups designated for the terminal time point of the tissue distribution study (264 h). Urine was collected in those groups over the following intervals: 0- to 2-, 2- to 4-, 4- to 6-, 6- to 8-, 8- to 12-, 12- to 24-h, and in intervals of 24 h until 264 h. Feces were collected in 24-h intervals until 264 h. Cage wash was performed at the end of each feces collection interval. From the other groups (time points 0.5-96 h) in the tissue distribution study excreta were collected cumulatively until the respective times of sacrifice. Urine and feces of dogs were collected through 0- to 2-, 2- to 4-, 4- to 6-, 6- to 8-, 8- to 10-, 10- to 12-, and 12- to 24-h intervals, and in intervals of 24 h until 336 h postdose.

Collection of bile. In the bile duct-cannulated rats excreta and bile were collected at room temperature in 0- to 2-, 2- to 4-, 4- to 6-, 6- to 8-, 8- to 24-, and 24- to 48-h intervals; urine and feces were collected as stated above. Feces were collected from 0 to 24 and 24 to 48 h. For collection of bile, the loop connecting the anterior and posterior catheter during reconvalescence was removed and the anterior catheter was connected to a silicon tube that was guided through a metal-protecting tether and connected with a two-channel swivel (Harvard Apparatus, South Natick, MA) on the top of the metabolic cages. Bile was then collected in a vial placed beside the cage.

Enteral reabsorption of radioactivity. Bile samples from four male donor rats collected in fractions (0-6 h) were pooled. Total radioactivity of the donor bile was determined and 5 ml/kg were administered as an infusion over 10 min via the distal (duodenal) tube corresponding to 8.5 µg-eq cetrorelix base/kg b.wt. to three male bile duct-cannulated recipient rats.

Tissue distribution in rats. After exsanguination by cardiac puncture under deep ether anesthesia the organs in thorax and abdominal cavity were dissected and transferred into vials. Gastrointestinal (GI) tract was sectioned, separated from its content, and cleaned with 2 ml of physiological saline solution added to its respective content. Skin was sampled from the thoracic region. Adipose tissue was taken from the abdominal cavity after removal of the GI tract. Muscle tissue was dissected from the thigh of a hind leg. Brain was removed as a whole and dissected into cerebrum, cerebellum, and hypothalamus. Pituitary gland was removed by the use of small rounded forceps. Bone was taken from the skull while opened for preparation of the brain. The site of administration was sampled by dissecting the area of skin and the underlying muscle tissue (2 × 2 cm) around the site marked after injection. Urinary bladder was rinsed with physiological saline to remove residual urine. Residual ¹⁴C concentrations in the remaining carcass were determined in the groups of the excretion balance study (terminal time point) from homogenized samples of the shredded carcass.

Workup of the samples. The dissected tissues were cleaned from blood (cellulose pads) and from any other tissues. Larger organs were homogenized in scintillation vials by an Ultra Turrax homogenizer without the addition of water. Approximately 120- to 150-mg aliquots were subjected to alkaline hydrolysis with NaOH (1 mol/liter, 1 ml) for 1 h at 80°C in a water bath under slight shaking. The hydrolysates were neutralized with HCl (2 mol/liter, 0.5 ml) and, if necessary, bleached with 50 µl of H₂O₂. Scintillation cocktail (3 ml) was added for LSC. Adipose tissue was treated with 0.5 ml NaOH, 0.25 ml HCl, and 4 ml of scintillation cocktail. Smaller organs (e.g., adrenals, pituitary gland) were sampled into small vials and taken as a whole for alkaline hydrolysis and LSC. Carcasses of the animals involved in the excretion study were frozen in liquid nitrogen for about 45 min and then shredded in a dry ice-cooled cutting mill (type SM-1; Retsch, Haan, Germany). Five aliquots of 240 to 260 mg of this material were taken for combustion. Aliquots of blood (500 µl, triplicate), bone (150-200 mg, duplicate), contents of GI tract, feces (120-150 mg, duplicate), and carcass (250 mg, five aliquots) were combusted using an oxidizer (OX-500; Zinsser). Resulting ¹⁴CO₂ was trapped in 20 ml of a CO₂-binding scintillation cocktail (Oxisolve 500; Zinsser) and directly used for LSC.

Metabolism.

In vitro assays with [¹⁴C]cetrorelix Pancreatin from porcine pancreas (activity at least equivalent to 3 × USP specifications) was purchased from Sigma (Deisenhofen, Germany). A stock solution was prepared with 50 mg of pancreatin in 5 ml of Sörensen phosphate buffer (pH 7.4). The incubation mixtures consisted of 32 µl of pancreatin stock solution, 2 µl (D-4-chloro[ring-U-¹⁴C]Phe)cetrorelix acetate salt stock solution (Batch no. CFQ9052-A181197, 2 mg in 450 µl of 30% acetic acid) and Sörensen phosphate buffer in a final volume of 200 µl. Incubations were carried out in a thermostatic shaker at 37°C. After 3, 5, or 24 h reaction was stopped by the addition of 200 µl of ice-cold methanol. After 20 min at 25°C denatured protein was removed by centrifugation (20 min, 14,000g, 4°C) and the supernatants were analyzed. Controls were made without the substance or with heat-inactivated protein (5 min, 95°C).

Preparation of urine, bile, and feces samples. Urine samples were analyzed directly by radiometric HPLC. Bile samples were diluted 1:2 (v/v) with 30% acetic acid for HPLC. Each HPLC run of a bile sample was followed by a water injection to achieve an optimal regeneration of the HPLC column. Homogenized fecal samples of approximately 5 g were subsequently extracted three times with 5 ml of methanol/ethyl acetate 2:1 (v/v) each time by using an ultra turrax (TP 18-10; IKA, Staufen, Germany) for 3 × 30 s (about 10,000 rpm) in 30-ml centrifuge tubes (Oak Ridge; Nalge Company, Rochester, NY). After centrifugation (15 min, 10,000g) the supernatants were combined and evaporated to dryness under vacuum at 40°C. The residue obtained was reconstituted in 2 ml of 30% acetic acid in water and filtered using minisart GF and minisart NML 0.8-µm devices (Sartorius, Göttingen, Germany). The stability of samples from urine, bile, and fecal extracts was checked by repeated analysis. Keeping these samples at room temperature for 12 h did not cause any detectable decomposition as measured by HPLC.

Analytics.

Measurement of radioactivity Samples from the rat studies were analyzed for total radioactivity after dephosphorescence in a liquid scintillation counter (Rackbeta 1219 or Wallac 1409; Pharmacia-LKB; now Wallac ADL, Freiburg, Germany) for 10 min or to reach 10⁵ counts, whichever came first. In the dog study the samples were measured with a liquid scintillation counter LSC Tri-Carb 2100 TR (Packard, Germany). The counting was performed for 10 min or until the 2  coincidence was reached, whichever came first.

Determination of cetrorelix. Cetrorelix plasma concentrations were analyzed by a specific radioimmunoassay (RIA) (Csernus et al., 1990).

Metabolite Isolation and Identification.

Analytical HPLC system HPLC system 400 (Kontron Instruments, Neufahrn, Germany) with data system 450-MT2 consisted of two HPLC pumps 420, gradient mixer M 800, autosampler 465, UV detector 432 (226 nm, range 0.5, response time 2 s), and radiomonitor LB 507 B equipped with a solid scintillator flow cell YG-150 U4D at a range of 10,000 cpm with a response time of 1 s and peak half-width of 30 s (Berthold, Bad Wildbad, Germany). Analyses were conducted on a LiChrospher WP 300 RP-18 column (250 × 3 mm, 5 µm) protected by a guard column (30 × 3 mm; Merck, Darmstadt, Germany). The flow rate was 0.5 ml/min and the injection volume used was 50 or 100 µl. Separations were performed using a linear gradient from 5 to 80% B in 180 min, followed by a linear increase to 100% B in 2 min, a hold for 10 min, and a decrease to 5% B in 3 min. The mobile phases used were A, water containing 0.1% trifluoroacetic acid and B, 90% acetonitrile/10% water containing 0.1% trifluoroacetic acid. Both mobile phases were adjusted with NaOH solution (1 mol/liter) to pH 2.0. In all cases the chromatography was performed at room temperature (21-23°C).

Semipreparative HPLC (SPHPLC). The system used for the semipreparative purification of the enriched fractions consisted of a gradient pump (Waters 600 E; Millipore, Watford, UK), a UV detector (432; Kontron Instruments, Neufahrn, Germany) adjusted to 226 nm, range 0.5, response time 2 s, a manual 1-ml loop injector 7125 (Rheodyne Inc., Cotati, CA), and a radiomonitor LB 507 B equipped with a solid scintillator flow cell YG-150 U4D at a range of 10,000 cpm with a response time of 1 s and peak half-width of 30 s. (Berthold, Bad Wildbad, Germany). Separations were performed on a LiChrospher WP 300 RP-18 column (250 × 10 mm, 5 µm), protected by a guard (30 × 10 mm; Merck, Darmstadt, Germany) by isocratic elution using a mobile phase containing 30% B/70% A (v/v) as described above at a flow rate of 4 ml/min.

Isolation of metabolites from dog bile. For the isolation and purification of metabolites from dog bile at first a special adsorption/desorption chromatography (A/D) was used. The enriched fractions were separated using SPHPLC. Finally, the isolated metabolite fractions were concentrated and purified by using solid phase extraction. To achieve a separation of inherent impurities from the biological matrix as well as to prevent matrix suppression of the electrospray ionization, an HPLC system on-line linked to the electrospray-mass spectrometry (MS) system was also necessary. For A/D enrichment of dog bile, a laboratory-made glass column (150 × 50 mm) was used filled with 50 g of Baker bond phase C18, flash material (7025-00; Malinckrodt Baker B.V., Deventer, the Netherlands) that was conditioned with 2 × 100 ml of methanol followed by 2 × 200 ml of water. Ten milliliters of bile from dog 3F, collecting interval 8 to 24 h, representing about 3.5 × 10⁶ dpm, diluted with water 1:1 (v/v), were loaded on the A/D column. After washing with 4 × 100 ml water, the column was air-dried under vacuum and afterward eluted with 100 ml of methanol. The eluate was evaporated to dryness under vacuum at 35-40°C. In this way four preparations were performed by A/D. To remove bile dyestuffs retained on the column, it was reconditioned with 2 × 100 ml 10% acetic acid in water after 4 × 100 ml of methanol and 4 × 100 ml of water after each purification cycle. The methanolic residue from each A/D was reconstituted in 1.3 ml of 30% acetic acid. After centrifugation (15 min, 7500g), 1 ml of the supernatant was injected on the semipreparative column and radioactive fractions of cetrorelix and four metabolites were collected. Combined individual fractions of each component were neutralized with a few drops of 25% ammonia in water and subsequently concentrated under vacuum at 35-40°C. Then the concentrated aqueous fraction was further worked-up by solid phase extraction. To this end reversed phase extraction columns C-18 (Bakerbond 7020-06, 6 ml; Baker Inc., Phillipsburg, NJ) were equilibrated with 10 × 5 ml of methanol after conditioning with 10 × 5 ml of water. Individual concentrated aqueous fraction from SPHPLC was loaded on the column. After washing with 3 × 10 ml of water, the column was air-dried by vacuum and eluted with 2 × 5 ml of methanol. The solvent was removed from the eluate under vacuum at 40°C and the individual metabolite fraction obtained was restored frozen up to LC/MS analysis. By using this procedure, the ¹⁴C recovery of bile, collecting interval 8 to 24 h, from dog 3F was about 87%. All metabolite fractions isolated were found to be chromatographically identical with those detected in the original samples.

HPLC/electrospray mass spectroscopy (ESI-MS/MS). Samples were analyzed using an HPLC system linked to a Quattro tandem mass spectrometer equipped with an Electrospray interface (Micromass, Manchester, UK). Samples of isolated metabolite fractions were reconstituted in 30% acetic acid. From that, 100 µl was injected on a LiChrospher WP 300 RP-18 column (250 × 2 mm, 5 µm) and eluted with 100% A during the first 10 min followed by a gradient profile of 0 to 30% B linear ramp from 10 to 15 min and 30 to 90% B linear ramp from 15 to 105 min at a flow rate of 0.25 ml/min. Mobile phase A was 0.1% formic acid (0.5% for cetrorelix) in water; and B was 0.1% formic acid (0.5% for cetrorelix) in acetronitrile/water, 90:10 (v/v). After UV (226 nm) and radiometric detection the effluent was split post column to 15 µl/min and introduced into the electrospray source (source temperature 80°C, capillary voltage +3.3 kV, cone voltage 20 and 35 V, skimmer offset 5 V). Nitrogen was used as nebulizer and drying gas. Under these conditions full scan spectra were acquired as profile data with 16 points/dalton (scan range 200-1500 m/z, scan speed 260 Da/s) in positive ion mode. Collision-induced dissociation spectra of the single and double protonated molecular ions ([M+H]⁺/[M + 2H]²⁺) were obtained with xenon as collision gas at 3 × 10⁴ mb and at a collision energy of 45 eV for [M+H]⁺ ions and 25 or 35 eV for [M + 2H]²⁺ ions and cone voltages of 35 or 20 V, respectively. All data were processed by MassLynx software (Micromass, Manchester, UK).

Calculations. The conversion of cpm into dpm for determination of the total radioactivity in rat tissues was done by use of the LSC-Software (Pharmacia-LKB, now Wallac ADL, Freiburg, Germany) on the basis of calibration curves. These calibration curves were performed with blank tissues and excreta of untreated rats spiked with ¹⁴C standard ([¹⁴C]n-hexadecane reference standard; Amersham International plc) using the external standard sample quench parameter (SQP E) and the efficiency (%) determined from the internal standard counting.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

All data are presented as means if not stated otherwise. Concentration values are expressed as ng-eq of [¹⁴C]cetrorelix peptide base per unit of volume or weight.

PB. Ex vivo PB of [¹⁴C]cetrorelix in plasma samples from dogs was determined with approximately 90% in both sexes. No time dependence could be observed in dog plasma sampled at 1.5, 4.0, and 8.0 h after s.c. application of the test article.

Pharmacokinetics.

Rats After a single s.c. injection of 0.1 mg/kg of [¹⁴C]cetrorelix t_max in both sexes was observed at 2 h postdose (Table 1). The plasma concentration/time course showed a similar decline in male and female rats. In both sexes the [¹⁴C]concentrations were below lower limit of quantification at 264 h. The terminal half-lives of ¹⁴C concentrations in plasma exceeded mean residence times nearly 3-fold. The area under the curve values in male and female rats were nearly identical. ¹⁴C concentration in blood (Table 2) reached C_max at 2 h postdose. The blood/plasma ratios varied between 0.58 and 0.65 in the first 24 h after dosing and increased slightly to 0.8 at 96 h. Erythrocyte binding increased slightly but not continuously with time from <10% in both sexes to >25% at 96 h. This demonstrates that initially after dosing up to 4 h radioactivity is preferably located in the noncorpuscular fraction of whole blood (serum), but with increasing time after dosing is distributed more in and/or on the corpuscular fraction. This might not result from a changing parent drug/metabolite ratio but seem to be a distributional phenomenon, because comparable results were found with differently labeled [¹⁴C-Arg]cetrorelix, the latter having a fraction of ¹⁴C metabolites in plasma.

Dogs. Peak plasma levels were reached between 1 and 2 h after application amounting in males to 196 ng-eq/ml and in females to 209 ng-eq/ml (Table 1). The plasma concentrations declined rapidly until 72 h postdose to <1 ng-eq/ml, then declined more slowly to reach values <0.2 ng-eq/ml at 336 h postdose in both sexes, respectively. The terminal half-lives of ¹⁴C concentrations in plasma were calculated with 129.4 h in males and 104.3 h in females.

View larger version (9K): [in this window] [in a new window]   Fig. 2.   Time course of mean 14C and RIA concentrations in plasma of Beagle dogs (n = 6; both sexes after a single s.c. injection of 0.1 mg/kg [14C]cetrorelix.

Absorption from s.c. Injection Site in Rats. The fraction of dose at the site of administration declined rapidly from >70% at 0.5 h to <10% at 24 h. In parallel more than 60% of the total radioactivity was found in tissues, GI tract, and excreta within the first 2 h after s.c. injection. Twenty-four hours after dosing more than 90% of the radioactive dose was recovered in excreta. These findings represent a rapid and extensive absorption from the s.c. site of administration. In a prolonged phase of decline a residual fraction of 2.6% of administered radioactivity was detected at the s.c. site of administration at 264 h after s.c. injection.

Tissue Distribution in Rats. Due to the well known binding of cetrorelix to the LH-RH receptor (Fekete et al., 1989) the distribution into pituitary gland, the target tissue of cetrorelix, was studied. In pituitary gland, ¹⁴C concentrations were 10- to 20-fold higher compared with other brain areas. The concentration time curves of pituitary gland versus plasma (Fig. 3) demonstrate the enrichment of radioactivity in the pituitary gland from plasma. Determination of radioactivity in cerebrum and hypothalamus showed that [¹⁴C]cetrorelix could not pass the blood-brain barrier in a marked extent.

View larger version (9K): [in this window] [in a new window]   Fig. 3.   Time course of mean total radioactivity concentrations in the pituitary gland versus plasma of Wistar rats (means from n = 8 per sampling time) after a single s.c. injection of 0.1 mg/kg [14C]cetrorelix.

Excretion. In rats ¹⁴C excretion was similar in both sexes (Table 3) with small differences in the urine/feces ratios. Urinary excretion of radioactivity occurred mainly within the first 24 h and the major part of fecal excretion lasted until 48 h. In the samples collected in the following intervals only very small residues of the radioactive dose were found. The fractions excreted between days 2 and 11 amounted cumulatively to 3 to 5% of the administered dose. Total radioactivity excreted in urine and feces in male rats was 88.3% and in female rats 99.7%. In comparison, the results from excreta collected cumulatively from the tissue distribution study at the time points between 24 and 96 h supplied high recovery of the administered dose (Table 4) and qualify the low values of the excretion balance study in male rats as an artifact.

Biliary Excretion. To study biliary excretion and enterohepatic circulation, eight male rats were s.c. injected with 0.1 mg/kg [¹⁴C]cetrorelix. A group of four rats was used for estimation of the biliary excretion balance. The rats excreted 69.6% of the administered dose via bile and 29.7% in urine. The total recovery in this group was 100.2%. The resulting bile/urine ratio was 2.3:1.

Metabolism.

Metabolite profiles Metabolite profiles obtained by radiometric HPLC from intact and bile duct-cannulated rats and dogs showed that in urine only the unchanged parent drug was excreted (Table 5). Fecal and biliary metabolite profiles from rats and dogs indicated the presence of up to four metabolites, which were found chromatographically identical in rats and dogs and showed that the bile is the principal route for the excretion of metabolites (Fig. 4). The (1-7) heptapeptide was the major metabolite excreted into bile in both species in the same amount (5% of dose). Rats eliminated about three times more radioactivity via bile than dogs. But the portion of cetrorelix cleared in rat bile (23% of dose) was considerably higher than that in dog bile (3%). The (1-7)heptapeptide was the most abundant radioactive fraction in dog feces. In contrast to dogs, the fecal profiles in rats were dominated by the (1-4)tetrapeptide, which was not determined in such high amounts in rat bile samples. To address the question whether this finding was caused by different intestinal metabolism [¹⁴C]cetrorelix was incubated with pancreatin from porcine pancreas. Indeed, the metabolite profiles from this incubate demonstrated that cetrorelix was converted by pancreatin (Fig. 5). During 24-h incubation, approximately 70% of the substrate was metabolized. The major degradation product of cetrorelix coeluted with the metabolite (1-4)tetrapeptide, isolated from dog bile. This assignment was confirmed by on-line HPLC/ESI-MS/MS analysis of the supernatant in comparison with data obtained from the synthetic standard (1-4)tetrapeptide. In all incubations with liver and kidney fractions, no significant differences in drug-related radioactivity between active assays and controls were found.

View larger version (16K): [in this window] [in a new window]   Fig. 4.   Metabolite profile from bile of dog 3F, collecting interval 4 to 8 h after s.c. single dose administration of 1 mg/kg [14C]cetrorelix, representing 0.81 MBq/kg, obtained by radiometric HPLC.

View larger version (13K): [in this window] [in a new window]   Fig. 5.   Metabolite profiles from in vitro incubation of [14C]cetrorelix (31 µM) with pancreatin from porcine pancreas obtained by radiometric HPLC. Top: Active protein. Bottom: Heat-inactivated protein. The conversion product (1-4)tetrapeptide was identified by HPLC/MS/MS in comparison with the synthetic standard.

Identification of metabolites. Enriched metabolite fractions from dog bile after administration of 1 mg/kg [¹⁴C]cetrorelix to dogs were characterized by on-line HPLC/ESI-MS/MS. Four metabolites could be identified as truncated peptides of the parent drug without any additional metabolic modification (Table 6). The obtained MS spectra of the major metabolite (1-7) heptapeptide are shown in Fig. 6. Information about the quasimolecular ion of metabolites and data from MS/MS peptide sequencing are summarized in Table 6. The individual charge state of the quasimolecular ions was derived from unit mass resolved mass spectra. The loss of HNCO in collision-induced dissociation experiments resulted from the cleavage of the citrulline side chain (Müller et al., 1994). The proposed structure for the metabolites (1-7)heptapeptide and (1-4)tetrapeptide was confirmed by chromatographic and mass spectroscopic data obtained from the respective synthetic standards.

View larger version (14K): [in this window] [in a new window]   Fig. 6.   Electrospray full scan mass spectrum and product ion spectrum of the [M + 2H]2+ ion of metabolite (1-7)heptapeptide from dog bile after s.c. administration of [14C]cetrorelix.

View larger version (12K): [in this window] [in a new window]   Fig. 7.   Synoptic representation of metabolic degradation pattern of cetrorelix after single s.c. administration in rats and dogs. The structures of metabolites are proposed on the basis of ESI-MS and MS/MS results of fractions isolated from dog bile. The metabolites detected in rat bile were identified by cochromatography.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

This paper summarizes the data on absorption, distribution, metabolism, and excretion of the decapeptide (D-4-chloro[ring-U-¹⁴C]Phe)cetrorelix, an LH-RH antagonist, after s.c. injection in rats and dogs of both sexes.

[¹⁴C]Cetrorelix was absorbed rapidly, mainly during the first 2 h in both species. Only a minor binding in/on erythrocytes was visible in both species as was also published for nafarelin (Chu et al., 1985) and ganirelix (Chan et al., 1991). Plasma PB was tested ex vivo from dog plasma and was measured to be 90%. This is in good correlation with the results reported for the in vitro plasma PB of nafarelin (78-84%) whereas LH-RH was reported to bind only with 22 to 25% (Chan and Chaplin, 1985). ¹⁴C pharmacokinetics of [¹⁴C]cetrorelix acetate salt in dog plasma could be described as being similar to pharmacokinetics of specifically analyzed parent drug (Fig. 2) within 24 h postdose, indicating that almost no ¹⁴C metabolites are circulating in plasma (note: at 48 h ¹⁴C concentration is <2 ng/ml; in the semilog presentation difference might be overestimated). Nevertheless, the apparent terminal ¹⁴C half-life was 104.3 and 129.4 h in male and female dogs, respectively, which is based on small fractions of residual radioactivity in plasma far below 1 ng/ml. These small fractions are due to the flip-flop effect (absorption is the rate-limiting step) characterizing pharmacokinetics after s.c. injection of cetrorelix in higher doses (Schwahn et al., 1997; M.S. and H. Derendorf, submitted). In principle the same similarity is visible in rats between ¹⁴C plasma concentrations at each time point of the tissue distribution study and Cetrorelix plasma levels of a PK study. The target organ of cetrorelix, the pituitary gland, showed high ¹⁴C tissue concentrations and slower elimination than from plasma (Fig. 3). The reason for this might be specific binding to LH-RH receptor sites (Fekete et al., 1989), which could also explain the high ¹⁴C tissue concentrations and slow elimination in ovaries where specific binding sites are reported (Hsueh and Schaeffer, 1985). Minimal concentrations in central nervous system except of pituitary gland indicate a very low permeability of the blood-brain barrier. The rapid decline and almost complete disappearance of radioactivity at the site of s.c. injection after administration of a dose of 0.1 mg/kg demonstrated rapid and total absorption. The prolonged terminal half-lives appear due to a slow absorption getting more pronounced with increased doses as visible from the study on biliary excretion in dogs. The fast and ubiquitous distribution within the tissues, except in the brain, is visible from the correlation between t_max in plasma and most of the tissues at 2 h postdose. High tissue-to-plasma ratios except of pituitary were observed in ovaries where specific binding is supposed to be the reason as well as in tissues dealing with the elimination of the ¹⁴C test article like kidneys, liver, GI tract, and urinary bladder. The reason for the high ratios found in adrenals and spleen remains unclear. Similar results concerning pituitary and excretory organs were published for zoladex (Barnfield et al., 1988). A reason for the surprisingly high ¹⁴C tissue concentration in aorta can only be interpreted speculatively as a more or less specific binding site for peptides at the inner aorta wall, e.g., to angiotensin-converting enzyme, which is known to metabolize in vitro endogenous LH-RH (Skidgel and Erdös, 1985) but not cetrorelix (K. Braun, R. Jankowsky, P. Kuhl, M. Bernd, and B. Kutscher, submitted).

Elimination from tissues was found to occur rapidly, yielding concentrations at the limit of detection mostly within 48 h after dosing. Only in tissues dealing with elimination prolonged terminal half-lives were visible. Excretion of total radioactivity in rats occurred with high preference on the biliary route, whereas urinary excretion contributed only a quarter of excreted radioactivity. In dogs both routes of excretion were used to equal parts. Similar excretion patterns were described for detirelix and ganirelix in rats (Chan et al., 1988, 1991). Excretion of buserelin in rats (Berger et al., 1993) seems to be different from this pattern, but 79% of the ³H label of the latter was located in the amino acid proline at position 9, an amino acid that is missing in all fecally excreted metabolites of cetrorelix, detirelix, and ganirelix containing the radiolabel in the most hydrophobic part of the antagonists, which is the (1-3)tripeptide.

The study on biliary excretion in rats shows that the ¹⁴C test article and/or its ¹⁴C metabolite(s) are excreted preferentially on the biliary route. In bile 70% of the administered dose was found. The remaining fraction of 30% was excreted via urine, demonstrating the complete absorption and excretion of the test article in the investigated time period of 48 h. The investigation on enterohepatic circulation showed virtually no enteral reabsorption of biliary excreted ¹⁴C material from the GI tract. Besides cetrorelix, the bile of rats and dogs contains four metabolites, namely the (1-4)tetrapeptide, the (1-6)hexapeptide, the (1-7)heptapeptide, and the (1-9)nonapeptide (Fig. 7). Other metabolites that could be formed by oxidative or conjugative metabolism were not observed. Therefore, peptidase-related products are the only metabolites identified for cetrorelix. It is interesting to note that the relative abundance of the metabolites of cetrorelix changes between bile and feces, especially in rats. This is due to additional intestinal breakdown of the peptide by enteral peptidases or bacteria. Our results from the digestion with porcine pancreatin are compatible with such a mechanism. The metabolism of cetrorelix is likely to be a sequential process. At early time points, cetrorelix and longer peptides dominate the metabolite profiles whereas the shorter peptides are more pronounced within the later phase. Our data suggest that, as with other LH-RH-related peptides (Chan et al., 1991), cetrorelix is attacked first on the C-terminal side with a subsequent progressive formation of truncated peptides. When [¹⁴C-Arg]cetrorelix (uniformly ¹⁴C-labeled arginine in position 8) is administered, ¹⁴C-labeled urea is the sole radioactive metabolite. As [¹⁴C-Arg]cetrorelix itself does not form [¹⁴C]urea together with arginase, it is likely, that arginine is first liberated from cetrorelix, and is then converted to urea (data not shown). This is additional evidence that the metabolic attack starts at the carboxy terminal side of cetrorelix. In plasma of rats and dogs metabolites of cetrorelix can only be discussed as a very minor fraction within 48 h after dosing in opposite to the findings for ganirelix in monkey plasma (Chan et al., 1991) and zoladex in rat and rabbit serum (Barnfield and Warrander, 1988). For LH-RH agonists a greater number of metabolites and a higher extent of metabolism in urine and plasma can generally be assumed from published data (Chu et al., 1985; Barnfield and Warrander, 1988; Ueno and Matsuo, 1991).

In summary, cetrorelix is rapidly and completely absorbed after s.c. injection to rats and dogs. No relevant sex differences were obvious throughout the studies in both species. Elimination took place via a moderate metabolism. We suggest that in the gut an additional peptidase-dependent digestion of the parent drug occurs, resulting in a higher quantity of the (1-4)tetrapeptide in feces than in bile. Because the dogs did not excrete high amounts of unchanged cetrorelix into the bile, this effect is more pronounced in rats, which secrete 23% of the unchanged cetrorelix into bile. No enteral reabsorption was observed in rats. Urinary ¹⁴C excretion contributing with half of the total excretion in dogs and 25 to 30% in rats contained only intact cetrorelix.

Modified LH-RH analogs are known to have improved stability against enzymatic degradation positively correlated with the number of D-amino acids. Cetrorelix showed superior stability against enzymatic degradation in vitro (Deger et al., 1993; Braun et al., submitted) and as presented herein also in vivo. Only under the influence of pancreatin from porcine pancreas a remarkable turnover was found in vitro, as described here. The reported data provide a good basis for the clinical development of cetrorelix.