Department of Drug Metabolism (S.-E.W.H., D.L.-A., B.J.D., E.M.M.,
S.-H.L.C.) and Laboratory Animal Resources (W.P.F.), Merck Research
Laboratories. Rahway, New Jersey
A water-soluble phosphoramidate prodrug (L-758,298, compound I) of
the potent and selective human Substance P receptor antagonist L-754,030 (compound II) is under development as an i.v. drug for treatment of emesis, migraine, and chronic pain. Compound I undergoes hydrolysis readily to II under acidic conditions. In the studies reported herein, we investigated the stability of I in blood and hepatic subcellular fractions from rats, dogs, and humans as well as
the conversion of I to II in rats and dogs after i.v. dosing. Compound
I was converted to II rapidly in rat blood but was stable in dog and
human blood. However, the conversion was rapid in liver microsomes
prepared from dogs and humans. As expected from the results of in vitro
studies, the in vivo conversion of I to II was rapid after i.v. dosing
of I to rats and dogs. The relative extent of exposure of II after i.v.
dosing of I was estimated by comparing the dose-adjusted area under the
plasma concentration versus time curve values of II after i.v. dosing
of I with those after i.v. dosing of II. In rats, the extent of
exposure was estimated to be ~90 and ~100% at 1 and 8 mg/kg,
respectively; in dogs, that was ~59% at 0.5 mg/kg. A nonproportional
increase in the area under the concentration versus time curve value of
II with dose was observed after i.v. administration of I in dogs from
0.5 to 32 mg/kg, suggesting that the elimination of II might have been saturated at higher doses.
 |
Introduction |
In recent years, the successful
cloning and expression of human tachykinin receptors have initiated an
intensive search for selective nonpeptide receptor antagonists in the
pharmaceutical industry using in vitro receptor binding assays (Takeda
et al., 1991
; Cascieri et al., 1992; Fong et al., 1992). The biological actions of the tachykinins are mediated through specific cell-surface receptors. Three subtypes, designated as NK1,
NK2, and NK3, have been
identified on the basis of marked differences in the rank order of
potencies of agonist peptides in different tissues, with Substance P
being the preferred agonist for NK1 receptors,
neurokinin A for NK2 receptors, and neurokinin B
for NK3 receptors (Maggi et al., 1993; Otsuka and
Yoshioka, 1993). Evidence to date suggests that Substance P acting via
the NK1 receptor may be involved in the
pathoetiology of emesis (Andrews et al., 1988; Bountra et al., 1993;
Tattersall et al., 1994, 1996; Hale et al., 1996; Kris et al., 1997),
migraine (Perianin et al., 1989; Moussaoui et al., 1993; Longmore et
al., 1995), pain (Otsuka and Yanigasawa, 1988), and depression (Kramer
et al., 1998).
Several pharmacophores have been used in the synthesis of
NK1 receptor antagonists, including quinuclidine
(Snider et al., 1991; Oury-Donat et al., 1993), piperidine (McLean et
al., 1993, 1996; Stevenson et al., 1995; Ward et al., 1995;
Armour et al., 1996; Gardner et al., 1996; Gonsalves et al., 1996;
Ladduwahetty et al., 1996; Rosen et al., 1998), tryptophan (MacLeod et
al., 1994), diacylpiperazine (Mills et al., 1993),
pyrido[3,4-b]pyridine (Natsugari et al., 1995), and morpholine (Hale
et al., 1996, 1998). L-754,030
(II)1 (Fig.
1) is a very potent reversible
NK1 receptor antagonist (M.A. Cascieri,
unpublished data) (Kd = 86 pM) of the
morpholine series (Cascieri et al., 1997); however, II exhibits limited
solubility in aqueous buffers (J.V. Pivnichney and D.A. Levorse,
unpublished data; H. Jahansouz and M.L. Bray, unpublished data) (~8
µg/ml at pH 8), which presents a challenge for formulation as an i.v. drug for antiemesis (Rupniak et al., 1997). A prodrug approach was thus
taken in chemical synthesis to increase the solubility of II (Benkovic
and Sampson, 1971; Anderson et al., 1985), and among prodrug
derivatives synthesized, L-758,298 (I), a phosphoramidate prodrug of
II, exhibits the best overall profile of a prodrug including solubility
and in vivo rate of conversion. Compound I is a relatively weak
antagonist for human NK1 receptor (M.A. Cascieri,
unpublished data) (Kd = 4 nM); it is freely
soluble in aqueous buffers at a solubility of ~55 mg/ml (free acid
equivalents at pH 8) (J.V. Pivnichney and D.A. Levorse, unpublished
data; H. Jahansouz and M.L. Bray, unpublished data), which is about a
7000-fold increase in aqueous solubility compared with II. Moreover, compound I can be hydrolyzed to II chemically under mild acidic conditions (J.V. Pivnichney and D.A. Levorse, unpublished data; H. Jahansouz and M.L. Bray, unpublished data) or enzymatically via the
action of alkaline phosphatase (M.A. Cascieri, unpublished data;
S.-E.W.H. and B.J.D., unpublished data). This report describes the in
vitro stability of I in blood and its metabolic stability in
subcellular fractions of preclinical species (rat and dog) and human.
In addition, studies conducted to investigate the pharmacokinetics of I
and its conversion after i.v. dosing in rats and dogs are presented in
this report.
| |
Materials and Methods |
Chemicals.
L-758,298 (I; bis-N-methyl-D-glucamine
salt or dipotassium salt), L-754,030 (II; free base), compound III
(bis-N-methyl-D-glucamine salt or
dipotassium salt), and compound IV (free base) were synthesized and the
synthetic procedures were published separately (Fig. 1) (Hale et al.,
1998). Compounds III and IV were used as the internal standard for the
quantification of I and II, respectively. Sodium vanadate, polyethylene
glycol (PEG400), and propylene glycol were purchased from Fisher
Scientific (Springfield, NJ). All other chemicals were reagent or HPLC
grade and were purchased from EM Science (Gibbstown, NJ).
Validation of a Procedure for Blood and Plasma Sample
Preparation.
Fresh heparinized blood from male Sprague-Dawley (SD) rats or
male beagle dogs was divided into 1- or 1.4-ml aliquots to which prodrug I was added in duplicate at concentrations ranging from 25 ng/ml to 25 µg/ml or 35 ng/ml to 70 µg/ml, respectively. Sodium vanadate (final 5 mM, 25 or 35 µl of 200 mM) was added immediately to
one set of blood samples whereas saline (35 µl) was added to the
other set of blood samples. The blood samples were kept on ice for 30 min, then centrifuged at 4°C for 15 min. The resulting plasma (0.2 ml) from each tube was transferred to clean tubes, IV (60 ng) was added
as the internal standard for II, and the samples were processed
immediately by solid-phase extraction and analyzed by liquid
chromatography-tandem mass spectrometry (LC-MS/MS) for II.
Stability of I in Blood from Rat, Dog, and Human.
Fresh heparinized blood from male SD rats, male beagle dogs, or humans
(from two male subjects) was divided into 0.5-ml aliquots to which
compound I was added in triplicate at a concentration of 1 or 10 µg/ml. Blood samples were incubated at 37°C for 15, 30, 60, and 120 min. At these specified time intervals, 12.5 µl of 200 mM vanadate
solution was added immediately to each incubate and the sample was kept
on ice for 5 min and centrifuged at 4°C for 10 min. The resulting
plasma from each tube (0.2 ml) was transferred to a clean tube
containing 50 ng of III and 50 ng of IV, the respective internal
standards for I and II; this was followed by the addition of 1.7 ml of
water and 0.5 ml of acetonitrile. The mixture was processed immediately
by solid-phase extraction and analyzed by LC-MS/MS for the
concentrations of I and II simultaneously. The storage time for samples
was less than 1 week.
Metabolic Stability of I in Subcellular Fractions of Dog and
Human Liver.
All liver microsomes and cytosolic fractions were prepared using the
following procedure. Thawed livers were homogenized with 2 volumes of
50 mM Tris-buffer (pH 7.5) containing 1.15% KCl. For microsomal and
cytosolic preparations, the homogenate was centrifuged for 20 min at
9000g and the resulting supernatant was centrifuged for 60 min at 105,000g. The resulting cytosolic fractions were
recentrifuged at 105,000g for 60 min. Fractions of cytosol
were aliquoted into small tubes and stored at 
70°C. Subsequently,
the microsomal pellets were washed with 10 mM EDTA containing 1.15%
KCl and were centrifuged at 105,000g for 60 min. Washed
microsomes were resuspended in 10 mM potassium phosphate buffer (pH
7.4) containing 250 mM sucrose, aliquoted into small tubes, and stored
at 70°C. Protein concentrations were determined by a modified Lowry
assay (Smith et al., 1985). The specific cytochrome P-450 content in
each microsomal preparation was measured as described by Omura and Sato
(1964).
Frozen cytosolic and microsomal fractions (stored at 70°C) of dog
(preparation 116) and human livers (Nos. 113, 115, and 118) were used
in the in vitro studies. Human liver was obtained from Professor W.G.
Levine (Department of Molecular Pharmacology, Albert Einstein College
of Medicine, Yeshiva University, Bronx, NY). The organ donors were a
46-year-old male (113), a 59-year-old male (115), and a 43-year-old
male (118), all with no known drug history. The microsomal or cytosolic
fractions from human liver were combined with equal amounts of proteins
from three subjects (Nos. 113, 115, and 118).
Compound I (final concentration 8.1 µM) was incubated in
triplicate with cytosolic or microsomal fractions (0.5 mg/ml) of dog or
human liver at 37°C for 15, 30, 60, and 120 min. The reaction was
quenched by the addition of acetonitrile (0.5 ml) and two internal
standards, kept on ice, followed by the addition of 1.7 ml of water
(final concentration of solvent ~18%). The mixture was processed
immediately by solid-phase extraction and analyzed by LC-MS/MS for I
and II simultaneously. The storage time for samples was less than 1 week.
Dose Preparation.
The doses for compound I were prepared by dissolving I (the
bis-N-methyl-D-glucamine salt,
molecular weight 1004.9) in a solution of lactose (50 mg/ml), potassium
carbonate (1.38 mg/ml), citric acid monohydrate (0.85 mg/ml), and
sodium chloride (4 mg/ml) (pH 7.0). Doses were filtered through a 0.45 µm filter before dosing. The doses for II were prepared in a solution
of ethanol/propylene glycol/water (15:60:25, v/v/v) or in a solution of
ethanol/PEG400/water (20:60:20, v/v/v). Doses of II were prepared and
stirred constantly at 25°C overnight before dosing.
Pharmacokinetics in Rats.
Male SD rats were obtained from Charles River Breeding Laboratories
(Wilmington, MA) or Taconic Laboratories (Germantown, NY). They were
housed under standard conditions and were maintained under a 12-h
light/dark cycle in the Laboratory Animal Resources facilities, Merck
Research Laboratories, Rahway, NJ. They were allowed access to
commercial rodent chow and water ad libitum. Rats were fasted overnight
before dosing and then until 1 h after dosing. Water was allowed
ad libitum during the fasting period.
Rats were cannulated at the femoral vein for serial bleeding and
jugular vein for dosing. The b.wt. of individual rats, ranging from 0.3 to 0.4 kg, were determined on the morning of the study. Four male rats
per group were dosed i.v. with I or II by bolus injection into the
jugular vein at 1, 8, or 25 mg/kg body weight or at 0.2, 2, or 5 mg/kg
b.wt., respectively. After dosing, 0.5-ml specimens of blood were
collected by serial sampling from the femoral cannula at 2 to 3, 5, 15, and 30 min, and 1-ml specimens of blood were drawn at 1, 2, 4, 6, 8, 10, 24, 30, 48, and 72 h. After 1 h, blood was replaced with
an equal volume of sterile heparinized saline and donor blood.
Pharmacokinetics in Dogs.
Six male Beagle dogs were housed under standard conditions and were
maintained under a 12-h light/dark cycle in the Laboratory Animal
Resources facilities, Merck Research Laboratories, Rahway, NJ. They
were allowed access to water ad libitum. Dogs were fasted overnight
before dosing and then until 4 h after dosing. Water was allowed
during the fasting period. The b.wt. of the individual dogs ranged from
9.6 to 14.3 kg.
The dogs were dosed i.v. with I or II by bolus injection into the
cephalic vein via an indwelling vascular catheter at 0.5 or 2 mg/kg
b.wt. or at 0.2, 0.5, and 2 mg/kg body weight, respectively. At the 32 mg/kg dose of I, the dogs were dosed by infusion into the cephalic vein
via an indwelling vascular catheter for 45 s at 3.2 ml/kg followed
by a saline flush for 15 s. After dosing, heparinized blood
samples (5 ml) were collected by serial bleeding from the jugular vein
at 2.5 (or 3), 5 (or 6), 15, 30 min, 1, 2, 4, 6, 8, 10, 24, 30 (only at
0.5 mg/kg), 48, and 72 h.
Sample Preparation.
To minimize enzymatic hydrolysis of I in blood during sample
preparation, 12.5 to 125 µl of 200 mM vanadate in saline (final concentration 5 mM) was added immediately to fresh blood samples, which
were kept on ice. Plasma was obtained by centrifugation at 4°C within
30 min. Each plasma sample (0.2 ml for rats and 0.5 ml for dogs) was
added to a test tube containing 60 to 250 ng of III and 60 to 250 ng of
IV, the respective internal standards for I and II, followed by 1.7 ml
of water and 0.5 ml of acetonitrile. The mixture then was loaded onto a
Varian BondElut C18 cartridge (500 mg). The
cartridge was washed with ~6 ml of water followed by elution with
~3 ml of methanol. The methanol eluent was evaporated to dryness and
stored at 4°C (storage time less than 1 week) before analysis using a
LC-MS/MS assay.
Quantification of I and II in Plasma by LC-MS/MS.
The quantification of I and II in plasma was performed on a SCIEX API
III tandem mass spectrometer using the ionspray interface. The
collision gas used for collision-induced dissociation was argon. The
HPLC system consisted of two Shimadzu 10AD pumps, SCL-10A controller
and SIL-10A autoinjector. Chromatographic separation was performed on a
BDS-Hypersil C18 column (4.6 × 250 mm)
using a mobile phase consisting of 72% acetonitrile and 28% 10 mM
ammonium acetate (adjusted to pH 7.4 with HPLC grade triethylamine).
The flow rate was 1.05 ml/min and the effluent was split such that 5%
of the flow entered the ionspray interface. In this system, I and its
internal standard, III, eluted at approximately 2.2 min; II and its
internal standard, IV, eluted at approximately 4.4 and 4.6 min, respectively.
A two-period acquisition method was used due to the significant
difference in retention time and peak shape of the two sets of
compounds. The first period used a dwell time for selected reaction
monitoring (SRM) of 475 ms with a 10-ms pause; the dwell time for the
second period was 400 ms with a 5-ms pause. Positive ion detection was
used during data acquisition for II and its internal standard; however,
both positive and negative ion detection were used in different cases
for the detection of I and its internal standard.
A two-period SRM assay (method 1) was developed, with negative ion
detection for I and III in the first period, followed by positive ion
detection for II and IV in the second period. In the first period, the
negative precursor/product ion pairs at m/z = 613/79 and 581/79 were used for quantification of I and III,
respectively; in the second period, the positive precursor/product ion
pairs at m/z = 535/277 and 517/259 were used
for quantification of II and V, respectively.
Alternatively, a two-period positive ion SRM assay (method 2) was used
subsequently to eliminate cross talk between the channels. In the first
period, the precursor/product ion pairs at
m/z = 615/277 and 583/259 were used for
quantification of I and III, respectively; in the second period, the
precursor/product ion pairs at m/z 535/179 and
517/161 were used for quantification of II and IV, respectively.
Two standard curves were generated for each assay by plotting the peak
area ratio of the response for either I or II to that of its respective
internal standard versus the amount of compound added to the control
plasma sample. The range of concentrations used to define the standard
curve was dependent on the expected plasma levels, and the amount of
internal standard used was chosen to be roughly at the mid-point of the
standard curve. An average of three replicates at each concentration
over the entire range was used in rat and dog plasma to establish the
standard curves. A power fit regression [ Y = kXn ] was used to quantify the unknowns. The
limits of quantification for I were 6.25 to 62.5 ng/ml of rat plasma
and 25 to 100 ng/ml of dog plasma; limits of quantification for II were
6.25 to 12.5 ng/ml of rat plasma and 5 to 20 ng/ml of dog plasma.
Calculations of Pharmacokinetic Parameters.
The area under the plasma concentration versus time curve (AUC) was
determined by the UNICUE program with linear trapezoidal interpolation
in the ascending slope and logarithmic trapezoidal interpolation in the
descending slope (Yeh and Small, 1989). The portion of the AUC from the
last measurable concentration of II in plasma to infinity was estimated
by Ct/
, where
Ct represents the last measurable
concentration in plasma and is the terminal rate constant
determined from the plasma concentration versus time curve by linear
regression at the elimination phase of the semilogarithm plot.
Concentrations below the quantifiable level were treated as zero for
the purpose of calculating mean concentrations.
| |
Results |
Validation of a Procedure for Blood Sample Preparation.
To determine whether vanadate can prevent the ex vivo conversion of
prodrug (compound I) to II in blood samples, compound I (25 ng/ml-25
µg/ml) was added to rat or dog blood in the presence of saline or
vanadate (5 mM, an inhibitor for alkaline phosphatase) and the
formation of II was quantified by LC-MS/MS. No significant increase of
concentrations of II was detected in rat blood while stored on ice up
to 1 h (data not shown). When rat blood samples were treated with
saline, 18 to 1800 ng/ml of II (or 10-13% of I added) was detected in
plasma. In comparison, only 16 to 170 ng/ml of II (or 1-3% of
conversion) was detected when vanadate was added immediately to rat
blood samples after mixing with I at various concentrations to rat
blood (data not shown). In dog blood, the ex vivo conversion of I to II
was low (1-3%), and the addition of vanadate did not achieve
significant reduction of the conversion.
In Vitro Conversion of Prodrug I to II.
When the prodrug, compound I, was added to blood at concentrations of 1 and 10 µg/ml, incubations were carried out immediately and processed
as described in Materials and Methods. Vanadate was included
in the sample preparation procedure after incubation to minimize the ex
vivo conversion of I to II. The concentrations of both compounds in
plasma at selected time intervals were determined simultaneously by
LC-MS/MS. The conversion of I to II in rat, dog, or human blood was
expressed by plotting the increase in molar concentration of II against
the decrease in molar concentration of I in plasma versus time (Fig.
2). The concentrations of I in rat plasma
diminished rapidly with a half-life of ~30 min when 10 µg/ml of I
was added (Fig. 2A). Similar results were obtained from incubations
with I at 1 µg/ml (data not shown). Compound I was more stable in dog
blood than in rat blood with a half-life (in dog blood) of ~230 and
~350 min for 1 and 10 µg/ml, respectively; only 20 to 30% of I was
converted to II during the 2-h incubation period (Fig. 2B). Compound I
was very stable in human blood with less than ~15% conversion
observed during the 2-h incubation period (Fig. 2C).
View larger version (16K):
[in this window]
[in a new window]
|
Fig. 2.
Stability of prodrug I in rat, dog, and
human blood at 10 µg/ml.
I (10 µg/ml; MW 614.4 as free base) was incubated with fresh
heparinized rat (A), dog (B), human (C) blood at 37°C for 0 to 120 min. After incubation, vanadate (5 mM) was added immediately and the
samples were kept on ice to minimize any ex vivo hydrolysis of I. Plasma was obtained by centrifugation at 4°C, and 0.2 ml of the
resulting plasma was mixed with internal standard, processed by
solid-phase extraction, and analyzed simultaneously for I and II by
LC-MS/MS.
|
|
Metabolic Stability of Compound I in Subcellular Fractions of Dog
and Human Liver.
Compound I (5 µg/ml) was incubated with microsomal and cytosolic
fractions followed by simultaneous quantification of compound I and its
dephosphorylated product (compound II) by LC-MS/MS. The decline
of its concentrations with the concomitant increase of the levels of
compound II with time is shown in Figs. 3
and 4. In dog liver microsomes, the
conversion was nearly complete in 30 min with only ~5% of the
substrate remaining at that time. The conversion of I to II was rapid
in human liver microsomes such that only ~2% of I was detectable
after the first time point (15 min); the conversion was complete at 30 min. The rate of conversion was slower in cytosolic fractions for both
species (55% remaining at 30 min in dog; 78% remaining in human).
After the 2-h incubation period, conversion by hepatic cytosolic
fractions was nearly complete (~ 6% remaining) in dog, and ~65%
complete (~35% remaining) in human.
View larger version (19K):
[in this window]
[in a new window]
|
Fig. 3.
Metabolic stability of I in subcellular
fractions of dog liver.
I (8.1 µM) was incubated with microsomal or cytosolic fractions of
dog liver at 37°C for 0 to 120 min. After incubation, samples were
processed by solid-phase extraction and analyzed simultaneously for I
and II by LC-MS/MS.
|
|
View larger version (19K):
[in this window]
[in a new window]
|
Fig. 4.
Metabolic stability of I in subcellular
fractions of human liver.
I (8.1 µM) was incubated with microsomal or cytosolic fractions of
human liver at 37°C for 0 to 120 min. After incubation, samples were
processed and analyzed by LC-MS/MS.
|
|
In Vivo Conversion of I in Rats.
The conversion of I to II was studied in rats dosed i.v. at 1, 8, and
25 mg/kg b.wt. The concentrations of I and II in plasma samples were
determined simultaneously by LC-MS/MS (method 2). The concentrations of
intact I were quantifiable only at the earliest time points (2-5 min)
for the lowest dose, and up to 1 h for the highest dose. After the
25 mg/kg dose, the concentrations of I in plasma declined rapidly from
1117 to 75 ng/ml between 3 and 60 min (data not shown).
The concentration of II in plasma was maximal at the first sampling
time point (2-3 min) after i.v. dosing of I at all three dose levels
(Fig. 5). Ten hours after dosing with
I at 1, 8, and 25 mg/kg, the plasma concentrations of II were ~14,
~140, and ~620 ng/ml, respectively; at 24 h, the levels were
measurable only in one rat given the highest dose (25 mg/kg).
View larger version (14K):
[in this window]
[in a new window]
|
Fig. 5.
Mean (± S.D.) concentrations of II in
plasma of male rats dosed i.v. with I.
Four male SD rats were dosed i.v. with I
(bis-N-methyl D-glucamine
salt; mw 1004.9 salt) prepared in a solution of lactose, potassium
carbonate, citric acid monohydrate, and sodium chloride (pH 7.0), as
described in Materials and Methods. Plasma samples (0.2 ml) were processed by solid-phase extraction and analyzed
simultaneously for I and II by LC-MS/MS. The limits of quantification
for I and II were 6.25 to 62.5 and 6.25 to 12.5 ng/ml, respectively.
|
|
A near proportional increase in the AUC values of II with dose
was observed after i.v. administration of I at 1 and 8 mg/kg (Table
1). At 25 mg/kg, the AUC value increased
~4-fold over that at 8 mg/kg. The elimination curve (Fig. 5) showed a
convex phase (2-10 h) at the highest concentration, and the factors
contributing to this phenomenon is uncertain.
Pharmacokinetics of II also were studied in rats dosed i.v. with II at
0.2, 2, and 5 mg/kg. As shown in Fig. 6,
a steady decline of II was observed after all three doses. The kinetics
appeared to be linear over the dosing range, with an increase in plasma AUC values nearly proportional to the 10- and 25-fold increase in dose
from 0.2 to 2 and 5 mg/kg, respectively. Plasma clearance was ~15
ml/min/kg, the volume of distribution at steady state (Vdss) was ~3 liters/kg, and the terminal
half-life was ~3 h (Table 1).
View larger version (15K):
[in this window]
[in a new window]
|
Fig. 6.
Mean (± S.D.) concentrations of II in
plasma of male rats dosed i.v. with II.
Male SD rats (n = 4) were i.v. dosed with II (mw
534.4 free base) prepared in a solution of ethanol/propylene
glycol/water (15:60:25, v/v/v) or in a solution of PEG400/water/ethanol
(60:20:20, v/v/v). Plasma samples (0.2 ml) were processed by
solid-phase extraction and analyzed for II by LC-MS/MS. The limit of
quantification for II was 0.5 to 1.25 ng/ml.
|
|
The AUC values of II in rat plasma after dosing of I or II were
compared to estimate the relative extent of exposure of II. Due to the
large differences in the molecular weights (I, mw 1004.9 salt; II, mw
534.4 free base), the AUC values were normalized to per mole of dose.
When rats were dosed with I at 1 mg/kg, the relative extent of exposure
of II in plasma was estimated to be ~91% by comparison with the
average AUC calculated from the 0.2 and 2 mg/kg i.v. doses of II.
Similarly, when the dose was 8 mg/kg, relative extent of exposure of II
in plasma was ~100%, estimated by comparing the average AUC values
calculated from the 2 and 5 mg/kg i.v. doses of II. At the
highest dose (25 mg/kg), the relative extent of exposure was not
determined because plasma concentrations and AUC values of II exceeded
those from the highest i.v. dose (5 mg/kg) of II and therefore could
not be compared with one another.
In Vivo Conversion of I in Dogs.
The in vivo conversion and pharmacokinetics of I were studied in beagle
dogs dosed i.v. at 0.5, 2, and 32 mg/kg. The concentrations of I and II
in plasma samples were determined simultaneously by LC-MS/MS (method
1). After i.v. dosing of prodrug I to dogs, conversion to II was very
rapid and intact I levels were measurable only at 2 to 3 min for the
two lower doses. For the highest dose (32 mg/kg), concentrations of I
ranged from 32 to 77 µg/ml (mean 57 µg/ml) and declined rapidly to
1.3 µg/ml by 5 to 6 min. The corresponding concentrations of II were
~20 and 17 µg/ml, indicating a much slower decline compared with
compound I. For all three doses studied, II levels in plasma were the
highest at the first time point and declined slowly thereafter (Fig.
7). At 24-h post dosing, plasma levels of
II were 7 to 155 ng/ml for the two lower doses; at 72 h, the
levels of II (~2.6 µg/ml) remained in the quantifiable range only
in dogs treated with the highest dose.
View larger version (15K):
[in this window]
[in a new window]
|
Fig. 7.
Mean (± S.D.) concentrations of II in
plasma of male dogs dosed i.v. with I.
Three male beagle dogs were dosed i.v. with I
(bis-N-methyl-D-glucamine
salt; mw 1004.9 as salt) prepared in a solution of lactose, potassium
carbonate, citric acid monohydrate, and sodium chloride (pH 7.0).
Plasma samples (0.5 ml) were processed by solid-phase extraction and
analyzed simultaneously for I and II by LC-MS/MS with III and IV as the
respective internal standards. The limits of quantification for I and
II were 25 to 100 and 5 to 20 ng/ml, respectively.
|
|
A nonproportional increase in AUC values of II with dose was observed
after i.v. administration of I. For the doses of 2 and 32 mg/kg, the
AUC value was estimated only through 48 or 72 h due to nonlinear
kinetics and the uncertainty involved in extrapolation from the last
measurable time point to infinity. The AUC value of II exhibited an
8-fold increase as the dose increased 4-fold from 0.5 to 2 mg/kg, and a
greater than 40-fold increase between 2 and 32 mg/kg (Table
2). These results suggest that the
elimination of II might have been saturated at the higher doses.
Pharmacokinetics of II also were studied in dogs dosed i.v. with
II at 0.2, 0.5, and 2 mg/kg b.wt. A steady decline in the concentration
of II in plasma was observed after i.v. dosing at 0.2 or 0.5 mg/kg
(Fig. 8). The kinetics appeared to be
linear over the 0.2 to 0.5 mg/kg dosing range with a nearly
proportional increase in AUC values when the dose increased 2.5-fold.
The half-life was 6 to 7 h, plasma clearance was ~2.5 ml/min/kg,
and the Vdss was ~1 liter/kg for the two
lower doses (Table 2). When plasma concentrations from the 2 mg/kg dose
were plotted as a function of time, a convex phase was detected,
indicating that the rate of elimination of II was nonlinear in this
species at this dose (Fig. 8). The increase in the plasma
AUC0-72h value was ~10-fold and the decrease
in clearance was ~2-fold when the dose was increased from 0.5 to 2 mg/kg (Table 2). These results suggest that the elimination of II may
have been saturated at the 2 mg/kg dose. The relative extent of
exposure of II in plasma was ~59% at the 0.5 mg/kg dose, as
determined by comparing the plasma AUC value with the average AUC value
obtained from i.v. doses of II at 0.2 to 0.5 mg/kg, at which linear
kinetics is followed. Due to nonlinear kinetics at high concentrations,
the relative extent of exposure of II in plasma at the higher doses (2 and 32 mg/kg) could not be determined (Table 2).
View larger version (16K):
[in this window]
[in a new window]
|
Fig. 8.
Mean (± S.D.) concentrations of II in
plasma of male dogs dosed i.v. with II.
The same three dogs were dosed i.v. with II (mw 534.4 free base) in two
studies and three different dogs (4549, 2804, and 0939) were dosed i.v.
with II in one study. All of the doses were prepared in
ethanol/propylene glycol/water (1:6:3, v/v/v). Plasma samples were
processed by solid-phase extraction and analyzed for II by LC-MS/MS.
The limit of quantification for II was 1.0 to 2.5 ng/ml.
|
|
| |
Discussion |
To establish assay methods for I in plasma, a procedure was
validated for the preparation of blood samples from rats and dogs. Vanadate, an inhibitor of alkaline phosphatase, was used to inhibit the
ex vivo hydrolysis of I to II by alkaline phosphatase (Hagerstrand et
al., 1976; Hatoff and Hardison, 1982). The addition of vanadate was
effective in reducing the ex vivo conversion of I to II from 10 to 13%
to 1 to 3% in rat blood, however, its use did not eliminate the
conversion of 1 to 3% in dog blood. Therefore, under the sample preparation conditions used in our studies, about 1 to 3% of
conversion of I to II is expected to take place in both rat and dog blood.
The conversion of compound I, a phosphoramidate prodrug, to the potent
NK1 receptor antagonist, II, was essential for
its in vivo biological activity as a prodrug. This conversion was studied in rat, dog, and human blood. The conversion was rapid in rat
blood, somewhat slower in dog blood, and very slow in human blood. The
conversion of I to II was further investigated in subcellular fractions
from dog and human liver. Results indicate that compound I was
hydrolyzed rapidly in human and dog liver microsomes. Based on the
above in vitro results, it is anticipated that the conversion of I to
II also will be rapid in preclinical species (rat and dog) and humans,
when compound I is administered i.v.
As expected, the conversion of I to II in vivo was rapid in rats. A
near proportional increase in the AUC values of compound II with
increase in doses of I was observed after i.v. administration to rats
at 1, 8, and 25 mg/kg. Pharmacokinetics of compound II appeared to be
linear in rats when it was dosed at 0.2, 2, and 5 mg/kg. As shown in
Table 1, the relative extent of exposure of II in plasma after i.v.
dosing of I to rats was ~91 to 100% of that after i.v. dosing of II
at two lower doses. The results indicate that I is a suitable prodrug
that is effectively converted to II in vivo in the rat.
Likewise, the conversion of I to II in vivo was rapid in dogs such that
the level of I was not quantifiable 15 min post dosing of compound I at
three doses studied. However, a nonproportional increase in AUC values
of II with dose was observed after i.v. administration of I at 2 and 32 mg/kg, suggesting the elimination of II might have been saturated at
the high doses (Table 2). In comparison, pharmacokinetics of II
appeared to be linear in dogs from 0.2 to 0.5 mg/kg. Deviation from
linear kinetics was observed at 2 mg/kg, in that the apparent plasma
clearance of II was decreased (from 2.3-0.9 ml/min/kg). The relative
extent of exposure of II in plasma was less in dogs, ~59% at the 0.5 mg/kg dose. Due to nonlinear kinetics observed in dogs at high concentrations, the relative extent of exposure of II after i.v. dosing
of I at higher doses could not be determined. Taken all together, the
results suggest that I was effectively converted to II in vivo despite
an apparent saturation of the elimination of II.
Based on the results of in vitro stability of I, it is
anticipated that the conversion of I to II will be rapid in preclinical species (rat and dog) and humans when I is administered i.v. As illustrated in this report, the conversion of I to II in vivo was rapid
in rats and dogs, therefore, it is feasible to predict that the
conversion of I to II will be rapid in humans as well. This remains to
be seen in clinical trails of I in the near future.
We thank Professor W.G. Levine (Department of Molecular Pharmacology,
Albert Einstein College of Medicine, Yeshiva University, Bronx, NY) for
supplying human liver samples; Dr. R. Stearns, Dr. S. Vincent, Dr. J. Zagrobelny, and C. Lin for helpful discussions; Dr. J. Jahansouz and M. Bray for supplying formulation solution; M. Bray and J. Vandrilla for
determination of water content of I and II; P. Cunningham and D. Hora
for technical support; S. Painter for discussion and technical support;
and Drs. T. Baillie, M. Rowland, and K.C. Kwan for the critical review
and helpful discussions.
![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
Top
Abstract
Introduction
