In vitro-in vivo correlation and translation to the clinical outcome for CJ-13,610, a novel inhibitor of 5-lipoxygenase

Article abstract of DOI:10.1124/dmd.110.032706

The metabolism of the 5-lipoxygenase inhibitor, 4-(3-(4-(2-methyl- 1Himidazol-1-yl)phenylthio)phenyl)-tetrahydro-2H-pyran-4-carboxamide (CJ-13,610), was investigated in liver microsomes from human and preclinical species in an effort to compare metabolite profiles and evaluate the in vitro-in vivo correlation for metabolic clearance. Overall, the metabolite profile of CJ-13,610 was comparable across the species tested with multiple oxidative metabolites observed, including sulfoxidation. The sulfoxidation kinetics characterized in rat, dog, and human liver microsomes (HLM) indicated a low apparent Michaelis-Menten constant (K_{m, app}) of 4 to 5 μM. Results from cDNA-expressed cytochrome P450 (P450) studies indicated that the metabolism in HLM was primarily mediated by CYP3A4 and 3A5. A subsequent in vitro study using ketoconazole as an inhibitor of CJ-13,610 sulfoxidation corroborated the CYP3A4/5-mediated pathway (IC₅₀ = 7 nM). Assessment of multiple methods for predicting the human pharmacokinetic profile observed with CJ-13,610 after a 30-mg single oral dose indicated that clearance scaled from human liver microsomes yielded a better prediction when coupled with a Vd_ss term that was scaled from dog [area under the concentration-time curve (AUC) and half-life within 1.3-fold of actual] versus a Vd_ss term obtained from rat. Single-species allometric scaling of clearance and Vd_ss from dog pharmacokinetic studies was equally predictive, whereas scaling from rat resulted in underpredictions of both AUC and maximal concentration (C_max). Results from these studies support the strategy of predicting human pharmacokinetics using human liver microsomal intrinsic clearance data. More importantly, results from the present investigation enabled the selection of alternative drug candidates from the chemical series via in vitro screening, while subsequently eliminating costly routine preclinical in vivo studies. Copyright

Full text of DOI:10.1124/dmd.110.032706

0090-9556/10/3807-1113–1121$20.00
DRUG METABOLISM AND DISPOSITION
Copyright © 2010 by The American Society for Pharmacology and Experimental Therapeutics
DMD 38:1113–1121, 2010
Vol. 38, No. 7
32706/3597364
Printed in U.S.A.
In Vitro-In Vivo Correlation and Translation to the Clinical
Outcome for CJ-13,610, a Novel Inhibitor of 5-Lipoxygenase
J. Matthew Hutzler,1 Collette D. Linder, Roger J. Melton,2 John Vincent, and J. Scott Daniels
Pharmacokinetics, Dynamics and Metabolism, Pfizer Global Research and Development, St. Louis Laboratories, St. Louis,
Missouri (J.M.H., C.D.L., R.J.M., J.S.D.); and Clinical Sciences, Pfizer Global Research and Development, New York,
New York (J.V.)
Received February 10, 2010; accepted April 7, 2010
ABSTRACT:
The metabolism of the 5-lipoxygenase inhibitor, 4-(3-(4-(2-methyl-1H- human pharmacokinetic profile observed with CJ-13,610 after a
imidazol-1-yl)phenylthio)phenyl)-tetrahydro-2H-pyran-4-carboxamide 30-mg single oral dose indicated that clearance scaled from human
(CJ-13,610), was investigated in liver microsomes from human and liver microsomes yielded a better prediction when coupled with a
preclinical species in an effort to compare metabolite profiles and Vdss term that was scaled from dog [area under the concentration-
evaluate the in vitro-in vivo correlation for metabolic clearance. time curve (AUC) and half-life within 1.3-fold of actual] versus a Vdss
Overall, the metabolite profile of CJ-13,610 was comparable
term obtained from rat. Single-species allometric scaling of clearance
across the species tested with multiple oxidative metabolites ob- and Vdss from dog pharmacokinetic studies was equally predictive,
served, including sulfoxidation. The sulfoxidation kinetics character- whereas scaling from rat resulted in underpredictions of both AUC and
ized in rat, dog, and human liver microsomes (HLM) indicated a low maximal concentration (Cmax). Results from these studies support the
apparent Michaelis-Menten constant (Km, app) of 4 to 5 ␮M. Results strategy of predicting human pharmacokinetics using human liver mi-
from cDNA-expressed cytochrome P450 (P450) studies indicated that crosomal intrinsic clearance data. More importantly, results from the
the metabolism in HLM was primarily mediated by CYP3A4 and 3A5. present investigation enabled the selection of alternative drug candi-
A subsequent in vitro study using ketoconazole as an inhibitor of dates from the chemical series via in vitro screening, while subsequently
CJ-13,610 sulfoxidation corroborated the CYP3A4/5-mediated path-
eliminating costly routine preclinical in vivo studies.
way (IC50 ؍ 7 nM). Assessment of multiple methods for predicting the
4-(3-(4-(2-Methyl-1H-imidazol-1-yl)phenylthio)phenyl)-tetrahydro- Kamada, 1996). Although efficacious in the treatment of asthma,
treatment with zileuton is fraught with issues such as a high daily dose
(2400 mg/day) and deleterious effects on the liver, requiring repeated
monitoring of liver enzyme levels. In addition, zileuton has been
shown to be a mechanism-based inactivator of CYP1A2 (Lu et al.,
2003), resulting in the potential for drug-drug interactions with coad-
ministered CYP1A2 substrates such as theophylline (Granneman et
al., 1995). An alternative therapeutic strategy includes use of LTB4
receptor antagonists, such as montelukast (Singulair), which has been
shown to be effective and safe in the treatment of asthma, including
the pediatric population (Reiss et al., 1996; Becker, 2000). In lieu of
the favorable safety profile of LTB4 receptor antagonists such as
montelukasst, upstream inhibition of the 5-LOX enzyme, theoretically
preventing formation of all of the aforementioned leukotrienes, pre-
sents a potential opportunity for enhanced clinical efficacy in
multiple therapeutic areas. Thus, there remains a desire for a potent
5-LOX inhibitor that can be administered to patients at a lower
therapeutic dose while maintaining the desired safety of LTB4
receptor antagonists.
2H-pyran-4-carboxamide (CJ 13,610) (Fig. 1) is a novel reversible
inhibitor of 5-lipoxygenase (5-LOX) (Fischer et al., 2004), a critical
enzyme involved in the initial step of the arachidonic acid cascade that
results in the ultimate formation of numerous proinflammatory bio-
active leukotrienes such as LTA4, B4, C4, D4, and E4 (Samuelsson,
1983). Leukotrienes have been clearly shown to be potent chemoat-
tractants for neutrophils, eosinophils, and monocytes in response to
inflammatory stimuli, and thus it is expected that inhibition of the
5-LOX enzyme may lead to successful therapeutic intervention in
several inflammation-based diseases, including asthma, rheumatoid
arthritis, and cardiovascular disease (Werz and Steinhilber, 2006).
Zileuton (Zyflo) is currently the only 5-LOX inhibitor available on
the market for the treatment of asthma (Liu et al., 1996; Wenzel and
1 Current affiliation: Boehringer-Ingelheim Pharmaceuticals Inc., Ridgefield,
Connecticut.
2 Current affiliation: Seventh Wave Laboratories, Chesterfield, Missouri.
Article, publication date, and citation information can be found at
http://dmd.aspetjournals.org.
The use of in vitro biotransformation systems to predict the phar-
macokinetic behavior of drug molecules remains an intense area of
doi:10.1124/dmd.110.032706.
ABBREVIATIONS: CJ 13,610, 4-(3-(4-(2-methyl-1H-imidazol-1-yl)phenylthio)phenyl)-tetrahydro-2H-pyran-4-carboxamide; 5-LOX, 5-lipoxygen-
ase; LT, leukotriene; P450, cytochrome P450; CP-680179, 4-(3-(4-(2-methyl-1H-imidazol-1-yl)phenylsulfinyl)phenyl)-tetrahydro-2H-pyran-4-car-
boxamide; LC, liquid chromatography; MS/MS, mass spectrometry; FMO, flavin monooxygenase; HPLC, high-performance liquid chromatogra-
phy; AUC, area under the concentration-time curve; MRT, mean residence time; hFMO, human flavin monooxygenase; HLM, human liver
microsome(s); SSS, single-species allometric scaling; CLh, hepatic clearance.
1113

1114
HUTZLER ET AL.
Reactions were initiated by addition of NADPH (1 mM) and incubated on an
Eppendorf Thermomixer at 37°C for 30 min. Incubations with FMO3 were
subjected to a 5-min preincubation period in the presence of NADPH, and the
reactions were initiated by addition of CJ-13,610 or benzydamine as a positive
control. Incubations were terminated by addition of ice-cold acetonitrile con-
taining 0.2 ␮M tolbutamide and centrifuged at 3000 rpm for 10 min. Samples
were subsequently analyzed by LC-MS/MS for determination of substrate
depletion after transfer of supernatant to a separate 96-well plate. Positive
control probe substrates for each individual P450 were included to confirm
enzyme activity. Inhibition of CJ-13,610 sulfoxidation by the CYP3A inhibitor
ketoconazole was tested with incubation conditions identical to those of
enzyme kinetic studies described above in human liver microsomes, except
that ketoconazole (0–5 ␮M) was also included in the incubation, with each
concentration in triplicate. CJ-13,610 was incubated at a single concentration
equal to the estimated Km (5 ␮M), and the total organic in each incubation
was 1% (v/v), and the sulfoxide metabolite (CP-680179) was measured by
LC-MS/MS.
Plasma and Microsomal Protein Binding. Plasma protein binding and
nonspecific microsomal binding in human, rat, and dog were performed in
triplicate using a 96-well equilibrium dialysis apparatus according to published
methods (Banker et al., 2003). To prepare the dialysis block, dialysis mem-
branes were soaked in phosphate-buffered saline, pH 7.4, for 60 min. After
hydration, membranes were rinsed with 100 mM phosphate, pH 7.4, buffer (1
mM MgCl2), separated, and allowed to soak for a minimum of 1 h. Human, rat,
FIG. 1. Structure of CJ-13,610.
research within the drug discovery continuum, with much of the
seminal work initiated by Houston (1994) and Iwatsubo et al. (1997).
In particular, the use of hepatic microsomes as an in vitro tool for
quantitatively predicting metabolic clearance, as well as the contribu-
tion of specific drug-metabolizing enzymes, is routinely applied in
drug discovery to diagnose liabilities associated with drug disposition
before nomination of drug candidates for development. To this end,
the primary objectives of the present investigations were to 1) profile
the in vitro metabolic pathways of CJ-13,610 in liver microsomes
from multiple species, 2) compare the kinetics of liver microsomal
metabolism in vitro and the in vivo distribution of CJ-13,610 across
species in an effort to assess in vitro-in vivo correlation, and and dog liver microsomes were diluted to a final concentration of 0.8, 0.5, and
0.5 mg/ml in 100 mM phosphate buffer, pH 7.4, respectively. Heparinized
human, rat, and dog plasma (Bioreclamation, Hicksville, NY) was collected
from at least three male and female fasted donors and frozen at Ϫ70°C until
use. The pH of the plasma was determined and adjusted to pH 7.4 by drop-wise
addition of dilute phosphoric acid. The final concentration of CJ-13,610 was 1
␮M in both microsomal binding (ketoconazole and midazolam as controls) and
plasma protein binding (S-warfarin as control) studies. The conditioned mem-
brane strips were then placed into the 96-well dialysis apparatus. The dialysate
side of the 96-well equilibrium dialysis block was loaded with 150 ␮l of
phosphate buffer (100 mM), pH 7.4, to prevent dehydration of the membrane.
The same volume of plasma or diluted microsomes spiked with compound was
placed into the sample side. Once the dialysis block was loaded with plasma
or microsomal samples and buffer, an adhesive sealing film was used to cover
the block to prevent evaporation. The equilibrium block was incubated at 37°C
for 4 h in a 5% CO2 incubator to ensure that equilibrium conditions were
achieved, as verified in preliminary experiments. After the incubation, a 50-␮l
aliquot of the postdialysis plasma (or microsomes) and buffer samples were
placed in a 96-well microtiter plate and then precipitated with 3 volumes of an
acetonitrile-containing internal standard. The extraction plate was centrifuged
for 5 min at 3000 rpm, and 90 ␮l was transferred to a fresh analytical plate for
LC-MS/MS analysis.
In Vitro Metabolite Profiling of CJ-13,610. The in vitro metabolite profile
of CJ-13,610 was investigated in rat, dog, and human liver microsomes. A
potassium phosphate-buffered 1-ml reaction (100 mM, pH 7.4) with CJ-13,610
(20 ␮M), liver microsomes (1 mg/ml), and MgCl2 (1 mM) was initiated by
addition of NADPH (2 mM) and incubated at 37°C for 60 min in 13 ϫ
100-mm borosilicate glass test tubes. Protein was precipitated by addition of 2
volumes of cold acetonitrile, and the resulting mixture underwent centrifuga-
tion at 3000 rpm (4°C) for 15 min. The supernatants were drawn off, placed
into clean glass test tubes, and subsequently dried under a gentle stream of
nitrogen (N2) gas. Dried samples were then reconstituted in 200 ␮l of mobile
phase [85:15 (v/v) ammonium formate (10 mM, pH 4.1)-acetonitrile] for
LC-MS/MS analysis.
3) estimate the contribution of the specific human enzymes involved
in the metabolism of CJ-13,610. This work was a retrospective anal-
ysis to investigate potential methodologies for predicting the disposi-
tion properties of CJ-13,610 as it relates to the potential for pharma-
cokinetic variability and drug-drug interaction risk of this potential
therapeutic agent in humans. Results from these analyses will prove
useful for any drug discovery program evaluating new chemical
entities within this chemical class of compounds.
Materials and Methods
Chemicals. CJ-13,610 and the sulfoxide metabolite standard (CP-680179)
were obtained from the Pfizer Global Research and Development chemical
bank. Potassium phosphate buffer, NADPH, magnesium chloride, tolbutamide,
bupropion, coumarin, dextromethorphan, diclofenac, midazolam, omeprazole,
paclitaxel, astemizole, tacrine, ketoconazole, and benzydamine were purchased
from Sigma-Aldrich (St. Louis, MO). Pooled human liver microsomes (50
donors), dog liver microsomes, and recombinant human CYP2B6, 2C8, 2J2,
and 3A5 Supersomes were purchased from BD Biosciences (San Jose, CA).
Rat liver microsomes were purchased from XenoTech, LLC (Kansas City,
MO), and recombinant human CYP1A2, 2D6, 2C9, 2C19, and 3A4 Baculo-
somes were purchased from Invitrogen (Carlsbad, CA). All other chemicals
were obtained from commercial sources and were of the highest purity avail-
able.
In Vitro Incubations. The intrinsic clearance of CJ-13,610 (1 ␮M) was
compared in human (0.8 mg/ml), rat (0.5 mg/ml), and dog (0.5 mg/ml) liver
microsomes using substrate depletion methodologies. Incubations were per-
formed in 100 mM potassium phosphate, pH 7.4, buffer (1 mM MgCl2) at
37°C and were initiated after a 5-min preincubation period by addition of
NADPH (1 mM). At selected time intervals (0, 2, 5, 10, 20, and 30 min), 50-␮l
aliquots were taken and subsequently placed into a 96-well plate containing
150 ␮l of cold acetonitrile with internal standard (0.2 ␮M tolbutamide). Plates
were then centrifuged at 3000 rpm (4°C) for 10 min, and the supernatant was
transferred to a separate 96-well plate for LC-MS/MS analysis. The in vitro
half-life (t1/2) was calculated by linear regression of the natural log percentage
remaining of substrate as a function of incubation time using GraphPad Prism
Enzyme Kinetics. The kinetics of CJ-13,610 sulfoxidation was investigated
in rat, dog, and human liver microsomes. Preliminary experiments established
the linear conditions for CJ-13,610 sulfoxidation with respect to incubation
(GraphPad Software Inc., San Diego, CA) software according to the equation time (0–20 min) and protein concentration (0–0.5 mg/ml microsomal protein).
t1/2 ϭ ln 2/slope. In an effort to identify the enzymes involved in the metab-
In brief, multiple concentrations of CJ-13,610 (0–100 ␮M) in triplicate were
olism of CJ-13,610 (1 ␮M), incubations with recombinant heterologously incubated in 100 mM potassium phosphate, pH 7.4, buffer (1 mM MgCl2) with
expressed P450 enzymes (150 pmol/ml and 15 pmol/ml for 3A4), and recom-
binant human flavin monooxygenase 3 (FMO3) (50 ␮g/ml) were conducted in
rat liver microsomes (0.05 mg/ml), human liver microsomes (0.05 mg/ml), and
dog liver microsomes (0.1 mg/ml) for 15 min at 37°C in a shaking water bath
the presence of 100 mM potassium phosphate, pH 7.4, buffer (1 mM MgCl2) (total volume 200 ␮l) after addition of 2 mM NADPH. Incubation reactions
and NADPH (1 mM) in a total volume of 1 ml (0.2 ml for FMO3 incubation). were subsequently quenched by addition of 200 ␮l of ice-cold acetonitrile

IN VITRO-IN VIVO CORRELATION OF CJ-13,610
1115
containing tolbutamide (0.2 ␮M) as internal standard. Incubation plates were
then centrifuged at 3000 rpm (4°C), and supernatant was transferred to an
analytical 96-well plate for LC-MS/MS analysis.
158.3) using a TurboIonSpray source in positive ionization mode (3.5 kV spray
voltage). The limit of quantitation (signal/noise ratio, 3:1) for CJ-13,610 was
0.0015 to 0.0024 ␮M for in vivo studies (rat, dog, and human) and 0.0049 ␮M
Pharmacokinetic Studies. Male Sprague-Dawley rats were purchased from for in vitro studies. The limit of quantitation of CP-680179 for enzyme kinetic
Charles River Laboratories, Inc. (Wilmington, MA) and acclimated to their studies was 0.0024 ␮M. Low, medium, and high quality control samples were
included in all bioanalyses, and each quality control sample was calculated to
be within 15% of nominal concentrations (data not shown). All data were
analyzed using PE Sciex Analyst 1.4.2 software. For in vitro profiling of
CJ-13,610 metabolites, an Agilent 1100 HPLC system (Agilent Technologies,
Palo Alto, CA) was coupled to a Discovery C18 column (5-␮m, 3.8 ϫ 150-mm;
Supelco, Bellefonte, PA). Solvent A was 10 mM ammonium formate (pH 4.1)
and solvent B was acetonitrile. The initial mobile phase was 85:15 A-B (v/v)
and by linear gradient transitioned to 20:80 A-B over 20 min. The flow rate
was 0.40 ml/min. The HPLC eluent was introduced via electrospray ionization
directly into a Finnigan LCQ Deca XPPLUS ion trap mass spectrometer
(Thermo Fisher Scientific, Waltham, MA) operated in the positive ion mode.
Ionization was assisted with sheath and auxiliary gas (nitrogen) set at 70 and
15 psi, respectively. The electrospray voltage was set at 5 kV with the heated
ion transfer capillary set at 350°C and 15 V. Relative collision energies of 25
to 40% were used when MS/MS operations were performed with the ion trap.
Enzyme Kinetic and Pharmacokinetic Analysis. In vitro intrinsic clear-
ance (CLint) was calculated from rat, dog, and human liver microsomes using
substrate depletion data and the standard values of 45 mg of microsomal
protein/g of liver for all species, and 40, 32, and 20 g of liver/kg b.wt. for rat,
dog, and human, respectively. In turn, hepatic clearance (CLh) was predicted
using the well stirred model, factoring in plasma protein binding (fu), nonspe-
cific microsomal binding [fu(microsomes)], and liver blood flow (Q) (70, 40, and
20.7 ml/min/kg for rat, dog, and human, respectively), according to eq. 1:
surroundings for approximately 1 week with food and water provided ad
libitum. On the day before the study, animals were anesthetized with isoflurane
(to effect) and then implanted with BASi vascular catheters in the carotid
artery and jugular vein. Animals were acclimated in Culex cages overnight
before dosing. Patency of the carotid artery catheter was maintained using the
“tend” function of a Culex ABS. Three rats were dosed intravenously with a
10% ethanol-56% PEG-400-34% phosphate-buffered saline (pH 7.4) at 0.5
mg/kg via the jugular vein catheter at a dose volume of 0.5 ml/kg. Blood
collections were performed by the Culex ABS predose and at 2, 5, 15, and 30
min and 1, 2, 4, 6, 8, 12, 18, and 24 h postdose; samples were collected into
chilled heparinized tubes and centrifuged for 10 min at 3000 rpm, and the
resulting plasma was aliquoted to 96-well plates for LC-MS/MS analysis. Male
beagle dogs (approximately 1–3 years of age and weighing between 9 and 12
kg at study initiation) were obtained from Marshall Farms (North Rose, NY).
Dogs (n ϭ 3) were dosed intravenously using a butterfly tubing set and syringe
into the cephalic vein. The intravenous dose was administered at 0.1 mg/kg (1
mg/ml) in 10% ethanol-50% PEG-400-40% phosphate-buffered saline, pH 7.4.
The dogs were placed in metabolism cages after dosing and for 24 h postdose.
All dogs were fasted overnight, but water was allowed ad libitum. Food was
returned 4 h postdose. All dogs were weighed before dosing. A catheter was
then inserted into the cephalic vein in one leg of each dog for blood collection.
Blood was collected predose and at 5, 15, and 30 min and 1, 2, 4, 6, 8, 10, and
24 h postdose. Blood was obtained from the catheter for collection times up to
2 h. After 2 h, the catheters were removed, and blood was collected from the
jugular vein for the remaining time points using a needle and syringe. Ap-
proximately 0.6 ml of blood was collected at each time point into lithium
heparin blood tubes, mixed thoroughly, and chilled. Plasma was obtained by
centrifugation at 5200 rpm for 15 min at 4°C. In addition, urine samples were
collected at 0 to 10 and 10 to 24 h postdose. Plasma and urine samples were stored
at Ϫ10°C until analysis by LC-MS/MS. All animal studies were approved by the
St. Louis Pfizer Institutional Animal Care and Use Committee. The animal care
and use program is fully accredited by the Association for Assessment and
Accreditation of Laboratory Animal Care, International.
CLint
Q ⅐ fu ⅐
fu(microsomes)
CLh ϭ
(1)
CLint
Q ϩ fu ⅐
fu(microsomes)
Enzyme kinetic parameters for CJ-13,610 sulfoxidation were estimated using
nonlinear regression within GraphPad Prism software, and the standard
Michaelis-Menten velocity equation (eq. 2):
Vmax ⅐ ͓S͔
Dosing of CJ-13,610 in Humans. The phase I first-in-human study was a
placebo-controlled, randomized parallel group study of increasing doses of
CJ-13,610 administered sequentially, at least 1 week apart. The study protocol
and informed consent documents were approved by the institutional review
board. All subjects signed an approved written informed consent form before
any study-related activities for this clinical study. A higher dose was given
only if the preceding dose did not result in any safety concerns. There were six
subjects in each dose group, with four randomized to active study drug and two
to a matching placebo. CJ-13,610 was administered as a solution to fasted male
subjects at doses of 1, 3, 10, 30, 100, and 300 mg. A light meal was provided
4 h after dosing. Blood samples were collected at intervals for up to 96 h
postdose for the assay of plasma CJ-13,610 and plasma LTB4. Urine was
collected from 0 to 24 and 25 to 96 h, the volume was measured, and aliquots
were sent to Pfizer Laboratories for bioanalysis of CJ-13,610. Blood pressure,
pulse rate, respiratory rate, and oral temperature were measured at intervals for
24 h and subsequently every 24 h for the duration of the study.
Liquid Chromatography-Mass Spectrometry Analysis. CJ-13,610 and
sulfoxide (CP-780179) were both analyzed on a PE Sciex API-3000 triple
quadrupole instrument. The mass spectrometer was equipped with an electro-
spray ionization interface connected in line with a Shimadzu LC20AD pump
and a CTC PAL (Leap Technologies, Carrboro, NC) autosampler. Analytes
were separated using a Zorbax 3.5-␮m Eclipse Plus C18 2.1 ϫ 50-mm column
with a gradient elution profile. The mobile phase was flowing at 0.45 ml/min,
and the gradient was initiated and held for the first 0.8 min at 95% A-5% B (A,
0.1% formic acid in H2O; B, 0.1% formic acid in acetonitrile) and was then
ramped linearly to 5% A-95% B over the next 2.3 min and held for 1.2 min.
The profile was then immediately returned to initial conditions and allowed to
reequilibrate for 2 min. The source temperature was set to 400°C, and mass
spectral analyses were performed using multiple reaction monitoring, with
transitions for CJ-13,610 (m/z 393.9/319.3) or sulfoxide metabolite (m/z 410.1/
v ϭ
(2)
Km ϩ ͓S͔
Intrinsic clearance from enzyme kinetic studies (CLЈint) was defined according
to eq. 3:
Vmax
CLЈint ϭ
(3)
Km
Inhibition of CJ-13,610 sulfoxidation in human liver microsomes was ex-
pressed as IC50 and estimated according to eq. 4, where Top is the maximum
percent control activity and Bottom is the minimum percent control activity:
Top Ϫ Bottom
Y ϭ Bottom ϩ
(4)
͑XϪLogIC50
͒
1 ϩ 10
Pharmacokinetic parameters after intravenous and oral dosing to rats and dogs
and oral dosing to humans were estimated using noncompartmental analysis
methods (WinNonlin version 5.2; Pharsight, Mountain View, CA). Total body
clearance was calculated using the standard equation (eq. 5), where AUC is
area under the plasma concentration-time curve, calculated using the linear
trapezoidal rule:
Dose
Clearance ϭ
(5)
AUC
͑0Ϫϱ͒
Volume of distribution at steady-state (Vdss) was calculated using the follow-
ing equation (eq. 6), where MRT is mean residence time and clearance is total
body clearance:
Vdss ϭ MRT ⅐ Clearance
(6)

1116
HUTZLER ET AL.
Human pharmacokinetic simulations were conducted using WinNonlin and a
one-compartment first-order absorption and elimination (no lag) pharmacoki-
netic model. Human clearance was predicted using CLh estimated either from
in vitro human liver microsomal experiments as described above or from
single-species scaling from rat and dog pharmacokinetic studies (Hosea et al.,
2009) using eq. 7:
ing the metabolic profile of CJ-13,610 in human liver microsomes is
provided in Fig. 2A. The protonated [M ϩ H]ϩ molecular ion for
CJ-13,610 was observed at m/z 394. Fragmentation of CJ-13,610
produced a key fragment ion at m/z 349 that corresponded to the loss
of the amide moiety (Fig. 2B). A subsequent ion produced from m/z
349 was observed at m/z 320 and corresponded to the further frag-
mentation of the tetrahydropyran moiety. The fragment ion at m/z 189
corresponded to the cleavage of the arylsulfide and proved useful in
the identification of metabolites of CJ-13,610 (e.g., M5). Overall, the
metabolite profile was consistent across the species tested (rat, dog,
and human; data not shown). The proposed metabolic pathways of
CJ-13,610 in human is depicted in Fig. 3.
0.75
BWhuman
BWanimal
Fuanimal
Fuhuman
CLi.v.predicted
ϭ
⅐
CLi.v.
⅐
⅐
(7)
ͩ ͪ ͩ
ͪ ͩ ͪ
animal
BWanimal
BWhuman
where CLi.v. is clearance after intravenous administration, BW is body weight,
and Fu is fraction unbound in plasma. Vdss was scaled and predicted using
pharmacokinetic data generated from intravenous studies in rat and dog using
eq. 8:
Metabolite M1. Oxidation of the imidazole moiety resulted in the
proposed acetamidine metabolite, M1 ([M ϩ H]ϩ at m/z 370), based
in part on the major fragmentation ion at m/z 353 that corresponded to
a loss of 17 Da (NH3). The loss of ammonia was not observed in the
parent CJ-13,610 mass spectrum; furthermore, the presence of an ion
at m/z 325 (Ϫ45 Da) indicated that the amide moiety was intact.
Metabolite M2. The oxidized metabolite producing a [M ϩ H]ϩ at
m/z 410 was proposed as the sulfoxide, M2. A loss of both 18 Da
(ϪH2O) and 17 Da (ϪOH) is consistent with sulfoxidation. The
presence of an ion at m/z 365 indicated that the amide moiety was
intact. The fragmentation pattern of M2 was consistent with the
synthetically prepared sulfoxide standard (CP-680179).
Fuhuman
Vdpredicted ϭ Vdanimal
⅐
(8)
Fuanimal
Absorption rate (ka, hourϪ1) for human pharmacokinetic simulations was
estimated using modeled rat oral pharmacokinetic data, consistent with re-
ported data suggesting use of rats for predicting human absorption of orally
administered drugs (Chiou and Barve, 1998).
Results
In Vitro Metabolism of CJ-13,610. The rates of in vitro liver
microsomal metabolism of CJ-13,610 in rat, dog, and human liver
microsomes, as well as plasma and microsomal binding are presented
in Table 1. CLint determined from substrate depletion methodologies
was highest for rat and lowest for dog (rat Ͼ human Ͼ dog), whereas
plasma protein binding appeared to be highest in rat (fu ϭ 0.07). The
predicted CLh using the well stirred model overestimated plasma
clearance when neither plasma nor nonspecific microsomal protein
binding was considered. Nonspecific binding to liver microsomes in
vitro appears to be low [fu(microsomes) ϳ0.8] and thus, did not have a
major impact on the calculated hepatic clearance values (Table 1).
However, factoring in both plasma and microsomal binding resulted
in CLh estimates that more closely resembled total plasma clearance
observed after intravenous dosing to rats and dogs (Table 1).
Metabolite M3. The metabolite producing an [M ϩ H]ϩ at m/z 410
was proposed to bear a hydroxylation at the tetrahydropyran moiety
(M3). The major dehydrated ion at m/z 392 (50% relative abundance)
was indicative of an aliphatic hydroxylation. The presence of an ion
at m/z 365 indicated that the amide moiety was intact.
Metabolite M4. The metabolite bearing an addition of 32 Da over
that of parent CJ-13,610 produced an [M ϩ H]ϩ at m/z 426. The
presence of fragment ions at 408 and 407 Da was indicative of a loss
of H2O and hydroxyl (–OH) and similar to that observed with M2.
The ion at m/z 220 represented an oxidation of the fragment ion of
CJ-13,610 at m/z 189. Together these data indicate M4 to be a sulfone
metabolite (a secondary metabolite of M2).
Characterization of CJ-13,610 and Metabolites In Vitro in
Liver Microsomes. The LC-mass spectrometry-UV data demonstrat-
Metabolite M5. The metabolite bearing an addition of 34 Da over
that of parent CJ-13,610 produced an [M ϩ H]ϩ at m/z 428. Proposed
as the dihydrodiolimidazole metabolite, M5, the collision-induced
fragment at m/z 370 (Ϫ58 Da) indicated a biotransformation event at
the 4,5-position of the imidazole moiety, incidentally producing a
protonated fragment ion isobaric to that of M1. A secondary fragment
of the ion at m/z 370 was observed at m/z 353 and corresponded to the
loss of ammonia (Ϫ17 Da) from the resulting acetamidine moiety.
Metabolite M6. The dihydropyran metabolite M6 produced an [M ϩ
H]ϩ at m/z 392. The collision-induced fragmentation of this metabo-
lite produced an ion at m/z 364, indicating a facile loss of ethylene
(Ϫ28 Da) from the dihydropyran moiety. The loss of formamide at
m/z 348 and the cleavage of the arylsulfide linkage indicated that the
remainder of the compound had been unaltered.
TABLE 1
Summary of substrate depletion kinetics of CJ-13,610 and enzyme kinetic
parameters for the sulfoxidation metabolic pathway in rat, dog, and human liver
microsomes
Numbers in parentheses indicate S.E. of nonlinear regression analysis. Reaction velocities
were estimated as described under Materials and Methods.
Rat
Dog
Human
Substrate depletion
In vitro half-life (min)
fu(plasma)
26
0.07
0.81
54
71
0.12
0.79
19
28
0.13
0.83
fu(microsomes)
CLint (␮l/min/mg microsomal
protein)
31
Metabolite M7. The metabolite producing an [M ϩ H]ϩ at m/z 408
bore biotransformation events of the arylsulfide (sulfoxidation) and
the tetrahydropyran moieties. Consistent with that of M2, a major
fragment ion observed was at m/z 391 and corresponded to the loss of
17 Da (–OH), a pattern consistent with sulfoxidation. The overall
mass of the remainder scaffold being 2 Da less than that of M2 and
parent CJ-13,610 implicated the corresponding oxidation of the pyran
moiety to the dihydropyran oxidation state.
CJ-13,610 Sulfoxidation Kinetics. The sulfoxide metabolite of
CJ-13,610 (CP-680179) was available as a synthetic standard, which
afforded the ability to compare the enzyme kinetics for this predom-
inant metabolic pathway across species. The enzyme kinetics of
Predicted CLh (ml/min/kg)a
Predicted CLh (ml/min/kg)b
Predicted CLh (ml/min/kg)c
CLplasma (ml/min/kg)
Enzyme kinetics (CJ-13,610
sulfoxidation)
40.6
6.2
7.4
16.5
3.1
3.8
4.4
11.7
3.1
3.6
11.5
Km, app (␮M)
4.1 (0.55)
5.4 (0.32)
4.8 (0.59)
Vmax (pmol/min/mg protein)
CLЈint (␮l/min/mg
247 (8.1)
60
77.9 (1.1)
14
184 (5.4)
38
microsomal protein)d
a No plasma or microsomal binding correction (eq. 1).
b Corrected for plasma protein binding only (eq. 1).
c Corrected for both plasma and nonspecific microsomal binding (eq. 1).
d Calculated using eq. 3.

IN VITRO-IN VIVO CORRELATION OF CJ-13,610
1117
FIG. 2. A, representative LC-mass spectrometry (I) and HPLC/UV
(II) profiles of CJ-13,610 metabolites obtained from human liver
microsomes. B, mass spectrum of CJ-13,610 obtained by LC-MS/
MS. Conditions are as described under Materials and Methods.
FIG. 3. Proposed metabolic scheme of CJ-13,610
in human liver microsomes.

1118
HUTZLER ET AL.
CYP3A4 and 3A5 play a predominant role in the metabolism of
CJ-13,610, whereas CYP2C8 and 2D6 demonstrated minimal activity
toward CJ-13,610 (84 Ϯ 0.07 and 85 Ϯ 0.21 remaining, respectively).
Negligible activity was observed for the remaining P450 enzymes
tested. Considering that the sulfoxidation of CJ-13,610 was observed
as a predominant pathway of metabolism, the role of human flavin
monooxygenase (hFMO3) was investigated. We were surprised to
find that hFMO3 did not show activity toward CJ-13,610, whereas the
positive control substrate, benzydamine, was rapidly metabolized
(data not shown). Last, ketoconazole, an established potent inhibitor
of CYP3A in vitro, was coincubated with CJ-13,610 in human liver
microsomes (Fig. 5B) and was shown to potently inhibit the formation
of the sulfoxide metabolite from CJ-13,610 (IC50 ϭ 7 nM).
Preclinical and Clinical Pharmacokinetics of CJ-13,610. The
pharmacokinetics of CJ-13,610 in rats and dogs were modeled using
a noncompartmental analysis approach (Table 2). Pharmacokinetics in
the rat were described by total body clearance of 11.5 ml/min/kg, Vdss
of 3.6 l/kg, and MRT of 5.2 h. Pharmacokinetics in the dog were
described by total body clearance of 4.4 ml/min/kg, Vdss of 2.9 l/kg, and
MRT of 11 h. Less than 1% of parent CJ-13,610 was recovered in urine
after intravenous dosing in both rats and dogs, which indicated negligible
urinary clearance in the disposition of CJ-13,610. Pharmacokinetic pa-
rameters after single oral doses of CJ-13,610 in humans (n ϭ 4) are
shown in Table 3. Exposure as estimated by AUC and the maximal
clearance after an oral dose (Cmax) was linear and proportional up to a
dose of 300 mg (Fig. 6), and the overall mean plasma elimination half-life
(t1/2) was comparable at all dose levels (Table 3).
FIG. 4. Substrate saturation plots of CJ-13,610 sulfoxidation in rat (RLM, f), dog
(DLM, Œ), and human (HLM, F) liver microsomes. Incubation conditions consis-
tent with Michaelis-Menten steady-state enzyme kinetics are described under Ma-
terials and Methods.
CJ-13,610 sulfoxidation in human liver microsomes was described by
the classic Michaelis-Menten equation (Fig. 4), with the apparent
kinetic constants shown in Table 1 for rat, dog, and human. The
apparent Michaelis-Menten constant (Km, app) was similar across spe-
cies (4.1–5.4 ␮M). However, maximal rates (Vmax) varied by greater
than 3-fold, with the highest Vmax observed in rat liver microsomes.
CLЈint was determined by dividing Vmax by Km, app (eq. 3), and the
rank order was rat Ͼ human Ͼ dog (Table 1). CLЈint estimates from
sulfoxidation kinetics closely resembled CLint values derived from
substrate depletion studies (Table 1).
Cytochrome P450 Identification. Data illustrating substrate de-
pletion (percent remaining) of CJ-13,610 after incubation for 30 min
with a battery of recombinant human P450s (1A2, 2A6, 2B6, 2C8,
2C9, 2C19, 2D6, 2J2, 3A4, and 3A5), recombinant human FMO3, and
Pharmacokinetic Simulations. In an effort to model the human
pharmacokinetic profile of CJ-13,610, multiple methods were applied,
ranging from in vitro scaling to single-species allometric scaling from
human liver microsomes are shown in Fig. 5A. It is apparent that rat and dog. The primary objective of this investigation was to
FIG. 5. In vitro characterization of the enzymes involved in the
metabolism of CJ-13,610. A, activity across a panel of recombi-
nantly expressed P450 enzymes, recombinant human FMO3, and
HLM. B, inhibition of CJ-13,610 sulfoxidation in human liver
microsomes by the CYP3A inhibitor ketoconazole (IC50 ϭ 7 nM).

IN VITRO-IN VIVO CORRELATION OF CJ-13,610
1119
TABLE 2
Summary of estimated pharmacokinetic parameters after intravenous
administration of CJ-13,610 to rats and dogs (n ϭ 3) using noncompartmental
pharmacokinetic analysis
Numbers in parentheses are S.D.
Pharmacokinetic Parameter
Rat
Dog
Dose (mg/kg)
AUC (␮M ⅐ h)
Clearance (ml/min/kg)
Vdss (l/kg)
0.5
0.1
1.83 (0.11)
11.5 (0.69)
3.6 (0.17)
5.2 (0.4)
0.94 (0.04)
4.4 (0.17)
2.9 (0.29)
11 (1.5)
MRT (h)
demonstrate that scaling in vitro intrinsic clearance using human liver
microsomes would accurately estimate clearance in human subjects.
In support of this approach, an in vitro-in vivo correlation using liver
microsomal intrinsic clearance was assessed in rat and dog. As dem-
onstrated in Table 1, predicted CLh in rat and dog very closely
resembled total plasma clearance observed in pharmacokinetic studies
(within 2-fold). When clearance was scaled from either human liver
microsomes (Vdss scaled from dog) or single-species allometric scal-
ing from dog (CL ϭ 3.2 ml/min/kg), the pharmacokinetic profile in
humans was closely predicted, with the AUC after a 30-mg single oral
dose predicted within 10%, roughly comparable to the observed S.D.
in exposure observed among four subjects (Table 4; Fig. 7). Likewise,
the predicted Cmax and half-life were within 1 to 1.5 S.D. of observed
pharmacokinetic parameters (Table 4). When Vdss scaled from rat
pharmacokinetics studies (6.7 l/kg) was combined with clearance
FIG. 6. Linearity of pharmacokinetic parameters AUC and Cmax observed after
scaled from human liver microsomes, the AUC was closely predicted, single-dose oral administration of CJ-13,610 to human subjects (n ϭ 4).
but the Cmax was underpredicted and the half-life was overpredicted.
in vivo behavior of drugs have been proposed for many years (Hous-
ton, 1994; Iwatsubo et al., 1997; Obach, 2001), and there are numer-
ous examples of successful predictions of the in vivo pharmacokinetic
properties of drugs using in vitro data (Gilissen et al., 2000; Obach,
2000; Kuperman et al., 2001; Obach et al., 2007). When this type of
in vitro-in vivo relationship is investigated with liver microsomes,
there are some requisite assumptions: 1) clearance must be primarily
mediated by phase I metabolism in the liver (e.g., P450 or FMO) and
2) clearance must not be susceptible to other non-P450-mediated
metabolism or excretion. Indeed, results of metabolism studies of
CJ-13,610 in humans demonstrated that this compound possesses the
aforementioned properties. [Results from rat studies indicated a neg-
ligible role for hepatobiliary transport in the disposition of CJ-13,610
(data not shown).] In particular, our data indicated that the in vitro
biotransformation of CJ-13,610 results primarily in metabolites aris-
ing from S-oxidation (M2, Fig. 2a) and imidazole oxidation (M1, Fig.
2a), whereas oxidation of the tetrahydropyran moiety was also ob-
served (M3) as a principal biotransformation pathway in CJ-13,610
metabolism (Fig. 3). More importantly, metabolic profiling studies in
rat and dog microsomes also indicate that the aforementioned bio-
transformation pathways significantly contribute to the hepatic me-
tabolism of CJ-13,610 (data not shown).
Intrinsic clearance data generated using the traditional substrate deple-
tion approach and the well stirred model (eq. 1) demonstrated that
predicted CLh, when corrected for both plasma protein and nonspecific
microsomal binding, was comparable to total plasma clearance (within
1.6-fold) observed for both rat and dog (Table 1). Microsomal binding
has been reported to potentially affect estimates of hepatic clearance
when microsomal fractions are used, especially for basic lipophilic
amines (Obach, 1997, 1999). Although binding to plasma proteins was
moderately high across the species tested (fu ϭ 0.07–0.13), nonspecific
binding to microsomes was low [fu(microsomes) ϳ0.8] and did not have a
dramatic impact on the calculated hepatic clearance (still within 2-fold of
Single-species allometric scaling of clearance and Vdss from rat
pharmacokinetic studies was less accurate, with the predicted AUC
accounting for roughly 60% of that observed in human subjects and the
predicted Cmax being ϳ33% of the observed value (Table 4; Fig. 7).
Discussion
CJ-13,610 is a novel reversible inhibitor of 5-LOX, a critical enzyme
in the arachidonic acid cascade that catalyzes the initial step in the
ultimate formation of numerous proinflammatory leukotrienes. Preclini-
cal pharmacology data suggest that chemical modulation of the 5-LOX
pathway by CJ-13,610 has potential therapeutic benefits in disease areas
such as asthma, liver fibrosis, and pain (Werz and Steinhilber, 2006;
Horrillo et al., 2007; Zweifel et al., 2008; Cortes-Burgos et al., 2009). In
the investigations described herein, in vitro studies were conducted with
CJ-13,610 to delineate the in vitro hepatic metabolism of this compound
in an effort to evaluate the in vitro-in vivo correlation for metabolic
clearance and enable the prediction of the pharmacokinetic profile of
CJ-13,610 observed in humans after single-dose administration.
Methodologies for scaling of in vitro metabolism data to predict the
TABLE 3
Summary of pharmacokinetics of CJ-13,610 in healthy fasted male subjects after
single oral doses administered as solution
a
a
Dose
AUC0–ϱ
Cmax
Tmax
t1/2
ng ⅐ h/ml
ng/ml
h
1 mg
3 mg
10 mg
30 mg
100 mg
300 mg
N.C.
249 (73)
783 (279)
1970 (220)
6240 (2520)
19,700 (4000)
2.1 (0.6)
11 (3)
34 (10)
144 (34)
684 (263)
2110 (1000)
6.8 (3.0)
5.0 (3.4)
4.5 (1.0)
1.8 (1.5)
1.5 (0.6)
3.5 (0.6)
N.C.
14.8 (2.0)
16.2 (2.8)
13.4 (2.1)
13.1 (2.9)
14.6 (0.9)
N.C., not calculated because of insufficient data.
a Mean (S.D.) are geometric for AUC and Cmax
.

1120
HUTZLER ET AL.
TABLE 4
Comparison of simulated human pharmacokinetic parameters using various scaling methodologies for clearance and Vdss and pharmacokinetics of CJ-13,610 after a
30-mg single oral dose to healthy male subjects
All pharmacokinetic simulations were conducted using WinNonlin, and an absorption rate constant (ka) of 0.83 hϪ1 (estimated by modeling rat oral pharmacokinetic data) was used. High
oral bioavailability was also assumed. n ϭ 4.
a
Scaling Methodology
Clearance
Vdss
AUC0–ϱ
Cmax
t1/2
h
ml/min/kg
l/kg
ng ⅐ h/ml
ng/ml
Human liver microsomes (Vdss scaled from rat)
Human liver microsomes (Vdss scaled from dog)
Scaled from rat pharmacokinetics
Scaled from dog pharmacokinetics
Single oral dose (30 mg)
3.6b
3.6b
5.3c
3.2c
6.7
3.2
6.7
3.2
1802
1802
1222
2014
51
97
48
98
144 (34)
21.7
10.4
14.7
11.6
1970 (220)
13.4 (2.1)
a Predicted Vdss in human from either rat or dog using eq. 8.
b Predicted clearance in human using eq. 1.
c Predicted clearance in human using eq. 7.
total plasma clearance). Nonetheless, in correcting for both plasma and al., 2003). With CYP3A being the predominant metabolic pathway, as
nonspecific microsomal binding, estimates of hepatic clearance were
more representative of total in vivo clearance (Table 1).
suggested by our in vitro studies, CJ-13,610 could be at risk for victim
drug-drug interactions if coadministered with potent inhibitors of
CYP3A4 such as azole antifungals or antimicrobials such as erythro-
mycin. Clinical drug-drug interaction studies would have to be conducted
to determine the magnitude of this potential interaction. hFMO3, a
microsomal enzyme that typically oxidizes nucleophilic heteroatoms, did
not appear to be involved in the sulfoxidation of CJ-13,610 (Fig. 5A),
whereas benzydamine, a reported probe substrate for FMO3 (Sto¨rmer et
al., 2000), was efficiently metabolized (data not shown). This result was
somewhat surprising in consideration of the rapid oxidation of the
sulfur by CYP3A4 and 3A5 and literature reports confirming that
sulfur-containing drug molecules, including cimetidine, methima-
zole, and methyltolylsulfide, may be substrates for FMO3 (Cash-
man, 2000). It is unfortunate that detailed structure-activity relationship
differences between cytochrome P450 and FMO enzymes have not been
thoroughly investigated, so the chemical properties of CJ-13,610 that
preclude a role for hFMO3 are not evident at this time. Recombinant
human FMO1 and FMO5 were not tested for CJ-13,610 sulfoxidation
activity, because hFMO3 is the primary isoform present in human liver.
The pharmacokinetic properties of CJ-13,610 were characterized in
an early clinical program with healthy subjects after single-dose
administration. Exposure was linear and proportional up to a dose of
300 mg (Fig. 6). With these data available, we chose to retrospectively
evaluate various scaling methods to see which may be the most
optimal for predicting the pharmacokinetic profile of CJ-13,610 in
humans, information that would be tremendously valuable for any
discovery programs working with this chemical class of compounds.
First, based on the favorable in vitro-in vivo correlation for clearance
observed with rat and dog (Table 1), we chose to scale clearance from
human liver microsomes, and use volume of distribution from rat and
dog single-species allometric scaling to compare simulations to the
observed pharmacokinetic profile after a 30-mg single oral dose. This
dose was arbitrarily selected for simulation, considering that dose
linearity was observed with CJ-13,610 up to 300 mg. Because of a
larger Vdss scaled from rat (6.7 l/kg), the half-life when HLM were
used to predict clearance was longer (21.7 h) than observed (13.4 h),
even though the AUC was closely predicted (Table 4). This resulted
in a projected Cmax (51 ng/ml) that was roughly one-third the ob-
served Cmax (144 ng/ml). However, when clearance from HLM was
scaled to humans and Vdss was scaled from dog, the pharmacokinetic
profile of CJ-13,610 was predicted within 10% for AUC, resulting in
a favorable half-life projection of 10.4 h (Table 4). In this case, it seems
that dog would be a preferred species for pharmacokinetic characteriza-
tion and, more specifically, for predicting Vdss in humans. A report that
was recently published by colleagues at Pfizer suggested that allometric
CLЈint for the sulfoxidation of CJ-13,610 calculated by eq. 3 after
enzyme kinetic analysis closely resembled total intrinsic clearance
calculated from substrate depletion methodologies. Although sulfoxi-
dation was not the only metabolic pathway observed in liver micro-
somes after incubations with CJ-13,610 (20 ␮M) for metabolite pro-
filing (Fig. 3), the observation that intrinsic clearance calculated by
Vmax/Km for the sulfoxidation pathway was comparable to that calculated
by substrate depletion methodologies (e.g., total metabolism) suggests
that sulfoxidation is the predominant metabolic pathway at lower and
more relevant concentrations (e.g., ϳ1 ␮M). In fact, the sulfoxide has
been reported to be the predominant circulating metabolite in plasma
from humans after oral administration (Dalvie et al., 2009).
Experiments were also conducted in an effort to identify which
liver microsomal enzymes were involved in the metabolism of CJ-
13,610. When incubated with a battery of recombinantly expressed
cytochrome P450 enzymes, it appeared that the CYP3A family
(CYP3A4 and 3A5) was predominantly involved in the overall me-
tabolism of CJ-13,610. In addition, when coincubated with the potent
CYP3A inhibitor ketoconazole, sulfoxidation of CJ-13,610 was in-
hibited with an estimated IC50 of 7 nM, closely resembling the
inhibitory potency of ketoconazole toward known selective probe
substrates of CYP3A4 such as midazolam and testosterone (Racha et
FIG. 7. Simulated human pharmacokinetic profile of CJ-13,610 using various scaling
methodologies compared with the clinical human pharmacokinetic profile after a single
30-mg oral dose to normal healthy subjects (n ϭ 4). An absorption rate constant (ka) of
0.83 hϪ1 (estimated by modeling rat oral pharmacokinetic data) was used for human
pharmacokinetic simulations, and clearance and Vdss parameters used in simulations are
summarized in Table 4. All simulations were conducted using WinNonlin 5.2.

IN VITRO-IN VIVO CORRELATION OF CJ-13,610
1121
Fischer L, Steinhilber D, and Werz O (2004) Molecular pharmacological profile of the nonredox-
type 5-lipoxygenase inhibitor CJ-13,610. Br J Pharmacol 142:861–868.
Gilissen RA, Ferrari L, Barnaby RJ, and Kajbaf M (2000) Human hepatic metabolism of a novel
2-carboxyindole glycine antagonist for stroke: in vitro-in vivo correlations. Xenobiotica
30:843–856.
Granneman GR, Braeckman RA, Locke CS, Cavanaugh JH, Dube LM, and Awni WM (1995)
Effect of zileuton on theophylline pharmacokinetics. Clin Pharmacokinet 29 (Suppl 2):77–83.
Horrillo R, Planaguma` A, Gonza´lez-Pe´riz A, Ferre´ N, Titos E, Miquel R, Lo´pez-Parra M,
Masferrer JL, Arroyo V, and Cla`ria J (2007) Comparative protection against liver inflamma-
tion and fibrosis by a selective cyclooxygenase-2 inhibitor and a nonredox-type 5-lipoxygen-
ase inhibitor. J Pharmacol Exp Ther 323:778–786.
Hosea NA, Collard WT, Cole S, Maurer TS, Fang RX, Jones H, Kakar SM, Nakai Y, Smith BJ, Webster
R, and Beaumont K (2009) Prediction of human pharmacokinetics from preclinical information:
comparative accuracy of quantitative prediction approaches. J Clin Pharmacol 49:513–533.
Houston JB (1994) Utility of in vitro drug metabolism data in predicting in vivo metabolic
clearance. Biochem Pharmacol 47:1469–1479.
Iwatsubo T, Hirota N, Ooie T, Suzuki H, Shimada N, Chiba K, Ishizaki T, Green CE, Tyson CA,
and Sugiyama Y (1997) Prediction of in vivo drug metabolism in the human liver from in vitro
metabolism data. Pharmacol Ther 73:147–171.
Kuperman AV, Kalgutkar AS, Marfat A, Chambers RJ, and Liston TE (2001) Pharmacokinetics and
metabolism of a cysteinyl leukotriene-1 receptor antagonist from the heterocyclic chromanol series
in rats: in vitro-in vivo correlation, gender-related differences, isoform identification, and compar-
ison with metabolism in human hepatic tissue. Drug Metab Dispos 29:1403–1409.
Liu MC, Dube´ LM, and Lancaster J (1996) Acute and chronic effects of a 5-lipoxygenase
inhibitor in asthma: a 6-month randomized multicenter trial. Zileuton Study Group. J Allergy
Clin Immunol 98:859–871.
Lu P, Schrag ML, Slaughter DE, Raab CE, Shou M, and Rodrigues AD (2003) Mechanism-based
inhibition of human liver microsomal cytochrome P450 1A2 by zileuton, a 5-lipoxygenase
inhibitor. Drug Metab Dispos 31:1352–1360.
Obach RS (1997) Nonspecific binding to microsomes: impact on scale-up of in vitro intrinsic
clearance to hepatic clearance as assessed through examination of warfarin, imipramine, and
propranolol. Drug Metab Dispos 25:1359–1369.
Obach RS (1999) Prediction of human clearance of twenty-nine drugs from hepatic microsomal
intrinsic clearance data: an examination of in vitro half-life approach and nonspecific binding
to microsomes. Drug Metab Dispos 27:1350–1359.
Obach RS (2000) Metabolism of ezlopitant, a nonpeptidic substance P receptor antagonist, in
liver microsomes: enzyme kinetics, cytochrome P450 isoform identity, and in vitro-in vivo
correlation. Drug Metab Dispos 28:1069–1076.
Obach RS (2001) The prediction of human clearance from hepatic microsomal metabolism data.
Curr Opin Drug Discov Dev 4:36–44.
Obach RS, Margolis JM, and Logman MJ (2007) In vitro metabolism of CP-122,721 ((2S,3S)-
2-phenyl-3-[(5-trifluoromethoxy-2-methoxy)benzylamino]piperidine), a non-peptide antago-
nist of the substance P receptor. Drug Metab Pharmacokinet 22:336–349.
Racha JK, Zhao ZS, Olejnik N, Warner N, Chan R, Moore D, and Satoh H (2003) Substrate
dependent inhibition profiles of fourteen drugs on CYP3A4 activity measured by a high
throughput LCMS/MS method with four probe drugs, midazolam, testosterone, nifedipine and
terfenadine. Drug Metab Pharmacokinet 18:128–138.
Reiss TF, Altman LC, Chervinsky P, Bewtra A, Stricker WE, Noonan GP, Kundu S, and Zhang
J (1996) Effects of montelukast (MK-0476), a new potent cysteinyl leukotriene (LTD4)
receptor antagonist, in patients with chronic asthma. J Allergy Clin Immunol 98:528–534.
Samuelsson B (1983) Leukotrienes: mediators of immediate hypersensitivity reactions and
inflammation. Science 220:568–575.
scaling from a single species [e.g., single-species allometric scaling
(SSS)] may be just as successful as full allometric scaling using multiple
species (Hosea et al., 2009). Therefore, these methods were also inves-
tigated for their predictive ability. In an attempt to predict the pharma-
cokinetic profile of CJ-13,610 using rat SSS, the AUC and Cmax were
underpredicted, because of the higher projected clearance and Vdss,
although the projected half-life closely predicted the observed half-life
(Table 4). When the same single-species approach was taken using dog,
the pharmacokinetic profile for CJ-13,610 was predicted with compara-
ble success compared with use of in vitro HLM data, attributable to
similar scaled clearance (3.2 ml/min/kg from dog SSS).
In summary, CJ-13,610 seems to be metabolized in vitro exclu-
sively by the cytochrome P450 family of drug-metabolizing enzymes
and specifically the CYP3A subfamily. Oxidation of the sulfur het-
eroatom was the predominant metabolic pathway and was consistently
observed across the species tested. Overall, good in vitro-in vivo
correlation was observed in rat and dog and the human pharmacoki-
netic profile of CJ-13,610 observed in the clinic after a single oral
dose was closely predicted using traditional scaling methods from
human liver microsomes. These data provide confidence moving
forward that scaling in vitro intrinsic clearance data from HLM may
be a successful approach for predicting clearance in humans for this
chemical series. It is anticipated that structurally related analogs
within the same chemical class will probably possess similar meta-
bolic clearance routes, and thus the use of predicted hepatic clearance
derived from human liver microsomes may result in a high probability
of successful pharmacokinetic predictions in humans and serve as an
efficient method for selection of additional drug candidates with
suitable pharmacokinetic properties.
Acknowledgments. We thank the in vivo group (Kathy Hotz, Steve
Wene, and Lesley Albin) for conducting pharmacokinetic studies for
CJ-13,610; Michael Baratta and I. Rochelle Riley for determining the
plasma and microsomal protein binding for CJ-13,610; and Connie
Wagner for conducting phenotyping experiments for CJ-13,610.
Sto¨rmer E, Roots I, and Brockmo¨ller J (2000) Benzydamine N-oxidation as an index reaction
reflecting FMO activity in human liver microsomes and impact of FMO3 polymorphisms on
enzyme activity. Br J Clin Pharmacol 50:553–561.
Wenzel SE and Kamada AK (1996) Zileuton: the first 5-lipoxygenase inhibitor for the treatment
of asthma. Ann Pharmacother 30:858–864.
Werz O and Steinhilber D (2006) Therapeutic options for 5-lipoxygenase inhibitors. Pharmacol
Ther 112:701–718.
Zweifel BS, Hardy MM, Anderson GD, Dufield DR, Pufahl RA, and Masferrer JL (2008) A rat
air pouch model for evaluating the efficacy and selectivity of 5-lipoxygenase inhibitors. Eur
J Pharmacol 584:166–174.
References
Banker MJ, Clark TH, and Williams JA (2003) Development and validation of a 96-well
equilibrium dialysis apparatus for measuring plasma protein binding. J Pharm Sci 92:967–974.
Becker A (2000) Clinical evidence with montelukast in the management of chronic childhood
asthma. Drugs 59 (Suppl 1):29–34.
Cashman JR (2000) Human flavin-containing monooxygenase: substrate specificity and role in
drug metabolism. Curr Drug Metab 1:181–191.
Chiou WL and Barve A (1998) Linear correlation of the fraction of oral dose absorbed of 64
drugs between humans and rats. Pharm Res 15:1792–1795.
Cortes-Burgos LA, Zweifel BS, Settle SL, Pufahl RA, Anderson GD, Hardy MM, Weir DE, Hu G,
Happa FA, Stewart Z, Muthian S, Graneto MJ, and Masferrer JL (2009) CJ-13610, an orally active
inhibitor of 5-lipoxygenase is efficacious in preclinical models of pain. Eur J Pharmacol 617:59–67.
Dalvie D, Obach RS, Kang P, Prakash C, Loi CM, Hurst S, Nedderman A, Goulet L, Smith E,
Bu HZ, and Smith DA (2009) Assessment of three human in vitro systems in the generation
of major human excretory and circulating metabolites. Chem Res Toxicol 22:357–368.
Address correspondence to: Dr. J. Matthew Hutzler, Boehringer-Ingelheim Phar-
maceuticals Inc., Drug Discovery Support (DDS), 175 Briar Ridge Rd., R&D 10578,
Ridgefield, CT 06877. E-mail: matt.hutzler@boehringer-ingelheim.com