The syntheses and in vitro biotransformation studies of [ 14C]apixaban, a highly potent, selective, efficacious and orally bioavailable inhibitor of blood coagulation Factor Xa

Article abstract of DOI:10.1002/jlcr.1890

Apixaban is a potent inhibitor of blood coagulation Factor Xa in the late stages of development. [¹⁴C]apixaban was synthesized with the 14C label in the two different lactam ring systems within the molecule for various in vitro and in vivo meta

Full text of DOI:10.1002/jlcr.1890

Research Article
Received 27 September 2010,
Revised 16 March 2011,
Accepted 18 March 2011
Published online 2 May 2011 in Wiley Online Library
(wileyonlinelibrary.com) DOI: 10.1002/jlcr.1890
The syntheses and in vitro biotransformation
studies of [14C]apixaban, a highly potent,
selective, efficacious and orally bioavailable
inhibitor of blood coagulation Factor Xa
Ã
Brad D. Maxwell,a Scott B. Tran,a Shiang-Yuan Chen,a Donglu Zhang,b
Bang-Chi Chen,c Huiping Zhang,c and Samuel J. Bonacorsi Jra
Apixaban is a potent inhibitor of blood coagulation Factor Xa in the late stages of development. [14C]apixaban was
synthesized with the 14C label in the two different lactam ring systems within the molecule for various in vitro and in vivo
metabolism studies. A nine-step synthesis of [14C]apixaban, 10, with the label in the central lactam ring was completed in
14% overall yield. A second synthesis of [14C]apixaban, 14, with the 14C label in the outer lactam ring was completed in
three steps in a 14% overall yield. No significant differences were observed between the metabolite profiles of 10 and 14,
[
14C]apixaban, from both rat and human hepatocyte or microsomal incubations.
Keywords: apixaban; Factor Xa; microsomes; hepatocytes; ADME
at 254 nm. HPLC gradient 2 = Vydac protein and peptide C18,
4.6 Â 250 mm column. Mobile phase A = 0.05% TFA in water,
mobile phase B = 0.05% TFA in acetonitrile, 0 min 90% A, 16 min
90% B, 23 min 95% B, 25 min 90% A, UV detection at 254 nm. HPLC
Gradient 3 = Phenomenex, Luna, C18 (2), 3 mm, 150 Â 4.6mm
column with Brownlee, RP-18 C18, 3 mm, 15 Â 3.2 mm guard
column, column temperature = 301C, autosampler temperature =
41C, mobile phase A = 0.4% formic acid in water adjusted to pH 3.2
with NH4OH, mobile phase B = acetonitrile, mobile phase
C = MeOH, 0 min 100% A, 25 min 20% A and 80% B, 27 min 50%
B and 50% C, 32 min 50% B and 50% C, 34 min 100% A,
flowrate = 1.0 ml/min at 323 nm. HPLC Gradient 4 = Ace 3, C18,
3mm, 150 Â 4.6 mm column with Ace 3, C18, 3 mm, 10 Â 3.0 mm
guard column, column temperature = 351C, autosampler tempera-
ture = ambient, mobile phase A = 0.4% formic acid in water,
adjusted to pH 3.2 with NH4OH, mobile phase B = acetonitrile,
0 min 100% A, 3 min 100% A, 5 min 90% A, 20 min 75% A, 50 min
75% A, 60 min 50% A, 65 min 100% B, 70 min 100% B, 72 min 100%
A, flowrate = 0.7 ml/min, UV detection at 240 nm. Preparative HPLC
condition 1: Phenomenex Luna 5 mm, 250 Â 21.2 mm column.
Introduction
Thrombotic diseases are the leading cause of death among
people living in developed countries. Although the oral anti-
coagulant warfarin is effective in both venous and arterial
thrombosis, it suffers from a narrow therapeutic index, slow
onset of therapeutic effects, several drug and dietary inter-
actions, the need for monitoring and dose adjustments. Thus,
the discovery and development of a highly potent, selective,
efficacious and orally bioavailable inhibitor of blood coagulation
Factor Xa as a replacement for warfarin would be extremely
valuable.1 Apixaban represents such an agent and the prepara-
tion of [14C] apixaban was required for various preclinical and
clinical studies. This paper describes the syntheses, purifications,
analyses and characterizations of [14C]apixaban and its use in
microsomal and hepatocyte in vitro and in vivo metabolism
studies.2
Experimental
General Experimental: All reagents used were of ACS grade or
higher. Flash chromatography was performed using either a
Biotage Flash Chromatography System or an AnaLogix BSR
Solvent Pump System with silica gel cartridges. HPLC analyses
for the syntheses were performed on an Agilent 1100 HPLC system
including solvent degasser, pump, automated injector and variable
UV detector connected to either an IN/US BetaRam Flow Detector
with a 0.25 ml detector cell or a Berthold Scintillation Detector with
a 0.5 ml detector cell. HPLC gradient 1 = Phenomenex Luna C18,
5mm, 4.6 Â 150 mm column. Mobile phase A = 0.05% TFA in water,
mobile phase B = 0.05% TFA in acetonitrile, 0 min 95% A, 6 min
50% A, 11 min 95% B, 13 min 95% B, 15 min 95% A, UV detection
aRadiochemistry-Discovery Chemical Synthesis, Bristol-Myers Squibb Company,
P. O. Box 4000, Princeton, NJ 08543, USA
bBiotransformation, Bristol-Myers Squibb Company, P. O. Box 4000, Princeton,
NJ 08543, USA
cDiscovery Chemical Synthesis, Bristol-Myers Squibb Company, P. O. Box 4000,
Princeton, NJ 08543, USA
*Correspondence to: Brad D. Maxwell, Radiochemistry-Discovery Chemical
Synthesis, Bristol-Myers Squibb Company, P. O. Box 4000, Princeton, NJ 08543,
USA.
E-mail: brad.maxwell@bms.com
J. Label Compd. Radiopharm 2011, 54 418–425
Copyright r 2011 John Wiley & Sons, Ltd.

B. D. Maxwell et al.
Mobile phase A = 0.1% TFA in water, mobile phase B = 0.1% TFA fractions was removed under reduced pressure to give 766 mg,
in acetonitrile, 0 min 70% A, 18 min 95% B, 22 min 95% B, 25 min 200 mCi, (100% yield) of a light yellow solid. HPLC analysis of the
70% A, flowrate = 15 ml/min. UV detection at 254 nm. Preparative product using gradient 1 showed the product to be 100%
HPLC condition 2: Zorbax SB-C18 5 mm, 250 Â 21.2 mm column. radiochemically pure. Co-injection of the radiolabeled product
Mobile phase A = 0.05% TFA in water, mobile phase B = 0.05% TFA with an unlabeled standard of product 2 produced a single peak.
in acetonitrile, 0 min 90% A, 16 min 90% B, 23 min 95% B, 25 min LC/MS showed a single peak with m/z ESI (1) = 221(4%),
90% A, Flowrate = 9.9 ml/min, UV detection at 254 nm. LC/MS 223(100%), 224(9%) as expected. 1H-NMR (300MHz, DMSO-D6)
analysis was performed on a Finnigan LXQ Mass Spectrometer d ppm 8.30(d, J = 9.06Hz, 2H), 7.68(d, J = 9.06Hz, 2H), 3.78(t,
System. NMR spectra were recorded on a Bruker AVANCE II J = 5.57Hz, 2H), 2.49–2.54(m, 2H, partially buried under DMSO
300 MHz Spectrometer with an UltrashieldTM Magnet or a Varian peak), 1.86–2.01(m, 4H).
Unity/Inova 300 MHz Spectrometer. Specific activities were per-
formed gravimetrically. Liquid scintillation counting was per-
3,3-Dichloro-1-(4-nitrophenyl)piperidin-2-[14C]one, 3
formed on either a PerkinElmer Tri-Carb Model 2900Tr or a
Wallac Model 1409 Liquid Scintillation Counter.
To a dry flask was charged 2 (192 mg, 0.872 mmol), chloroben-
zene (2.0 ml) and PCl5 (599 mg, 2.88 mmol). The reaction mixture
was heated to 531C for 1.5 h at which time all of the solid PCl5
had dissolved and the reaction was complete by HPLC analysis
using gradient 1. The retention time of the product was
10.25 min. The reaction mixture was cooled to room tempera-
ture and water (3.0 ml) was added slowly while cooling in an ice
water bath. The layers were separated and the aqueous layer
was extracted with CH2Cl2 (2 Â 10 ml). The CH2Cl2 extracts from
a second reaction on the same scale were combined with the
first and were washed with brine, dried over MgSO4 filtered and
solvent was removed under reduced pressure. Further drying on
a vacuum line produced 397 mg (79% yield) of a yellow solid, 3.
The product was 98.6% radiochemically pure by HPLC using
gradient 1. 1H-NMR (300 MHz, DMSO-D6) d ppm 8.37(d, J =
8.88 Hz, 2H), 7.69(d, J = 8.88 Hz, 2H,) 3.90(t, J = 5.95 Hz, 2H),
2.98–3.06(m, 2H) 2.21(quin, J = 5.76 Hz, 2H).
Materials and equipment used in microsomal and hepato-
cyte studies
Cryopreserved human hepatocytes (mixed gender, pooled from
nX3) and rats (Sprague-Dawley, mixed gender) hepatocytes
were obtained from Celsis In Vitro Technologies, Inc. (Baltimore,
MD). Liver microsomes from humans (mixed gender, pool of 50)
and rats (Sprague-Dawley; pooled from mixed gender) were
obtained from XenoTech, LLC (Kansas City, KS). Hepatocyte
incubation media (InVitroGRO HI media), obtained from Celsis
was used for hepatocyte incubations, and 0.1 M potassium
phosphate buffer, pH 7.4, was used for microsomal incubations.
7-ethoxycoumarin (7-EC) and 7-hydroxycoumarin (7-HC) were
obtained from Sigma-Aldrich (St. Louis, MO). Incubations were
analyzed by HPLC with a Waters 2695 Separations Module with
Waters 486 UV Detector or 996 PDA Detector with Millennium
v 4.0 Data System. Deepwell Luna PlateTM-96 was from
PerkinElmer Company. SpeedVac Concentrator AES2010 was
3-Chloro-1-(4-nitrophenyl)-5,6-dihydropyridin-2(1H)-[14C]one, 4
from Savant Instrument Inc. Fraction Collector Model ISCO Foxy To a dry flask was charged 3 (391 mg, 1.35 mmol), LiCl (29.4 mg,
200 was from ISCO, Inc. Packard Topcount NXTTM Microplate 0.692 mmol) and DMF (1.1 ml). The reaction was warmed to
Scintillation & Luminescence Counter was from the Packard 1051C for 20 min and Li2CO3 (54 mg, 0.73 mmol) was added in
Instrument Company. Radioactivity was measured with a b-RAM two equal portions over 40 min. The mixture was heated to
radioactivity detector, Model 3 from IN/US Systems Inc. and 1051C for 4 h at which time HPLC analysis using gradient 1
Liquid Scintillation Counters; Models LS 6000LL, LS 6000TA and showed the reaction to be 97% complete The retention time of
LS 6000IC from Beckman Instruments, Inc.
product was 9.55 min. As the reaction cooled, a precipitate
formed. At room temperature, water (1.5 ml) was added and
the large solid particles were broken up with a spatula. The
suspension was centrifuged at 3000 rpm for 10 min, the super-
natant was decanted and this was repeated with another 1.0 ml
of water, followed by washing with isopropyl alcohol (1.0 ml) to
give an off-white solid. The aqueous washings were extracted
with CH2Cl2 (2 Â 10 ml). The combined CH2Cl2 extracts were
washed with brine, dried over MgSO4 and solvent removed
under reduced pressure to give a solid that was combined with
the solid from the iPrOH washings after solvent removal to give
341 mg of an off white solid (100% crude yield). The product
was 96% radiochemically pure by HPLC using gradient 1. The
product was used in the next step of the synthesis without
additional purification.
1-(4-Nitrophenyl)piperidin-2-[14C]one, 2
To a small flask was charged K2CO3 (200mg, 1.45mmol) in 0.30ml
water. To this was added 4-nitroaniline (276.3mg, 2.00mmol),
THF (0.68 ml) and chlorobenzene (0.66 ml) resulting in a biphasic
solution. To this was added [1-14C]5-bromovaleryl chloride
(346.2mg, 1.72mmol, 100 mCi, specific activity= 58.0mCi/mmol,
source= GE Healthcare) dropwise and the reaction was stirred
at room temperature for 15min. To the heterogeneous solution
was added tetrabutylammonium bromide (TBAB, 12.8mg,
0.040 mmol) and KOH (132mg, 4.0 mmol, 85%) dissolved in
0.28ml H2O. Progress of the reaction was monitored by HPLC
using gradient 1. Product retention time= 8.25min. After 3 h, the
organic solvents were removed under reduced pressure leaving
the aqueous solution. The aqueous layer was extracted with
EtOAc (2 Â 15ml). The pooled EtOAc extracts were washed with
water, brine, dried over Na2SO4, filtered and solvent removed
Ethyl 1-(4-methoxyphenyl)-6-(4-nitrophenyl)-7-[14C]oxo-4,5,6,7-
tetrahydro-1H-pyrazolo[3,4-c]pyridine-3-carboxylate, 6
under reduced pressure to give 780 mg of crude product. The To a dry flask was added 4 (40.0mg, 0.158 mmol), ethyl-2-chloro-2-
reaction was repeated on the same scale and the combined crude (2-(4-methoxyphenyl)hydrazono)acetate 5 (42.3 mg, 0.165 mmol)
products were purified by flash chromatography on silica gel and toluene (0.346 ml). The suspension was heated to 961C.
eluting with 10% EtOAc : 90% hexane to 100% EtOAc. The To this was added Et3N (46 ml, 0.33 mmol) in two equal portions
product eluted at 60–80% EtOAc. The solvent from the pooled over 20 min. The reaction continued to be heated at 961C for 2 h
J. Label Compd. Radiopharm 2011, 54 418–425
Copyright r 2011 John Wiley & Sons, Ltd.
www.jlcr.org

B. D. Maxwell et al.
Ethyl 1-(4-methoxyphenyl)-7-[14C]oxo-6-(4-(2-oxopiperidin-
1-yl)phenyl)-4,5,6,7-tetrahydro-1H-pyrazolo[3,4-c]pyridine-
3-carboxylate, 9
and then cooled to room temperature. A second batch starting
with 300 mg of 4 was also completed. Both batches were
combined. Water (10 ml) was added and the aqueous layer was
extracted with CH2Cl2 (3 Â 15 ml). The CH2Cl2 layers were
combined, washed with brine, dried over MgSO4, filtered and
solvent removed under reduced pressure to give 433 mg of a light
brown solid (74% crude yield). HPLC analysis using gradient 1
showed the product was 90% pure with a retention time = 11.33 -
min. The product was used in the next step of the synthesis
without additional purification.
To a dry small flask with stirbar under nitrogen was charged 8
(197 mg, 0.372 mmol), NMP (2.0 ml), triethylorthoformate
(0.156 ml, 0.938 mmol) and TFA (5.6 ml, 0.075 mmol). The solution
was stirred at room temperature for 25 min and 21% NaOEt in
EtOH (0.350 ml, 0.938 mmol) was added slowly over 5 min. After
30 min, HPLC analysis using gradient 1 showed the reaction was
complete. Water (7.0 ml) was added along with 3.5 mg crude
product from a previous small scale reaction. The pH was
adjusted with saturated NH4Cl solution to pH 7. The mixture
was extracted with CH2Cl2 (3 Â 10 ml). The combined organic
extracts were washed with brine, dried over MgSO4, filtered and
the solvent removed under reduced pressure to produce a red
paste. The crude product was purified by silica gel flash column
chromatography using a gradient starting with 30:70 EtOAc:
hexane to 100% EtOAc to 5:95 MeOH:EtOAc. Pure fractions
were pooled and the solvent was removed under reduced
pressure. The sample was further dried on a vacuum line to
produce an oil. The oil was dissolved in ether (2.5 ml) and the
ether was removed under reduced pressure. The resulting solid
was dried on a vacuum line to produce 142 mg of a white solid
(78% yield). The product was 99.1% radiochemically pure by
HPLC using gradient 1. LC/MS showed ESI (1) m/z = 489(5%),
491(100%), 492(27.5%), 493(5%). 1H-NMR (300 MHz, DMSO-D6)
d ppm: 7.56 (d, J = 9.06 Hz, 2H), 7.39–7.45 (m, 2H), 7.32–7.38
(m, 2H), 7.07(d, J = 9.06 Hz, 2H), 4.41(q, J = 7.05 Hz, 2H), 4.15(t,
J = 6.52 Hz, 2H), 3.87(s, 3H), 3.66(t, J = 5.38 Hz, 2H), 3.27(t, J =
6.42 Hz, 2H), 2.45(t, J = 6.04 Hz, 2H), 1.87–2.01(m, 4H), 1.40(t,
J = 7.08 Hz, 3H).
Ethyl 6-(4-aminophenyl)-1-(4-methoxyphenyl)-7-[14C]oxo-4,5,6,7-
tetrahydro-1H-pyrazolo[3,4-c]pyridine-3-carboxylate, 7
To a 30 ml high pressure glass vessel with stirbar was charged 6
(433 mg, 0.992 mmol), NMP (2.25 ml) and 43 mg of 5% Pd on
Al2O3. The black mixture was purged with nitrogen gas for
15 min, then pressurized with hydrogen gas to 15 psi, pressure
released and repeated four additional times. Finally, it was
pressurized with hydrogen gas to 30 psi and then heated to
521C for 30 min. The pressure dropped to 20 psi and the reaction
was repressurized with hydrogen gas to 30 psi. After 1 h, the
reaction showed only 5% product by HPLC using gradient 1.
The retention time of 7 was 7.52 min. The reaction was stopped,
the Pd was filtered away and the reaction was restarted with
43 mg of fresh 5% Pd on Al2O3. The vessel was repressurized
with hydrogen gas to 30 psi using the same procedure as
described above and then heated to 521C. After 2 h, the reaction
mixture was 88% radiochemically pure product 7 by HPLC. The
reaction mixture was cooled to room temperature and filtered
to remove the Pd. The resulting NMP solution was dried over
MgSO4, filtered and the NMP removed under vacuum to give
400 mg (99% yield) of product that was used in the next step
without additional purification.
1-(4-Methoxyphenyl)-7-[14C]oxo-6-(4-(2-oxopiperidin-1-yl)-
phenyl)-4,5,6,7-tetrahydro-1H-pyrazolo[3,4-c]pyridine-3-
carboxamide, 10, [14C]apixaban
Ethyl 6-(4-(5-chloropentanamido)phenyl)-1-(4-methoxyphe-
nyl)-7-[14C]oxo-4,5,6,7-tetrahydro-1H-pyrazolo[3,4-c]pyri-
dine-3-carboxylate, 8
To a dry flask was charged 9 (20.4 mg, 0.042 mmol), DMF
(0.15 ml) and formamide (46 ml, 1.2 mmol). The solution was
To a dry flask with stirbar under nitrogen was charged 7 warmed to 501C and trimethylorthoformate (3.2 ml, 29 mmol)
(330 mg, 0.812 mmol) in NMP (4.0 ml). 5-Chlorovaleryl chloride and TFA in DMF (2.0 ml of solution containing 0.61 ml TFA,
was added slowly and stirred at room temperature for 15 min. 8.2 mmol) were added. After stirring for 15 min, NaOMe in
HPLC analysis showed the formation of 83% of the amide MeOH (25 wt%, 18 ml, 78 mmol) was added and the solution
product. The retention time of product 8 using HPLC gradient 1 stirred for 30 min. To the reaction at 501C was added water
was 10.98 min. The reaction was stirred for 15 additional minutes (0.5 ml). The reaction was cooled to room temperature
and water (8.0 ml) was added slowly to form a cloudy solution. overnight. The suspension was centrifuged at 3000 rpm for
The aqueous solution was extracted with CH2Cl2 (3 Â 10 ml). The 15 min and the supernatant removed. Water (0.5 ml) was
combined organic extracts were washed with brine, dried over added and centrifugation was repeated. The centrifugation
MgSO4, filtered and solvent removed under reduced pressure step was repeated twice more with isopropyl alcohol. The
resulting in a brown paste. The crude product was purified by solid product was dried under high vacuum to produce
flash column chromatography on silica gel using a gradient 10.1 mg of white solid, 10. The water and isopropyl alcohol
starting from 10:90 EtOAc:hexane to 100% EtOAc. Fractions washes were recovered and were purified by preparative HPLC
containing the desired product were pooled and the solvent using gradient 1 to produce an additional 5.8 mg of product
removed under reduced pressure to produce 200 mg of an off (83% yield, 14% overall yield). HPLC analysis using gradient 1
white solid (47% yield). HPLC analysis using gradient 1 showed showed the product to be 99.3% radiochemically pure and
the product to be 90% radiochemically pure. The product was 99.9% chemically pure at 254 nm. The specific activity was
used in the next step without additional purification. LC/MS measured gravimetrically at 56.4 mCi/mmol or 122 mCi/mg.
analysis showed ESI (1) m/z = 525(5%), 527(100%), 529(35%). LC/MS showed ESI (1) 460(5%), 462(100%), 463(28%) and
1H-NMR (300 MHz, DMSO-D6) d ppm: 10.03(s, 1H), 7.65(d, 464(5%). 1H-NMR (300 MHz, DMSO-D6) d ppm: 7.72(s, 1H),
J = 8.31 Hz, 2H), 7.54(d, J = 8.88 Hz, 2H), 7.33(d, J = 8.31 Hz, 2H), 7.50(d, J = 8.88 Hz, 2H), 7.44(s, 1H), 7.32–7.38(m, 2H), 7.25–7.32
7.06(d, J = 8.88 Hz, 2H), 4.40(q, J = 7.20 Hz, 2H), 4.10(t, J = 6.80 Hz, (m, 2H), 7.00(d, J = 8.88 Hz, 2H), 4.05(t, J = 6.42 Hz, 2H), 3.80(s, 3H),
2H) 3.87(s, 3H), 3.73(t, J = 6.04 Hz, 2H) 3.25(t, J = 6.80 Hz, 2H) 3.60(t, J = 5.57 Hz, 2H), 3.20(t, J = 6.61 Hz, 2H), 2.39(t, J = 6.33 Hz,
2.40(t, J = 6.89 Hz, 2H) 1.80(m, 4H), 1.39(t, J = 7.10 Hz, 3H).
2H), 1.80–1.91(m, 4H).
www.jlcr.org
Copyright r 2011 John Wiley & Sons, Ltd.
J. Label Compd. Radiopharm 2011, 54 418–425

B. D. Maxwell et al.
Ethyl 6-(4-(5-bromopentan-[14C]amido)phenyl)-1-(4-methoxy-
phenyl)-7-oxo-4,5,6,7-tetrahydro-1H-pyrazolo[3,4-c]pyridine-
3-carboxylate, 12
prepared by bubbling ammonia gas through 1,2-propanediol for
20 min. The reaction was heated to 1201C for 16 h, cooled to
room temperature and diluted with water (20 ml). The aqueous
solution was extracted with CH2Cl2 (3 Â 20 ml) and then with
EtOAc (3 Â 20 ml). The combined organic extracts were dried
over Na2SO4, filtered and the solvent removed under reduced
pressure to give a solid. To the solid was added acetonitrile
(5 ml) and the mixture warmed to 801C to produce a clear
solution. After cooling to room temperature over an hour, a
solid formed. The sample was filtered and dried to give 134 mg
of a white solid, (57% yield, 14% overall yield). HPLC analysis
using gradient 3 showed the sample was 100% radiochemically
pure. The specific activity of [14C]apixaban, 14, was measured
gravimetrically at 49.5 mCi/mmol or 108 mCi/mg. The specific
activity of the product was reduced to 2.49 mCi/mmol or 5.4 mCi/
mg with unlabeled apixaban for use in the human ADME clinical
study.3 1H-NMR (300 MHz, CDCl3) d ppm: 7.49(d, J = 8.80 Hz, 2H),
7.37(d, J = 9.10 Hz, 2H), 7.26(d, J = 8.80 Hz, 2H), 6.98(s, 1H), 6.95(d,
J = 9.20 Hz, 2H), 6.28(s, 1H), 4.14(t, J = 6.60 Hz, 2H), 3.81(s, 3H),
3.61(m, 2H), 3.39(t, J = 6.60 Hz, 2H), 2.63(t, J = 6.20 Hz, 2H),
1.96(m, 4H).
To a flask with stirbar was charged 11 (432.7 mg, 1.065 mmol) in
THF (7.0 ml) and 25% K2CO3 (aq) (0.90 ml). This material was
sonicated to completely dissolve all solids and form a pale
brown solution. To this was added [1-14C]5-bromovaleryl
chloride (50 mCi, 51 mCi/mmol, in 2 ml THF, 498% radio-
chemically pure, source = ViTrax) dropwise and then it was
stirred overnight at room temperature. The THF was removed
under reduced pressure. Sterile water (20 ml) and CH2Cl2 (20 ml)
were added. The two layers were separated. The water layer was
extracted with CH2Cl2 (20 ml). The CH2Cl2 extracts were dried
over Na2SO4, filtered and solvent removed under reduced
pressure. To the residue was added EtOAc (20 ml) to form a
solid. The solid was collected by filtration and dried under full
vacuum overnight to give 330 mg of product (54% yield).
HPLC analysis using gradient 2 showed the product to be 499%
radiochemically pure. The product was used in next step with-
out additional purification. A second reaction was performed on
the same scale to give an additional 510 mg of product (84%
yield) for an average yield of 69%.
Incubations
Ethyl 1-(4-methoxyphenyl)-7-oxo-6-(4-(2-[14C]oxopiperidin-
1-yl)phenyl)-4,5,6,7-tetrahydro-1H-pyrazolo[3,4-c]pyridine-
3-carboxylate, 13
As positive controls, 7-ethoxycoumarin (7-EC) and 7-hydroxy-
coumarin (7-HC) were dissolved in methanol. The solutions were
separately spiked to cell suspension and liver microsomes with
a final concentration of ca. 100 mM (containing 1% methanol). The
mixtures were incubated for 1 and 4 h in a manner similar to
apixaban incubations. At 0 and 4 hr, an aliquot of the incubation
mixtures was extracted with two volumes of ice-cold methanol,
followed by centrifugation. The resultant extracts were analyzed
by HPLC using gradient 3. At the completion of a 4-hour
incubation, 83 and 92% of cells remained viable for rat and
human hepatocytes, respectively. After 4-hour incubation, 44.1
and 51.0% of 7-EC and 42.1 and 97.8% of 7-HC were metabolized
by rat and human hepatocytes, respectively. In addition,
significant amounts of 7-HC sulfate and 7-HC glucuronide were
also observed. After 1-hour incubation, 71.2 and 90.2% of 7-EC
was metabolized by rat and human liver microsomes, respec-
tively. A significant amount of 7-HC, a metabolite of 7-EC, was
also observed in 1-hour incubation samples.
As negative controls, [14C]apixaban 10 at 122 mCi/mg and 14
at 108 mCi/mg, both 1 and 10 mM, were separately incubated
with the incubation media for 1 and 4 h without enzyme
systems. At the end of incubations, the incubations were
extracted with two volumes of ice-cold acetonitrile, followed by
centrifugation.
The cell viability was checked by the trypan blue exclusion
method. [14C]apixaban (both 10 and 14), at 1 and 10 mM in
incubation media, were separately incubated with cryopre-
served rat and human hepatocyte suspensions (ca. 1.0 Â 106
viable cells/ml) or 1 mg/ml liver microsomes in the presence of
5 mM MgCl2 and 1 mM NADPH in 0.1 M potassium phosphate
buffer (pH 7.4). Incubations were carried out in a 24-well cell
culture plates with each well containing 0.5 ml of the incubation
mixture. At 1, and 4 h, the entire incubation was quenched and
extracted with 2 volumes of ice-cold acetonitrile. The super-
natants from the incubation mixtures were concentrated using
an N-Evap to dryness, and reconstituted with a mixture of
To a dry small flask with stirbar was charged 12 (330 mg,
0.581 mmol) and THF (3.0 ml). To this slurry was added KOEt (24%
in EtOH, 0.30 ml) dropwise and then the reaction was stirred for 5 h.
The reaction was acidified with glacial acetic acid (0.10 ml). The
solvent was removed under reduced pressure and the residue was
partitioned between water (10 ml) and CH2Cl2 (10 ml). The aqueous
phase was extracted with CH2Cl2 (3 Â 10 ml). The combined CH2Cl2
extracts were dried over Na2SO4, filtered and the solvent removed
under reduced pressure to give an oil. The oil was taken up in EtOAc
(1 ml) and left overnight to induce crystallization. The solid was
collected by filtration to give 150 mg (53% crude yield) of solid
product 13. HPLC analysis using HPLC gradient 2 showed the
product to be 88% radiochemically pure and 90% pure by UV area
% at 254 nm. A second reaction starting with 510 mg of 12 was
completed. The crude product was purified by silica gel flash
chromatography eluting with 4% MeOH and 96% CH2Cl2 in place of
the crystallization with EtOAc to produce 180 mg (41% crude yield)
of product 13. HPLC analysis of the second batch using gradient 2
showed the product to be 94% radiochemically pure with a UV area
% purity of 91% at 254 nm. The crude products from both reactions
were combined and subjected to preparative HPLC using gradient 2.
Pure fractions were pooled and the acetonitrile was removed
under reduced pressure. The remaining aqueous layer was extracted
with CH2Cl2 (3 Â 100 ml) and then with EtOAc (2 Â 100 ml). The
combined organic extracts were dried over Na2SO4, filtered and
the solvent removed under reduced pressure to give 250 mg of a
white solid, (35% yield). The product was 499.9% radiochemically
pure and 499.9% pure by UV area % at 254 nm using gradient 2.
1-(4-Methoxyphenyl)-7-oxo-6-(4-(2-[14C]oxopiperidin-1-yl)-
phenyl)-4,5,6,7-tetrahydro-1H-pyrazolo[3,4-c]pyridine-3-
carboxamide, 14 or [14C]apixaban
To a pressure tube was added product 13 (250 mg, 0.512 mmol) acetonitrile/water (2:1) prior to analysis by HPLC. The 14C
and a saturated solution of ammonia in 1,2-propanediol (4.5 ml) profiles in microsomal and hepatocyte incubation samples were
J. Label Compd. Radiopharm 2011, 54 418–425
Copyright r 2011 John Wiley & Sons, Ltd.
www.jlcr.org

B. D. Maxwell et al.
determined by HPLC radiochromatography using HPLC Gradient 4. in quantitative yield. Product 2 was reacted with PCl5 to produce
Fractions of chromatography effluents were collected by a,a- dichlorinated lactam 3 in 79% yield. Product 3 was reacted
time (15 s/fraction) to Deepwell LumaPlateTM-96 plates. The with Li2CO3 at 1051C to produce a,b-unsaturated lactam 4 in
plates were subsequently dried by a SpeedVac concentrator for 96% yield. Product 4 was reacted with ethyl 2-chloro-2-(2-(4-
up to 8 h. The radioactivity in each fraction was determined methoxyphenyl)hydrazono)acetate 5 synthesized by the Bristol-
by Packard TopCount NXTTM Microplate Scintillation and Myers Squibb Process Chemistry Group to generate the pyrazole
Luminescence Counter technology. HPLC radiochromatograms ring system product 6, in 67% yield. The aromatic nitro group
were reconstructed using ARC Convert and Evaluation software. was reduced to aniline 7 with hydrogen gas and 5% Pd on Al2O3
Radioactivity peaks were integrated to determine the percent in 99% yield. The second lactam ring was prepared in four steps
distribution of individual radioactivity peaks or regions in each by reacting aniline product 7 with chlorovaleryl chloride to
sample. Selected incubation sample extracts from microsomal generate amide 8 in 42% yield followed by base promoted
and hepatocyte incubations were analyzed using HPLC gradient lactam ring formation in 78% yield. The final step involved
4. The eluate was collected and the total volume was measured. the conversion of ethyl ester 9 to the primary amide with
Duplicate 2.0 ml aliquots of each HPLC eluate collection were formamide to generate [14C]apixaban, 10, in 83% yield with the
taken for LSC. The data obtained with or without the use of an 14C label located in the central lactam ring system. The overall
HPLC column were compared.
yield for the nine-step synthesis was 14%. The product was
99.3% radiochemically pure and 99.9% chemically pure by UV
at 254 nm. MS and 1H-NMR matched those expected for
[14C]apixaban. The specific activity was measured gravimetrically
at 56.4 mCi/mmol or 122 mCi/mg.
Results and discussion
[14C]Apixaban was synthesized with the 14C label located in two
different positions in the molecule. In the first synthesis shown
in Scheme 1, 4-nitroaniline was reacted with [1-14C]5-bromova-
leryl chloride under basic conditions to generate nitro lactam 2
A separate synthesis of [14C]apixaban with the 14C label in
the outer lactam ring system was completed and is shown in
Scheme 2. In this synthesis, compound 11 synthesized by the
O
Cl
O
O
Br
PCl5
PhCl
NO2
Cl
*
Cl
*
*
N
NO2
N
NO2
K2CO3, H2O
THF, PhCl
TBAB, KOH
* = 14C
53˚C
79%
H2N
1
2
3
100%
Cl
EtO2C
N
O
N
Li2CO3
LiCl
H
Et3N
Toluene
N
*
N
Cl
+
NO2
DMF
105˚C
96%
N
N
67%
CH3O
CO2Et
4
5
O
NO2
6
CH3O
EtO2C
EtO2C
N
O
H2
N
Cl
N
8
N
7
5% Pd/Al 2O3
Cl
N
N
O
NMP, 52˚C
99%
O
O
NMP
42%
NH
NH2
Cl
CH3O
CH3O
H2NOC
N
EtO2C
N
HCONH2
DMF
1. CH(OEt)3
TFA, NMP
N
N
9
N
N
CH(OMe)3
TFA
NaOMe in MeOH
83%
O
O
2. 21% NaOEt
N
N
in EtOH
78%
O
O
10
14C]apixaban
CH3O
CH3O
[
Scheme 1. Synthesis of 10, [14C]apixaban.
www.jlcr.org
Copyright r 2011 John Wiley & Sons, Ltd.
J. Label Compd. Radiopharm 2011, 54 418–425

B. D. Maxwell et al.
EtO2C
N
EtO2C
N
O
Br
N
N
Cl
N
N
O
K2CO3,
THF
69%
O
O
NH
NH2
12
11
Br
* = 14
C
CH3O
CH3O
EtO2C
N
H2NOC
N
KOEt
THF
NH3
1,2 Propandiol
N
N
N
N
35%
57%
O
O
N
N
13
O
O
14
CH3O
CH3O
[
14C]apixaban
Scheme 2. Synthesis of 14, [14C]apixaban.
apixaban
ARC(CPM)
Compound 14
180
160
140
120
100
80
M7
M3
M6
60
M2
M4
40
M1
20
0
0.00
10.00
20.00
30.00
40.00
50.00
60.00
70.00 Min
apixaban
ARC(CPM)
Compound 10
180
160
140
120
100
80
M7
M3
M6
60
M2
M4
40
M1
20
0
0.00
10.00
20.00
30.00
40.00
50.00
60.00
70.00 Min
Figure 1. HPLC radiochromatograms of 4-hour human hepatocyte incubations with 10 mM [14C]apixaban, compounds 10 and 14.
J. Label Compd. Radiopharm 2011, 54 418–425
Copyright r 2011 John Wiley & Sons, Ltd.
www.jlcr.org

B. D. Maxwell et al.
apixaban
ARC(CPM)
240
Compound 14
220
200
180
160
140
120
100
80
M7
M2
M6
M3
60
M4
40
20
0
0.00
10.00
20.00
30.00
40.00
50.00
60.00
70.00 Min
apixaban
ARC(CPM)
240
Compound 10
220
200
180
160
140
120
100
80
M7
M2
M6
M3
60
40
20
0
0.00
10.00
20.00
30.00
40.00
50.00
60.00
70.00 Min
Figure 2. HPLC radiochromatograms of 1-hour human liver microsomal incubations with 10 mM [14C]apixaban, compounds 10 and 14.
Bristol-Myers Squibb Process Chemistry Group was reacted with
The 1- or 4-hour microsomal or hepatocyte incubation
[1-14C]bromovaleryl chloride to generate amide 12 in 69% yield. mixtures of [14C]apixaban (both 10 and 14, at 1 and 10 mM),
Product 12 was converted with potassium ethoxide to the were extracted with ice-cold acetonitrile. The extraction
penultimate lactam 13 in 35% yield after purification by recoveries of the total radioactivity from the incubation mixtures
preparative HPLC. The penultimate ester was converted to were quantitative for rat and human hepatocyte and micro-
[14C]apixaban, 14, with ammonia in 57% yield. The three-step somes for 14 and 10. [14C]apixaban (both 10 and 14) appeared
sequence was completed in 14% overall yield. The final product to be relatively stable in hepatocyte incubation media and
was 100% radiochemically pure. MS and 1H-NMR matched phosphate buffer at 10 mM. Greater than 97% of [14C]apixaban
those expected for the labeled product. The specific activity remained unchanged after a 1- or 4-hour incubation control. No
of [14C]apixaban, 14, was measured at 49.5 mCi/mmol or detectable degradation products were observed in the 1 mM
108 mCi/mg. The specific activity of a portion of the product negative control incubations of both 14 and 10.
was reduced to 2.49 mCi/mmol or 5.4 mCi/mg with unlabeled
The metabolism of apixaban in rat and human hepatocyte or
microsomal incubations was very limited, with 492 and 95% of
apixaban for use in the human ADME clinical study.3
The in vitro biotransformation profiles of both labels, 10 and the total radioactivity being attributed to apixaban after a 1- or
14, in cryopreserved hepatocytes and liver microsomes from rats 4-hour incubation. Six metabolites, M1–M4, M6 and M7, were
and humans were determined. Both hepatocyte preparations tentatively identified in hepatocytes and five metabolites,
used in this study were enzymatically active for the 4-hour M2–M4, M6 and M7 were tentatively identified in microsomal
incubation. Also both microsomal preparations used in this incubations. These in vitro metabolites were similar to the in vivo
study were enzymatically active for the 1-hour incubation.
metabolites.3–5 Structures for all of the metabolites have been
www.jlcr.org
Copyright r 2011 John Wiley & Sons, Ltd.
J. Label Compd. Radiopharm 2011, 54 418–425

B. D. Maxwell et al.
reported previously in references 4 and 5. M2 was identified as the Process Chemistry Group at Bristol-Myers Squibb for
the O-desmethyl metabolite.4,5 M1 was a sulfate conjugate of providing valuable intermediates, 5 and 11, and unlabeled
M2.4,5 M4 and M7 were hydroxylated metabolites.4,5 M3 was apixaban.
the ring-opened metabolite4 and M6 was the amide hydrolysis
metabolite.4 There were no significant differences observed
between the metabolite profiles of both 10 and 14, [14C]apix-
aban, from both rat or human hepatocyte or microsomal
incubations, see Figures 1 and 2. Metabolite profiles were
similar for 1 and 10 mM incubations. Similar metabolic profiles
of 10 and 14 supported the use of 14 for ADME studies and
References
[1] D. J. P. Pinto, M. J. Orwat, S. Koch, K. A. Rossi, R. S. Alexander,
A. Smallwood, P. C. Wong, A. R. Rendina, J. M. Luettgen, R. M. Knabb,
K. He, B. Xin, R. R. Wexler, P. Y. S. Lam, J. Med. Chem. 2007, 50,
that experiments with compound 14 adequately assessed the
metabolic profile of apixaban.
5339–5356. DOI: 10.1021/jm070245n and references therein.
[2] This research was partially presented as an oral paper at the 10th
International Symposium on the Synthesis and Applications of
Isotopes and Isotopically Labelled Compounds held June 14-18,
2009 in Chicago, Illinois, USA and published as an abstract in the
Proceedings of the Symposium. See B. D. Maxwell, S. B. Tran,
S. Y. Chen, D. Zhang, B. C. Chen, H. Zhang, S. J. Bonacorsi Jr.
J. Label Comp. Radiopharm 2010, 53, 363–367.
[3] N. Raghavan, C. E. Frost, Z. Yu, K. He, H. Zhang, W. G. Humphreys,
D. Pinto, S. Chen, S. Bonacorsi, P. C. Wong, D. Zhang, Drug Metab.
Dispos. 2009, 37, 74–81. DOI: 10.1124/dmd.108.023143.
[4] D. Zhang, K. He, N. Raghavan, L. Wang, J. Mitroka, B. D. Maxwell,
R. M. Knabb, C. Frost, A. Schuster, F. Hao, Z. Gu, W. G. Humphreys,
S. J. Grossman, Drug Metab. Dispos. 2009, 37, 1738–1748. DOI:
10.1124/dmd. 108.025981.
Conclusion
Two separate syntheses of [14C]apixaban, 10 and 14, labeled in
two different lactam ring systems within the molecule were
completed for various in vitro studies and in vivo studies. There
were no significant differences observed between the meta-
bolite profiles of 10 and 14 [14C]apixaban from both rat or
human hepatocyte or microsomal incubations.
Acknowledgements
[5] L. Wang, D. Zhang, N. Raghavan, M. Yao, M. Li, C. A. Frost,
B. D. Maxwell, S. Y. Chen, K. He, T. C. Goosen, W. G. Humphreys,
S. J. Grossman, Drug Metab. Dispos. 2010, 38, 448–458. DOI:
10.1124/dmd. 109.029694.
The authors thank members of the Bristol-Myers Squibb
Radiochemistry Group for helpful suggestions and members of
J. Label Compd. Radiopharm 2011, 54 418–425
Copyright r 2011 John Wiley & Sons, Ltd.
www.jlcr.org