Discovery of a factor Xa inhibitor (3R,4R)-1-(2,2-difluoro-ethyl)- pyrrolidine-3,4-dicarboxylic acid 3-[(5-chloro-pyridin-2-yl)-amide] 4-{[2-fluoro-4-(2-oxo-2H-pyridin-1-yl)-phenyl]-amide} as a clinical candidate

Article abstract of DOI:10.1016/j.bmcl.2010.06.126

A series of (3R,4R)-pyrrolidine-3,4-dicarboxylic acid amides was investigated with respect to their factor Xa inhibitory activity, selectivity, pharmacokinetic properties, and ex vivo antithrombotic activity. The clinical candidate from this series, R1663

Full text of DOI:10.1016/j.bmcl.2010.06.126

Bioorganic & Medicinal Chemistry Letters 20 (2010) 5313–5319
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Discovery of a factor Xa inhibitor (3R,4R)-1-(2,2-difluoro-ethyl)-pyrrolidine-3,4-
dicarboxylic acid 3-[(5-chloro-pyridin-2-yl)-amide] 4-{[2-fluoro-4-(2-oxo-2H-
pyridin-1-yl)-phenyl]-amide} as a clinical candidate
Lilli Anselm, David W. Banner, Jörg Benz, Katrin Groebke Zbinden, Jacques Himber, Hans Hilpert,
Walter Huber, Bernd Kuhn, Jean-Luc Mary, Michael B. Otteneder, Narendra Panday , Fabienne Ricklin,
*
Martin Stahl, Stefan Thomi, Wolfgang Haap
F. Hoffmann-La Roche Ltd, Pharma Research, Grenzacherstr. 124, CH-4070 Basel, Switzerland
a r t i c l e i n f o
a b s t r a c t
Article history:
A series of (3R,4R)-pyrrolidine-3,4-dicarboxylic acid amides was investigated with respect to their factor
Xa inhibitory activity, selectivity, pharmacokinetic properties, and ex vivo antithrombotic activity. The
clinical candidate from this series, R1663, exhibits excellent selectivity against a panel of serine proteases
and good pharmacokinetic properties in rats and monkeys. A Phase I clinical study with R1663 has been
finalized.
Received 27 April 2010
Revised 23 June 2010
Accepted 24 June 2010
Available online 1 July 2010
Ó 2010 Elsevier Ltd. All rights reserved.
Keywords:
Factor Xa inhibitors
Anticoagulation
Structure based design
X-ray analysis
Cardiovascular events caused by thrombosis are still the leading
cause of death in developed countries.1 Anticoagulant therapy has
been dominated for decades by parenteral drugs (e.g., low molec-
ular weight heparin) or by vitamin K antagonists such as warfarin.
Both therapies have well documented limitations.2 Extensive ef-
forts have therefore been undertaken in the pharmaceutical indus-
try to discover and develop new antithrombotic agents with less
limitations resulting in a better patient compliance.3
from modest pharmacokinetic properties (e.g., low bioavailability).
In search of alternative scaffolds our attention was attracted to the
pyrrolidine scaffold 1 which provides a very good vector into the
S1b-pocket (Fig. 1). This pocket is formed by Arg-143, Gln-192
and the disulfide bridge between Cys-191 and Cys-220 of fXa, and
has been shown to provide additional binding affinity if occupied
with favorably interacting ligand atoms.11 Molecular modeling into
the binding site of fXa (Fig. 1) of a suitably decorated scaffold 1 re-
vealed favorable interactions of the potential inhibitors with the
binding site of fXa and in particular with the S1b-pocket. Another
advantage of this scaffold is the flexibility of the five-membered
ring system which allows a conformational accommodation of the
inhibitor to the active site of fXa. A preference for the chlorophenyl
moiety instead of the chlorothiophenyl as S1-pocket motif was sug-
gested. In addition, the chiral synthesis of the orthogonally pro-
tected scaffold 1 has been developed at Roche12 and rapid access
to larger quantities of 1 was available. Later, similar series have
been described by Qiao et al. (BMS)13 and Kohrt et al. (Pfizer) result-
ing in the identification of PD0348292 (Eribaxaban).14
Over the past decade, factor Xa (fXa) has been the focus of in-
tense research activity.4 At the convergence of the intrinsic and
extrinsic coagulation pathways it plays a pivotal role in the coagu-
lation cascade.5 The research efforts to identify fXa inhibitors have
culminated in the discovery of rivaroxaban (Bayer),6 which has
been approved in the EU for the prevention of venous thromboem-
bolism. In addition, YM150 (Astellas),7 edoxaban tosilate (Daiichi
Sankyo)8 and apixaban (BMS/Pfizer)9 are currently undergoing
Phase III clinical trials.
In our previous work,10 we described the identification of 3-
aminopyrrolidines as potent and selective factor Xa inhibitors
(Ki = 3 nM; 2xPT = 1.5
l
M, Fig. 1). However, this series suffered
Here we describe the synthesis, SAR, pharmacokinetic proper-
ties and X-ray structure of potent and selective fXa inhibitors
based on the pyrrolidine scaffold 1.
* Corresponding author. Tel.: +41 61 6888693; fax: +41 61 6885964.
E-mail address: wolfgang.haap@roche.com (W. Haap).
Present address: European Patent Office, Bayerstrasse 34, D-80298 Munich,
Germany.
The first residue which we intended to optimize was the P1b-
substituent. The synthesis of the compounds described in Table 1
is depicted in Schemes 1 and 2. To synthesize compounds 2–23,
0960-894X/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2010.06.126

5314
L. Anselm et al. / Bioorg. Med. Chem. Lett. 20 (2010) 5313–5319
Boc
N
Boc
N
a
b, c
H
N
COOEt
HOOC
COOEt
O
1
R2
N
Cl
O
H
N
d-f
H
H
N
H
N
H
N
F
F
N
O
O
O
O
N
N
Cl
Cl
O
Boc
N
2-23
S1β-pocket
S1β-pocket
Scheme 1. Reagents and conditions: (a) MeCN, DIEA, BOP-Cl, 4-chloroaniline, 0 °C,
2 h, 69%; (b) THF/water, LiOH, 25 °C, 18 h, 87%; (c) SOCl2, 1-(4-amino-3-fluoro-
phenyl)-1H-pyridin-2-one, 25 °C, 18 h, 99%; isolation of free base by extraction
(compounds 3). Modification on the pyrrolidine nitrogen with one of the methods
d–g: (d) MeCN, sulfonylchlorides or acidchlorides, DIEA, 25 °C, 18 h, 15–45%
(compounds 2, 4, 5, 7, 10, and 12–18); (e) MeCN, alkylbromides, K2CO3, Ag2O, 80 °C,
18 h, 19–24% (compounds 6, 8, 19, 20, 22, and 24); (f) CH2Cl2, trifluoro-methane-
sulfonic acid 2,2,2-trifluoro-ethyl ester, DIEA, 25 °C, 72 h, 11% (compound 11); (g)
MeOH, NaCNBH3, acetone, acetic acid, 80 °C, 18 h, 5–23% (compounds 9, 21, and
23).
Me
O
HOOC
COOEt
R2
N
1
N
O
HN
O
HN
H
N
H
N
F
R4
R1
S
O
O
N
Cl
S1-Pocket
Ki (fXa) = 3 nM
O
S4-Pocket
S1-Pocket
S4-Pocket
Boc
N
Boc
N
2xPT = 1.7 μM
a
b
F
Figure 1. Illustration of design principle of fXa inhibitors using 2vvc.pdb10 as
H
N
example.
OH
EtOOC
EtOOC
O
O
compound 1 was reacted with 4-chloro aniline to afford the corre-
sponding amide (Scheme 1).
N
O
Saponification of the ester and treatment of the obtained acid
with thionylchloride in presence of the P4-substituent yielded the
Boc-deprotected pyrrolidine bis-amide. After isolation of the free
base by extraction, variations on the pyrrolidine nitrogen substitu-
ent were achieved by reactions with sulfonyl chlorides, acid chlo-
rides, alkyl bromides, alkyl triflates, and by reductive amination.
To introduce the corresponding 2-amino-5-chloro pyridine
moiety as replacement for the 4-chloro aniline, the reaction se-
quence started first by amide coupling of the P4-substituent using
BOP-Cl as coupling reagent, followed by Weinreb amide coupling.
Boc-cleavage using HCl in dioxane and subsequent variations on
the pyrrolidine nitrogen as described in Scheme 1 afforded com-
pounds 25–31.
Boc
N
H
N
c
H
N
H
N
F
H
H
F
N
N
N
O
O
N
O
O
N
Cl
N
O
Cl
O
R2
N
d-f
H
H
F
N
N
As listed in Table 1, a large variety of S1b-pocket residues are
tolerated, leading to low nanomolar binding affinities against fXa.
For example, the sulfonamide 2 showed a Ki of 15 nM and a favor-
N
O
O
25-31
N
Cl
able in vitro anticoagulant activity (2xPT = 2.2
lM). In contrast, the
O
unsubstituted pyrrolidine showed only moderate binding
3
(Ki = 120 nM), underlining the importance of an appropriate P1b-
substituent for good activity. The strongest binding affinity and
best anticoagulant activity was demonstrated by the dimethyl sulf-
amide 13 with Ki = 8 nM and 2xPT = 1.7 lM. In general, compounds
with comparable binding affinity showed better anticoagulant
Scheme 2. Reagents and conditions: (a) MeCN, DIEA, BOP-Cl, 1-(4-amino-3-fluoro-
phenyl)-1H-pyridin-2-one, 25 °C, 4 d, 58%; (b) 5-chloro-pyridin-2-ylamine, toluene,
AlMe3, 1 h at 25 °C, then reflux for 2 h, 72%; (c) HCl, 1,4-dioxane, 25 °C, 50 min, 60%;
isolation of free base by extraction (compound 28). Modification on the pyrrolidine
nitrogen with one of the methods d–f: (d) MeCN, sulfonylchlorides or methylchlo-
roformate DIEA, 25 °C, 18 h, 25–90% (compounds 25, 26, and 30); (e) CH2Cl2,
trifluoro-methanesulfonic acid 2,2-difluoro ethyl ester or trifluoro-methanesulfonic
acid 2,2,2-trifluoro-ethyl ester, DIEA, 25 °C, 72 h, 46–64% (compounds 27 and 31);
(f) MeOH, NaCNBH3, acetone, acetic acid, 80 °C, 18 h, 59% (compound 29).
activity if their log D was lower.
We had concerns about using the 4-chloro aniline moiety as the
final P1-substituent, since severe toxicity is associated with this
motif should the amide bond be cleaved in vivo. For this reason
we switched to the 2-amino-5-chloro pyridine moiety. Combina-
tions with the most interesting S1b-pocket residues (compounds
25–31) revealed that the binding affinity was in general main-
tained. In addition, due to the increased polarity, the anticoagulant
activity was improved (e.g., compound 27 vs 8).

L. Anselm et al. / Bioorg. Med. Chem. Lett. 20 (2010) 5313–5319
5315
Table 1
P1b-substituent SAR
R2
N
H
N
H
N
F
X
O
O
2-31
N
Cl
O
R2
X
fXa Ki
(
lM)
2xPT (
l
M)
log D (pH 7.4)
a,b
Compound
2
3
4
5
6
7
8
9
SO2Me
H
Ac
SO2Et
CH2–CH2–OH
C(O)OMe
CH2–CHF2
iPropyl
SO2NH2
CH2–CF3
SO2CH2CF3
SO2NMe2
SO2iPropyl
C(O)OEt
C(O)Pyrrolidine
C(O)O(n)-Propyl
SO2n-Propyl
CH2-Cyclopropyl
CH2–CH2F
CH2-4-Pyridyl
CH2–CH2–OMe
CH2-2-Thiophenyl
CH2–CN
CH
CH
CH
CH
CH
CH
CH
CH
CH
CH
CH
CH
CH
CH
CH
CH
CH
CH
CH
CH
CH
CH
CH
N
0.015
0.120
0.028
0.011
0.024
0.022
0.010
0.017
0.010
0.009
0.017
0.008
0.010
0.054
0.071
0.047
0.018
0.046
0.021
0.034
0.044
0.011
0.042
0.020
0.024
0.007
0.055
0.015
0.027
0.008
2.2
5.3
1.8
2.0
2.7
2.6
2.9
2.0
1.8
2.9
3.2
1.7
2.0
5.2
5.2
6.3
2.2
8.1
5.4
5.9
4.7
2.1
7.6
1.7
3.8
1.5
6.9
1.7
1.6
1.4
2.17
n.d.c
1.80
2.44
1.80
2.43
2.69
n.d.c
1.83
3.20
2.68
2.62
2.81
2.57
2.78
3.06
2.94
n.d.c
n.d.c
n.d.c
n.d.c
n.d.c
n.d.c
1.46
1.77
2.20
0.63
1.86
2.02
2.62
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
SO2Me
SO2Et
CH2–CHF2
H
iPropyl
N
N
N
N
N
N
C(O)OMe
CH2–CF3
a
Values are means of three experiments.
All compounds exhibit high selectivity versus thrombin (Ki > 25 lM).
n.d., not determined.
b
c
We note that all compounds listed in Table 1 demonstrated
excellent selectivity versus thrombin, with Ki values greater than
25 M.
In search of alternative P1-motifs, we elaborated a further alter-
native synthetic route (Scheme 3). In analogy to Scheme 2, the
reaction sequence started with the introduction of the P4-substitu-
ent, followed by Boc-cleavage, reaction with methanesulfonyl
chloride or alternatively with 2,2-difluoroethyl triflate, saponifica-
tion of the ester to the corresponding acid, and finally coupling of
various aniline derivatives using BOP-Cl as coupling reagent. The
aniline derivatives which were introduced by this method are
listed in Table 2 in comparison to the 4-chlorophenyl (compound
2) and 5-chloropyridin-2-yl (compound 25) analogs.
With the exception of the 4-chloro-3-fluorophenyl (compound
39; Ki = 62 nM) and the 5-chlorothiophenyl moieties (compound
42; Ki = 36 nM), significantly weaker binding affinity towards fXa
l
Boc
N
R2
N
R2
N
c,d
a, b
H
N
F
H
N
H
N
H
EtOOC
F
F
N
EtOOC
R1
O
O
O
O
N
N
N
32-42
O
O
O
Scheme 3. Reagents and conditions: (a) 6 N HCl in isopropanol, 25 °C, 2 h, 100%; (b) MeCN, methanesulfonyl chloride, DIEA, 25 °C, 18 h, 99%; or CH2Cl2, trifluoro-
methanesulfonic acid 2,2-difluoro ethyl ester, DIEA, 25 °C, 72 h, 46%; (c) 1,4-dioxane/water, LiOH, 25 °C, 24 h, 45%; (d) MeCN, BOP-Cl, DIEA, anilines, 25 °C, 24 h, 3–78%.

5316
L. Anselm et al. / Bioorg. Med. Chem. Lett. 20 (2010) 5313–5319
Table 2
P1-substituent SAR
F
O2S
N
F
N
H
N
H
N
H
N
H
F
F
N
R1
O
O
O
O
S
N
N
Cl
32-41
42
O
O
Compound
R1
4-Chlorophenyl
5-Chloropyridin-2-yl
5-Indolyl
fXa Kia
(lM)
2xPT (
l
M)
log D (pH 7.4)
2
25
32
33
34
35
36
37
38
39
40
41
42
0.015
0.020
0.345
0.463
0.810
0.192
17.30
0.130
16.59
0.062
1.247
0.421
0.036
2.2
1.7
2.17
1.46
1.07
1.19
1.90
0.56
2.36
n.d.b
n.d.b
2.37
0.50
n.d.b
n.d.b
22.4
12.5
35.7
5.9
4-Methoxyphenyl
3-Chloro-4-methoxypheny
3-Fluoro-4-methoxyphenyl
4-Trifluormethyloxyphenyl
4-Chloro-2-fluorophenyl
4-Chlorophenylmethyl
4-Chloro-3-fluorophenyl
5-Indazolyl
n.d.b
4.7
n.d.b
7.9
n.d.b
12.3
3.6
2-Amino-4-chlorophenyl
5-Chlorothiophenyl
a
Values are means of three experiments.
n.d., not determined.
b
was observed. This is also reflected by a reduction of the anticoag-
ulant activity.
ficient of Peff = 0.46 ꢀ 10ꢁ6 cm sꢁ1, which was threefold lower than
for compound 2 with a PSA of 96 Å2. Therefore, oral bioavailability
of compound 55 was expected to be lower than for compound 26,
which is analogous to compound 2 (see Table 4).
A direct comparison of the 2,2-difluoroethyl derivatives 8, 27,
and 42 demonstrated that the binding affinities and anticoagulant
activities are clearly in favor of the six-membered (hetero)aromatic
moieties of compounds 8 and in particular 27 as suggested by our
initial modeling studies. Furthermore, the amide coupling of the 2-
amino-5-chlorothiophene residue turned out to be problematic, as
a rapid decomposition of the thiophene derivative was observed.
Finally, optimization of the P4-substituent was investigated. For
this purpose the sequence outlined in Scheme 1 was adapted
slightly: the P1b-motif was introduced first and the P4-substituent
was attached by BOP-Cl mediated amide coupling at the last reac-
tion step (Scheme 4).
In support of the modeling and chemistry efforts, X-ray co-crys-
tal structures of fXa with compounds 2, 11, 12, and 27 were deter-
mined. The novel pyrrolidine scaffold proved to be an excellent
framework to direct substituents into the S1, S1b, and S4 pockets
of factor Xa. While the S1 pocket is deeply buried, the S4 and S1b
sites are more solvent-exposed, explaining the much steeper SAR
for P1-motifs. The chlorine atom of the P1 pyridine substituent sits
on top of the aromatic ring of Tyr-228 and the P4 phenylpyridone
group is favorably aligned with the aromatic residues in the S4
pocket as observed by others (e.g., PDB code 2p93).13 The P1b-sub-
stituent clearly offered better opportunities to optimize molecular
properties without sacrificing binding affinity. The crystal struc-
ture (Fig. 2) shows that the polar atoms of 27 interact through sev-
eral direct or water-mediated hydrogen bonds with fXa binding
site residues. The difluoroethyl moiety attached to the pyrrolidine
nitrogen nicely fits into the S1b-pocket and the terminal fluorine
atoms are engaged in both hydrophobic (sulfur of Cys-191-220
bridge, Cb of Gln-192) and cation-dipole (guanidine of Arg-143)
interactions. The S1b-pocket in factor Xa with its electropositive
character appears to be a fluorophilic binding region as also seen,
for example, in neprilysin.15
Table 3 summarizes all evaluated S4-pocket residues with their
corresponding activities. When compared to compound 2 no sig-
nificant improvement in terms of binding affinity was achieved.
Compounds 47 and 55 showed a better anticoagulant activity than
compound 2 with 2xPT = 1.8 lM and 2xPT = 1.4 lM, respectively.
Both compounds, however, where not profiled further. Compound
47 showed a twofold higher clearance in rat microsomes (data not
shown) than compound 2. Compound 55 was characterized by a
high polar surface area of 106 Å2 and an associated low passive
permeation through artificial membranes with a permeation coef-
Finally, the 2-amino-5-chloro pyridine derivatives 26, 27, 29,
and 30 were chosen for further in-depth profiling (Table 4). All
compounds showed good binding affinities and anticoagulant
activities as well as excellent selectivity versus a panel of serine
proteases. The clearance after incubation in human microsomes
or human hepatocytes was low for all compounds and no issues
were detected in the Ames, MNT and phototoxicity assays. Large
differences were observed for the protein binding determined in
human plasma. The basic compound 29 with the isopropyl P1b-
motif showed a high protein binding with less than 1% free fraction
Boc
N
O2S
N
O2S
N
c,d
a, b
H
H
H
H
N
N
N
N
COOEt
COOEt
R4
O
O
O
O
43-55
Cl
Cl
Cl
Scheme 4. Reagents and conditions: (a) 6 N HCl in isopropanol, 25 °C, 2 h, 100%; (b)
MeCN, methanesulfonyl chloride, DIEA, 25 °C, 18 h, 55%; (c) 1,4-dioxane/water,
LiOH, 25 °C, 24 h, 84%; (d) MeCN, BOP-Cl, DIEA, anilines, 25 °C, 24 h, 14–61%.
and also exhibited hERG inhibition with IC50 of 12
lM and further
caused marginal phospholipidosis in the in vitro assay. In contrast,

L. Anselm et al. / Bioorg. Med. Chem. Lett. 20 (2010) 5313–5319
5317
M)
Table 3
P4-substituent SAR
Table 3 (continued)
Compound
R4
fXa Ki (
l
M)
2xPT (
l
O2S
N
N
S
O
51
0.201
12.0
H
N
H
N
O
O
R4
O
O
43-55
Cl
F
52
53
0.022
0.046
2.1
3.9
N
Compound
R4
fXa Ki (
lM)
2xPT (
2.2
lM)
F
2
0.015
N
N
N
N
O
O
O
O
O
N
O
43
44
0.092
0.031
8.5
2.2
54
55
0.040
0.024
4.5
1.4
N
F
F
F
N
O
O
the moderately basic difluoroethyl derivative 27 showed only
modest protein binding with a free fraction of 23%, no significant
hERG inhibition and no phospholipidosis in vitro. When tested in
rat PK studies, compound 26 showed only low oral bioavailability
and a relatively short half-life. This trend was even more pro-
nounced when tested in a cynomolgus monkey PK study. Bioavail-
ability was decreased down to 9%, disqualifying the compound
from further development. In contrast to compound 26, all other
derivatives showed good oral bioavailability in cynomolgus mon-
keys and half-lives in a favorable range. In addition, all compounds
exhibited a relatively low volume of distribution.
Compounds 27 and 30 were profiled in more detail. fXa and fVa
form on charged phospholipid surfaces the prothrombinase com-
plex, which is responsible for the rapid thrombin generation neces-
sary for clot formation. Compound 27 inhibited peptidase activity
of the prothrombinase complex16 with Ki = 1.3 nM, and compound
30 with Ki = 6.6 nM. Interestingly, compound 27 showed one order
of magnitude faster on- and off-rates when compared to com-
pound 30, resulting in similar KD values of 3.7 or 4.3 nM, respec-
tively. The faster binding kinetics of 27 may be associated with
the more flexible difluoroethyl P1b-motif when compared with
the more rigid methylcarbamate P1b-motif of 30. Both molecules
showed full reversibility of binding. When tested in a human
whole blood flow system for ex vivo anticoagulant activity17 com-
pound 27 demonstrated inhibition of fibrin deposition in a dose-
dependent manner with IC50 values in the range of 310 nM when
coagulation was triggered by collagen or 840 nM when coagulation
was triggered by J82 cells expressing tissue factor. Compound 30
turned out to be slightly weaker with an IC50 value of 1260 nM
in the latter assay.
45
1.779
n.d.
N
N
O
F
46
47
48
0.124
0.014
0.042
4.5
1.8
4.4
N
N
O
O
N
O
S
O
F
F
49
50
0.015
0.040
3.5
8.1
N
O
O
Both compounds showed comparable fXa activity for the hu-
man and rabbit enzymes. Compounds 27 and 30 were therefore
profiled in rabbit in vivo models of anticoagulation. The results will
be reported elsewhere.
S
O

5318
L. Anselm et al. / Bioorg. Med. Chem. Lett. 20 (2010) 5313–5319
Table 4
Pharmacological, safety, and pharmacokinetic data
Compound
26
27 R1663
29
30
fXa Ki (nM)
24
7
15
27
2xPT (
l
M)
3.8
1.5
1.3
3.3
100
1.7
1.8
1.6
6.6
Prothrombinase Ki (nM)
fXa Ki (nM) rabbit
fXa Ki (nM) rat
n.d.b
n.d.b
n.d.b
n.d.b
n.d.b
n.d.b
n.d.b
n.d.b
n.d.b
n.d.b
n.d.b
17
n.d.b
4.3
fXa KD (nM) human
3.7
kon (Mꢁ1 sꢁ1
)
3.6 ꢀ 107
9.8 ꢀ 105
koff (sꢁ1
)
1.3 ꢀ 10ꢁ1
3.9 ꢀ 10ꢁ2
Human whole blood ex vivo anticoagulation
IC50 (nM) (collagen)
IC50 (nM) (J82)
n.d.b
n.d.b
310
840
n.d.b
n.d.b
n.d.b
1260
Selectivitya
fu human (%)
(lM)
>33
14
>33
23
>33
<0.6
>33
6.8
CL human microsomes (ml/min/kg)
CL human hepatocytes (ml/min/kg)
5.3
0.09
4.7
1.1
10
0.01
0.2
n.d.b
hERG IC50
(lM)
>10
>10
12
>10
Ames/MNT/Phototox.
pKa/phospholipidosis
Negative
2.99/negative
Negative
4.79/negative
Negative
7.55/marginal
Negative
2.75/negative
Rat PK
F (%)
24
1.1
1.3
0.040
3300
100
1.3
2.1
0.18
10,000
39
100
1.5
7.3
0.42
2300
t1/2 (h) po
CL (ml/min/kg)
Vss (L/kg)
2.9
4.2
0.47
900
AUCpo/D (ng h/ml)
Cyno. monkey PK
F (%)
9
65
80
64
t1/2 (h)
CL (ml/min/kg)
Vss (L/kg)
1.9
5.5
0.46
230
6.3
6.4
0.54
1700
4.7
17
1.6
820
5.0
1.8
0.44
6700
AUCpo/D (ng h/ml)
a
Selectivity was determined versus a panel of serine proteases consisting of TF/fVIIa, thrombin, fIXa, trypsin, fXIa, aPC, plasmin, kallikrein.
n.d., not determined.
b
Acknowledgments
The authors thank Christiane Wohlgensinger and Olivier Kuster
for testing the compounds in the human whole blood ex vivo anti-
coagulation assay and Martine Stihle for crystallization.
References and notes
1. CDC/NCHS/AHA, 2003 data.
2. (a) Eriksson, B. I.; Quinlan, D. J.; Weitz, J. I. Clin. Pharmacokinet. 2009, 48, 1; (b)
Golino, P.; Loffredo, F.; Riegler, L.; Renzullo, E.; Cocchia, R. Curr. Opin. Invest.
Drugs 2005, 6, 298; (c) Carreiro Zikria, J.; Ansell, J. Curr. Opin. Hematol. 2009, 16,
347; (d) Laux, V.; Perzborn, E.; Heitmeier, S.; von Degenfeld, G.; Dittrich-
Wengenroth, E.; Buchmueller, A.; Gerdes, C.; Misselwitz, F. Thromb. Haemost.
2009, 102, 892.
3. (a) Verheugt, F. W. A. Nat. Rev. Cardiol. 2010, 7, 149; (b) Turpie, A. G. G. Eur.
Heart J. 2008, 29, 155; (c) Blondon, M.; Bounameaux, H.; Righini, M. Expert Opin.
Pharmacother. 2009, 10, 1159.
4. de Candia, M.; Lopopolo, G.; Altomare, C. Expert Opin. Ther. Patents 2009, 19,
1535.
5. (a) Mann, K. G.; Jenny, R. J.; Krishnaswamy, S. Annu. Rev. Biochem. 1988, 57, 915;
(b) Mann, K. G.; Butenas, S.; Brummel, K. Arterioscler. Thromb. Vasc. Biol. 2003,
23, 17.
6. (a) Roehrig, S.; Straub, A.; Pohlmann, J.; Lampe, T.; Pernerstorfer, J.; Schlemmer,
K.-H.; Reinemer, P.; Perzborn, E. J. Med. Chem. 2005, 48, 5900; (b) Perzborn, E.;
Strassburger, J.; Wilmen, A.; Pohlmann, J.; Roehrig, S.; Schlemmer, K.-H.;
Straub, A. J. Thromb. Haemost. 2005, 3, 514; (c) Perzborn, E.; Roehrig, S.; Straub,
A.; Kubitza, D.; Mueck, W.; Laux, V. Arterioscler. Thromb. Vasc. Biol. 2010, 30,
376.
7. (a) Eriksson, B. I.; Turpie, A. G. G.; Lassen, M. R.; Prins, M. H.; Agnelli, G.;
Kaelebo, P.; Gaillard, M. L.; Meems, L. J. Thromb. Haemost. 2007, 5, 1660; (b)
Turpie, A. G. G. Arterioscler. Thromb. Vasc. Biol. 2007, 27, 1238.
8. (a) Escolar, G.; Diaz-Ricart, M. Drugs Future 2009, 34, 861; (b) Mendell, J.;
Basavapathruni, R.; Swearingen, D.; Draves, A.; Zhang, D. Clin. Pharmacol. Ther.
2009, 85. Abst PI-79.
9. (a) Pinto, D. J.; Orwat, M. J.; Koch, S.; Rossi, K. A.; Alexander, R. S.; Smallwood,
A.; Wong, P. C.; Rendina, A. R.; Luettgen, J. M.; Knabb, R. M.; He, K.; Xin, B.;
Wexler, R. R.; Lam, P. Y. J. Med. Chem. 2007, 50, 5339; (b) Bates, S. M.; Weitz, J. I.;
Serradell, N.; Rosa, E.; Bolos, J. Drugs Future 2008, 33, 293; (c) Lassen, M. R.;
Figure 2. X-ray co-crystal structure of compound 27 bound to the active site of fXa
(resolution 1.66 Å, PDB code 2xbv).18 Hydrogen bonds of the ligand are shown in
red and protein-ligand interactions in the S1b-pocket in blue dashed lines.
After carefully reviewing all data, compound 27 (R1663) was se-
lected for further clinical development.
In summary, we have discovered potent, selective and orally
bioavailable fXa inhibitors based on the pyrrolidine scaffold 1. By
modulating in particular the S1b-pocket residue, the pharmacoki-
netic properties of the previously published series could be signif-
icantly improved. Within this series compound 27 (R1663) was
selected for further clinical development and a Phase I trial has
been finalized.

L. Anselm et al. / Bioorg. Med. Chem. Lett. 20 (2010) 5313–5319
5319
Raskob, G. E.; Gallus, A.; Pineo, G.; Chen, D.; Portman, R. J. New Engl. J. Med.
2009, 361, 594.
for 5 min at 25 °C. The velocity of the reaction is determined by the autoreader
from the slope of the linear regression fit of 7 times points (1 min). The initial
velocity for each inhibitor concentration is determined with the slope of at
least 4 times points in the linear phase by a linear regression fit (mOD/min2)
(5–10–15–20 min time points). Kiꢂ is calculated from the IC50 with the Cheng–
Prusoff equation ½Kꢂi ¼ IC50=ð1 þ ðS=KmÞÞꢃ (Km prothrombin/PTase complex:
51.2 nM). The effect of inhibitors on thrombin activity is determined separately
in an assay using all components of the reaction mixture except fXa. All
reactants are diluted in 0.1 M Hepes, 0.14 M NaCl, 0.5 mg/ml BSA (fatty acid
free) pH 7.5 except S-2238 (in H2O).
10. Groebke Zbinden, K.; Anselm, L.; Banner, D. W.; Benz, J.; Blasco, F.; Decoret, G.;
Himber, J.; Kuhn, B.; Panday, N.; Ricklin, F.; Risch, P.; Schlatter, D.; Stahl, M.;
Thomi, S.; Unger, R.; Haap, W. Eur. J. Med. Chem. 2009, 44, 2787.
11. Nazaré, M.; Will, D. W.; Matter, H.; Schreuder, H.; Ritter, K.; Urmann, M.;
Essrich, M.; Bauer, A.; Wagner, M.; Czech, J.; Lorenz, M.; Laux, V.; Wehner, V. J.
Med. Chem. 2005, 48, 4511.
12. Rodriguez Sarmiento, R. M.; Wirz, B.; Iding, H. Tetrahedron: Asymmetry 2003,
14, 1547.
13. (a) Qiao, J. X.; Chang, C.-H.; Cheney, D. L.; Morin, P. E.; King, S. R.; Wang, G. Z.;
Wang, T. C.; Rendina, A. R.; Luettgen, J. M.; Knabb, R. M.; Wexler, R. R.; Lam, P.
Y. S. Bioorg. Med. Chem. Lett. 2007, 17, 4419; (b) Qiao, J. X.; Wang, T. C.; Wang, G.
Z.; Cheney, D. L.; He, K.; Rendina, A. R.; Xin, B.; Luettgen, J. M.; Knabb, R. M.;
Wexler, R. R.; Lam, P. Y. S. Bioorg. Med. Chem. Lett. 2007, 17, 5041.
14. (a) Van Huis, C. A.; Bigge, C. F.; Casimiro-Garcia, A.; Cody, W. L.; Dudley, D. A.;
Filipski, K. J.; Heemstra, R. J.; Kohrt, J. T.; Narasimhan, L. S.; Schaum, R. P.;
Zhang, E.; Bryant, J. W.; Haarer, S.; Janiczek, N.; Leadley, R. J., Jr.; McClanahan,
T.; Peterson, J. T.; Welch, K. M.; Edmunds, J. J. Chem. Biol. Drug Des. 2007, 69,
444; (b) Kohrt, J. T.; Bigge, C. F.; Bryant, J. W.; Casimiro-Garcia, A.; Chi, L.; Cody,
W. L.; Dahring, T.; Dudley, D. A.; Filipski, K. J.; Haarer, S.; Heemstra, R.; Janiczek,
N.; Narasimhan, L.; McClanahan, T.; Peterson, J. T.; Sahasrabudhe, V.; Schaum,
R.; Van Huis, C. A.; Welch, K. M.; Zhang, E.; Leadley, R. J.; Edmunds, J. J. Chem.
Biol. Drug Des. 2007, 70, 100; (c) Van Huis, C. A.; Casimiro-Garcia, A.; Bigge, C.
F.; Cody, W. L.; Dudley, D. A.; Filipski, K. J.; Heemstra, R. J.; Kohrt, J. T.; Leadley,
R. J.; Narasimhan, L. S.; McClanahan, T.; Mochalkin, I.; Pamment, M.; Thomas
Peterson, J.; Sahasrabudhe, V.; Schaum, R. P.; Edmunds, J. J. Bioorg. Med. Chem.
2009, 17, 2501.
17. Kirchhofer, D.; Tschopp, T. B.; Hadvàry, P.; Baumgartner, H. R. J. Clin. Invest.
1994, 93, 2073.
18. X-ray structural analysis: Crystals of short form factor Xa (wt or the Arg150Glu
mutant) were produced and data measured as previously described10 except
that data were also collected at the Swiss Light Source. Data were processed
with XDS19 and data reduction used the CCP4 package.20 The first structure
was solved by molecular replacement using 1fax.pdb as model. Model building
was performed with Moloc21 and refinement with CNX 2000.22 Later structures
were refined with Refmac523 and for deposition Coot24 and Buster25 were also
used.
Coordinates have been deposited at the Protein Data Bank with the PDB codes:
2xbv (27), 2xbw (10), 2xbx (2), 2xc0 (35), 2xc4 (11), 2xby, and 2xc5 (other
compounds).
19. Kabsch, W. J. Appl. Crystallogr. 1993, 26, 795.
20. Collaborative Computational Project Number 4. Acta Crystallogr., Sect. D 1994,
D50, 760.
21. (a) Gerber, P. R.; Müller, K. J. Comput. Aided Mol. Des. 1995, 9, 251; Gerber
Molecular Design. Available from: <http://www.moloc.ch/>.
15. Morgenthaler, M.; Aebi, J. D.; Grüninger, F.; Mona, D.; Wagner, B.; Kansy, M.;
Diederich, F. J. Fluorine Chem. 2008, 129, 852.
22. Brünger, A. T.; Adams, P. D.; Clore, G. M.; DeLano, W. L.; Gros, P.; Grosse-
Kunstleve, R. W.; Jiang, J.-S.; Kuszewski, J.; Nilges, N.; Pannu, N. S.; Read, R. J.;
Rice, L. M.; Simonson, T.; Warren, G. L. Acta Crystallogr., Sect. D 1998, D54, 905.
23. Murshudov, G. N.; Vagin, A. A.; Dodson, E. J. Acta Crystallogr., Sect. D 1997, D53,
240.
24. Emsley, P.; Cowtan, K. Acta Crystallogr., Sect. D 2004, D60, 2126.
25. (a) Bricogne, G. Acta Crystallogr., Sect. D 1993, D49, 37; (b) Bricogne, G. Methods
Enzymol. 1997, 276, 361; (c) Roversi, P.; Blanc, E.; Vonrhein, C.; Evans, G.;
Bricogne, G. Acta Crystallogr., Sect. D 2000, D56, 1313; (d) Blanc, E.; Roversi, P.;
Vonrhein, C.; Flensburg, C.; Lea, S. M.; Bricogne, G. Acta Crystallogr., Sect. D
2004, D60, 2210.
16. Prothrombinase assay test description: First,
complex formed for 10 min at 25 °C (final concn: fXa 2 pM, fVa 1 nM,
phospholipids PC/PS (70%:30%) 25 M, CaCl2 5 mM) is incubated for 10 min
with the test compound before adding the substrate prothrombin (40 nM
final). At different time points (1, 2, 3, 5, 10, 15, 20, and 30 min), 50 l aliquots
of the incubation mixture are transferred into a 96-wells plate containing
150 l EDTA (12 mM final) to stop the reaction. The concentration of thrombin
generated is determined with specific chromogenic substrate, S-2238
(100 M final). The cleavage of S-2238 by thrombin is followed at 405 nm
a prothrombinase (PTase)
l
l
l
a
l