The design, synthesis, and biological evaluation of potent receptor tyrosine kinase inhibitors

Article abstract of DOI:10.1016/j.bmcl.2012.06.029

Variously substituted indolin-2-ones were synthesized and evaluated for activity against KDR, Flt-1, FGFR-1 and PDGFR. Extension at the 5-position of the oxindole ring with ethyl piperidine (compound 7i) proved to be the most beneficial for attaining both biochemical and cellular potencies. Further optimization of 7i to balance biochemical and cellular potencies with favorable ADME/ PK properties led to the identification of 8h, a compound with a clean CYP profile, acceptable pharmacokinetic and toxicity profiles, and robust efficacy in multiple xenograft tumor models.

Full text of DOI:10.1016/j.bmcl.2012.06.029

Bioorganic & Medicinal Chemistry Letters 22 (2012) 4979–4985
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
The design, synthesis, and biological evaluation of potent receptor tyrosine
kinase inhibitors
⇑
Moon H. Kim, Amy Lew Tsuhako , Erick W. Co, Dana T. Aftab, Frauke Bentzien, Jason Chen, Wei Cheng,
Stefan Engst, Levina Goon, Rhett R. Klein, Donna T. Le, Morrison Mac, Jason J. Parks, Fawn Qian,
Monica Rodriquez, Thomas J. Stout, Jeffrey H. Till, Kwang-Ai Won, Xiang Wu, F. Michael Yakes,
Peiwen Yu, Wentao Zhang, Yeping Zhao, Peter Lamb, John M. Nuss, Wei Xu
Exelixis, 210 E. Grand Avenue, South San Francisco, CA 94080, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Variously substituted indolin-2-ones were synthesized and evaluated for activity against KDR, Flt-1,
FGFR-1 and PDGFR. Extension at the 5-position of the oxindole ring with ethyl piperidine (compound
7i) proved to be the most beneficial for attaining both biochemical and cellular potencies. Further opti-
mization of 7i to balance biochemical and cellular potencies with favorable ADME/ PK properties led to
the identification of 8h, a compound with a clean CYP profile, acceptable pharmacokinetic and toxicity
profiles, and robust efficacy in multiple xenograft tumor models.
Received 4 May 2012
Revised 6 June 2012
Accepted 11 June 2012
Available online 16 June 2012
Keywords:
Receptor tyrosine kinase
Angiogenesis
KDR
Ó 2012 Elsevier Ltd. All rights reserved.
VEGFR
Flt
FGFR
PDGFR
Oxindole
Indolin-2-one
Vascular endothelial growth factor receptor 1 (VEGFR1 or fms-
like tyrosine kinase receptor, Flt-1) and vascular endothelial
growth factor receptor 2 (VEGFR2 or kinase insert domain receptor,
KDR) are receptor tyrosine kinases (RTKs) that regulate endothelial
cell function.1 Upon binding to members of the VEGF family of
growth factors, these receptors undergo dimerization and auto-
phosphorylation, and initiate a cascade of intracellular down-
stream signaling that promotes angiogenesis. Induction of
angiogenesis is thought to be required for the ongoing growth of
human tumors and consequently, targeting angiogenesis via inhi-
bition of the VEGF pathway has been a prevalent approach to can-
cer therapy.2,3
N
N
H
N
H
O
N
H
O
HN
HN
2
1
KDR:
FGFR1
: 95 nM
KDR:
FGFR1
: >4880 nM
97 nM;
>4880 nM;
α
PDGFR : 261 nM
Flt-1: 405 nM
Figure 1. Lead Compound 1 and the corresponding des-phenyl analog.
Selective inhibition of VEGF signaling, best exemplified by the
use of bevacizumab, has shown limited efficacy as a single thera-
peutic approach, but has been highly successful when used in com-
bination with chemotherapy.4–6 One possible explanation for the
limited single agent activity is the presence of alternate mecha-
nisms for the promotion of angiogenesis.7,8 Studies in pre-clinical
tumor models suggest that both platelet-derived growth factor
R1
Sunitinib: R1=CONH(CH2)N(C2H5)2; X = F
SU5416: R1 = H; X = H
3
4
5
H
SU6668: R1 = (CH2)2COOH; X = H
N
H
O
X
HN
⇑
Corresponding author. Tel.: +1 650837 8042.
E-mail addresses: amy.lew@gmail.com, alew@exelixis.com (A.L. Tsuhako).
Figure 2. Oxindole VEGFR-2 inhibitors.
0960-894X/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmcl.2012.06.029

4980
M. H. Kim et al. / Bioorg. Med. Chem. Lett. 22 (2012) 4979–4985
Table 1
Biochemical inhibitory activity for A, B, C ring oxindole modifications
5
4
R1
6
A
B
N
R2
N
H
O
5
HN
C
R3
6
7
a
ID
R1
R2
R3
Kinase inhibitory activity IC50 (nM)
KDR
37
135
97
119
88
633
125
106
290
19903
2180
327
9876
68
Flt-1
FGFR1
3
5
—
—
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
1
28
618
681
599
1064
1950
205
814
187
950
79
4000
628
4899
725
86
6293
57
446
4880
4880
75
6a
6b
6c
6d
6e
6f
6g
6h
6i
o-Me
m-Me
p-Me
p-F
H
H
H
H
H
H
H
H
H
H
H
H
2308
522
m-F
3,5-Di-F
2,4-Di-F
p-CF3
2,4-Di-Cl
p-OMe
p-OEt
p-NH2
H
H
H
H
H
H
H
p-Me
p-Me
H
2551
1796
24000
24400
984
10543
371
118
4544
4880
39
582
20716
201
6j
6k
6l
6m
6n
6o
6p
6q
6r
6s
6t
5-Br
6-Br
7-Br
5-COOH
6-COOH
7-COOH
5-NH2
5-COOH
5-NH2
5-COOH
5-NH2
51
4880
4880
4
213
23633
32
1586
24400
13
5
82
3
60
45
473
77
377
67
51
39
42
6u
6v
6w
5-OMe
5-OMe
H
a
Values reported are the average of at least two independent dose-response curves; variation was generally 15%.35
receptor (PDGFR) and fibroblast growth factor receptor (FGFR) acti-
vation can contribute to angiogenesis and promote resistance to
VEGF-pathway targeting agents.9,10
Fig. 1). Similar to compound 2, sunitinib (3), in the absence of the
phenyl substitution, was also inactive against FGFR1 (Table 1).
With such dramatic differences in KDR and FGFR activities in the
presence/absence of the phenyl ring substitution on the alkene,
we felt that there was unexplored SAR in this region. Herein, we
describe optimization efforts of 1 that led to the discovery of com-
pound 8h.
Multiple first generation small molecule inhibitors of the VEGFR
pathway are currently on the market or in clinical trials.2,3 A num-
ber of these combine inhibition of VEGFRs with other RTKs, includ-
ing PDGFRs, FGFRs or EGFRs. AG-013736 (Axitinib)11,12 CHIR-258/
TKI258 (Dovitinib),13 BIBF 1120 (Vargatef),14 AZD2171 (Cedira-
nib),15 and BMS-540215 (Brivanib)16 are examples of small mole-
cule inhibitors that target VEGFRs in combination with other
RTKs. Emerging clinical evidence suggests that these single agents
may have a broader spectrum of activity than highly selective
VEGF targeting agents.17 Given the limited activity of selective
VEGF-pathway inhibitors, we sought to optimize a second genera-
tion angiogenesis inhibitor that would target multiple pro-angio-
genic pathways simultaneously through balanced inhibition of
VEGFRs, PDGFRs and FGFRs.
Efforts towards establishing SAR around 1 were focused on
varying substituents in the B and C rings, two areas less explored
in oxindoles targeted against KDR. In the phenyl B ring, a variety
of ortho, meta, and para substituents were explored (Table 1). Small
substituents were tolerated in this region, as ortho-, meta-, and
para- methyl and fluoro substitutions retained KDR activity (6a–
6e). However, once the substituents were larger, such as the di-
chloro (6i), or p-ethoxy (6k) groups, KDR activity was dramatically
decreased. In addition, the para-position was less tolerant of elec-
tron withdrawing functional groups as the p-F (6d) was less potent
than the m-F (6e) analog and replacement of the active p-Me with
p-CF3 (6h) rendered the analog inactive. However, none of these B
ring modifications were significantly more potent than HTS lead 1.
As opposed to B ring modifications, a clear trend emerged from
modifications in the C ring (Table 1). Placing substituents in the 5-
position of the oxindole ring improved KDR potency. For example,
in the bromide series (6m–6o), a 5-bromo substitution was 100-
fold more active than a 6- or 7-bromo substitution. Likewise, a 5-
carboxylate substitution was 50 to 1000-fold more potent than
the corresponding 6- or 7- carboxylate substitution (6p–6r).
Interestingly, both anilines and carboxylic acids were tolerated in
Our efforts began with high throughput screening (HTS) of our
internal small molecules library against KDR, leading to the identi-
fication of oxindole 1 with modest activities against KDR, FGFR1,
and PDGFR
a
(Fig. 1). Several oxindoles14,18–23 or oxindole-like
compounds24–26 have been documented in the literature as RTK
inhibitors, with sunitinib,27 currently approved to treat kidney,
pancreatic, and gastrointestinal cancers (Fig. 2). However, as op-
posed to HTS hit 1, sunitinib (3) does not have a phenyl group on
the substituted alkene (Comparison of Fig. 2, compound 3 with
Fig. 1, compound 1). Removal of this phenyl ring on HTS hit 1, ren-
dered the analog inactive against VEGFR and FGFR1 (Compound 2,

M. H. Kim et al. / Bioorg. Med. Chem. Lett. 22 (2012) 4979–4985
4981
Table 2
Biochemical and cellular inhibitory activity for 5-amide, sulfonamide and amine oxindoles
R1
N
R2
R4
N
H
O
HN
R1
R2
R4
Kinase inhibitory activity IC50 (nM)
Cell based IC50 (nM)
KDR/ERKb FGFR1/ERKc
a
a
ID
KDR
Flt-1
FGFR1
PDGFR
13
a
O
O
O
7a
7b
H
H
H
H
15
374
326
145
128
30000
9666
N
H
O
N
69
381
199
70
55
—
N
H
7c
H
H
157
4322
1804
983
N
O
O
S
7d
7e
H
H
H
H
1134
660
24400
1950
300
287
844
614
—
—
N
H
O
O
O
S
O
30000
30000
N
H
O
S
7f
H
H
1950
1950
1950
1950
30000
30000
N
N
O
7g
7h
5-OMe
5-OMe
H
H
22
691
20
17
213
496
736
N
N
H
N
279
>1950
216
431
1010
H
N
7i
7j
5-OMe
H
H
H
10
9
163
184
11
46
6
27
17
41
50
N
N
H
N
10
O
a
b
c
Values reported are the average of at least two independent dose-response curves; variation was generally 15%.35,38,39
VEGF-stimulated ERK phosphorylation in human endothelial cells (HUVEC).
FGF-stimulated ERK phosphorylation in SK-N-MC neuroblastoma cells.
the 5-position; although carboxylic acid derivatives were more po-
tent than aniline derivatives (6p vs 6s, 6t vs 6u, 6v vs 6w).
With encouraging data from the 5-amino and 5-carboxylate
analogs, we focused on expanding the 5-position substitutions of
the oxindole ring. Table 2 summarizes the amides, sulfonamides,
and amines explored in the 5-position. In general, bulky amides
were not tolerated, and primary amides were better than second-
ary amides. In addition, amide analogs were more potent in KDR,
an ether linkage such as analog 7g, still maintained KDR potency,
but in general was less potent, especially against Flt-1.
To better understand the SAR observed for the 5-aminopiperid-
inyl series, 7i was co-crystallized in EphB4 (IC50 = 765 nM), a surro-
gate receptor tyrosine kinase, since access to the crystal structure
of KDR, FGFR or PDGFR was not available at the time of this study.
The crystal structure showed that in addition to the triad of hydro-
gen bonds between the indolin-2-one core of 7i and the hinge re-
gion, there was an additional hydrogen bond between the
piperidine nitrogen and Asp234, the catalytic acid responsible for
chelating magnesium in the reaction complex (Fig. 3). This interac-
tion is also predicted to be present in KDR and could account for
the increase in potency observed for 7g, 7i, and 7j. To the best of
our knowledge, the interaction of this aminopiperidinyl functional
group with the kinase had not previously been described for oxin-
dole-based inhibitors of KDR. In addition, the phenyl B ring is
tightly packed, forming primarily hydrophobic interactions with
the protein. This tight packing likely contributes to the high
FGFR1 and PDGFRa than the corresponding sulfonamide analogs
(7a vs 7d, 7b vs 7e, 7c vs 7f). However, despite the fact that amide
7a showed good biochemical potency, it was only modestly potent
at inhibiting endogenously expressed VEGF and FGF receptor activ-
ity in human endothelial cells and human umbilical vein endothe-
lial cell (HUVEC), respectively (Table 2). In contrast, the 5-amino
piperidinyl oxindole series was very potent in cells (7i, 7j). Regio-
chemistry of the amine was important as the 3-amino piperidine
analog 7h was significantly less potent than the 4-amino piperi-
dine analog 7i. Switching the 5-amino oxindole linkage to that of

4982
M. H. Kim et al. / Bioorg. Med. Chem. Lett. 22 (2012) 4979–4985
Vehicle sunitinib
Compound 8h, mg/kg
V
100
1
3
10
30
60
100
32
FGFR1
pFGFR1
Normalized
to V
100
104
0
92
8
92
8
57
52
34
pFGFR1
inh. %
-
-
-
43
48
66
68
plasma
conc. μM
Tumor
<LOQ 0.01
0.04
0.20
1.40
0.26
3.60
0.80
8.80
conc. μM
<LOQ <LOQ 0.34
Figure 5b. Pharmacodynamic studies for 8h in MDA-MB231T xenograft model
measuring inhibition of phosphorylation of FGFR1.
using an antibody that recognizes the endothelial cell marker
CD31. Compound 7i inhibits VEGF-induced tubule formation
(Fig. 4) with an IC50 of 110 nM. Likewise, FGF-induced tubule for-
mation was also tested in this system; and compound 7i potently
inhibits FGF-induced tubule formation with an IC50 of 10 nM
(Fig. 4). Collectively, these data demonstrate that inhibition of
VEGFR and FGFR by 7i translates into potent inhibition of endothe-
lial tube formation function in response to the key angiogenic fac-
tors VEGF and FGF.
Figure 3. Crystal structure of EphB4, a surrogate receptor tyrosine kinase, co-
crystallized with compound 7i (PDB code 4AW5). Unique to the oxindole class
targeting VEGFR, the 5-amino-piperidinyl moiety of 7i is able to form a hydrogen
bond with the catalytic ASP-234.
With biochemical and cellular potencies optimized for the C-
ring, and clear demonstration of anti-angiogenic properties in cells,
we were now ready to re-optimize the A and B rings with the goal
of improving ADME/PK properties. The 5-aminopiperidinyl oxin-
dole analog 7i showed CYP3A4 liabilities (CYP3A4 IC50: 5.6 lM),
high clearance (6497 mL/h/kg) and low oral availability (F = 19%)
in rats. Holding the C ring constant as the 5-aminopiperidinyl oxin-
dole, a variety of substituted and unsubstituted benzimidazoles
and imidazoles were synthesized as the A ring in combination with
a variety of substituted B rings. Highlights of these analogs are
shown in Table 3.
VEGF
FGF
VEGF + 7i
FGF + 7i
No VEGF
No FGF
With the 5-aminopiperidinyl substitution installed on the C
ring, a majority of the analogs showed single digit nM potencies
against KDR, FGFR, and PDGFR, and were below 100 nM in VEGF
and FGF-driven cellular signaling assays (Table 3). As predicted,
the benzimidazole A ring contributed significantly to the CYP3A4
activity. Surprisingly, when replaced with an imidazole or methyl
imidazole, the CYP3A4 activity was attenuated (Table 3). For exam-
ple, when the A ring benzimidazole of analog 8a was replaced with
an imidazole (8f), the CYP3A4 activity was abolished. Likewise,
when the A ring benzimidazole of analog 8d was replaced with
an imidazole (8g) or methyl imidazole (8i), the CYP3A4 activity im-
proved. In addition, we discovered that halogens on the B ring,
especially a Cl or di-F, aided in improving oral bioavailability in
rat pharmacokinetic studies (Table 3). Four compounds emerged
(8g, 8h, 8i, and 8k) with high potency against KDR, Flt-1, FGFR1
Figure 4. Effect of 7i on capillary-like tube formation induced by VEGF and FGF.
Vehicle
Compound 8h, mg/kg
V-
V+
1
3
10
30
60
100
KDR
pKDR
Compared to V+ 39
100
90
10
95
5
86
14
80
20
48
52
39
61
pKDR inh. %
plasma conc. μM
Lung conc. μM
-
-
-
-
0.004 0.01 0.02 0.01 0.01 0.62
<LOQ 0.07 0.28 0.47 0.81 3.55
-
and PDGFRa, good cell activity, a clean CYP profile (>10 lM) and
Figure 5a. Pharmacodynamic studies for 8h measuring inhibition of phosphory-
lated KDR in mouse lung tissue. Vꢁ is the basal level of KDR and pKDR while V+ is
the level of KDR and pKDR before compound treatment, but after VEGF stimula-
tion.28 Lung tissue was analyzed 4 h post dose.
reasonable PK profile.
Of the four compounds, 8h was chosen for in vivo pharmacody-
namic studies due to its more favorable oral bioavailability (F = 63%
in rats; Table 3). As shown in Figure 5a, single oral dose adminis-
tration of 8h resulted in a dose dependent decrease in VEGF-stim-
ulated phosphorylation of KDR in lung lysates 4 h post dose. The
plasma concentration dose response relationship predicts 50%
inhibition of phosphorylation of KDR in vivo to occur at
potency observed against KDR in this series and also explains the
intolerance to larger substituents on the phenyl ring.
To ascertain whether the in vitro profile of 7i correlated with
anti-angiogenic properties, endothelial tube formation assays were
performed. When plated on a confluent layer of normal human
diploid fibroblast cells, human microvascular endothelial cell
(HMVECS) form extensive networks of tubules in response to VEGF
over a period of 7–10 days. Tubules are stained and quantitated
ꢀ0.28
relationship indicates that 50% inhibition of phosphorylation of
KDR is predicted to occur at a lung concentration of 1.6 M. Note
lM. The corresponding lung lysate concentration-response
l
that the 100-mg/kg dose caused a reduction in phosphorylation
of the receptor to levels below the basal phosphorylation state.

M. H. Kim et al. / Bioorg. Med. Chem. Lett. 22 (2012) 4979–4985
4983
Table 3
Biochemical, cellular, and cytochrome P450 inhibitory activity for 5-amino ethyl piperidine oxindole analogs
R1
R1
N
N
R2
R2
N
N
H
O
H
O
H
N
H
N
HN
HN
N
N
(8a-8e)
(8f-8k)
R1
R2
Kinase inhibitory activity IC50 (nM)
Cell based IC50 (nM)
Rat PK dosed at 10 mg/kg (PO)
a
a
ID
KDR
Flt-1
FGFR1
PDGFR
a
KDR/ERKb
FGFR1/ERKc
CYP3A4 IC50
(lM)
AUC(0–1)
(
lMꢂh)
t1/2 (h)
%F
8a
8b
8c
8d
8e
8f
8g
8h
8i
H
H
H
H
H
H
H
H
H
6
161
112
96
62
66
166
25
7
5
3
5
7
7
6
20
6
3
2
3
5
<10
17
59
44
38
46
154
17
24
20
24
15
3
12
9
1.44
2.36
0.33
1.99
3.14
1.50
1.12
5.04
1.90
5.19
1.03
5.7
4.6
7.5
4.4
3.8
11.6
5.4
10.7
4.6
7.2
2.3
37
55
27
63
72
16
15
63
20
59
25
p-Me
p-OMe
m-F
3,5-DiF
H
m-F
p-Cl
m-F
8
10
14
5
<10
13
<10
<10
235
20
11
6
7
13
3
4
6
4
8
16
7
12
13
6
4
7
25
23
>50
25
10
26
4
4
3
3
Me
Me
Me
20
11
29
8j
8k
3,5-DiF
o-F
8
16
a
b
c
Values reported are the average of at least two independent dose-response curves; variation was generally 15%.35,38,39
VEGF-stimulated ERK phosphorylation in human endothelial cells (HUVEC).
FGF-stimulated ERK phosphorylation in SK-N-MC neuroblastoma cells.
In a separate experiment, 8h was administered to mice bearing
MDA-MB-231 xenograft tumors transfected with a stable construct
expressing FGFR1. Dose dependent inhibition of phospho-FGFR1
was observed with the corresponding dose response relationship
predicting 50% inhibition of phosphorylation of FGFR1 to occur at
occurred during the dosing period. Antitumor efficacy in both
models was associated with cyclical plasma exposure with peak
concentrations of 1–5 lM.
The analogs in Tables 1–3 were prepared by heating benzimid-
azole 9b together with benzoyl chloride 10 at 135 °C in the pres-
ence of triethyl amine and acetonitrile (Scheme 1).29–31 Although,
both N- and C- acylation of the benzimidazole occurred, refluxing
the diacylated intermediate with 7% aqueous HCl for 30 min led
to hydrolysis at the N-acylated site, and yielded the desired C-acyl-
ated benzimidazole ketone 11b. The corresponding imidazole ke-
tone 11a was also synthesized in a similar manner. Subsequent
aldol condensation of 11a or 11b with oxindole 12 in a sealed tube
of ammonia in ethanol afforded the desired analogs, with high ste-
reoselectivity.32 X-ray analysis (data not shown) and 2D NMR (data
not shown) also confirmed the E/Z stereoselectivity shown in
Scheme 1. Non-commercially available oxindoles such as 5-sulfon-
amide oxindoles were synthesized via regioselective sulfonylation
with chlorosulfonic acid, followed by amine addition to the corre-
a plasma concentration of ꢀ0.14
lM. Note that despite achieving
significantly higher plasma concentrations when compared to 8h,
a 100-mg/kg dose of sunitinib did not result in inhibition of phos-
phorylation of FGFR1 (Fig. 5b).
The encouraging pharmacodynamic data for 8h supported addi-
tional in vivo characterization. Mice bearing MDA-MBA-231
(breast adenocarcinoma) and HT29 (colon adenocarcinoma) xeno-
graft tumors were orally administered 8h once daily for 14 days,
starting when tumors were approximately 100 mg in size. Dose
dependent tumor growth inhibition was observed in both models,
(Fig. 6), and was associated with significant reductions in tumor
mean vessel density, consistent with an anti-angiogenic mecha-
nism of action (data not shown). The estimated ED50 values were
35 and 28 mg/kg for MDA-MB-231 and HT29 models, respectively.
The compound was well tolerated as no loss of body weight
sponding
2-oxo-2,3-dihydro-1H-indole-5-sulfonyl
chloride
(Scheme 2).33 The 5-aminopiperidinyl oxindole was synthesized
A
B
800
800
600
400
200
0
600
400
200
0
0
5
10
15
0
5
10
15
Days of Treatment
Days of Treatment
Figure 6. Xenograft efficacy for 8h.40 Female athymic mice bearing established MDA-MB-231 (A) and HT29 (B) xenograft tumors were orally administered 8h at 10 mg/kg (),
30 mg/kg (}), and 100 mg/kg (s) or saline vehicle (h) for 14 days. Tumor weights were determined twice weekly. Data points represent the mean tumor weight (in
milligrams) and standard error for each treatment group.

4984
M. H. Kim et al. / Bioorg. Med. Chem. Lett. 22 (2012) 4979–4985
R1
Table 4
Selectivity profile for 8h
R1
N
N
9a
a
Kinase
Inhibitory activity for 8h IC50 (nM)
N
H
11a
N
H
O
VEGFR family
KDR (VEGFR-2)
Flt-1 (VEGFR-1)
Flt-4 (VEGFR-3)
PDGFR family
c-kit
PDGFR-
PDGFR-b
Flt-3
Other tyrosine kinases
FGFR1
EGFR
EphB4
EphA2
IGF-1R
R3
OR
OR
R2
4.2 0.5
7.2 0.8
3.2 0.4
R2
Cl
N
N
R3
10
N
N
H
11
2
H
a-c
O
a
3.0 0.3
1.3 0.1
0.85 0.19
11b
9b
R4
O
5.7 0.5
137 27
N
H
1250 60
1700 400
2700 400
4100 200
12
d or e
IRK
R2
OR
R1
Scr-family tyrosine kinases
N
N
Src
33
—
3
2
R3
Yesb
N
H
N
Serine threonine kinases
MAP4K3
Rsk2
H
O
25
8f-8k
3.0
HN
R4
a
Values reported are the average of at least two independent dose-response
curves with the exceptions of values for Yes, MAP4K3, and Rsk where only one IC50
6a-6w, 7a-7j, 8a-8e
value was obtained.
b
Compound 8h inhibited Yes with 99% inhibition when tested at 1 lM.
Scheme 1. Reagents and conditions: (a) Et3N, CH3CN, 135 °C, 2 h (b) 7% HCl,
acetone, charcoal, reflux 30 min (c) NH4OH (d) NH3, EtOH, 90 °C (e) N,N-dimeth-
ylethylene diamine, EtOH, 90 °C.
biochemical activity. Further optimization for favorable ADME/PK
properties led to the development of 8h,41 a compound with an
acceptable MDCK value (68 nM/sec), a clogP of 3.3, and a polar sur-
face area of 73 Å.
X
a
O
O
N
H
N
H
This compound demonstrated potent activities against KDR, Flt-
1, PDGFR, and FGFR1, which translated into robust pharmacody-
namic and antitumor activity in vivo. When screened against a
wider panel of kinases, 8h also exhibited low nM inhibitory activity
against several Src family tyrosine kinases (Src, and Yes), and sev-
eral cell cycle and mitogen activated protein kinases (MAP4K3, and
the Rsk family members), but was inactive against EGFR, EphB4/
A2, IGFR, or IRK kinases. (Table 4). Although, preliminary hERG
binding studies showed modest affinity for the ion channel (IC50
3367 nM), we felt that 8h’s high potency, warranted further inves-
tigation. On the basis of its favorable in vitro profile and in vivo
efficacy, 8h was advanced into further preclinical in vivo safety
studies.
13
14
X= SO2Cl
b
c
15
X
X= SO2NR5R6
O
d
N
H
16
X= NO2
H
N
O
N
e
N
H
17
18
X= NH2
Acknowledgments
Scheme 2. Reagents and conditions: (a) ClSO3H (b) NHR5R6, DIEA, THF (c) KNO3,
H2SO4 (d) Pd/C, EtOH, H2, 10–20 psi (e) 4-ethyl piperidone, NaBH(OAc)3, HOAc, 1,2-
DCE, 4 Å mol sieves.
We greatly appreciate the contributions to the biochemical, cel-
lular, ADME/PK, and in vivo PD and efficacy data from members of
the Exelixis Genome Biochemistry, New Lead Discovery, Cell
Culture Facility, Pharmacology and Compound Repository.
via a regioselective nitration34 of oxindole using potassium nitrate
and sulfuric acid to give 5-nitro oxindole (16, Scheme 2), followed
by reduction of the nitro group with palladium on carbon in etha-
nol to give the desired 5-amino oxindole 17. Reductive amination
of 17 with 4-ethyl piperidone and sodium triacetoxyborohydride
then gave the desired 5-piperidinyl amino oxindole 18. Analogs
using the piperidinyl oxindole 18 or sulfonamide 15 were then
made via an ammonia ethanol aldol condensation with the appro-
priate benzimidazole or imidazole ketones as previously described.
In summary, we obtained a potent, small molecule inhibitor of
multiple angiogenic RTKs, starting from a lead oxindole identified
by screening an in-house library for inhibitors of KDR Systematic
References and notes
1. Ferrara, N.; Gerber, H. P.; LeCouter, J. Nat. Med. 2003, 9, 669.
2. Paz, K.; Zhu, Z. Front. Biosci. 2005, 10, 1415.
3. Ivy, S. P.; Wick, J. Y.; Kaufman, B. M. Nat. Rev. Clin. Oncol. 2009, 6, 569.
4. Hurwitz, H.; Fehrenbacher, L.; Novotny, W.; Cartwright, T.; Hainsworth, J.;
Heim, W.; Berlin, J.; Baron, A.; Griffing, S.; Holmgren, E.; Ferrara, N.; Fyfe, G.;
Rogers, B.; Ross, R.; Kabbinavar, F. N. Eng. J. Med. 2004, 350, 2335.
5. Ferrara, N.; Hillan, K. J.; Novotny, W. Biochem. Biophys. Res. Commun. 2005, 333,
328.
6. Cohen, M. H.; Gootenberg, J.; Keegan, P.; Pazdur, R. Oncologist 2007, 12, 713.
7. Bergers, G.; Hanahan, D. Nat. Rev. Cancer 2008, 8, 592.
8. Gordon, M. S. Expert Rev. Anticancer Ther. 2011, 11, 485.
9. Thurnher, A.; Katsen, A. D.; Groth, G.; Kerger, H.; Hammes, H-P.; Menger, M. D.;
Ullrich, A.; Vajkoczy, P. FASEB J. 2004, 18, 338.
modification around the core led to the identification of
a
5-amino-piperidinyl functionality that significantly improved
10. Casanovas, O.; Hicklin, D. J.; Bergers, G.; Hanahan, D. Cancer Cell 2005, 8, 299.

M. H. Kim et al. / Bioorg. Med. Chem. Lett. 22 (2012) 4979–4985
4985
11. Hu-Lowe, D. D.; Zou, H. Y.; Grazzini, M. L.; Hallin, M. E.; Wickman, G. R.;
incubated with 3 nM biotinylated poly(Glu,Tyr) peptide, 5 or 3
60 or 30 min, respectively.
36. Gendreau, S. B.; Venture, R.; Keast, P.; Laird, A. D.; Yakes, F. M.; Zhang, W.;
Bentzien, F.; Cancilla, B.; Lutman, J.; Chu, F.; Jackman, L.; Shi, Y.; Yu, P.; Wang,
J.; Aftab, D. T.; Jaeger, C. T.; Meyer, S. M.; De Costa, A.; Engell, K.; Chen, J.;
Martini, J.; Joly, A. H. Clin. Cancer Res. 2007, 13, 3713.
37. Qian, F. E. S.; Yamaguchi, K.; Yu, P.; Won, K.; Mock, L.; Lou, T.; Tan, J.; Li, C.; Tam,
D.; Lougheed, J.; Yakes, F. M.; Bentzien, F.; Xu, W.; Zaks, T.; Wooster, R.;
Greshock, J.; Joly, A. H. Cancer Res. 2009, 69, 8009.
l
M ATP for
Amundson, K.; Chen, J. H.; Rewolinski, D. A.; Yamazaki, S.; Wu, E. Y.; McTigue,
M. A.; Murray, B. W.; Kania, R. S.; O’Connor, P.; Shalinsky, D. R.; Bender, S. L.
Clin. Cancer Res. 2008, 14, 7272.
12. Rugo, H. S.; Herbst, R. S.; Liu, G.; Park, J. W.; Kies, M. S.; Steinfeldt, H. M.;
Pithavala, Y. K.; Reich, S. D.; Freddo, J. L.; Wilding, G. J. Clin. Oncol. 2005, 23,
5474.
13. Sarker, D. Clin. Cancer Res. 2008, 14, 2075.
14. Hilberg, F.; Roth, G. J.; Krssak, M.; Kautschitsch, S.; Sommergruber, W.;
Tontsch-Grunt, U.; Garin-Chesa, P.; Bader, G.; Zoephel, A.; Quant, J.; Heckel, A.;
Rettig, W. J. Cancer Res. 2008, 68, 4774.
15. Wedge, S. R.; Kendrew, J.; Hennequin, L. F.; Valentine, P. J.; Barry, S. T.; Brave, S.
R.; Smith, N. R.; James, N. H.; Dukes, M.; Curwen, J. O.; Chester, R.; Jackson, J. A.;
Boffey, S. J.; Kilburn, L. L.; Barnett, S.; Richmond, G. H. P.; Wadsworth, P. F.;
Walker, M.; Bigley, A. L.; Taylor, S. T.; Cooper, L.; Beck, S.; Jürgensmeier, J. M.;
Ogilvie, D. J. Cancer Res. 2005, 65, 4389.
38. FGF-stimulated ERK phosphorylation ELISA: SK-N-MC (ATCC) cells were seeded
at 4 ꢃ 104cells/well onto 96-well microtiter plates (Costar 3904), in MEME
(Cellgro) containing 10% FBS (heat-inactivated, hyclone) and 1% penicillin–
streptomycin (Cellgro). The cells were then incubated for 48 h at 37 °C in a
humidified, 5% CO2 atmosphere. DMSO or serial dilutions of test compounds in
fresh serum-free medium were added to the cells and incubated for 1 h prior to
recombinant human fibroblast growth factor basic FGF157 stimulation (25 ng/
ml, R&D Systems, 234-FSE-025/CF) for 30 min. Negative control wells were not
treated with growth factor. After treatment, medium was removed and cells
were fixed followed by quenching of endogenous peroxidases with 0.6% H2O2
(Sigma). Fixed cells were then incubated in PBS containing 0.1% Triton X-100
and 10% FBS (heat-inactivated, Hyclone) for 1 h followed by incubation with
anti-phosphorylated p44/42 MAPK, E10 (1:2,000, Cell Signaling, 9106) diluted
in PBS containing 0.1% Triton X-100 and 1% BSA. Cells were then incubated for
1 h in PBS containing 0.1% Triton X-100 and 1% BSA with HRP-conjugated, goat
anti-mouse (1:3,000, Jackson ImmunoResearch) antibody. A luminol-based
substrate solution (Pierce, 37075) was added and measurements were taken in
a Victor Wallac V plate reader (Perkin Elmer). The percentage inhibition of
FGF157-mediated Erk phosphorylation was calculated based on the ratio of the
phosphorylated Erk luminescence measurement for compound-treated and
FGF157-stimulated samples to the phosphorylated Erk luminescence
measurement for the DMSO-treated and FGF157-stimulated samples.
Percentage inhibition values at eight different compound concentrations
were used to calculate IC50 values.
16. Huynh, H.; Ngo, V. C.; Fargnoli, J.; Ayers, M.; Soo, K. C.; Koong, H. N.; Thng, C. H.;
Ong, H. S.; Chung, A.; Chow, P.; Pollock, P.; Byron, S.; Tran, E. Clin. Cancer Res.
2008, 14, 6146.
17. Arnold, D.; Seufferlein, T. Gut 2010, 59, 838.
18. Motzer, R. J.; Michaelson, M. D.; Redman, B. G.; Hudes, G. R.; Wilding, G.; Figlin,
R. A.; Ginsberg, M. S.; Kim, S. T.; Baum, C. M.; DePrimo, S. E.; Li, J. Z.; Bello, C. L.;
Theuer, C. P.; George, D. J.; Rini, B. I. J. Clin. Oncol. 2006, 24, 1.
19. Demetri, G. D.; Van Oosterom, A. T.; Garrett, C. R.; Blackstein, M. E.; Shah, M. H.;
Verweij, J.; McArthur, G.; Judson, I. R.; Heinrich, M. C.; Morgan, J. A.; Desai, J.;
Fletcher, C. D.; George, S.; Bello, C. L.; Huang, X.; Baum, C. M.; Casali, P. G. Lancet
2006, 368, 1329.
20. Sun, L.; Tran, N.; Liang, C.; Tang, F.; Rice, A.; Schreck, R.; Waltz, K.; Shawver, L.
K.; McMahon, G.; Tang, C. J. Med. Chem. 1999, 42, 5120.
21. Laird, A. D.; Vajikoczy, P.; Shawver, L. K.; Thurnher, A.; Liang, C.; Mohammadi,
M.; Schlessinger, J.; Ullrich, A.; Hubbard, S. R.; Blake, R. A.; Fong, T. A. T.;
Strawn, L. M.; Sun, L.; Tang, C.; Hawtin, R.; Tang, F.; Shenoy, N.; Hirth, K. P.;
McMahon, G.; Cherrington, J. M. Cancer Res. 2000, 60, 4152.
22. Sun, L.; Tran, N.; Liang, C.; Hubbard, S.; Tang, F.; Lipson, K.; Schreck, R.; Zhou,
Y.; McMahon, G.; Tang, C. J. Med. Chem. 2000, 43, 2655.
39. VEGF-stimulated ERK phosphorylation ELISA: Primary human umbilical vein
endothelial cells (HUVEC, ATCC) were seeded at 2 ꢃ 104cells/well onto 96-well
microtiter plates (Costar 3904), in EBM-2 (Clonetics) containing supplements
(EGM-2 BulletKit, Clonetics). The cells were then incubated for 24 h at 37 °C in
a humidified, 5% CO2 atmosphere. Growth medium was replaced with serum-
free EBM-2 and cells were incubated for an additional 24 h. DMSO or serial
dilutions of test compounds in fresh serum-free medium were added to the
cells and incubated for 1 h prior to recombinant human vascular endothelial
growth factor VEGF165 stimulation (20 ng/mL, R&D Systems, 293-VE-010) for 5
minutes. Negative control wells were not treated with growth factor. After
treatment, medium was removed and cells were fixed followed by quenching
of endogenous peroxidases with 0.6% H2O2 (Sigma). Fixed cells were then
incubated in PBS containing 0.1% Triton X-100 and 10% FBS (heat-inactivated,
Hyclone) for 1 h followed by incubation with anti-phosphorylated p44/42
MAPK, E10 (1:2,000, Cell Signaling, 9106) diluted in PBS containing 0.1% Triton
X-100 and 1% BSA. Cells were then incubated for 1 h in PBS containing 0.1%
Triton X-100 and 1% BSA with HRP-conjugated, goat anti-mouse (1:3000,
23. Khanwelkar, R. R.; Chen, G. S.; Wang, H. C.; Yu, C. W.; Huang, C. H.; Lee, O.;
Chen, C. H.; Hwang, C. S.; Ko, C. H.; Chou, N. T.; Lin, M. W.; Wang, L. M.; Chen, Y.
C.; Hseu, H.; Chang, C. N.; Hsu, H. C.; Lin, H. C.; Shih, Y. C.; Chou, S. H.; Tseng, H.
W.; Liu, C. P.; Tu, C. M.; Hu, T. L.; Tsai, Y. J.; Chern, J. W. Bioorg. Med. Chem. Lett.
2010, 18, 4674.
24. Tripathy, R.; Reiboldt, A.; Messina, P. A.; Iqbal, M.; Singh, J.; Bacon, E. R.;
Angeles, T. S.; Yang, S. X.; Albom, M. S.; Robinson, C.; Chang, H.; Ruggeri, B. A.;
Mallamo, J. P. Bioorg. Med. Chem. Lett. 2006, 16, 2158.
25. Tripathy, R.; Ghose, A.; Singh, J.; Bacon, E. R.; Angeles, T. S.; Yang, S. X.; Albom,
M. S.; Aimone, L. D.; Herman, J. L.; Mallamo, J. P. Bioorg. Med. Chem. Lett. 2007,
17, 1793.
26. Han, S. Y.; Choi, J. W.; Yang, J.; Chae, C. H.; Lee, J.; Jung, H.; Lee, K. H.; Ha, J. D.;
Kim, H. R.; Cho, S. Y. Bioorg. Med. Chem. Lett. 2012, 22, 2837.
27. Prakash, C. R.; Theivendren, P.; Raja, S. Pharmacol. Pharm. 2012, 3, 62. and
references within for Sutent, SU5416, and SU6668.
28. Basal levels of KDR phosphorylation are relatively low in mouse lung tissue.
Consequently, VEGF was administered i.v. 30 min prior to termination of the
experiment to stimulate phosphorylation of KDR.
29. Khristich, B. I.; Simonov, A. M.; Shepelenko, E. N.; Yatsimirskii, V. A. Chem.
Heterocycl. Compd. 1983, 19, 87.
Jackson immuno research) antibody. A luminol-based substrate solution
(Pierce, 37075) was added and measurements were taken in a Victor Wallac
V plate reader (Perkin Elmer). The percentage inhibition of VEGF165-mediated
Erk phosphorylation was calculated based on the ratio of the phosphorylated
Erk luminescence measurement for compound-treated and VEGF165
-
30. Sun, L.; Tran, N.; Tang, F.; App, H.; Hirth, P.; McMahon, G.; Tang, C. J. Med. Chem.
1998, 41, 2588.
31. Detailed experimentals for this manuscript are found in the following patent:
Bannen, L. C.; Brown, S. D.; Cheng, W.; Co, E.W.; Nuss, J. M.; Kim, M. H.; Klein, R.
R.; Le, D. T.; Lew, A.; Mac, M. B.; Parks, J. J.; Wen, Z.; Xu, W. WO 2004050681 A2
2004.
32. Tang, P. C.; Sun, L.; McMahon, G.; Shawver, L. K.; Hirth, K. P. WO 9850356 A1
1998.
33. Guan, H.; Laird, A. D.; Blake, R. A.; Tang, C.; Liang, C. Bioorg. Med. Chem. Lett.
2004, 14, 187.
34. Walsh, D. A.; Moran, H. W.; Shamblee, D. A.; Uwaydah, I. M.; Welstead, W. J.;
Sancilio, L. F.; Dannenburg, W. N. J. Med. Chem. 1984, 27, 1379.
35. Kinase Inhibition Assays Kinase activities of KDR and PDGFRa were measured as
the percent of ATP consumed following the kinase reaction using luciferase-
luciferin-coupled chemiluminescence as described previously.36 Briefly, kinase
reactions were initiated by combining test compound, ATP, kinases and
substrates in a 20 mL volume using 384-well microtiter plates. For KDR, the
final reaction mixture contained 3 mM ATP, 1.6 mM poly(Glu, Tyr) 4:1 (Sigma,
St. Louis, MO) and 1.5 nM KDR of residues D807-V1356 with an N-terminal GST
tag (ProQinase, Germany). For PDGFRa, the final reaction mixture contained
2 mM ATP, 10 mM MBP (Sigma) and 14 nM PDGFRa of residues Q551-L1089
with an N-terminal GST tag (ProQinase). The reaction mixture was incubated at
room temperature for 4 h (KDR) or 2 h (PDGFRa) before a 20 mL aliquot of
Kinase Glo (Promega, Madison, WI) was added and luminescence signal was
measured using a Victor2 plate reader (Perkin Elmer). Total ATP consumption
was limited below 50%.
stimulated samples to the phosphorylated Erk luminescence measurement
for the DMSO-treated and VEGF165-stimulated samples. Percentage inhibition
values at eight different compound concentrations were used to calculate IC50
values.
40. Solid tumor efficacy studies: Female nu/nu mice (Taconic, Hudson, NY) were
housed according to the Exelixis Institutional Animal Care and Use Committee
guidelines. On day 0, cells were inoculated intradermally into the hind flank
(2 ꢃ 106 cells, HT29) or subcutaneously into a noncleared mammary fat pad
(1 ꢃ 106 cells, MBA-MB-231). When tumors reached ꢀ100 mg (12-14 days
after implantation), mice were randomized (n = 10/group) and treated orally
once daily for 14 or 28 days with test article or saline vehicle. Body weights
were collected daily, and tumor weights were collected twice weekly. Percent
tumor growth inhibition values were expressed as follows: (1-[(mean treated
tumor weight on the final day ꢁ mean tumor weight on day 0)/(mean vehicle
tumor weight on the final day ꢁ mean tumor weight on day 0)] ꢃ 100).
Statistical analysis of 8h-treated tumors versus vehicle-treated tumors was
done by one-way analysis of variance (ANOVA) with significance defined as
P <0.05.
41. Spectral data for (Z)-3-((4-chlorophenyl)(1H-imidazol-2-yl)methylene)-5-(1-
ethylpiperidin-4-ylamino)indolin-2-one (8h).
1H NMR (400 MHz, d6-DMSO): d 10.92 (s, 1H), 7.57 (d, 2H), 7.54 (s, 1H), 7.30 (d,
2H), 7.16 (s, 1H), 6.59 (d, 1H), 6.43 (d, 1H), 4.92 (s, 1H), 4.87 (d, 1H), 2.75 (d,
2H), 2.34 (q, 2H), 2.29 (m, 1H), 1.94 (t, 2H), 1.56 (d, 2H), 1.11 (m, 2H), 1.00 (t,
3H); Analytical HPLC, retention time = 6.69 min, 99% purity (conditions: YMC
C18 150 ꢃ 4.6 column, gradient 5–95% MeCN/H2O, in the presence of 0.1% TFA,
25 min run at 1.5 mL/min flow rate, k = 254 nM); MS(EI) for C25H26ClN5O:
448.2 (MH+); Elemental analysis, Calcd for C25H26ClN5O: C, 60.89; H, 6.35; N,
14.37; Cl, 7.13. Found: C, 61.08, H, 5.86, N, 13.96, Cl, 7.73.
Activities of Flt1 and FGFR1 were measured using AlphaScreen as described
previously.37 Briefly, 0.13 nM N-terminal GST-tagged cytoplasmic Flt1
(ProQinase), or 0.09 nM N-terminal GST-tagged cytoplasmic FGFR1 was