Identification of aminopyrazolopyridine ureas as potent VEGFR/PDGFR multitargeted kinase inhibitors

Article abstract of DOI:10.1016/j.bmcl.2007.10.018

Tumor angiogenesis is mediated by KDR and other VEGFR and PDGFR kinases. Their inhibition presents an attractive approach for developing anticancer therapeutics. Here, we report a series of aminopyrazolopyridine ureas as potent VEGFR/PDGFR multitargeted k

Full text of DOI:10.1016/j.bmcl.2007.10.018

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry Letters 18 (2008) 386–390
Identification of aminopyrazolopyridine ureas as potent
VEGFR/PDGFR multitargeted kinase inhibitors
Yujia Dai,* Kresna Hartandi, Niru B. Soni, Lori J. Pease, David R. Reuter,
Amanda M. Olson, Donald J. Osterling, Stella Z. Doktor, Daniel H. Albert,
Jennifer J. Bouska, Keith B. Glaser, Patrick A. Marcotte, Kent D. Stewart,
Steven K. Davidsen and Michael R. Michaelides
Cancer Research, Global Pharmaceutical Research and Development, Abbott Laboratories, 100 Abbott Park Road,
Abbott Park, IL 60064-6100, USA
Received 30 August 2007; revised 5 October 2007; accepted 8 October 2007
Available online 17 October 2007
Abstract—Tumor angiogenesis is mediated by KDR and other VEGFR and PDGFR kinases. Their inhibition presents an attractive
approach for developing anticancer therapeutics. Here, we report a series of aminopyrazolopyridine ureas as potent VEGFR/
PDGFR multitargeted kinase inhibitors. A number of compounds have been identified to be orally bioavailable and efficacious
in the mouse edema model.
Ó 2007 Elsevier Ltd. All rights reserved.
Angiogenesis, the formation of new vessels from pre-
existing vasculature, is a required process in tumor
growth and metastasis.1 Angiogenesis is primarily med-
iated by vascular endothelial growth factor receptor
(VEGFR) tyrosine kinases, a subfamily of receptor tyro-
sine kinases (RTKs), which include Fms-like tyrosine ki-
nase 1 (FLT1; VEGFR1), kinase insert domain-
containing receptor (KDR; VEGFR2), and FLT4
(VEGFR3).2 Aberrant activation of VEGFR family
RTKs and especially of KDR by vascular endothelial
growth factors (VEGFs) has been linked to the progres-
sion of various human cancers.3
tively, with the progression of acute myeloid leukemia
(AML)5 and gastrointestinal stromal tumor (GIST).6
Due to the vital roles of VEGFR/PDGFR signaling in
tumor angiogenesis, inhibition of VEGFR/PDGFR
activation has become a compelling approach for devel-
oping targeted cancer therapies. Both selective KDR
agents and multitargeted kinase inhibitors have been
developed.7
In our efforts to identify potent and novel small-mole-
cule kinase inhibitors, we discovered that 3-aminoindaz-
ole could serve as a novel kinase hinge-binding template
Although VEGFR kinases play the major role in tumor
angiogenesis, platelet derived growth factor receptor
(PDGFR) tyrosine kinases, a structurally related RTK
subfamily containing PDGFRa, PDGFRb, cKit, col-
ony-stimulating factor 1 receptor (CSF1R), and FLT3,
are believed to indirectly promote tumor angiogenesis
and also directly contribute to tumor growth through
various mechanisms.4 Additionally, it is also well docu-
mented that the constitutive activation of FLT3 and
cKIT through mutation is directly associated, respec-
F
O
O
HN
N
HN
N
H
H
H2N
N
H2N
N
4
3
N
H
N
H
2 (ABT-869)
1
KDR: IC50 = 4 nM
Keywords: Pyrazolopyridines; KDR; VEGFR/PDGFR kinase
inhibitors.
KDR: IC50 = 64 nM
*
Figure 1. Aminoindazole urea VEGFR/PDGFR kinase inhibitors.
Corresponding author. E-mail: yujia.dai@abbott.com
0960-894X/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2007.10.018

Y. Dai et al. / Bioorg. Med. Chem. Lett. 18 (2008) 386–390
387
for kinase inhibitors. By incorporating a diaryl urea unit
Cl
Cl
Cl
NC
Cl
at the C4-position, a series of potent and orally bioavail-
able VEGFR/PDGFR inhibitors were generated
(Fig. 1). Optimization of this series led to the identifica-
tion of a clinical candidate (2, ABT-869)8,9 that is cur-
rently in phase II trials. Encouraged by the promising
in vitro and in vivo properties of this series of com-
pounds, we expanded our investigation into 3-aminopy-
razolopyridine systems. We were, on one hand,
interested in how the incorporated pyridine nitrogen
would impact the activity of these inhibitors; on the
other hand, we reasoned that the increased basicity
would likely make these compounds more soluble than
their indazole analogs, which exhibited very poor aque-
ous solubility. In this paper, we report in brief the syn-
thesis, structure–activity relationships (SARs), and
preliminary characterization of these new inhibitors.
d
H2NOC
N
a, b, c
N
N
Cl
Cl
17
16
15
NH2
NH2
e
NC
NC
Cl
N
Cl
N
19
18
f
O
O
R
R
HN
N
H
HN
N
H
The chemical synthesis of the 3-aminopyrazolo[3,4-c]pyr-
idine ureas 11a–e and 12a–d is outlined in Scheme 1. The
Suzuki coupling reaction between chloride 6 and 4-ami-
nophenyl boronic acid pinacol ester (7) afforded aniline
8. The aniline was then heated with hydrazine monohy-
drate to form 3-aminopyrazolo[3,4-c]pyridine 9. Simi-
larly, N-methyl 3-aminopyrazolo[3,4-c]pyridine 10 was
obtained by using methylhydrazine instead of hydrazine.
Reaction of 9 and 10 with various isocyanate generated
ureas 11a–e and 12a–d, respectively.
NC
Cl
NC
Cl
N
O
N
20
21
g
O
R
R
HN
N
HN
N
H
H
H2N
H2N
N
N
N
As shown in Scheme 2, 3-aminopyrazolo[3,4-b]pyridine
ureas 13a–n and 3-aminopyrazolo[4,3-c]pyridine ureas
14a–e were synthesized utilizing a strategy similar to
N
H
N
H
N
13a-n
14a-e
Scheme 2. Reagents and conditions: (a) LDA, THF, ꢀ78 °C, 1 h; then
ground dry ice, ꢀ78 °C ! rt, 70%; (b) oxalyl chloride, cat. DMF, rt,
3 h; (c) concentrated aqueous NH4OH, THF, 100%; (d) POCl3,
100 °C, 4 h, 54%; (e) 2, Pd(dppf)Cl2, Na2CO3, DME, H2O; (f) RNCO,
CH2Cl2, 50 °C; (g) NH2NH2–H2O, n-BuOH, 110 °C, 5 h.
Cl
Cl
Cl
H2NOC
Cl
b, c
HO2C
Cl
a
N
N
N
Cl
5
4
3
that used in the preparation of 3-aminopyrazolo[3,4-
c]pyridine analogs. The Suzuki coupling between chlo-
ride 17 and 7 yielded a mixture of 18 and 19 in a ratio
of 2:1. Attempts to separate them using flash column
chromatography were unsuccessful. Consequently, the
mixture was directly used for the subsequent urea for-
mation and cyclization. The final products 13a–n and
14a–e were then separated using reverse phase HPLC.
NH2
O
B
NH2
Cl
O
NC
Cl
d
7
NC
N
e
N
Cl
O
6
8
R
NH2
HN
N
Expected to be ATP-competitive KDR inhibitors, the
synthesized 3-aminopyrazolopyridine ureas were evalu-
ated in the presence of a high concentration (1.0 mM)
ATP.10 We first looked at the 3-aminopyrazolo[3,4-
b]pyridine ureas and the results are shown in the upper
portion of Table 1. These compounds proved to be
highly potent KDR inhibitors and compared favorably
to their aminoindazole counterparts. For instance, both
13a and 13j are, respectively, more potent than 1 and 2
(Fig. 1). Introduction of a range of mono- or bis-substit-
uents on the urea terminal phenyl is generally well
tolerated. In fact, all the substituted ureas of the 3-
aminopyrazolo[3,4-b]pyridine except o-Me analog 13b
in Table 1 showed an IC50 value of <7 nM. The o-Me
analog 13b (KDR IC50 = 35 nM) is 7-fold less potent
H
H2N
N
g
H2N
N
f
N
N
N
R'
N
R'
11a-e R' = H
12a-d R' = CH3
9 R' = H
10 R' = CH3
Scheme 1. Reagents and conditions: (a) LDA, THF, ꢀ78 °C, 1 h; then
ground dry ice, ꢀ78 °C ! rt, 56%; (b) oxalyl chloride, cat. DMF, rt,
3h; (c) concentrated aqueous NH4OH, THF, 86%; (d) POCl3, 100 °C,
4h, 87%; (e) Pd(dppf)Cl2, Na2CO3, DME, H2O, 35%; (f) NH2NH2–
H2O or CH3NHNH2, n-BuOH, 110 °C, 5 h; (g) RNCO, DMF.

388
Y. Dai et al. / Bioorg. Med. Chem. Lett. 18 (2008) 386–390
Table 1. Inhibition of KDR in vitro and mouse uterine edema in vivo
O
R
HN
N
H
H2N
3
5
Z
Y
N
1
N
R'
X
7
a
KDR IC50 (nM)
a
KDR (cell) IC50 (nM)
Compound
R0
X
Y
Z
R
H
UEb % inhibition
13a
13b
13c
13d
13e
13f
13g
13h
13i
H
N
CH
CH
CH
CH
CH
CH
CH
CH
CH
CH
CH
CH
CH
CH
N
CH
CH
CH
CH
CH
CH
CH
CH
CH
CH
CH
CH
CH
CH
CH
CH
CH
CH
CH
CH
CH
CH
CH
N
4.5
35
4.0
2.2
1.0
16
39
<5
H
N
o-Me
m-Me
p-Me
m-Cl
H
N
2.0
2.9
1.0
1.7
3.8
6.3
1.0
2.0
2.4
1.1
2.0
1.8
0.7
1.2
2.1
1.0
0.7
4.6
1.5
7.7
3.7
12
H
N
<5
23
H
N
1.5
2.6
3.9
8.1
1.1
1.0
7.9
1.2
3.8
5.1
0.6
3.1
1.0
1.3
0.7
1.8
8.8
3.0
13
H
N
m-CF3
H
N
m-OMe
3,5-di-F
3,5-di-Me
2-F-5-Me
2-F-5-CF3
4-F-3-Me
4-F-3-CF3
3-Cl-4-F
m-Me
H
N
H
N
77
66
13j
H
N
13k
13l
H
N
H
N
108
13m
13n
11a
11b
11c
11d
11e
12a
12b
12c
12d
14a
14b
14c
14d
14e
H
N
H
N
55
50
<5
98
35
19
79
52
72
59
H
CH
CH
CH
CH
CH
CH
CH
CH
CH
CH
CH
CH
CH
CH
H
N
m-Cl
H
N
2-F-5-Me
2-F-5-CF3
4-F-3-Me
m-Me
H
N
H
N
Me
Me
Me
Me
H
N
N
m-Cl
N
2-F-5-Me
4-F-5-CF3
m-Cl
N
CH
CH
CH
CH
CH
46
99
H
N
m-CF3
7.1
18
<5
<5
<5
<5
H
N
2-F-5-Me
3,5-di-Me
4-F-3-Me
61
65
H
N
12
20
H
N
34
a Each IC50 determination was performed with seven concentrations, and each assay point was determined in duplicate.
b Dosed orally at 10 mg/kg.
than 13a and >10-fold less potent than its meta- and
para-analogs (13c and 13d). The negative impact on
KDR potency by introduction of a substituent other
than a fluorine at the ortho-position is consistent with
the SAR observed in the aminoindazole series.
to their corresponding analogs of the 3-aminopyrazol-
o[3,4-b]pyridine both enzymatically and cellularly.
Additionally, introduction of a methyl group on the pyr-
azole ring NH is well tolerated. Compounds 12a–d all
exhibited a single-digit nanomolar IC50 value against
KDR.
The potent inhibitory activity of these 3-aminopyrazol-
o[3,4-b]pyridine ureas against KDR is well reflected at
the cellular level. Table 1 also displays the activity of
these compounds in the inhibition of VEGF-induced
KDR phosphorylation in 3T3 murine fibroblast cells
that were engineered to express human KDR.11 Except
13d, all the 3-aminopyrazolo[3,4-b]pyridine ureas in Ta-
ble 1 exhibited a single-digit nanomolar IC50 value
against cellular KDR phosphorylation.
While almost all the tested diaryl ureas of both the
pyrazolo[3,4-c]pyridine and the pyrazolo[3,4-b]pyridine
displayed low nanomolar potency against KDR and
compared well to the indazole analogs, the corre-
sponding analogs of pyrazolo[4,3-c]pyridine are gener-
ally less active. For instance, 14c is about 9-fold less
potent than both 13j and 11c, and about 4-fold less
potent than 2.
Moving the pyridine nitrogen from the 7-position to the
6-position (see the numbering on the structure in Table
1) has little impact on the KDR affinity. All the ureas of
the 3-aminopyrazolo[3,4-c]pyridine system compare well
To interpret the difference in potency observed for the
regioisomeric pyrazolopyridines, we looked into the
binding mode of these compounds to KDR. Figure 2
is a model of 14a bound to KDR generated by molecu-

Y. Dai et al. / Bioorg. Med. Chem. Lett. 18 (2008) 386–390
Table 2. SARs of replacement of the urea terminal phenyl group
389
O
R
HN
N
H
H2N
N
N
H
N
Compound
R
KDR
a
IC50 (nM)
KDR (cell)
a
IC50 (nM)
13a
22
4.5
12.7
115
64
4.0
37
Figure 2. Binding model of 14a in an inactive form of KDR (DFG-
out). Two hinge H-bonds to Glu 917 and Cys 919 and urea H-bond to
Glu 885 are shown as dotted lines.
S
lar modeling.12 As this model suggests, these 3-aminopy-
razolopyridine ureas bind to the ATP-site of KDR in
the same fashion as 3-aminoindazole ureas.8 Specifi-
cally, the 3-amino pyrazolopyridine core is anchored
to the KDR hinge region via two H-bonds. The kinase
adopts an inactive confirmation (DFG-out) so that the
urea portion occupies the back hydrophobic pocket with
additional H-bond interactions. This model also indi-
cates that the pyrazole ring NH does not form a direct
H-bond interaction with the kinase, which is confirmed
by the observation that the N-methylated compounds
are still very potent against KDR as discussed before.
23
24
25
240
a Each IC50 determination was performed with seven concentrations,
and each assay point was determined in duplicate.
activity of KDR inhibitors. The percent inhibition of
edema at a 10 mg/kg oral dose is given in Table 1.
A number of compounds were found to be orally effi-
cacious in this model. For instance, compounds 13i,
13j, 13l, 13n, 11c, and 12a–d exhibited >50% inhibi-
tion at the 10 mg/kg dose. Interestingly, all orally
potent compounds are ureas of the pyrazolo[3,4-b]pyr-
idines and pyrazolo[3,4-c]pyridines; the analogs of the
pyrazolo[4,3-c]pyridine did not show measurable oral
activity at the same dose. Although the pyrazolo[4,3-
c]pyridines exhibited inferior enzymatic and cellular
KDR inhibitory activity in comparison with their ana-
logs of pyrazolo[3,4-b]pyridines and pyrazolo[3,4-
c]pyridines, their lack of activity in the UE model is
likely due to their poor oral systemic exposure.
Several compounds were evaluated for their mouse
pharmacokinetic properties (Table 3). 14d, a pyrazol-
Based on this model, the 7- and 6-positions of pyrazol-
opyridine cores project toward solvent accessible region.
Consequently, insertion of a hydrophilic pyridine nitro-
gen at either of these two positions should be favored.
This is consistent with the improved potency displayed
by both the pyrazolo[3,4-b]pyridines and pyrazolo[3,4-
c]pyridines relative to the indazoles. However, a hydro-
philic nitrogen at the 5-position would have a negative
impact on potency because the 5-position projects to a
hydrophobic region comprised of Phe 1047 of the
DFG-motif. This prediction is indeed in line with the
weaker activity observed for the pyrazolo[4,3-c]pyridine
sub-series.
We also briefly investigated the replacement of the
urea terminal aromatic residue with an aliphatic group
(Table 2). Consistent with the SARs in other series of
urea class KDR inhibitors,8,12 a diaryl urea moiety is
favored for optimal interaction with the hydrophobic
back pocket of KDR kinase.
Table 3. Mouse pharmacokinetic profiles of 12a, 12c, 13j, and 14d
Compound
iva
pob
t1/2
Vd
Cl
(ml/min kg)
AUC
(lM h)
F
Given their impressive enzymatic and cellular KDR
inhibitory activity, these compounds were subse-
quently evaluated in an estradiol-induced mouse uter-
ine edema (UE) model, which is based on the finding
that the enhanced vascular permeability is a direct re-
sponse to VEGF-stimulated KDR signaling. Since the
tested compounds were given orally, this in vivo func-
tional assay enabled a rapid screening of the oral
(h)
(L/kg)
(%)
12a
12c
13j
0.6
1.5
1.0
0.45
0.66
1.48
1.50
1.71
12.7
11.5
17.3
43.7
25.0
30.2
15.7
0.89
71
81
62
9
14d
a Dosed at 3 mg/kg.
b Dosed at 10 mg/kg.

390
Y. Dai et al. / Bioorg. Med. Chem. Lett. 18 (2008) 386–390
Table 4. Kinase inhibitory profiles of 12a and 13f
Compound
KDR
FLT1
FLT3
cKit
CSF1R
FYN
SRC
a
IC50 (nM)
12a
13f
4.6
4
7
5
4
2
7
7
2
3
>12,500
11,800
>12,500
—
a Each IC50 determination was performed with seven concentrations, and each assay point was determined in duplicate.
o[4,3-c]pyridine analog displaying little oral UE activ-
ity at 10 mg/kg, is in fact barely orally available with
an AUC value of less than 1 lM h. In contrast, com-
pounds 12a, 12c, and 13j, which were all potent in the
UE model, possess significantly better oral PK profiles
(AUCs > 15.7 lM h and Fs > 60%).
References and notes
1. (a) Carmeliet, P.; Jain, R. K. Nature 2000, 407, 249; (b)
Zetter, B. R. Annu. Rev. Med. 1998, 49, 407.
2. (a) Ullrich, A.; Schlessinger, J. Cell 1990, 61, 203; (b)
Hubbard, S. R.; Till, J. H. Annu. Rev. Biochem. 2000, 69,
373.
3. Bergland, E. K. Am. J. Health-Syst. Pharm. 2004,
61(Suppl. 5), S4.
Although the discussion so far has been primarily focused
upon the inhibition of KDR kinase, these compounds
proved to be potent VEGFR/PDGFR multitargeted
RTK inhibitors. To demonstrate this, Table 4 gives the ki-
nase inhibitory profiles of 12a and 13f. Clearly, 12a and
13f not only possess impressive activity against KDR,
but also potently inhibit other kinases of VEGFR family
as well as the kinases of PDGFR family. However, they
were much less active in the inhibition of other structur-
ally non-related kinases such as FYN and SRC.
4. (a) Skobe, M.; Fusenig, N. E. Proc. Natl. Acad. Sci. USA
1998, 95, 1050; (b) Lindahl, P.; Johansson, B. R.; Leveen,
P.; Betsholtz, C. Science 1997, 277, 242; (c) Lin, E. Y.;
Nguyen, A. V.; Russell, R. G.; Pollard, J. W. J. Exp. Med.
2001, 193, 727.
5. Stirewalt, D. L.; Radich, J. P. Nat. Rev. Cancer 2003, 3,
650.
6. Demetri, G. D. Semin. Oncol. 2001, 28, 19.
7. (a) Ferrara, N.; Kerbel, R. S. Nature 2005, 438, 967; (b)
Bilodeau, M. T.; Fraley, M. E.; Hartman, G. D. Expert
Opin. Investig. Drugs 2002, 11, 737; (c) Bergsland, E. K.
Am. J. Health-Syst. Pharm. 2004, 61(Suppl. 5), S12.
8. Dai, Y.; Hartandi, K.; Ji, Z.; Ahmed, A. A.; Albert, D. H.;
Bauch, J. L.; Bouska, J. J.; Bousquet, P. F.; Cunha, G. A.;
Glaser, K. B.; Harris, C. M.; Hickman, D.; Guo, J.; Li, J.;
Marcotte, P. A.; Marsh, K. C.; Moskey, M. D.; Martin, R. L.;
Olson, A. M.; Ostering, D. J.; Pease, L. J.; Soni, N. B.; Stewart,
K. D.; Stoll, V. S.; Tapang, P.; Reuter, D. R.; Davidsen, S. K.;
Michaelides, M. R. J. Med. Chem. 2007, 50, 1584.
9. Albert, D. H.; Tapang, P.; Magoc, T. J.; Pease, L. J.;
Reuter, D. R.; Wei, R.-Q.; Li, J.; Guo, J.; Bousquet, P. F.;
Ghoreish-Haack, N. S.; Wang, B.; Bukofzer, G. T.; Wang,
Y.-C.; Stavropoulos, J. A.; Hartandi, K.; Niquette, A. L.;
Soni, N.; Johnson, E. F.; McCall, J. O.; Bouska, J. J.; Luo,
Y.; Donawho, C. K.; Dai, Y.; Marcotte, P. A.; Glaser, K.
B.; Michaelides, M. R.; Davidsen, S. K. Mol. Cancer Ther.
2006, 5, 995.
10. Kinase enzymatic assays were performed utilizing the
homogeneous time-resolved fluorescence (HTRF) proto-
col in the presence of 1.0 mM ATP. For the detailed
procedure, see Ref. 8.
11. For the procedure of the KDR cellular assay, see Ref. 8.
12. Dai, Y.; Guo, Y.; Frey, R. R.; Ji, Z.; Curtin, M. L.;
Ahmed, A. A.; Albert, D. H.; Arnold, L.; Arries, S. S.;
Barlozzari, T.; Bauch, J. L.; Bouska, J. J.; Bousquet, P. F.;
Cunha, G. A.; Glaser, K. B.; Guo, J.; Li, J.; Marcotte, P.
A.; Marsh, K. C.; Mosley, M. D.; Pease, L. J.; Stewart, K.
D.; Stoll, V. S.; Tapang, P.; Wishart, N.; Davidsen, S. K.;
Michaelides, M. J. Med. Chem. 2005, 48, 6066.
A brief study on the aqueous solubility of selected com-
pounds was also conducted. The measured solubility for
11c, 13e, and 14c at pH 7.2 was 5.8 lM, 1.6 lM, and
9.5 lM, respectively. At the same pH, the measured sol-
ubility for ABT-869 (2) was 0.07 lM. The incorporated
pyridine nitrogen led to an improved aqueous solubility
for these compounds.
In summary, we have identified a series of potent mult-
itargeted RTK inhibitors by expanding our investigation
of the aminoindazole series into the aminopyrazolopyri-
dines. The diaryl ureas of the pyrazolo[3,4-b]pyridine
and pyrazolo[3,4-c]pyridine potently inhibited KDR
and other RTKs of the VEGFR family as well as the
kinases of the PDGFR family. Their potent activity
against KDR is also reflected by their inhibition of
VEGF-induced cellular KDR phosphorylation. Further
evaluation of these compounds in a VEGF-induced
mouse edema model resulted in the identification of a
number of orally active compounds. Mouse pharmaco-
kinetic studies of 12a, 12c, and 13j demonstrated that
some of these compounds possess reasonable pharmaco-
kinetic profiles and are promising as oral agents. More-
over, a brief investigation of selected compounds
revealed that these inhibitors possess improved aqueous
solubility in comparison to their aminoindazole analogs.