Design and SAR of thienopyrimidine and thienopyridine inhibitors of VEGFR-2 kinase activity

Article abstract of DOI:10.1016/j.bmcl.2003.10.030

Novel classes of thienopyrimidines and thienopyridines have been identified as potent inhibitors of VEGFR-2 kinase. The synthesis and SAR of these compounds is presented, along with successful efforts to diminish EGFR activity present in the lead series.

Full text of DOI:10.1016/j.bmcl.2003.10.030

Bioorganic & Medicinal Chemistry Letters 14 (2004) 21–24
Design and SAR of thienopyrimidine and thienopyridine inhibitors
of VEGFR-2 kinase activity
Michael J. Munchhof,* Jean S. Beebe, Jeffery M. Casavant, Beth A. Cooper,
Jonathan L. Doty, R. Carla Higdon, Stephen M. Hillerman, Catherine I. Soderstrom,
Elisabeth A. Knauth, Matthew A. Marx, Ann Marie K. Rossi, Susan B. Sobolov
and Jianmin Sun
Pfizer Global Research and Development, Groton Laboratories MS8220-2405, Eastern Point Road, Groton, CT 06340, USA
Received 19 August 2003; revised 14 October 2003; accepted 15 October 2003
Abstract—Novel classes of thienopyrimidines and thienopyridines have been identified as potent inhibitors of VEGFR-2 kinase.
The synthesis and SAR of these compounds is presented, alongwith successful efforts to diminish EGFR activity present in the lead
series.
#
2003 Elsevier Ltd. All rights reserved.
Angiogenesis is a critical component of human tumor
growth and metastasis, and the VEGF receptor-2
In order to explore the series potential, 7-chloro-2-
bromo-theino[3,2-b]pyrimidine (5, Scheme 1) was pre-
pared by modification of a literature procedure. This
6
(
nent of the signaling pathway responsible for the sprout-
KDR, VEGFR-2) has been identified as a key compo-
compound could be regioselectively functionalized at
the 2 or 7- positions. Heatingwith amines in t-BuOH/
DCE resulted in displacement of the chloride, whereas
palladium-catalyzed couplings with boronic acids or
organotin reagents, occurred exclusively at the 2-posi-
tion. The 2-position could also be functionalized by Li–
halogen exchange and quenching with an appropriate
1
,2
ingand maturation of new blood vessels from tumors.
Additionally, disruption of VEGF signaling by a number
of different approaches has resulted in inhibition of
3
,4
angiogenesis and tumor growth. Moreover, with the
exceptions of the female reproductive system and wound
healing, angiogenesis is a relatively quiescent process,
makingit a very attractive tar ge t for therapeutic inter-
vention. Our goal was to discover a small-molecule
7
electrophile.
5
inhibitor of VEGF receptor kinase signaling. In this
We began by exploring the indole portion of 1 utilizing
medium speed chemistry techniques (Table 1, examples
8–18). Profilingof substituted phenyl and heterocyclic
report, we describe the structure–activity relationships
of a novel series of thienopyrimidines and thienopyr-
idines, which are potent VEGF kinase inhibitors.
Thienopyrimidine 1 (Fig. 1) was identified from high-
throughput screening. Compound 1 possessed promis-
ingpotency and was structurally attractive in that it
contained three motifs that could be easily modified: the
indole, thienopyrimidine and phenyl portions. A key
driver for early chemistry efforts was to achieve selec-
tivity versus epidermal growth factor receptor kinase
(EGFR), since 1 was 26-fold more potent against this
target.
*
0
Correspondingauthor. Tel.:+1-860-441-6189; fax:+1-860-686-1176;
e-mail: michael_j_munchhof@groton.pfizer.com
Figure 1. Thienopyrimidine lead.
960-894X/$ - see front matter # 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2003.10.030

22
M. J. Munchhof et al. / Bioorg. Med. Chem. Lett. 14 (2004) 21–24
Table 1. Thienopyrimidine SAR
Compd
R
1
R
2
VEGFR kinase
EGFR kinase
IC50 mM
IC50 mM
1
8
9
5-Indolyl
H
CH Ph
4-PhCONH
4-PhNHPh
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
H
Br
0.08
>1.25
>2.50
0.25
0.39
0.19
0.5
0.62
0.47
>2.50
0.14
0.29
1.41
0.22
0.77
2.21
2
0.13
0.056
0.043
0.036
0.025
0.022
0.018
0.003
ND
ND
>1.00
ND
ND
0.094
1.6
ND
ND
0.1
0.005
0.6
ND
0.009
ND
ND
0.001
0.001
0.001
0.001
0.001
0.005
0.004
2
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
4-PhNHMe
4-PhNMe
2
5-Indazolyl
5-Benztriazolyl
1-Methyl-5-indolyl
2-Methyl-5-indolyl
3-Methyl-5-indolyl
5-Indolyl
5-Indolyl
5-Indolyl
5-Indolyl
5-Indolyl
5-Indolyl
5-Indolyl
5-Indolyl
5-Indolyl
COPh
C(CH)OHPh
CONHnPr
2-Thiophene
4-PhCH
4-PhNH(CH -N-pyrrolidine
4-Ph(CH N-1-imidazole
4-PhCH NHC(CH CH OH
4-Ph(CH NMe
4-PhCH NH
2
OH
2 2
)
2 3
)
5-Indolyl
5-Indolyl
5-Indolyl
2
3
)
2
2
2
)
2
2
2
2
Scheme 1. Synthesis of 7-chloro-2-bromo-thieno[3,2-b]pyrimidine.
replacements revealed that a hydrogen bond donor or
acceptor, in the region occupied by the indole nitrogen,
was required. Consistent with this, alkylation of the
indole nitrogen abolished activity. Attempts to sub-
stitute the indole ringwere also met with limited suc-
cess. Methyl groups at the 2- or 3-positions were
tolerated, but larger substituents (data not shown)
reduced potency considerably.
Furthermore, indole and its methyl substituted ana-
logues were the only analogues that maintained potency
in cellular assays (IC ’s<1 mM). In light of the tight
SAR and lack of cellular potency, no suitable replace-
ments for the indole were identified in these early efforts.
5
0
In contrast, modifications at the 7-position of the thie-
nopyrimidine were better tolerated. While replacement

M. J. Munchhof et al. / Bioorg. Med. Chem. Lett. 14 (2004) 21–24
23
of the phenyl group with a variety of functional groups
lead to modest losses in potency (Table 1, examples 19–
24), exploration of substitution on the phenyl ringpro-
vided an excitingfindin .g The para position of the
phenyl ringwas amenable to introduction of a diverse
array of functionality, both in terms of size and physical
properties. Presumably this portion of the pharmacophore
extends out of the bindingsite into a solvent exposed
region (Table 1, examples 25–30). These modifications led
to a 4-fold increase in potency, and allowed the introduc-
tion of a basic amine to enhance physical properties.
the quinazoline N-3 nitrogen resulted in a 10-fold decrease
in EGFR potency. We were interested in exploringif a
similar modification of the thienopyrimidine to the corre-
spondingthienop yridine would improve EGFR selectivity,
either by further increasing potency against our target, or
by decreasingpotency a ga inst EGFR.
The thienopyridine was accessible through the same
amino carboxylic ester used in the thienopyrimidine
6
synthesis (Scheme 2). Saponification followed by
decarboxylation of methyl 3-amino-thiophene-2-car-
boxylate provided 3-amino thiophene. Condensation
with Mander’s reagent, and subsequent heating in
Although we had achieved moderate success in increas-
ingpotency, little improvement in EGFR selectivity was
observed. As can be seen in Table 1, the compounds
were still 5- to 60-fold more potent against EGFR than
VEGFR. In an attempt to address this issue, we turned
our attention to the thienopyrimidine core of these
molecules. SAR observations in our EGFR kinase pro-
gram with quinazoline leads revealed that removal of
TM
Dowtherm,
afforded pyridone 32. Treatment with
POCl , followed by lithiation and treatment with iodine
3
yielded multigram quantities of 7-chloro-2-iodo-[3,2-b]
thienopyridine (33). Intermediate 33 could be regiose-
lectively functionalized at the 2- or 7- positions. Heating
with amines in t-BuOH/DCE resulted in displacement
of the chloride, whereas palladium-catalyzed couplings
Scheme 2. Synthesis of 7-chloro-2-iodo-thieno[3,2-b]pyridine.
Table 2. Thienopyrimidine versus thienopyridine
Compd
R
2
X=C
Compd
X=N
VEGFR kinase EGFR kinase Ratio
IC50, mM IC50, mM VEGFR/ EGFR
VEGFR kinase EGFR kinase Ratio
IC50, mM IC50, mM VEGFR/ EGFR
1
Ph
2-Thiophene
0.08
0.003
0.001
0.002
0.001
0.001
0.001
26
132
12
17
36
39
40
41
42
43
44
0.048
0.055
0.004
0.002
0.004
0.003
0.068
0.039
0.005
0.007
0.005
0.008
1.4
0.7
1.3
3.5
1.3
2.7
36
37
38
27
28
0.132
0.024
0.017
0.036
0.025
4-PhCH
4-PhCH N-pipiperdine-4-CONH
4-PhCH NH (CH N-1-imidazole
4-PhCH NHC(CH CH OH
2 2 2
NH(CH ) NHAc
2
2
2
)
2 3
2
3
)
2
2
25

24
M. J. Munchhof et al. / Bioorg. Med. Chem. Lett. 14 (2004) 21–24
Table 3. Methyl versus hydrogen
Compd
R
1
2
R =H
Compd
R
2
=Me
VEGFR kinase EGFR kinase
IC50, mM
Ratio
EGFR/VEGFR
VEGFR kinase EGFR kinase
IC50, mM
Ratio
EGFR/VEGFR
IC50, mM
IC50, mM
45
46
47
48
49
50
CH
CH NH(CH
CH NH(CH
NH(CH O(CH
CH NH(CH OH
CH -N-morpholine
2
NHCH
3
(CH
-N-piperazine
OH
2
)
2
OH
0.004
0.007
0.004
0.004
0.006
0.005
0.003
0.003
0.009
0.004
0.002
0.03
0.8
2.3
3.3
1
0.3
5.6
51
52
53
54
55
56
0.003
0.003
0.005
0.005
0.004
0.005
0.14
0.25
0.5
0.657
1
1.25
47
83
100
135
250
250
2
2
)
2
2
2
)
3
CH
2
2
)
2
2
)
2
OH
2
2
)
2
2
with boronic acids, organotin, and organozinc reagents
occurred exclusively at the 2-position.
In summary, a new series of potent VEGFR kinase
inhibitors has been developed. The synthetic routes to
access these are brief and scalable, allowingfor rapid
analogue preparation. Interestingly, two minor modifi-
cations—addinga methyl gr oup and removinga nitro-
gen atom—have inverted the selectivity of the series
between EGFR and VEGFR.
The resultingthienopyridines did have improved selec-
tivity versus EGFr (Table 2). This was a result of both
increased potency against VEGFR, and diminished
activity against EGFR. However, even with this
improvement, most compounds were still equipotent
against both targets.
References and notes
Compound 17 (Table 1) suggested that the presence of a
-methyl substituent on the indole could further dimin-
2
1. Fan, T. P. D.; Jaggar, R.; Bicknell, R. Trends Pharmacol.
Sci. 1995, 16, 57.
ish EGFR potency. In the thienopyrimidine series this
was accompanied with an approximately 2-fold decrease
in VEGFR potency as well. Much to our surprise and
delight, use of the 2-methylindole group in the thieno-
pyridine series decreased EGFR activity more than 100-
fold, while havinglittle effect on VEGFR activity
2. Folkman, J. Nature Medicine 1995, 1, 27.
3. Asano, M.; Yukita, A.; Matsumoto, T.; Kondo, S.;
Suzuki, H. Cancer Res. 1995, 55, 5296.
4
5
6
. Millauer, B.; Shawver, L. K.; Plate, K. H.; Risau, W.;
Ulrich, A. Nature 1994, 367, 576.
. For a review of small molecule VEGFR-2 inhibitors see:
Connell, R. D. Ex. Opin. Therap. Pat. 2002, 12, 1763.
. Robba, M., et al. Bul. Soc. Chim. Fr. 1970, 3630.
(Table 3). The resultingcompounds were 50- to 250-
fold more selective for VEGFR, demonstratingthat the
effect of the methyl group on EGFR potency is ampli-
fied by removal of the thienopyrimidine nitrogen.
7. For full experimental details see Munchhof, M. J.;
Sobolov, S. B.; Marx, M. A. U.S. Patent 6,492,383.