Potent quinoxaline-based inhibitors of PDGF receptor tyrosine kinase activity. Part 2: The synthesis and biological activities of RPR127963 an orally bioavailable inhibitor

Article abstract of DOI:10.1016/S0960-894X(03)00655-3

RPR127963 demonstrates an excellent pharmacokinetic profile in several species and was found to be efficacious in the prevention of restenosis in a Yucatan mini-pig model upon oral administration of 1-5 mg/kg. The in vitro selectivity profile and SAR of t

Full text of DOI:10.1016/S0960-894X(03)00655-3

Bioorganic & Medicinal Chemistry Letters 13 (2003) 3097–3100
Potent Quinoxaline-Based Inhibitors of PDGF Receptor
Tyrosine Kinase Activity.Part 2: The Synthesis and Biological
Activities of RPR127963 an Orally Bioavailable Inhibitor§
y
y
y
{
Wei He,* Michael R. Myers, Barbara Hanney, Alfred P. Spada, Glenda Bilder,
{
{
{
x
{
Helen Galzcinski, Dilip Amin, Saul Needle, Ken Page, Zaid Jayyosi
y
and Mark H. Perrone
Aventis Pharmaceuticals, Route 202/206, Brigdewater, NJ 08807, USA
Received 14 April 2003; accepted 20 May 2003
Abstract—RPR127963 demonstrates an excellent pharmacokinetic profile in several species and was found to be efficacious in the
prevention of restenosis in a Yucatan mini-pig model upon oral administration of 1–5 mg/kg. The in vitro selectivity profile and
SAR of the highly optimized PDGF-R tyrosine kinase inhibitor are highlighted.
#
2003 Elsevier Ltd. All rights reserved.
Overexpression of PDGF and PDGF-R have been
observed in patients with several diseases such as rest-
enosis, certain types of cancers, atherosclerosis, lung
fibrosis, kidney fibrosis, liver cirrosis, palmar fibrosis as
kinase inhibitors. In this paper we highlight the activity
and selectivity profile of RPR127963 that has been
shown to have an excellent pharmacokinetic profile in
several species. RPR127963 successfully demonstrated
efficacy in a Yucatan mini-pig model of restenosis and
was ultimately was selected for advanced preclinical
studies.
2
5
well as rheumatoid arthritis. Therefore, PDGF recep-
tor tyrosine kinase inhibitors may provide new treat-
ments of many chronic diseases. For example, the recent
approval of Gleevec (CGP57148B or STI-571) for the
treatment of GI cancers is partially due to its activity
against PDGF and c-kit receptor tyrosine kinases, in
addition to its original indication for CML related to its
As elucidated in the preceeding paper, aniline analogues
such as 1 were about 100-fold more potent than
RPR101511A in both the ELISA and Mitogenesis
assays. Unfortunately, these analogues suffered a similar
pharmacokinetic profile to RPR101511 in that they
induced p450 enzymes (in rat) which led over the course of
several days dosing to greatly diminished oral exposure.
3
activity against abl tyrosine kinase. The challenge for
us was to identify orally bioavailable compounds with
sustained plasma level of compounds upon chronic oral
4
administrations. In the preceeding paper we disclosed
two novel series of hetereoatom linked quinoxaline and
quinoline derivatives as potent PDGF receptor tyrosine
*Corresponding author at present address: Drug Discovery, Johnson
&
&
Johnson Pharmaceutical Research & Development, L.L.C., Welsh
McKean Roads, Spring House, PA 19477, USA. E-mail: whe@
prdus.jnj.com
§
This work was presented in part of the conferences shown in ref 1.
Address at time of study: Department of Medicinal Chemistry, Col-
y
legeville, PA, USA.
{
Address at time of study: Department of Cardiovascular Biology,
Collegeville, PA, USA.
x
Address at time of study: Department of Metabolism and Pharma-
cokinetics, Dagenham, UK.
Address at time of study: Department of Drug Safety, Collegeville,
PA, USA.
k
However, compound 2 less potent than 1 did not induce
upregulation and more importantly was observed with
0
960-894X/$ - see front matter # 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/S0960-894X(03)00655-3

3098
W. He et al. / Bioorg. Med. Chem. Lett. 13 (2003) 3097–3100
Table 1. PDGF-R Activity
Compd
R
PDGF-R (IC50, mM)
RPR101511
1
0.461
0.006
Scheme 1. Synthesis of RPR127963.
2
0.124
0.076
1
0
1
1
4
5
14% @ 10 mM
0.041
a
1
1
1
1
2
2
2
6a
ꢁ9% @ 1mM
15% @ 1mM
0.049
7a
Scheme 2. Synthesis of functionalized quinoxaline analogues.
the same exposure levels in plasma between days 1 and 4
upon oral administration once daily over a period of 4
days, albeit with low exposure as shown in Table 2.
With these positive results and the fact that 2 was only
8a
9
0.028
4
-fold more potent than RPR101511 led us to explore
further the SAR of substituted cycloalkyl derivatives.
Our designing strategies were to incorporate functional
groups Y for exploring the potential improvement of
physicochemical properties of the inhibitors as well as
the enhancement of potency.
0
0.345
1a
0.93
The general methods of preparation for these com-
pounds were exemplified in Scheme 1. Compounds were
typically prepared by heating a>5-fold excess of the
2a
3.1
appropriate amines with the 2-chloroquinoxaline for 3–
ꢀ
1
2 h at 180 C. For the synthesis of RPR127963 (com-
pound 10) this method proved to be fairly robust and all
the starting materials were readily available and inex-
pensive. RPR127963 has been prepared at a very large
scale (kg).
23a
10% @ 1mM
0.068
2
4
5
However, the thermal coupling reaction failed with
cyclohexylamines that are substituted with functional
groups such as acids or esters (Scheme 2). This problem
was circumvented by designing an alternative approach.
Thus, 4-amino-cyclohexanyl carboxylic acid was
2
5% @ 1mM
0.018
6
cyclized first to give the corresponding lactam. The
sodium salt of such lactam, 13, coupled smoothly in
THF with 2-chloro-6,7-dimethoxyl-quinoxaline to a
versatile intermediate 14. Treatment of intermediate 14
with various nucleophiles provided aminoquinoxaline
derivatives 12 that possess acid, ester, amide functional
groups (17, 18, 21, 22, and 23). Ester 18 was separated
into cis-isomer 19 and trans-isomer 20. Compounds 25,
26
2
7
0.024
aMixture of cis- and trans-isomers.

W. He et al. / Bioorg. Med. Chem. Lett. 13 (2003) 3097–3100
3099
2
alcohols.
6 and 27 were made from the corresponding amino
7
The cellular functional activity of a certain number of
analogues is presented in Table 3. The most potent
compounds in our ELISA assay tended to be the most
active in the mitogenesis assay. Compound 25 demon-
strated only very weak activity in our PDGF-R assay
and was not significantly active in the mitogenesis assay.
The biological data for these analogues are included in
Table 1. The replacement of the methyl group of com-
pound 15 with a basic amino group dramatically
decreased the inhibitory activity against PDGF-R tyr-
osine kinase (16 vs 15). For unknown reasons, a basic
group is not tolerated at this region. This is a general
trend with other compounds (not shown) that also pos-
sess basic functional groups at the 3- or 4-position of the
cyclohexyl ring. The substitution at the 4-position with
a carboxylic acid group also resulted in the loss of
activity (17 vs 2) while the methyl ester substitution at
the same position slightly increased the potency (18 vs
The selectivity profile and functional activities of
RPR127963 are presented in Tables 4 and 5. In most
cases we evaluated direct enzyme inhibition. However,
for some targets (i.e., HGF) functional activity was
measured. The only significant activity discovered for
RPR127963 beyond PDGF-R was for EGF-R and
lck
p56 . From our experience it is anticipated that the low
micromolar level of in vitro activity against these
particular kinases would not translate into significant
functional activity. We also measured the inhibition of
PDGF-BB or PDGF-AA stimulation of mitogenesis
and inhibition of PDGF-BB stimulated chemotaxis and
collagen production. Nearly identical IC s of about 300
2
). Although the cis-isomer 19 was 10-fold more active
than the trans-isomer 20 in the ELISA assay, neither
these esters (18, 19 and 20) nor the acid (17) showed any
significant activity in the cellular PDGF-induced mito-
genesis assay. Such results were not really surprising.
These esters were probably hydrolyzed to the acid by cel-
lular esterases under the functional assay conditions.
Unfortunately, more stable amide analogues 21, 22 and
5
0
nM were found. These results coupled with the lack of
anti-mitogenic activity of compounds like 25 (Table 3)
support our contention that RPR127963 is a very selec-
tive pharmacological tool.
23 are much less active against PDGF-R tyrosine kinase
with micromolar potency at best.
As a consequence of the excellent in vitro and in-cell
profile as well as an excellent ADME profile in rat we
chose to evaluate the PK profile of RPR127963 further.
Its superior PK profiles were observed in many species
including dogs and mini-pigs. RPR127963 was found to
have a linear pharmacokinetic profile in several PK
studies in Yucatan min-pigs with an oral bioavailability
of 100% and a very long half-life. Subsequently,
RPR127963 was evaluated in the Yucatan mini-pig
model of restenosis and it was found to be orally active
in two separate trials (PTCA and implanted stent) at
doses of 1–5 mg/kg/day for 28 days as determined by
Our breakthrough was achieved with hydroxylation of
the saturated rings. Interestingly, we discovered that a
hydroxy group substitution to the cyclohexyl and nor-
bonyl rings maintains the potency (10, 24, 26 and 27 vs
2
). trans-Isomer 10 and cis-isomer 24 were equipotent in
the ELISA assay. More importantly, compound 10 was
observed to have a greatly improved PK profiles in a rat
PK study as shown in Table 2. Gratifyingly, oral
administration of the most potent isomers in both the
substituted cyclohexyl and norbonyl series (26, and 27)
in rats resulted in an overall exposure that was sig-
nificantly improved relative to compound 10. Ulti-
mately, RPR127963 was selected for evaluation in our
Yucatan mini-pig model of restenosis despite the
improved in vitro potency and in vivo ADME proper-
ties of 26 and 27 due to the relative ease of preparing the
large quantities of material required for a chronic study.
Table 4. Selectivity profiles of RPR127963
Kinase
IC50 or I% (N) (mM)
PDGFr
EGFr
0.076ꢂ0.013 (7)
3.7
CSF-1r
VEGF
2% @ 30
0% @ 30 (3)
33ꢂ5 (4)
ꢃ10 (3)
Insulin r (glucose transporter)
IGF (mitogenesis)
SYK
ZAP70
JAK2
JAK3
JNK
LCK
HGFr (chemotaxis)
PKA
PKC
Table 2. Bioavailability and p450 Induction
54% @ 10
0% @ 10
>50
Compd
Day 1
(
Day 4
(%)
P450 upregulation
(in vitro, rat)
%)
>50
>50
1.4ꢂ0.1 (3)
17% @ 30
3% @ 100
7% @ 100
RPR101511
2
49
17
82
21
16
72
Yes
No
No
10
Table 3. Comparison of inhibition of PDGF-R autophosphorylation
and mitogenesis in human aortic smooth muscle cells
Table 5. Additional functional activity of RPR127963
Compd
PDGF-R
HASMC
(IC50, mM)
Ratio activity
mitogenesis/enzyme
(IC50, mM)
Assay
IC50 or I% (N) (mM)
10
25
26
27
0.076
5% @ 1 mM
0.018
0.353
>10.0
0.134
5
—
7
Mitogenesis (PDGF-BB)
Mitogenesis (PDGF-AA)
Chemotaxis
0.353ꢂ0.071 (11)
0.299ꢂ0.055 (3)
0.292ꢂ0.075 (3)
0.389ꢂ0.144 (3)
0.024
0.164
7
Collagen production

3100
W. He et al. / Bioorg. Med. Chem. Lett. 13 (2003) 3097–3100
angiographic, intravascular ultrasound and histological
criteria. The experimental details and results of these
studies will be published elsewhere.
H. Abstracts of Papers, 219th ACS National Meeting, San Fran-
cisco, CA, March 26–30, 2000; MEDI-307. (b) Myers, M. R. The
SCI-Fine Chemicals sponsored conference Protein Tyrosin
Kinases, Good Targets for Drugs?, London, May 22, 2001.
2
. (a) Heldin, C.-H.; Ostman, A.; Ronnstrand, L. Biochim.
In conclusion, highly optimized PDGF-R tyrosine
kinase inhibitors have been identified. The initial ani-
lino-leads suffered from significant issues with bioavail-
ability and induction of p450 enzymes. The optimized
analogues have been shown to have superior profiles
making them useful compounds for further in vivo pro-
filing. The oral activity of compound 10 at a relatively
low dose and a chronic administration in animal models
represent a major improvement over the prototype
compound RPR101511. The results of additional stud-
ies of RPR127963 will be reported in due course.
Biophys. Acta 1998, 1378, F79. (b) Heldin, C.-H.; Westermark,
B. Physiol. Rev. 1999, 79, 1283. (c) Bassett, P. Drug Market
Develop. 2000, 327. (d) Boschelli, D. H. Drugs Future 1999, 24,
5
3
15.
. (a) Buchdunger, E.; Cioffi, C. L.; Law, N.; Stover, D.;
Ohno-Jones, S.; Druker, B. J.; Lydon, N. B. J. Pharm. Exp.
Ther. 2000, 295, 139. (b) Tsao, A. S.; Kantarjian, H.; Talpaz,
M. Brit. J. Haematol. 2002, 119, 15. (c) Dematteo, R. P.;
Heinrich, M. C.; El-Rifai, W. M.; Demetri, G. Human Path.
2
4
M. P.; Zulli, A.; Bilder, G.; Galzcinski, H.; Amin, D.; Needle,
S.; Spada, A. P. Bioorg. Med. Chem. Lett. See preceding
paper. doi:10.1016/S0960-894X(03)00654-1.
002, 33, 466.
. Myers, M. E.; He, W.; Hanney, B.; Setzer, N.; Maguire,
References and Notes
5
. Bilder, G. E.; Wentz, T.; Leadley, R.; Amin, D.; Byan, L.;
O’Conner, B.; Needle, S.; Galczenski, H.; Bostwick, J.;
Kasiewski, C.; Myers, M. R.; Spada, A. P.; Merkel, L.; Ly, C.;
Persons, P.; Page, K.; Perrone, M.; Dunwiddie, C. Circulation
1999, 99, 3292.
6. Pearlman, W. M. Org. Syn. 1969, 49, 75.
7. The synthesis of the corresponding amino alcohols will be
published later.
1
. (a) He, W.; Myers, M. R.; Spada, A. P.; Hanney, B.; Setzer,
N.; Mailliet, P.; Gontier, S.; Guegen, J.-C.; Cans, P.; Orton,
E.; Cheve, M.; Kubiak, G. G.; Shah, H. C.; O’Brien, M. K.;
Bilder, G. E.; Allen, E.; Bissery, M.-C.; Page, K.; Natarajan,
C.; Jayyosi, Z.; Kelley, M. F.; Toutain, H.; Constan, A.;
Amin, D.; Needle, S.; Galczenski, H.; Wang, W.; Perrone, M.