Discovery and optimization of a potent and selective triazolopyridinone series of c-Met inhibitors

Article abstract of DOI:10.1016/j.bmcl.2012.04.072

Deregulation of the receptor tyrosine kinase c-Met has been implicated in several human cancers and is an attractive target for small molecule drug discovery. Herein, we report the discovery of a structurally diverse series of carbon-linked quinoline triazolopyridinones, which demonstrates nanomolar inhibition of c-Met kinase activity. This novel series of inhibitors exhibits favorable pharmacokinetics as well as potent inhibition of HGF-mediated c-Met phosphorylation in a mouse liver pharmacodynamic model.

Full text of DOI:10.1016/j.bmcl.2012.04.072

Bioorganic & Medicinal Chemistry Letters 22 (2012) 4089–4093
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Discovery and optimization of a potent and selective triazolopyridinone
series of c-Met inhibitors
Christiane M. Bode a, , Alessandro A. Boezio a, , Brian K. Albrecht a, Steven F. Bellon a, Loren Berry a,
Martin A. Broome b, Deborah Choquette a, Isabelle Dussault b, Richard T. Lewis a, Min-Hwa Jasmine Lin a,
Karen Rex b, Douglas A. Whittington a, Yajing Yang b, Jean-Christophe Harmange a
⇑
⇑
a Amgen Inc., 360 Binney St., Cambridge, MA 02142, USA
b Amgen Inc., One Amgen Center Drive, Thousand Oaks, CA 91320, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Deregulation of the receptor tyrosine kinase c-Met has been implicated in several human cancers and is
an attractive target for small molecule drug discovery. Herein, we report the discovery of a structurally
diverse series of carbon-linked quinoline triazolopyridinones, which demonstrates nanomolar inhibition
of c-Met kinase activity. This novel series of inhibitors exhibits favorable pharmacokinetics as well as
potent inhibition of HGF-mediated c-Met phosphorylation in a mouse liver pharmacodynamic model.
Ó 2012 Elsevier Ltd. All rights reserved.
Received 3 February 2012
Revised 12 April 2012
Accepted 16 April 2012
Available online 25 April 2012
Keywords:
Tyrosine kinase
c-Met
Triazolopyridinones
The receptor tyrosine kinase c-Met and its endogenous ligand,
hepatocyte growth factor (HGF), are implicated in several cellular
processes relevant to cancer, including cell proliferation, cell
migration, and invasive growth. They also play an important role
in embryonic development and wound healing.1 However, deregu-
lation of the c-Met/HGF pathway can lead to tumorigenesis and
metastasis.2 Amplification of the MET gene, overexpression of c-
Met and/or HGF, and constitutive activation conferred by sequence
mutations are some of the mechanisms of deregulation which have
been found in human cancers.3
describe the design, synthesis and improved in vivo efficacy for this
class of compounds.
Based on the evolution of the O- and N-linked triazolopyrid-
azine series (1 and 2), we hypothesized that intramolecular hydro-
gen bonds between the oxygen or nitrogen tether and the
quinoline or naphthyridine moiety were important for maintaining
a favorable binding conformation, and hence, enzyme and cellular
potencies. Analysis of the co-crystal structure of 2 with c-Met
showed that the inhibitor adopts a U-shaped conformation envel-
oping Met1211, and the 4-aminonaphthyridine makes key hydro-
There are several approaches to inhibiting the HGF/c-Met path-
way currently being tested in the clinic.4 The appeal of an ATP-
competitive, small molecule inhibitor acting via the intracellular
kinase domain is its potential to block both ligand-dependent
and ligand independent activity of c-Met.5 We previously reported
the discovery of a potent, orally active c-Met inhibitor for the treat-
ment of cancer,6 and more recently disclosed two series of oxygen-
and nitrogen-linked triazolopyridazines (1 and 2) as potent and
selective c-Met inhibitors (Fig. 1).7,8 However, these compounds
displayed poor overall solubility and exhibited short duration of
coverage in our pharmacodynamic model. In efforts to improve
these limitations, further exploration of these structural chemo-
types led to the discovery of a triazolopyridinone series of c-Met
inhibitors (3) that are highly potent and selective. Herein, we
O
O
7
N
1
N
N
3
N
F
H
O
5
F
F
O
H N
N
S
N
N
N
N
N
N
5'
N
N
6'
6'
7'
N
N
2"
7'
1'
N
N
1
2
3
cMET IC50: 4 nM
PC3 cell IC50: 8 nM
Solubilitya:
cMET IC50: 6 nM
cMET IC50: 612 nM
PC3 cell IC50: ND
Solubilitya:
PC3 cell IC50: 3 nM
Solubilitya:
85.1/3.5/16.5
≥200/3.4/12.7
≥200/≥200/60.4
a0.01N HCl/PBS/SIF (μg/mL)
⇑
Corresponding authors. Tel.: +1 617 444 5104; fax: +1 617 621 3907.
E-mail address: aboezio@amgen.com (A.A. Boezio).
Figure 1. Core modifications that lead to the triazolopyridinone chemotype.9
0960-894X/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmcl.2012.04.072

4090
C. M. Bode et al. / Bioorg. Med. Chem. Lett. 22 (2012) 4089–4093
and 2, and therefore provided a good starting point for lead
optimization.
A co-crystal structure of 3 within the catalytic domain of
unphosphorylated c-Met was obtained (overlaid with
2 in
Fig. 2),13 indicating that the inhibitor adopted the desired U-
shaped binding mode around Met1211. A favorable intramolecular
interaction was also observed between the carbonyl oxygen and
the C-5 hydrogen of the quinoline (3.2 Å). The quinoline nitrogen
atom participates in a single-point hydrogen bond with linker res-
idue Met1160, while N-10 of the dihydro-triazolopyridinone inter-
acts with the backbone –NH of Asp 1222. In addition, a face-to-face
p-stacking interaction is observed between the triazolopyridinone
and Tyr1230.
Having identified a structurally unique scaffold, our focus
shifted to improving the c-Met potency of 3. The SAR developed
in our previously reported triazolopyridazine series indicated that
variation of the C-60 substituent of 1 and 2 significantly impacted
potency in both enzymatic and cellular assays. Accordingly, a syn-
thetic route was developed that enabled rapid analoging at the cor-
responding N-50 position of 3 (Scheme 1). HATU amide coupling
between the alkynyl acid 4 and the desired N-50 amine provided
Figure 2. Crystal structure overlay of c-Met + 2 (green) and 3 (pink).
intermediate 5. Dipolar cycloaddition of
5 with 6-(azido-
gen bond interactions with linker residue Met1160 (Fig. 2, green
structure). In the design of a new scaffold, we hypothesized that
the 4-aminonaphthyridine of compound 2 could be replaced with
a carbon-linked quinoline (3), as the quinoline nitrogen should re-
side in approximately the same position for interaction with
Met1160. However, this change could disrupt the U-shaped confor-
mation enforced by the intramolecular hydrogen bonds present in
1 and 2. In order to maintain that structural feature in the new
scaffold and to introduce non-planar elements that could improve
solubility,10 the triazolopyridazine core was modified to a dihydro-
triazolopyridinone. This placement of atoms may allow the car-
bonyl to favorably interact with the proton at the C-5 position of
the quinoline,11 and the saturation across C-60 and C-70 could dis-
rupt the packing of the crystal lattice. Based on this hypothesis,
the parent compound 3 (Fig. 1) with a phenyl group in the N-50 po-
sition was synthesized12 and although a loss of enzyme potency
was observed, this new core exhibited greater solubility than 1
methyl)quinoline (6) led to a 3:2 product ratio as observed by
LC/MS, favoring the desired regioisomer 7. After removing the silyl
group, Mitsunobu cyclization yielded the substituted dihydro-tria-
zolopyridinones 10.14
Consistent with SAR established in the O- and N-linked triazo-
lopyridazine series, a dramatic increase in enzyme potency was ob-
served upon replacement of the phenyl group of 3 with 3-methyl
isothiazole (11, 612 nM to 21 nM), and a cellular potency of
62 nM was also achieved. A positive interaction between the sulfur
of the isothiazole and the oxygen from the dihydro-triazolopyridi-
none15 may favor a coplanar orientation of the N-50 substituent,
resulting in optimal binding to c-Met.
In the case of compound 2, a two-point interaction was ob-
served between the hydrogens appended at C-70 and C-2’’ and
the lone pairs of the carbonyl of Arg1208, as well as a face-to-face
p-stacking interaction with the triazolopyridazine core and
Tyr1230 (Fig. 2). The saturation in the dihydro-triazolopyridinone
core at the C-60 and C-70 positions, incorporated to improve solubil-
ity, precluded 3 and 11 from fully engaging these interactions.
Introducing unsaturation at this position to afford triazolopyridi-
none 18 increased the enzymatic and cellular potencies to 3 and
5 nM, respectively.16
The unsaturation of the triazolopyridinone core was introduced
at a later stage in the synthesis, utilizing the same common inter-
mediate leading to either the saturated or unsaturated molecule.
The synthesis began with cycloaddition of the azidomethyl-quino-
line 12 with ethyl-5-hydroxypent-2-ynoate (13) yielding the regio-
isomeric triazoles 14 and 15 (3:2). Compound 14 was then
converted to the desired amide 16 (introducing the N-50 substitu-
tion), which upon exposure to Dess-Martin periodinane triggered
an imine/enamine cyclization to yield the desired triazolopyridi-
nones 17 (Scheme 2).
Although the introduction of the methyl isothiazole (compound
11) and the unsaturated core (compound 18) imparted enhanced
potency in comparison to 3, both changes led to a dramatic de-
crease in solubility (11: 43.2/2.2/<1 and 18: 80.1/<1/4.8).17 Further
modifications were made to address this issue. An overlay of the
crystal structures of 2 and 3 (Fig. 2) indicates that substitution with
a methoxy group at the 3-position of the quinoline in compound 3
could occupy the same solvent front as the methoxy group in com-
pound 2. Modification of this moiety could potentially increase sol-
ubility and potency, as was observed in the N-linked
triazolopyridazine series. Addition of the ethoxymethoxy group
in compound 20 (c-Met = 1 nM, PC3 cell = 2 nM) provided a slight
R
HN
HO
O
N
O
R
NH2
+
HATU, DiPEA
DMF, 23°, 2h
N3
OTBS
OTBS
6
4
48%
N
5
O
R
N
N
H
N
N
O
+
TBSO
PhMe, 150°C
R
N
H
N
N
3h, μW
N
46% (7)
TBSO
7
N
8
3:2
N
N
Aq. HCl (6N)
PPh3, DEAD
THF, 23°C, 2h
57%
O
O
R
MeOH, 23°C
2h
N
N
R
N
H
N
N
N
N
N
HO
100%
10
11: R = 3-methyl isothiazole
9
Scheme 1. Synthesis of dihydro-triazolopyridinones.

C. M. Bode et al. / Bioorg. Med. Chem. Lett. 22 (2012) 4089–4093
4091
N
O
R3
EtO
+
O
N
N
N
EtO
HO
R3
N
R2
+
O
PhMe, 150°C
R2
N
N
R2
EtO
HO
3h, μW
N
R3
N3
OH
20% (14)
13
R3
N
12
14
15
3:2
N
O
N
R3
R1 NH2
Dess-Martin
R2
R2
O
R1
DCM, 60°C, 1h
KO-tBu, MeOH
60°C, 1h
R1
N
N
N
N
N
N
N
H
N
100%
11-21%
HO
17
16
18: R1 = 3-methyl isothiazole
R2 = H
R3 = H
Scheme 2. Convergent synthesis of triazolopyridinones.
c-Met = 42 nM, PC3 cell = 29 nM).18 Compound 21 also showed im-
proved solubility compared to 18 (P200/4.5/22.4).
Table 1
Representative triazolopyridinones
To further explore the SAR of compounds 18–21, a variety of
heterocycles19 and fluorinated aromatic rings20 at the R1 position
were synthesized. Table 1 illustrates the biochemical and cellular
potencies of these compounds against c-Met. All of the R1 substit-
uents were tolerated when paired with the chiral (S)-methyl (R2)
and/or the ether groups (R3). A more noticeable improvement in
potency was observed when incorporating an (S)-methyl substitu-
ent at R2 (24 versus 25, and 30 versus 31). Compounds 29 and 30
indicated that the ethoxymethoxy substituent in the R3 position
provided a slightly more cell-potent compound than the smaller
methoxy group. Upon combining the best R2 and R3 substituents,
the fluorinated aromatic compound 32, along with the 3-methyl
isothiazole derivative 23, were highlighted as R1-diverse com-
pounds with superior solubility profiles compared to 1 and 2
(23: 37.0/8.7/39.7; 32: 127.7/11.1/37.7).
The metabolic stability and pharmacokinetic properties of 23
and 32 were evaluated (Table 2). For both compounds, moderate
intrinsic clearance was observed across species. The plasma pro-
tein binding was greater than 95% in rat, mouse and human. Con-
sistent with its turnover in rat liver microsomes, compound 32
showed a slightly higher in vivo clearance than 23, resulting in a
shorter half-life since both the volume of distribution and PPB
were comparable for the compounds. The oral bioavailability of
23 was also higher than that of 32.
N
R3
H
R2
O
R1
N
N
N
N
a
b
Compd. R1
R2
R3
cMET IC50
(nM)
PC3 IC50
(nm)
N
S
18
19
20
21
22
23
H
H
H
H
OMe
3
3
1
2
1
1
5
3
2
3
2
2
S
S
OCH2CH2OMe
H
Me
Me OMe
Me OCH2CH2OMe
24
25
H
OCH2CH2OMe 10
Me OCH2CH2OMe
102
12
2
26
27
28
Me
Me
Me
H
H
H
7
17
39
37
S
N
7
3
N
F
29
30
31
H
H
OMe
12
8
110
14
3
Crystallographic analysis of 23 confirmed many of the struc-
tural design aspects we had incorporated throughout the optimiza-
tion process, as well as those previously observed with 3 (Fig. 3).21
The two-point binding with the triazolopyridinone ring and the
OCH2CH2OMe
Me OCH2CH2OMe
1
F
F
32
Me OCH2CH2OMe
2
3
carbonyl of Arg1208 was restored, and more efficient
p-stacking
F
with Tyr1230 was observed. We also confirmed that the (S)-enan-
tiomer of the chiral methyl group was preferred by the protein22
and this methyl substituent fills a small, lipophilic pocket in c-
Met, defined by the side chains of Val1092, Leu1157, Ala 1226,
and Lys1110.
a
nP2, inhibition of kinase activity in a biochemical assay.
nP2, inhibition of HGF-mediated c-Met phosphorylation in PC3 cells.
b
improvement in potency; however, the solubility was not greatly
affected. (13.7/2.5/9.8).
Before assessing the in vivo pharmacological activity of our new
series, compound 23 was screened for inhibitory activity against a
broad panel of 100 kinases (Ambit screen), and displayed good
overall selectivity against a range of tyrosine and serine/threonine
kinases.23
To demonstrate that compound 23 inhibits c-Met activity
in vivo over time, a pharmacodynamic assay measuring changes
in HGF-induced phosphorylation of c-Met in the mouse liver was
Additionally, a methyl group was introduced at the methylene
bridge in order to improve solubility by integrating another non-
planar element as well as to maintain/increase potency by favoring
the desired rotameric disposition of the two ring systems. We
observed that the S enantiomer exhibited better potency than its
R counterpart (21: (S) c-Met = 2 nM, PC3 cell = 3 nM versus (R)

4092
C. M. Bode et al. / Bioorg. Med. Chem. Lett. 22 (2012) 4089–4093
Table 2
c-Met phosphorylation was inhibited at least 89% through 24 h
with an associated plasma concentration of less than 0.2
M,25
In vitro ADME and in vivo PK properties of selected compounds
l
a
much-improved result upon comparison to compounds 1 and 2,
both of which showed inhibition >90% only up to 6 h, with a larger
dose of 30 mg/kg.7,8
23
32
RLM Cla
MLM Cla
HLM Cla
(
(
(
l
l
l
L/min/mg)
L/min/mg)
L/min/mg)
89
68
84
98.2
97.3
98.8
105
98
120
98.0
95.6
95.5
In conclusion, through the use of structure-based design, a no-
vel series of triazolopyridinone c-Met inhibitors was developed.
The inhibitors, which incorporate key design elements present in
compounds 1 and 2, potently block the kinase activity of c-Met
in both biochemical and cellular assays at a single-digit nanomo-
lar level. Compound 23, which displayed a modest solubility
improvement in relation to the progenitor series, exhibits an im-
proved pharmacodynamic profile, significantly inhibiting HGF-
mediated c-Met phosphorylation for up to 24 h in a mouse liver
PD model.
PPBb (%) Rat
Mouse
Human
Rat phamacokineticsc
Cld (L/h/kg)
0.559
2.53
3.40
1560
43
1.13
2.38
1.81
638
36
d
Vss (L/Kg)
d
T1/2 (h)
e
AUC0?1 (ng h/mL)
Fe (%)
a
In vitro (RLM = rat liver microsomes; MLM = mouse liver microsomes;
HLM = human liver microsomes).
b
Separation method = equilibrium dialysis.
c
Acknowledgments
In vivo experiments with male Sprague–Dawley rats (n = 3).
d
IV, 0.25 mg/kg (DMSO).
e
PO, 2 mg/kg (2% HPMC, 1% Tween 80 in H2O, pH 2.2 w/MSA.
Special thanks to both Roman Shimanovich and Kavita Shah for
in vivo formulation support and to Earl Moore for analytical sup-
port of the PK and PD samples.
References and notes
1. Giordano, S.; Ponzetto, C.; Di Renzo, M. F.; Cooper, C. S.; Comoglio, P. M. Nature
1989, 339, 155.
2. Birchmeier, C.; Birchmeier, W.; Gherardi, E.; Vande Woude, G. F. Nat. Rev. Mol.
Cell Biol. 2003, 4, 915.
3. Dharmawardana, P. G.; Giubellino, A.; Bottaro, D. P. Curr. Mol. Med. 2004, 4,
855.
4. Small molecules: (a) Cui, J. J. Expert Opin. Ther. Patents 2007, 17, 1035;
Biologicals: (b) Cao, B.; Su, Y.; Oskarsson, M.; Zhao, P.; Kort, E. J.; Fisher, R. J.;
Wang, L. M.; Vande Woude, G. F. Proc. Natl. Acad. Sci. U.S.A. 2001, 98, 7443.
5. Porter, John Expert Opin. Ther. Patents 2010, 20, 159.
6. Liu, L.; Siegmund, A.; Xi, N.; Kaplan-Lefko, P.; Rex, K.; Chen, A.; Lin, J.;
Moriguchi, J.; Berry, L.; Huang, L.; Teffera, Y.; Yang, Y.; Zhang, Y.; Bellon, S. F.;
Lee, M.; Shimanovich, R.; Bak, A.; Dominguez, C.; Norman, M. H.; Harmange, J.-
C.; Dussault, I.; Kim, T.-S. J. Med. Chem. 2008, 51, 3688.
7. Albrecht, B. K.; Harmange, J.-C.; Bauer, D.; Berry, L.; Bode, C.; Boezio, A. A.;
Chen, A.; Choquette, D.; Dussault, I.; Fridrich, C.; Hirai, S.; Hoffman, D.; Larrow,
J. F.; Kaplan-Lefko, P.; Lin, J.; Lohman, J.; Long, A. M.; Moriguchi, J.; O’Connor, A.;
Potashman, M. H.; Reese, M.; Rex, K.; Siegmund, A.; Shah, K.; Shimanovich, R.;
Springer, S. K.; Teffera, Y.; Yang, Y.; Zhang, Y.; Bellon, S. F. J. Med. Chem. 2008,
51, 2879.
8. Boezio, A. A.; Berry, L.; Albrecht, B. K.; Bauer, D.; Bellon, S. F.; Bode, C. M.; Chen,
A.; Choquette, D.; Dussault, I.; Hirai, S.; Kaplan-Lefko, P.; Larrow, J. F.; Lin, J.;
Lohman, J.; Potashman, M. H.; Rex, K.; Santostefano, M.; Shah, K.; Shimanovich,
R.; Springer, S. K.; Teffera, Y.; Yang, Y.; Zhang, Y.; Harmange, J.-C. Bioorg. Med.
Chem. Lett. 2009, 19, 6307.
Figure 3. Crystal structure of c-Met + 23.
9. The hydrogen bond framework in 3, between the proton at C-5 and the
carbonyl, was not included for
compound.
a clearer structural representation of the
10. Ishikawa, M.; Hashimoto, Y. J. Med. Chem. 2011, 54, 1539.
11. Pierce, A. C.; Sandretto, K. L.; Bemis, G. W. Proteins Struct. Funct. Gen. 2002, 49,
567.
12. Phenyl substitution at the C-6 position was tolerated, but not optimal, in the O-
and N-linked series. The phenyl analog was synthesized first, based on
synthetic precedent: Pinto, D. J. P.; Orwat, M. J.; Quan, M. L.; Han, Q.; Galemmo,
R. A., Jr.; Amparo, E.; Wells, B.; Ellis, C.; He, M. Y.; Alexander, R. S.; Rossi, K. A.;
Smallwood, A.; Wong, P. C.; Luettgen, J. M.; Rendina, A. R.; Knabb, R. M.;
Mersinger, L.; Kettner, C.; Bai, S.; He, K.; Wexler, R. R.; Lam, P. Y. S. Bioorg. Med.
Chem. Lett. 2006, 16, 4141.
13. PDB deposition codes for the crystal structures of c-Met + 2 and c-Met + 3 are
4DEG and 4DEH, respectively.
14. Albrecht, B. K.; Bellon, S.; Bode, C. M.; Boezio, A.; Choquette, D.; Harmange, J.-C.
PCT Int. Appl. 2009, 109, WO 2009143477.
15. Ángyán, J. G.; Poirier, R. A.; Kucsman, Á.; Csizmadia, I. G. J. Am. Chem. Soc. 1987,
109, 2237.
16. See Figure 3 for example of unsaturated core binding to c-Met.
17. All solubility data reported as follows: 0.01 N HCl/PBS/SIF (lg/mL).
18. See Figure 3 for determination of stereochemical preference and explanation
for enantiomeric preference.
Figure 4. Compound 23 inhibits HGF-mediated c-Met phosphorylation in mouse
liver when dosed at 10 mg/kg.26
19. Sulfur-containing heterocycles were the focus of our efforts due to the
favorable S–O interaction described previously.
20. Fluorinated aromatic substitutions were chosen based on their potencies in the
N-linked triazolopyridazine series.
21. PDB deposition code number for the crystal structure of c-Met + 23 is 4DEI.
22. The absolute stereochemistry was assigned by crystallization of enantiopure
23 with unphosphorylated c-Met.
used (Fig. 4).24 Mice were administered a single oral dose of com-
pound 23 at 10 mg/kg. Human recombinant HGF was injected
intravenously at 1, 3, 6, 9, 12 or 24 h post-dose. Liver and blood
were harvested 5 min after administration of recombinant HGF.

C. M. Bode et al. / Bioorg. Med. Chem. Lett. 22 (2012) 4089–4093
4093
23. At a concentration of 1
l
M, compound 23 binds to c-Met (0.3 percent of control
25. The in vivo potency correlated well with the observed in vitro cellular IC50 of
2 nM when corrected for observed free fraction in mouse plasma.
(POC)) and the following kinases: VEGFR2 (34 POC), KIT (26 POC) and PDGFRB
(10 POC).
26. Data are mean SD (n = 3).
P values correspond to statistical difference
24. D’Angelo, N. D.; Bellon, S. F.; Booker, S. K.; Cheng, Y.; Coxon, A.; Dominguez, C.;
Fellows, I.; Hoffman, D.; Hungate, R.; Kaplan-Lefko, P.; Lee, M. R.; Li, C.; Liu, L.;
Rainbeau, E.; Reider, P. J.; Rex, K.; Siegmund, A.; Sun, Y.; Tasker, A. S.; Xi, N.; Xu, S.;
Yang, Y.; Zhang, Y.; Burgess, T. L.; Dussault, I.; Kim, T-S. J. Med. Chem. 2008, 51, 5766.
between groups treated with vehicle + HGF and compound 23 by Oneway
Anova followed by Bonferroni/Dunn post hoc test ⁄p <0.0001.