Pyrazolone-based anaplastic lymphoma kinase (ALK) inhibitors: Control of selectivity by a benzyloxy group

Article abstract of DOI:10.1016/j.bmcl.2011.10.055

Anaplastic lymphoma kinase (ALK) is transmembrane receptor tyrosine kinase, with oncogenic variants that have been implicated in ALCL, NSCLC and other cancers. Screening of a VEGFR2-biased kinase library resulted in identification of 1 which showed cross-reactivity with ALK. SAR on the indole segment of 1 showed that a subtle structural modification (the ethoxy group of 1 changed to a benzyloxy to generate 5a) enhanced potency (ALK), selectivity for VEGFR2 and IR along with improvement in metabolic stability. From docking studies of ALK versus VEGFR2 kinase, we postulated that the loss of entropy of the VEGFR2 in the bound form with 5a might be the origin of the reduced activity against that protein. Modification of the heterocyclic segment showed that thiazole-bearing pyrazolones preserved enzyme potency, and enhanced inhibition of NPM-ALK autophosphorylation in ALK-positive ALCL cells (Karpas-299). SAR of the benzyloxy group resulted in compounds which demonstrated good cellular potency in Karpas-299 cells. Compound 8 showed best overall profile for the series with broad kinome selectivity and liver micorsome stability. Compound 8 showed reasonable iv PK in rat, but with little oral exposure.

Full text of DOI:10.1016/j.bmcl.2011.10.055

Bioorganic & Medicinal Chemistry Letters 21 (2011) 7261–7264
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Pyrazolone-based anaplastic lymphoma kinase (ALK) inhibitors: Control
of selectivity by a benzyloxy group
⇑
Rabindranath Tripathy , Robert J. McHugh, Arup K. Ghose, Gregory R. Ott, Thelma S. Angeles,
Mark S. Albom, Zeck Huang, Lisa D. Aimone, Mangeng Cheng, Bruce D. Dorsey
Cephalon, Inc., Worldwide Discovery Research, 145 Brandywine Parkway, West Chester, PA 19380, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Anaplastic lymphoma kinase (ALK) is transmembrane receptor tyrosine kinase, with oncogenic variants
that have been implicated in ALCL, NSCLC and other cancers. Screening of a VEGFR2-biased kinase
library resulted in identification of 1 which showed cross-reactivity with ALK. SAR on the indole seg-
ment of 1 showed that a subtle structural modification (the ethoxy group of 1 changed to a benzyloxy
to generate 5a) enhanced potency (ALK), selectivity for VEGFR2 and IR along with improvement in met-
abolic stability. From docking studies of ALK versus VEGFR2 kinase, we postulated that the loss of
entropy of the VEGFR2 in the bound form with 5a might be the origin of the reduced activity against
that protein. Modification of the heterocyclic segment showed that thiazole-bearing pyrazolones pre-
served enzyme potency, and enhanced inhibition of NPM-ALK autophosphorylation in ALK-positive
ALCL cells (Karpas-299). SAR of the benzyloxy group resulted in compounds which demonstrated good
cellular potency in Karpas-299 cells. Compound 8 showed best overall profile for the series with broad
kinome selectivity and liver micorsome stability. Compound 8 showed reasonable iv PK in rat, but with
little oral exposure.
Received 13 September 2011
Accepted 14 October 2011
Available online 20 October 2011
Keywords:
ALK
NMP-ALK
ALCL
Pyrazolone
VEGFR2
Ó 2011 Elsevier Ltd. All rights reserved.
Inhibitor promiscuity remains a divisive topic in kinase-based
drug discovery due to the common mode of binding for most inhib-
itors at the conserved adenosine triphosphate (ATP) pocket.1
Despite this facet, several ATP-competitive compounds with activ-
ity against multiple kinase targets have been approved as drugs.
One positive aspect of this paradigm is that the parent molecular
scaffolds, in principle, have the potential to serve as launching
points for alternative kinase programs via judicious structural
modifications and structure–activity relationship (SAR). In this
communication, we have utilized this structural re-design process,
and successfully translated a member of the hetero-substituted
pyrazolones (A, Fig. 1), a known class of vascular endothelial
growth factor receptor 2 (VEGFR2) inhibitors2 into potent and
selective anaplastic lymphoma kinase (ALK) inhibitors via intro-
duction of a 4-benzyloxy substituent on the indole segment which
purportedly acts as a position specific selectivity determinant.3
ALK is a transmembrane receptor tyrosine kinase that was orig-
inally identified as a part of the nucleophosmin (NPM)-ALK fusion
protein derived from t(2,5)(p23,5q35) chromosomal translocation.
This translocation was detected in 60–75% of the ALK-positive
anaplastic large-cell lymphoma (ALCL) patients.4 Moreover,
additional ALK chimeras, as well as activating mutations in full
length ALK have been found in recent years.5
N
N
H
N
Cl
N
N
N
O
Het.
N
N
N
H
H
S
O
O
O
R
A
NVP-TAE684
H
N
H
N
N
Cl
S
O
N
N
F
N
N
Cl
O
N
O
H2N
N
1
Crizotinib
(PF-2341066)
ALK IC50 = 47 nM
ALK Cell IC50 = 300 nM
VEGFR2 IC50 = 36 nM
⇑
Corresponding author. Tel.: +1 610 738 6137; fax: 1 610 738 6558.
E-mail address: rtripath@cephalon.com (R. Tripathy).
Figure 1. Reported ALK inhibitors and pyrazolones.
0960-894X/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2011.10.055

7262
R. Tripathy et al. / Bioorg. Med. Chem. Lett. 21 (2011) 7261–7264
Table 1
OH
OR
CHO
CHO
SAR of the 4-methyl-1,2,3-thiadiazole series
RX/ Cs2CO3
H
N
N
N
Acetone
reflux
N
S
O
N
N
2
3
CH3
N
H
N
H
N
N
N
S
O
O
O
N
N
R
3
S
N
N
N
Piperidine
EtOH
5a-e
a
b
RO
Compd
R
ALK IC50
Karpas-299 cell IC50
Δ
4
5a
5b
5c
5d
5e
5f
Ph
23
158
34
650
—
2500
1600
—
5
2-Pyridyl
3-Pyridyl
4-Pyridyl
2-Me-thiazol-4-yl
Me2N–CH2–
Scheme 1. General synthesis of 4-hydroxy-1-methyl-1H-indole-3-carbaldehydes
(3) and subsequent condensation with the pyrazolone 4.
55
2999
291
—
a
IC50 values are reported as an average of 3 determinations; see Ref. 5d.
Karpas-299 cells; IC50 values are reported as a mean of at least two determi-
Importantly, echinoderm micro-tubule-associated protein-like
4 (EML4)-ALK fusion oncogene has also been identified in non-
small cell lung cancer (NSCLC).5a,b These ALK fusion genes encode
oncoproteins with constitutively active kinases which stimulate
anti-apoptotic and mitogenic signaling pathways associated with
survival and proliferation of ALCL, NSCLC, and other cancers.5e
Inhibiting oncogenic signaling through direct ALK kinase inhibi-
tion has emerged as an attractive approach to targeted cancer ther-
apy in recent years.6 Shown in Figure 1 are two prototypical
inhibitors in this area of research, a preclinical ALK inhibitor
NVP-TAE6846g and the c-Met/ALK inhibitor, crizotinib6i which re-
cently gained FDA approval for the treatment of ALK-positive
NSCLC.
In a program to design ALK inhibitors as a possible therapeutic
intervention against such cancers, we screened our internal kinase
library, which resulted in identification of a pyrazolone 1 (Fig. 1) as
an initial lead from this series. Pyrazolones were designed as a
class of potent and selective VEGFR2 kinase inhibitors.2 The design
concept for pyrazolones originated from their structural similarity
with the oxindole-based VEGFR2 inhibitors, a series which ulti-
mately provided the drug sunitinib for advanced renal cell carci-
noma and malignant gastrointestinal stromal tumor (GIST).7 The
pyrazolone class of compounds can be synthesized readily via con-
densation of simple, substituted pyrazolones with a suitable alde-
hyde (generally indole-3-aldehydes or pyrrole-2-aldehydes). As
described previously, oral PK has limited the development of this
series, ascribed to rapid metabolism, though select members dem-
onstrated acceptable values (cf., F = 29%).8
b
nations.; see Ref. 5d.
against IR and the broader kinome warranted further investigation.
The challenge, however, was to translate a molecular scaffold,
which was initially biased to inhibit a specific kinase (VEGFR2),
into another series that would be potent and selective with re-
duced (or no) affinity for the parent target. Based on the SAR of
the internal screening data of the pyrazolones, our initial focus
was to modify the 4-ethoxy indole segment of 1 to enhance
potency against ALK. As shown in Scheme 1 and 4-hydroxy-1-
methyl-1H-indole-3-carbaldehyde was alkylated with aryl or
heteroaryl methyl halides and the resulting compounds (3) were
condensed with the pyrazolone 4 in presence catalytic amounts
of piperidine at 90 °C. The Knoevenagel condensation products
(5) were isolated via simple filtration.
Table 1 shows ALK enzyme and cellular IC50 values for 5a–5f.
Compound 5a, with a benzyloxy substituent, was found to be the
most potent, demonstrating an IC50 of 23 nM against ALK.
Compound 5a was also found to be twofold more potent than 1
against the enzyme, but was less potent in cells. Benzene replace-
ment of 5a with pyridyl (5b–5d), 2-methyl-thiazole (5e) or with an
aliphatic segment (5f) were found to be less active against the en-
zyme and ALK-positive ALCL cells (Karpas-299).
Although, 5a bears an additional aromatic ring (as compared to
1) LMS data across the species improved dramatically (Table 2).
Selectivity against IR also improved considerably (IC50 >10 lM).
Our initial lead 1 bears a methyl 4-methyl-1,2,3-thiadiazole
subunit as the heterocyclic segment that is directly attached to
the pyrazolone ring (A, Fig. 1) along with a 4-ethoxy-N-methylin-
dole unit. It is potent (IC50 values of 47 nM against ALK enzyme
and 300 nM in ALK-positive Karpas-299 cell), and has a low molec-
ular weight (367 Da). Ambit KINOMEscan™9 across a panel of 235
kinases for 1 showed it to be a relatively selective compound,
Moreover, the 4-benzyloxy group in 5a also significantly reduced
the affinity for VEGFR2 kinase (Table 2, 5a vs 1). This result pro-
vided our first structural handle to control selectivity against ki-
nases including IR, VEGFR2 and others.
In order to understand the origin of such selectivity, these li-
gands (1 and 5a) were docked in the ALK (2XB7) and VEGFR2
(3CJG) structures available in the Protein Data Bank (PDB). All
the ALK structures in PDB had type I inhibitors with DFG-in struc-
tures. VEGFR2 had mostly DFG-out structures for both apo and
type I inhibitors. 3CJG is one of the few VEGFR2 structures that
inhibiting only 22 kinases >90% at 1 l
M (S90 = 0.09).10 Since ALK
is a member of insulin receptor (IR)11 and the inhibition of IR has
been documented to impact glucose homeostasis in preclinical
animal models, we desired selectivity against this closely related
family member.6h Importantly, counterscreening of 1 registered
21-fold selectivity against IR (IR IC50 = 998 nM). The compound 1
also inhibited VEGFR2 kinase with an IC50 value of 36 nM as re-
ported earlier.8 As expected, liver microsome stability (LMS) and
rat pharmacokinetics (PK) of 1 were found to be poor with no oral
bioavailability.8 Moreover, aqueous solubility of 1 at pH 7.4 and 2,
Table 2
Selectivity and liver microsome stability (LMS) of 1 and 5a
a
a
Compd VEGFR2 IC50
(nM)
ALK IC50
(nM)
LMS (t1/2 in min) human, dog, rat,
mouse
1
5a
36
>300
47
23
6, 5, 5,5
32, 29, >40, >40
respectively were low (<1 lg/mL).
Despite the poor aqueous solubility and PK profile, the unique
structure of 1 combined with the favorable selectivity profile
a
IC50 values are reported as an average of 3 determinations; see Refs. 8 and 5d.

R. Tripathy et al. / Bioorg. Med. Chem. Lett. 21 (2011) 7261–7264
7263
Table 4
Effect of heterocyclic substituents on the SAR of the 4-methyl-1,2,3-thiadiazole
replacement series
H
N
N
R
O
N
BnO
a
Compd
R
ALK IC50 (nM)
N
6
54
N
7
8
138
46
Figure 2. The binding mode of 5a in ALK is shown in cyan. The binding mode of 5a
in VEGFR2 (orange) was very similar to ALK with a small counter clockwise twist.
S
N
S
Table 3
Effect of benzene substituents on the SAR of the 4-methyl-1,2,3-thiadiazole series
9
10 lM
N
H
F3C
N
N
N
S
S
N
N
N
O
10
11
4.6 lM
N
N
S
104
O
a
IC50 values are reported as an average of 3 determinations; see Ref. 5d.
R
ALK IC50 (nM) Karpas-299 cellb (nM) IR (nM)
a
Compd
R
5a
5g
5h
5i
5j
5k
5l
H
2-F
3-F
4-F
23
13
20
12
12
12
21
650
400
1000
400
300
200
200
300
400
>10,000
>10,000
>10,000
>10,000
>1000
>10,000
720
Table 5
Effect of benzyloxy substituents on the SAR of the thiazole series
H
2,6-Di-F
2,4-Di-F
2-F-6-Cl
2-Cl-3,6-di-F 14
3-Cl-2,6-di-F
N
N
S
O
5m
5n
661
861
N
5
N
a
IC50 values are reported as an average of 3 determinations; see Ref. 5d.
Karpas-299 cells; IC50 values are reported as a mean of at least two determi-
O
b
nations.; see Ref. 5d.
R'
R
a
Compd
R/R0
ALK IC50 (nM)
Karpas-299 cellb (nM)
IR (nM)
had a DFG-in structure. The best scoring binding modes derived by
Glide12 for both 1 and 5a were consistent with similar ligands, like
Sutent,7 in the PDB. This binding mode is illustrated in Figure 2
(data shown only for 5a). Even though it was not obvious why
the activity of 5a dropped in VEGFR2 from the docking study with
a static DFG-in protein structure, it was not very difficult to inter-
pret the SAR in terms of the high entropy VEGFR2. Several DFG-out
VEGFR2 structures either had problem in docking 5a or had poor
docking score. The loss of entropy of the protein in the bound form
with 5a might be responsible for the loss of activity of 5a.13
Based on these encouraging findings, structural modifications of
the aromatic group of 5a were carried out to improve cellular and
enzyme potencies against ALK. Table 3 shows SAR for the methyl-
thiadiazole series. It was intriguing to find that a variety of halogen
substituents on the aromatic ring provided enhancement in po-
tency. A single fluoro substituent on the 2-, 3- and 4-positions of
the benezene ring (5g–5i) provided inhibitors with enhanced
potencies over 5a. However, ALK cellular potency increased only
modestly with 5g and 5i, and dropped with 5h. Compounds bear-
ing di- or tri-halo substituents (5j–5n) also demonstrated accept-
able potency against ALK and in one case (5n) an IC50 value in
single digit nM range. However, cellular potency remained in the
200–400 nM range. Moreover, except 5l–5n, all the compounds
showed good selectivity against IR.
8
H/H
2-F/H
3-F/H
4-F
46
46
25
63
42
9
13
14
17
229
300
400
120
—
500
100
200
90
9124
>10,000
>1000
—
>10,000
428
1415
6499
—
8a
8b
8c
8d
8e
8f
8g
8h
8i
H/Me
2,6-Di-F/H
2-Cl-6-F/H
2,3-Di-F/H
2,5-Di-F/H
2,4-Di-F/H
200
—
—
a
IC50 values are reported as an average of 3 determinations; see Ref. 5d.
Karpas-299 cells; IC50 values are reported as a mean of at least two determi-
nations; see Ref. 5d.
b
We, therefore, turned our attention toward modification of the
heterocyclic segment of the inhibitors that is directly attached to
the pyrazolone ring to assess impact on potency and physiochem-
ical properties. The condensation of the requisite pyrazolones with
4-benzyloxy-1-methyl-1H-indole-3-carbaldehyde was carried out
as per the standard protocol to make the final targets 6–11 (in very
similar fashion as shown in Scheme 1). The pyrazolone starting
materials were prepared as per the methods described earlier.2,8
As shown in Table 4, thiazole-bearing pyrazolone 8 provided
the best result (<50 nM) against ALK. The thiazole compound 8

7264
R. Tripathy et al. / Bioorg. Med. Chem. Lett. 21 (2011) 7261–7264
Table 6
recombinant ALK enzyme, Dr. Jean Husten and Ms. Mary Birchler
for the DLK and CDK selectivity data, Mr. Shi Yang, for VEGFR2 as-
say and Ms. Lihui Lu and Ms. Weihua Wan for cellular activity data.
Profile of the compounds 8, 8b, 8e and 8g
Compd ALK Cell
IC50 (nM)
Selectivity
Ambit S90
VEGFR2 (% Inh
LMS t1/2 (min) H,
@ 1
lM)
D, R, Ma
0.039b
0.031c
—
50%
50%
—
30, 22, 30, 25
10, 8, 8, 6
7, 10, 6, 6
References and notes
8
300
120
100
90
8b
8e
8g
1. For kinase inhibitor promiscuity and selectivity see: (a) Tanramluk, D.;
Schreyer, A.; Pitt, W. R.; Blundell, T. L. Chem. Biol. Drug Des. 2009, 74, 16; (b)
Brandt, P.; Jensen, A. J.; Nilsson, J. Bioorg. Med. Chem. Lett. 2009, 19, 5861; (c)
Bhagwat, S. S. Curr. Opin. Invest. Drugs 2009, 10, 1266. and references cited
therein.
2. (a) Tripathy, R.; Reiboldt, A.; Messina, P. A.; Iqbal, M.; Singh, J.; Bacon, 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; (b) Singh, J.; Tripathy, R.
US 6455525.
—
—
14, 12, 10, 8
a
b
c
H = Human, D = Dog, R = Rat, M = Mouse.
10/253 kinases showing 90% inh. at 1 M.
8/261 kinases showing 90% inh. at 1 M.
l
l
exhibited a cell IC50 value of 300 nM (Table 5), which is a slight
improvement over the 4-methyl-1,2,3-thiadiazole-based com-
pound 5a. Moreover, 8 also showed good selectivity against IR.
To improved cellular potency further, substituent effects on the
benzyloxy segment of 8 were examined. These results are shown in
Table 6. With a single 3-fluoro substituent, 8b showed ꢀ2-fold
improvement in enzyme potency and 2.5-fold increase in ALK cell
potency. Selectivity against IR, however, dropped slightly. Addition
of a methyl group at the aliphatic carbon of the benzyloxy was not
helpful (8d). Di-substituted compounds 8e–8g also showed
improvements in enzyme and cellar potencies against ALK. Except
for 8e, selectivity against IR dropped significantly.
3. A part of the work was presented at an American chemical society meeting.
McHugh, R. J.; Tripathy, R.; Angeles, T.; Albom, M.; Cheng, M.; Dorsey, B. 235th
ACS National Meeting, April, 2008.
4. (a) Morris, S. W.; Kirstein, M. N.; Valentine, M. B.; Dittmer, K. G.; Shapiro, D. N.;
Saltman, D. L.; Look, A. T. Science 1994, 263, 1281; (b) Webb, T. R.; Slavish, J.;
George, R. E.; Look, A. T.; Xue, L.; Jiang, Q.; Cui, X.; Rentrop, W. B.; Morris, S. W.
Expert Rev. Anticancer Ther. 2009, 9, 331.
5. (a) Soda, M.; Choi, Y. L.; Enomoto, M.; Takada, S.; Yamashita, Y.; Ishikawa, S.;
Fujiwara, S.; Watanabe, H.; Kurashina, K.; Hatanaka, H.; Bando, M.; Ohno, S.;
Ishikawa, Y.; Aburatani, H.; Niki, T.; Sohara, Y.; Sugiyama, Y.; Mano, H. Nature
2007, 448, 561; (b) Scott, J. R.; Shapiro, G. I. Curr. Opin. Invest. Drugs 2010, 11,
1477; (c) Mosse, Y. P.; Wood, A.; Maris, J. M. Clin. Cancer Res. 2009, 15, 5609;
Moreover for missense ALK mutants see: (d) Lu, L.; Ghose, A. K.; Quail, M. R.;
Albom, M. S.; Durkin, J. T.; Holskin, B. P.; Angeles, T. S.; Cheng, M. Biochemistry
2009, 48, 3600; For a review see: (e) Milkiewicz, K. L.; Ott, G. R. Curr. Opin.
Invest. Drugs 2010, 20, 1653.
6. (a) Lovly, C. M.; Heuckmann, J. M.; de Stanchina, E.; Chen, H.; Thomas, R. K.;
Liang, C.; Pao, W. Cancer Res. 2011, 71, 4920; (b) Ott, G. R.; Tripathy, R.; Cheng,
M.; McHugh, R.; Anzalone, A. V.; Underiner, T. L.; Curry, M. A.; Quail, M. R.; Lu,
L.; Wan, W.; Angeles, T. S.; Albom, M. S.; Aimone, L. D.; Ator, M. A.; Ruggeri, B.
A.; Dorsey, B. D. ACS Med. Chem. Lett. 2010, 1, 493; (c) Xianming, D.; Wang, J.;
Zhang, J.; Sim, T.; Kim, N. D.; Sasaki, T.; Luther, W., II; George, R. E.; Janne, P. A.;
Gray, N. S. ACS Med. Chem. Lett. 2011, 2, 379; (d) Zificsak, C. A.; Theroff, J. P.;
Aimone, L. D.; Angeles, T. S.; Albom, M. S.; Cheng, M.; Mesaros, E. F.; Ott, G. R.;
Quail, M. R.; Underiner, T. L.; Wan, W.; Dorsey, B. D. Bioorg. Med. Chem. Lett.
2011, 21, 3877; (e) Mesaros, E. F.; Burke, J. P.; Parrish, J. D.; Dugan, B. J.;
Anzalone, A. V.; Angeles, T. S.; Albom, M. S.; Aimone, L. D.; Quail, M. R.; Wan,
W.; Lu, L.; Zeqi, H.; Ator, M. A.; Ruggeri, B. A.; Cheng, M.; Ott, G. R.; Dorsey, B. D.
Bioorg. Med. Chem. Lett. 2011, 21, 463; (f) Kinosita, K.; Ono, Y.; Emura, T.; Asoh,
K.; Furuich, N.; Ito, T.; Kawada, H.; Tanaka, S.; Morikami, K.; Tsukaguchi, T.;
Sakamoto, H.; Tsukuda, T.; Oikawa, N. Bioorg. Med. Chem. Lett. 2011, 21, 3788;
(g) Galkin, A. V.; Melnick, J. S.; Kim, S.; Hood, T. L.; Li, N.; Li, L.; Xia, G.;
Steensma, R.; Chopiuk, G.; Jiang, J.; Wan, Y.; Ding, P.; Liu, Y.; Sun, F.; Schultz, P.
G.; Gray, N. S.; Warmuth, M. Proc. Natl. Acad. Sci. U.S.A. 2007, 104, 270; (h)
Sabbatini, P.; Korenchuk, S.; Rowand, J. L.; Groy, A.; Liu, Q.; Leperi, D.; Atkins, C.;
Dumble, M.; Yang, J.; Anderson, K.; Kruger, R. G.; Gontarek, R. R.; Maksimchuk,
K. R.; Suravajjala, S.; Lapierre, R. R.; Shotwell, J. B.; Wilson, J. W.; Chamberlain,
S. D.; Rabindran, S. K.; Kumar, R. Mol. Cancer Ther. 2009, 8, 2811; (i) Christensen,
J. G.; Zou, H.; Arango, M.; Li, Q.; Lee, J. H.; McDonnell, S. R.; Yamazaki, S.; Alton,
G.; Mroczkowski, B.; Los, G. Mol. Cancer Ther. 2007.
The Ambit selectivity profile (S90 values) and LMS data for se-
lected thiazole compounds are shown in Table 6. Compounds 8
and 8b had improved selectivity in comparison to our initial lead
(1, S90 = 0.09). As expected, selectivity improved against VEGFR2
kinase (IC50 value of 36 nM for 1 vs 50% inhibition at 1 lM for 8
and 8b). LMS did not dramatically improve, except for 8. Halo sub-
stituents on the aromatic ring of the benzyloxy group (8b, 8e, and
8g) did little to enhance stability. Despite acceptable LMS, rat PK
data for 8 showed negligible oral exposure, although the iv profile
was reasonable (1 mg/kg; t1/2 = 0.69 h, Vd = 1.2 L/Kg, Cl = 19.6
mL/min, AUC = 972 ng h/mL, oral at 5 mg/Kg, AUC = 13 ng h/mL).
In conclusion, we have demonstrated that a lead (1) from the
pyrazolone-based kinase inhibitor series (bearing a 4-methyl-
1,2,3-thiadiazole) has been successfully modified to provide potent
ALK inhibitors with improved enzyme and cellular potencies,
selectivity against IR and other kinases along with enhanced LMS
stability. Placement of a 4-benzyloxy group on the indole segment
was key structural element acting as a molecular switch to im-
prove selectivity against VEGFR2 kinase. To probe the origin of
such selectivity, we carried out modeling of 5a against ALK and
VEGFR2 kinases. Even though it was not obvious why the activity
of 5a dropped in VEGFR2 from the docking study, we postulate that
the loss of entropy of that protein in the bound form with 5a might
be responsible for the loss of activity. From the SAR on the hetero-
cyclic part of the inhibitors, thiazole was identified as a good sub-
stitute for the 4-methyl-1,2,3-thiadiazole. Benzyloxy modification
of the thiazole series resulted in potent compounds 8b, 8e, and
8g with improved cellular activities. The inhibitor 8 had best over-
all profile with improved kinase selectivity and metabolic stability.
The PK data for 8 in rat showed a reasonable iv profile, but with
very little oral exposure.
7. Rock, E. P.; Goodman, V.; Jiang, J. X.; Mahjoob, K.; Verbois, S. L.; Morse, D.;
Dagher, R.; Pazdur, R. Oncologist 2007, 12, 107.
8. 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.
9. Fabian, M. A.; Biggs, W. H., III; Treiber, D. K.; Atteridge, C. R.; Azimioara, M. D.;
Benedetti, M. G.; Carter, T. A.; Ciceri, P.; Edeen, P. T.; Floyd, M.; Ford, J. M.;
Galvin, M.; Gerlach, J. L.; Grotzfeld, R. M.; Herrgard, S.; Insko, D. E.; Insko, M. A.;
Lai, A. G.; Lelias, J.-M.; Mehta, S. A.; Milanov, Z. V.; Velasco, A. M.; Wodicka, L.
M.; Patel, H. K.; Zarrinkar, P. P.; Lockhart, D. J. Nat. Biotechnol. 2005, 23, 329.
10. S90 value = number of kinases showing 90% inhibition at 1 lM/total number of
kinases tested.
11. Palmer, R. H.; Vernersson, E.; Grabbe, C.; Hallberg, B. Biochem. J. 2009, 420, 345.
12. Methods (Computational). Docking Steps. The essential steps in the current
docking experiment are summarized below: (i) Prepare proteins using Maestro
protein preparation work flow; (ii) Build 3D structures of the inhibitors using
LigPrep; (iii) Glide/XP docking to keep top 10 binding poses; (iv) Selection of
the binding mode using our knowledge based approach13; most of these
modules are available in the Schrodinger molecular modeling package
(www.schrodinger.com).
Acknowledgments
13. Ghose, A. K.; Herbertz, T.; Pippin, D. A.; Salvino, J. M.; Mallamo, J. P. J. Med.
Chem. 2008, 51, 5149.
We thank Mr. Kurt Josef for his help with NMR experiments,
Dr. Sheryl L. Meyer and Ms. Beverly Holskin for supplying the