Discovery, synthesis, and in vivo activity of a new class of pyrazoloquinazolines as selective inhibitors of aurora B kinase

Article abstract of DOI:10.1021/jm061335f

The Aurora kinases have been the subject of considerable interest as targets for the development of new anticancer agents. While evidence suggests inhibition of Aurora B kinase gives rise to the more pronounced antiproliferative phenotype, the most clinically advanced agents reported to date typically inhibit both Aurora A and B. We have discovered a series of pyrazoloquinazolines, some of which show greater than 1000-fold selectivity for Aurora B over Aurora A kinase activity, in recombinant enzyme assays. These compounds have been designed for parenteral administration and achieve high levels of solubility by virtue of their ability to be delivered as readily activated phosphate derivatives. The prodrugs are comprehensively converted to the des-phosphate form in vivo, and the active species have advantageous pharmacokinetic properties and safety pharmacology profiles. The compounds display striking in vivo activity, and compound 5 (AZD1152) has been selected for clinical evaluation and is currently in phase 1 clinical trials.

Full text of DOI:10.1021/jm061335f

J. Med. Chem. 2007, 50, 2213-2224
2213
Discovery, Synthesis, and in ViWo Activity of a New Class of Pyrazoloquinazolines as Selective
Inhibitors of Aurora B Kinase
†
,†
†
‡
‡
‡
Andrew A. Mortlock, Kevin M. Foote,* Nicola M. Heron, Fr e´ d e´ ric H. Jung, Georges Pasquet, Jean-Jacques M. Lohmann,
‡
‡
†
†
†
†
†
Nicolas Warin, Fabrice Renaud, Chris De Savi, Nicola J. Roberts, Trevor Johnson, Cyril B. Dousson, George B. Hill,
David Perkins, Glenn Hatter, Robert W. Wilkinson, Stephen R. Wedge, Simon P. Heaton, Rajesh Odedra,
†
†
†
†
†
†
†
,§
†
†
†
†
†
Nicholas J. Keen, Claire Crafter, Elaine Brown, Katherine Thompson, Stephen Brightwell, Liz Khatri,
†
†
†
†
†
†
Madeleine C. Brady, Sarah Kearney, David McKillop, Steve Rhead, Tony Parry, and Stephen Green
AstraZeneca Pharmaceuticals, Alderley Park, Macclesfield, Cheshire SK10 4TG, United Kingdom, and AstraZeneca, Centre de Recherches,
Parc Industriel, Pompelle, BP1050, Chemin de Vrilly, 51689 Reims Cedex2, France
ReceiVed NoVember 17, 2006
The Aurora kinases have been the subject of considerable interest as targets for the development of new
anticancer agents. While evidence suggests inhibition of Aurora B kinase gives rise to the more pronounced
antiproliferative phenotype, the most clinically advanced agents reported to date typically inhibit both Aurora
A and B. We have discovered a series of pyrazoloquinazolines, some of which show greater than 1000-fold
selectivity for Aurora B over Aurora A kinase activity, in recombinant enzyme assays. These compounds
have been designed for parenteral administration and achieve high levels of solubility by virtue of their
ability to be delivered as readily activated phosphate derivatives. The prodrugs are comprehensively converted
to the des-phosphate form in ViVo, and the active species have advantageous pharmacokinetic properties
and safety pharmacology profiles. The compounds display striking in ViVo activity, and compound 5
(AZD1152) has been selected for clinical evaluation and is currently in phase 1 clinical trials.
been reported and/or reviewed elsewhere.4
,9-18
VX-680 was
9
Introduction
the first inhibitor reported to show potent antitumor activity in
The Aurora proteins are a small family of serine/threonine
kinases that are expressed during mitosis and have been
suggested to be attractive drug targets. They have roles in
16
17
ViVo; similar results for PHA-680632 and PHA-739358 have
also been reported recently. All of these drugs inhibit Aurora
A and Aurora B (while being more potent inhibitors of Aurora
A kinase activity) and have some activity against other, non-
1
chromosome segregation and cytokinesis and ensure that a
complete copy of the duplicated genome is precisely partitioned
into the two daughter cells. Humans express three Aurora kinase
paralogues, Aurora A, B, and C, the biology of which has been
9,16,17,19
Aurora, kinases.
To date, no reports of Aurora B-selective
inhibitors have appeared although the first disclosure of an
2
0
reviewed extensively.1,2 With their key role in mitosis and being
Aurora A-selective inhibitor has been reported.
In this paper we describe the discovery of a novel series of
Aurora kinase inhibitors that are specific for Aurora kinases
over other kinases and which show high selectivity for Aurora
B over Aurora A. Compounds from this series have attractive
physicochemical and pharmacokinetic properties and have been
demonstrated to possess potent antitumor activity in preclinical
models.
aberrantly overexpressed in tumor cells, there has been con-
siderable interest in developing specific Aurora kinase inhibitors
and in understanding the therapeutic value of selectively
inhibiting Aurora A and B.
While there are now considerable data supporting a link
between Aurora A and B kinase expression and cancer (the role
3
of Aurora C is currently poorly understood ), it is still unclear
whether inhibition of Aurora A and/or Aurora B, could be
advantageous in terms of providing therapeutic benefit in
oncology. Studies using both pharmacological and genetic
Chemistry. Compounds 5-39 were prepared as shown in
Scheme 1. To explore the SAR of the lipophilic pocket binding
acetanilide group, we first prepared the intermediate 43 by
displacing the chloro leaving group in 41 with 4-(hydroxy-
methyl)piperidine. The acid group in 43 was then further reacted
4
disruption of Aurora kinases in cells have shown that the mitotic
defects described following exposure of cells to Aurora kinase
1,5-8
a
inhibitors seem largely due to inhibition of Aurora B.
When
with a range of substituted anilines, using EDC and 2-hy-
Aurora B is inhibited in tumor cells, the cells are forced through
a catastrophic mitotic exit leading to polyploid cells that rapidly
lose viability. In contrast, inhibition of Aurora A kinase activity
causes a mitotic delay and abnormalities in centrosome separa-
tion leading to the formation of a monopolar spindle.
A number of small molecule inhibitors of Aurora kinase
activity, from structurally diverse chemical series, have already
droxypyridine 1-oxide to give the acetanilides 6, 7, 9, and 12-
17. To explore the SAR in the quinazoline C-7 side chain, we
first made the chloropropoxy acetanilide intermediates 44-48,
from the acids 41 and 42. Compounds 44-47 were prepared
using either EDC and 2-hydroxypyridine 1-oxide or POCl3 and
pyridine, in moderate to excellent yields. Compound 48 was
prepared by reacting 42 with pentafluorophenyl trifluoroacetate
to first give the pentafluorophenyl ester which was subsequently
reacted with 3-fluoroaniline to give 48 in excellent yield without
8
*
To whom correspondence should be addressed. E-mail: kevin.foote@
astrazeneca.com. Telephone: +44 (0)1625 518629. Fax: +44 (0)1625
10097.
5
†
a
AstraZeneca Pharmaceuticals, United Kingdom.
AstraZeneca, France.
Current address: AstraZeneca Pharmaceuticals LP, 35 Gatehouse Drive,
Abbreviations: DMA, N,N-dimethylacetamide; DMF, N,N-dimethyl-
‡
§
formamide; DMSO, dimethyl sulfoxide; EDC, 1-(3-dimethylaminopropyl)-
3-ethylcarbodiimide hydrochloride; NMP, 1-methyl-2-pyrrolidinone; TFA,
trifluoroacetic acid.
Waltham MA 02451.
1
0.1021/jm061335f CCC: $37.00 © 2007 American Chemical Society
Published on Web 03/21/2007

2
214 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 9
Mortlock et al.
Scheme 1a
a
Reagents: (a) 4-(hydroxymethyl)piperidine, DMA, 100 °C; (b) ArNH2, DMA, EDC, 2-hydroxypyridine 1-oxide, 50 °C; (c) ArNH2, pyridine, POCl3, 0
C f r.t. or ArNH2, pentafluorophenyl trifluoroacetate, pyridine, DMF, 0 °C f r.t.; (d) amine, DMA, 90 °C or amine, NMP, KI, 60 °C or amine, DMA,
°
KI, 50 °C; (e) di-tert-butyldiethylphosphoramidite, tetrazole, DMF, r.t.; H2O2, -10 °C f r.t.; (f) HCl, dioxane, r.t.; (g) cyclohexene oxide, MeOH.
Scheme 2a
Scheme 3a
a
Reagents: (a) AcOH, 50, reflux.
the need for chromatography and on a multigram scale. Finally,
the drug compounds 8, 10, 11, and 18-36 were prepared by
displacing the chloro leaving groups in 44-48 with the
appropriate amine, using a range of standard conditions. The
phosphate prodrugs 5, 37, 38, and 39 were prepared as described
previously.15 Thus the free hydroxyl group in compounds 18,
a
Reagents: (a) NaH, 1,3-propanediol, DMF, 60 f 110 °C; (b) SOCl2,
DMF, 85 °C; (c) 50, DMF, HCl (dioxane), 90 °C.
2
9, 33, and 34 was first reacted with di-tert-butyldiethylphos-
Scheme 4a
phoramidite and tetrazole in DMA to give the phosphite esters.
These were then oxidized in the same pot with an aqueous
solution of hydrogen peroxide to give the corresponding tert-
butyl phosphate esters. The tert-butyl groups were then cleaved
with hydrogen chloride in dioxane to afford the phosphate
prodrugs 5, 37, 38, and 39 as dihydrochloride salts. The
hydrochloride salts could be optionally desalted with cyclo-
hexene oxide in methanol to generate the free base.
a
Reagents: (a) 12 M NaOH, 100 °C; (b) H O, reflux.
2
The intermediate compounds 41 and 42 were prepared as
shown in Scheme 2 and Scheme 3, respectively. Compound 41
was prepared from the N,N-dimethylimidoformamide compound
dichloro compound 53. Finally, 53 was reacted with the
aminopyrazole 50 under acidic conditions to give 42 in good
yield and high purity without the need for chromatography.
9, which we have described previously,15 following Dimroth
4
cyclization and rearrangement with the aminopyrazole 50 in
refluxing acetic acid. The quinazoline C-6-unsubstituted, pyr-
azole acetic acid intermediate 42 was prepared by an alternative
sequence as shown in Scheme 3. The mono anion of 1,3-
propanediol was reacted with 7-fluoroquinazolinone 51, to give
the alcohol 52 in high yield. The alcohol functionality and
quinazolinone groups in 52 were chlorinated in a single step
using hot thionyl chloride and DMF as catalyst to give the
The aminopyrazole starting material 50 can be prepared in
multigram quantities in two steps from commercially available
5-amino-3-(cyanomethyl)-1H-pyrazole-4-carbonitrile 54 (Scheme
4). Hydrolysis of the nitrile groups in 54 was achieved with
concentrated sodium hydroxide at 100 °C followed by neutral-
ization with hydrochloric acid to give the diacid 55. Finally,
55 underwent facile decarboxylation in refluxing water to give
50 in high yield over the two stages.
21

SelectiVe Inhibitors of Aurora B Kinase
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 9 2215
Figure 1. Compounds from the AstraZeneca Aurora kinase inhibitor series.
Discussion
identify Aurora kinase inhibitors suitable for clinical evaluation.
In particular, we report the synthesis, structure-activity rela-
tionships (SAR), and much improved in ViVo activity of a novel
series of 5-acetanilide-substituted 3-aminopyrazoles. A com-
pound from this series, compound 5 (AZD1152), has been
In previous publications we have described a number of
quinazoline-based Aurora kinase inhibitor series.5,10,14,15 The
p-benzamidoanilinoquinazoline series, e.g., ZM447439 (1)
(Figure 1), was identified following a high throughput screening
2
2
selected for clinical evaluation.
campaign. Compound 1 was shown to inhibit Aurora A and
Aurora B equipotently but in a panel of other serine/threonine
and tyrosine kinases, showed high specificity for Aurora
The phosphate derivatives (e.g., compound 4) have very high
aqueous solubility (>10 mg/mL) in simple pH-adjusted saline
without the need for recourse to complex with potentially highly
toxic vehicles. This high level of solubility allowed us to explore
the effects of both short- (bolus) and longer-term exposure of
our compounds in preclinical models and hence to understand
the complex balance between concentration and duration of
5
kinases. Compound 1 has sub-micromolar activity in cellular
1
0
proliferation assays but relatively poor drug-like properties.
Investigation of the SAR around the anilino ring linking the
key hinge (quinazoline) and lipophilic pocket (benzamide)
binding groups in 1 showed that this ring could be replaced
with a number of heterocyclic systems to give improvements
in both potency and physicochemical properties.10 Furthermore,
it was found from both the testing of small libraries and from
in-house crystallographic studies that Aurora A is able to
accommodate highly polar functional groups within the side
chain at the quinazoline C-7 position. Compounds with elabo-
rated C-7 side chains show good cellular activity and benefit
2
3
exposure on efficacy and tolerability.
While a number of highly potent compounds were made in
this pyrazole series which lacked a suitable hydroxyl group for
2
4
attaching a solubilizing phosphate group, all the compounds
disclosed in this paper were designed for parenteral administra-
tion as the corresponding dihydrogen phosphate prodrugs.2
For clarity of discussion, data are presented for a limited,
representative set of compounds (further compounds have been
4-26
1
0
from much better solubility and free drug levels. The
structure-activity relationships of the pyridino- and pyrimi-
dinoquinazoline compounds have been described recently and
includes one of the first crystal structures of an Aurora kinase
with a bound small molecule inhibitor, the pyrimidinoquinazo-
24-26
described in patent applications).
It should be borne in mind
that the structure-activity relationships described in the paper
have come from analysis of the full set of compounds. Thus,
where trends are exemplified by a single pair of compounds in
the tables, the conclusions presented are typically supported by
data from additional compounds. For clarity we have chosen
to represent the pyrazole ring in this paper in a single tautomeric
form. It is likely that both tautomers can exist but the relative
contribution to binding of each form remains undetermined.
It was found that the pyrazole series are highly potent Aurora
B kinase inhibitors but generally show less activity against
Aurora A, in assays using recombinant enzymes (Table 1).
Knowledge of the SAR from earlier series suggested that small
and lipophilic groups were particularly favored in the terminal
1
0,14
line 2.
The X-ray structural determination of compound 2
in Aurora A was helpful in rationalizing the observed high
potency and high specificity for the Aurora kinases.
More recently we have described a series of thiazoleacet-
anilide inhibitors of Aurora kinase that show further improve-
ments in cellular activity and which included compound 3.
1
4
1
0,15
When compound 3 was administered as its dihydrogen phos-
phate derivative 4, a significant inhibition of histone H3
phosphorylation in human SW620 colorectal adenocarcinoma
tumors could be demonstrated.1
0,15
Histone H3 is a cellular
1
0,15
substrate of Aurora B and is therefore a useful marker of Aurora
lipophilic binding pocket.
It was found that the SAR in the
B kinase inhibition in ViVo.5
pyrazole series followed these same broad trends, so a small
focused library of compounds was made where a small number
of preferred substitution patterns were emphasized (Table 1).
Fluoro groups were found to be particularly favorable especially
in the ortho- and meta-positions (e.g., compounds 7, 8, 10, and
11 compared with compound 9; Table 1). The most favored
aniline substituents, in terms of Aurora B kinase activity and
inhibition of histone-H3 phosphorylation in SW620 cells are
While the identification of a compound with in ViVo activity
against a relevant biomarker represented significant progress,
this activity was somewhat variable, and antitumor studies with
this compound (SW620 tumors grown in nude mice) were
ultimately inconclusive. However, in parallel to the thiazole
series we were also evaluating alternative chemical starting
points. In this paper we describe some of our further efforts to

2
216 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 9
Mortlock et al.
Table 2. Optimization of the Quinazoline C-7 Side Chain
Table 1. Acetanilide SAR in the 3-Aminopyrazole Series
Aurora B-
Aurora A
Ki, µM
INCENP
Ki, µM
Cell
IC50, µM
a
a
a,b
No.
R
6
7
8
9
H
2-F
3-F
4-F
2,3-di-F
3,5-di-F
3-Cl
3-CN
3-NO2
3-OH
3-OMe
3-CF3
1.6
1.4
0.018
0.014
0.002
0.030
0.002
0.001
0.008
0.083
0.039
0.14
1.2
0.47
0.68
2.6
0.54
0.37
1.2
0.45
2.0
0.41
0.23
0.45
2.6
0.89
4.0
1.9
10
11
12
13
14
15
16
17
>10
>10
>10
0.009
0.013
5.5
1.8
0.51
a
Average of at least two independent dose-response curves. Variation
b
was generally <25%. Inhibition of histone-H3 phosphorylation following
4 h incubation in SW620 tumor cells.
2
3-fluoro, 2,3-difluoro, and 3,5-difluoro (compounds 8, 10, and
1
1, respectively). The 2,3-difluoro and 3-fluoro acetanilide
groups generally give rise to better physicochemical properties.
Larger and more polar functional groups are also tolerated but
27
are less preferred, especially in terms of cellular potency. These
data suggest the binding conformations of the thiazole and
pyrazole series are likely to be comparable;15 however, the
presence of a further hydrogen bond donor in the pyrazole ring
(compared to the thiazole ring) may allow additional binding
to the protein backbone although this is difficult to model given
that the pyrazole hydrogen could be on either of the two ring
nitrogens.
a
Measured using shake-flask methodology with a buffer:octanol volume
ratio of 100:1. The concentration of compound in the aqueous phase before
and after partitioning with octanol was determined by generic HPLC-UV
analysis. b Average of at least two independent dose-response curves.
Variation was generally <25%. Inhibition of histone-H3 phosphorylation
following 24 h incubation in SW620 tumor cells.
Despite the similarities in the putative binding conformation
between the thiazole and pyrazole series, there exists a
significant difference in cellular inhibition of phosphorylation
of histone H3 (at serine 10) between compound 3 (IC50 ) 0.005
c
1
5
µM) and the direct 3-aminopyrazole analogue compound 8
Table 1: IC50 ) 0.68 µM) in SW620 tumor cells. Both the
strategy requiring synthesis of in ViVo hydrolyzable phosphate
esters (to obtain high solubility for parenteral administration),
the C-7 side chain was also required to contain a pendent
hydroxyl functional group. This hydroxyl group plays a second
critical role in the drug as, depending on its proximity to the
basic nitrogen, it exerts a powerful role in moderating the
(
thiazole and pyrazole series are clearly highly potent inhibitors
of Aurora B kinase enzyme activity. The cellular difference
between compounds 3 and 8 may be attributable to a combina-
tion of subtle differences in binding conformation together with
differences in the biophysical characteristics of the two com-
pounds. The pyrazole compounds are generally much less
lipophilic than the corresponding thiazole analogues (e.g.,
measured log D7.4 ) 3.2 vs 2.1 for compounds 3 and 8
respectively) resulting in an improvement in general physico-
chemical properties and a decrease in protein binding (98% vs
28
2
9
basicity of the latter group.
In the related aniline, pyrimidine, and thiazole series we have
demonstrated that the length of the aliphatic spacer, between
the oxygen atom on the C-7 position of the quinazoline ring
and the basic nitrogen atom within the side chain, governs the
1
0,15
ability of the polar group to solvate.
Thus, if the spacer is
9
4% bound for compounds 3 and 8, respectively, in rat plasma).
too short or too sterically demanding, the polar group is brought
too close to the protein and the resultant clash causes a reduction
in potency. Two-carbon spacers are well-tolerated but the
presence of a second â-oxygen atom may reduce the basicity
of the nitrogen in the side chain too much so that, in general,
compounds with three carbon spacers were the most studied.
Table 2 shows the SAR of selected compounds substituted with
a range of basic side chains utilizing a three-carbon spacer.
Both cyclic and noncyclic amine side chains are tolerated by
the Aurora kinases, giving rise to excellent levels of cellular
potency, in most cases, without a significant increase in
The improved physical properties, together with reasonable
enzyme and cell activity, made the pyrazole series an attractive
proposition for further medicinal chemistry aimed at optimizing
cellular potency while at the same time maintaining drug-like
properties.
Knowledge from previous work suggested that optimization
of the quinazoline C-7 side chain might result in further
enhancement of cellular potency.
10,15
The presence of the amino
functional group increases potency while its basicity helps to
reduce the compound’s binding to plasma albumin. With a
1
0

SelectiVe Inhibitors of Aurora B Kinase
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 9 2217
Table 3. Quinazoline C-6 H SAR
lipophilicity (Table 2). Again, these compounds are highly
potent Aurora B kinase inhibitors but show less activity against
Aurora A kinase. A range of five- and six-membered cyclic
amines (pyrrolidine, piperidine, and piperazine) were prepared
(
(
e.g., compounds 18-23 in Table 2). The prolinol side chains
compounds 18 and 19) were found to have the greatest cellular
activity with little noticeable difference in terms of potency,
between the two enantiomers. Six-membered cyclic amines (e.g.,
compounds 20, 21, and 22) were found to be marginally less
active in cells. In the acyclic series, N-substituted ethanolamines
were found to be optimal with a clear relationship between the
lipophilicity of the N-substituent and cellular potency (com-
pounds 24-30 in Table 2) with the N-isobutyl-substituted
compound 30 having an IC50 in SW620 cells of less than 1
nM. However, we found that as lipophilicity and cellular potency
increased, protein binding increased concomitantly. In addition,
compounds with high lipophilicity were found to be more likely
to inhibit the major P450 isoforms, especially against 3A4 (e.g.,
compare compound 30: log D7.4 ) 3.6; 3A4 IC50 ) 0.4 µM vs
compound 28: log D7.4 ) 1.9; 3A4 IC50 >10 µM). Within this
set of compounds the 2,3-difluorophenyl group gave greater
cellular potency compared with the 3-fluorophenyl group. The
2,3-difluorophenyl group also resulted in lower protein binding
(compare compounds 18 vs 23 and 28 vs 26: 85% vs 95%
bound and 83% vs 95% bound in rat plasma, respectively). By
appropriate combination of the quinazoline C-7 side chain and
anilide groups, as described, it is possible to balance attractive
physicochemical-based properties (high free drug levels and no
P450 inhibition) with high potency in cell assays.
a
Average of at least two independent dose-response curves. Variation
b
was generally <25%. Inhibition of histone-H3 phosphorylation following
4 h incubation in SW620 tumor cells.
2
Table 4. Selectivity Profile (IC50, µM) of Compound 34 in a Panel of
Kinases
Important roles for the quinazoline C-7 side chain and the
lipophilic pocket binding group had been established in the
earlier series, and these roles and the associated SAR also
broadly applied to the new pyrazole series, albeit with some
detailed differences. In contrast, the role of the substituent at
the C-6 position on the quinazoline ring was less well
understood. It was hypothesized that this group was not involved
in direct binding to the protein (although it did have the potential
to influence the hydrogen bond acceptor properties of the
quinazoline ring in its key interaction with the hinge) and that
its removal would allow a reduction in molecular weight and
remove any risk of reactive metabolites being formed resulting
from oxidative metabolism of the methyl group. Surprisingly,
it was found that replacing the C-6 methoxy group with a proton
generally reduced potency versus Aurora A kinase but main-
tained very high potency versus Aurora B kinase; improved
potency was also observed in the SW620 cell assay (compounds
IC50, µMa
IC50, µMa
kinase
Aurora A
Aurora B-INCENP
Aurora C-INCENP
LCK
kinase
1.4
KDR
PHK
1.8
1.8
8.2
<0.001
0.017
0.17
ZAP70
othersb
>10
a
Non-Aurora kinase assays were carried out as described in ref 15.
JAK3, vABL, CSK, FAK, SRC, IGFR, EGFR, FGFR, P38A, PAK1,
CDK2, JNK1A, PKA, MEK, CHK1, PLK1.
b
Table 5. Physicochemical and Pharmacokinetic Properties of Selected
Compounds from the Pyrazoloquinazoline Series
%
rat Cl,
3A4 inhibition
hERG
IC50, µM
log D7.4a
boundb
mL/min/kgc
IC50, µMd
e
No.
1
2
8
9
33
2.0
2.3
2.1
2.3
85
81
89
96
63
16
15
14
7.5
3.4
6.4
>30
>30
>30
>30
3
4
>10
a
31 to 36 in Table 3).
Measured using shake-flask methodology with a buffer:octanol volume
ratio of 100:1. The concentration of compound in the aqueous phase before
Compounds such as 34 and 36 show 1000-fold greater
and after partitioning with octanol was determined by generic HPLC-UV
potency for Aurora B than for Aurora A, in the recombinant
enzyme assays. The cellular phenotype observed for these
Aurora B-selective compounds is identical to that described
previously for compound 1 and other Aurora kinase inhibi-
analysis. b Measured in rat serum albumin. c Compounds were dosed to male
Han Wistar rats at 5 mg/kg formulated in a mixture of 10% DMSO in
water. d >10 µM vs 1A2, 2C19, 2C9, and 2D6. e Activity against the human
ether-a-go-go-related gene (hERG)-encoded potassium channel was deter-
3
1
5
,6,9
mined using conventional or automated whole-cell electrophysiology.
tors.
The Aurora B-selective inhibitors, such as compound
3
4, inhibit phosphorylation of histone H3 on serine 10 and result
exposure, and absence of key safety pharmacology liabilities
such as inhibition of the hERG ion-channel and the major
in a failure of cell division and polyploidy without affecting
the core cell cycle timing through mitosis.30 The Aurora-B
selective compounds also retain the high specificity for Aurora
kinases demonstrated in our earlier series. For example, in a
panel of other serine/threonine and tyrosine kinases, compound
(
cytochrome P450 isoforms). Discriminating data for four
representative compounds are shown in Table 5. Overall, the
compounds in the pyrazole series have moderate lipophilicity
and excellent free drug levels and show little or no significant
inhibition of any of the five major cytochrome P450 isoforms.
Despite the presence of a basic side chain (pKa of side chain
∼9 for compounds in Tables 2 and 3), the compounds from
this class do not significantly inhibit the hERG ion-channel at
concentrations up to 30 µM when measured in a whole cell
3
4 showed very little other kinase activity (Table 4) with the
exception of Aurora C-INCENP.
A range of compounds, from this new series, spanning the
chemical diversity shown in Tables 1-3, were considered for
in ViVo evaluation based on their cellular activity, free drug

2
218 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 9
Mortlock et al.
Table 6. Phosphate Prodrugs of Compounds 18, 29, 33, and 34
Figure 2. Group mean plasma concentrations ((SE) of compound 5
and compound 34 in male and female Wistar Hannover rats (n ) 3 for
each sex) following single intravenous bolus dosing of compound 5 at
6
8 µmol/kg.
a
Phosphates 37, 38, 39, and 5 were administered to nude mice by
subcutaneous infusion using an osmotic mini-pump at 150 mg/kg/day for
48 h. Plasma concentration determined immediately at the end of the dosing
period. Average plasma concentration determined from three animals.
assay.31 The pyrazole compounds also have attractive pharma-
cokinetic properties displaying low to moderate clearance,
following intravenous bolus injections in rats, leading to high
systemic exposure in ViVo when dosed parenterally. These four
compounds also allow the in ViVo properties of compounds with
a range of Aurora A and Aurora B activities to be evaluated.
While all four compounds have high activity against Aurora B
kinase (Table 2 and 3), compounds 33 and 34 show much less
activity against Aurora A (Table 3).
The phosphate prodrugs of compounds 18, 29, 33, and 34
(
compounds 37, 38, 39, and 5, respectively) (Table 6) were
Figure 3. Pharmacodynamic activity of compounds 37, 38, 39, and
. Human SW620 colorectal tumors were established subcutaneously
found to be freely soluble in 0.3 M Tris buffer at pH 9 at a
concentration of 25 mg/mL. The phosphate compounds were
administered as described previously,15 at a dose of 150 mg/
kg/day over a period of 48 h to nude mice bearing SW620
tumors. The plasma concentrations of the corresponding active
drug compounds were measured at the end of the dosing period.
As expected, each of the phosphate prodrugs generated high
plasma concentrations of the corresponding active drug com-
pounds (Table 6). Generation of the active drug substances from
the corresponding phosphate prodrugs following a bolus iv-dose
was also monitored so as to allow the rate and extent of
conversion to be determined. For example, Figure 2 shows the
mean plasma concentration of compound 5 and compound 34
at the indicated time points after a single intravenous infusion
of 68 µmol/kg of compound 5. The peak mean plasma
concentrations (Cmax) of both compound 5 and compound 34
occurred at 2 min postdose (tmax), indicating that compound 5
was rapidly converted into the active drug compound 34. The
plasma concentrations of compound 5 declined in a multiphasic
manner following the peak concentration, and the terminal
elimination half-life of compound 34 ranged from 4.9 to 5.8 h.
There was no apparent difference in the plasma concentrations
of compound 5 or compound 34 between the sexes (data not
shown). These data, together with the data presented in Table
5
in male nude mice. Compounds (150 mg/kg/day) or vehicle were
administered as a constant infusion for 48 h using a subcutaneous
osmotic mini-pump (implanted on the side opposite to the tumor).
Following treatment the tumors were excised, disaggregated, and
analyzed by flow cytometric analysis for phospho-histone H3 (PhH3)
(percentage of PhH3 positive cells gated in G2/M phase of cell cycle)
and 4N DNA content (percentage 4N DNA cells). Data points represent
mean values from four mice per group ((SD).
the corresponding active drug substances is both rapid and
complete following parenteral administration in rats.
All four compounds gave rise to a highly significant decrease
in phosphorylation of histone-H3 in SW620 tumors (Figure 3).
In all cases inhibition of histone-H3 phosphorylation correlated
with an accumulation of cells with 4N DNA content compared
to vehicle-treated controls. These results are fully consistent with
inhibition of Aurora B kinase activity in the tumors.1
,5,23
The
in ViVo activity of the two quinazoline C-6-unsubstituted
compounds 39 and 5 was particularly noteworthy with both
compounds giving >75% inhibition of histone H3 phosphoryl-
ation in the tumor and up to a 3-fold increase in 4N DNA
2
3
content.
The robust pharmacodynamic activity of these pyrazole
compounds led to their evaluation in a tumor growth inhibition
study. In contrast to the earlier thiazole series, using the same
6
, demonstrate that conversion of the phosphate prodrugs to

SelectiVe Inhibitors of Aurora B Kinase
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 9 2219
Figure 4. In ViVo tumor growth inhibition activity of compounds 37, 38, and 5. Male nude mice bearing established human SW620 colorectal
tumor xenografts were dosed with either vehicle or a single cycle of compound (150 mg/kg/day) as a constant infusion for 48-h using a subcutaneous
3
osmotic mini-pump from day 6 to day 8. Data points represent mean tumor volume (cm ) of 10 mice ((SEM).
dosing protocol that had been employed to achieve activity
against the PD marker (infusion of the compound as its
phosphate derivative over a period of 48 h), compounds 37,
pharmacokinetics. When prepared as their phosphate derivatives,
the compounds possess excellent solubility in simple pH-
adjusted aqueous vehicles. When these phosphate prodrugs are
dosed to nude mice, using subcutaneous implanted osmotic
minipumps to effect continuous infusion, high exposure of the
corresponding drug compounds is observed. At well-tolerated
doses, inhibition of histone H3 phosphorylation is observed in
SW620 tumors along with an increase in 4N DNA content and
failure of cell division that leads to durable antitumor growth
activity. Compound 5 has been selected for clinical evaluation
and is currently in phase 1 clinical trials.
3
8, and 5 (compound 39 was not progressed) all demonstrated
significant and durable inhibition of SW620 tumor growth in
nude mice (Figure 4) at well tolerated doses. A single cycle of
dosing at 150 mg/kg/day for 2 days gave rise to a reduction in
tumor size of 50-87% at the end of the study (15 days after
cessation of dosing) compared to tumors in vehicle-treated
control animals.23 Each of the compounds tested was well
tolerated at doses where potent antitumor efficacy was ob-
2
3
served.
The pharmacodynamic effects (shown in Figure 3) for the
pyrazole compounds translate into significant antitumor activity
for each of the compounds tested. The highly Aurora B-selective
compound 5 shows equivalent or better activity to the com-
pounds retaining Aurora A kinase activity as would be predicted
from the inhibition of histone H3 phosphorylation and percent-
age 4N DNA in the tumor cells (Figure 3). The striking in ViVo
activity together with the attractive biophysical and pharmaco-
kinetic properties of the pyrazole series led us to conduct
additional efficacy and safety studies. In particular, activity at
lower doses using alternative dosing schedules have been
explored. These data, together with activity in a number of other
Experimental Section
General Methods. All experiments were carried out under an
inert atmosphere and at room temperature unless otherwise stated.
Chromatographic purifications were by flash chromatography using
40-63 µM Merck silica gel. Purification by preparative HPLC/
MS was done on a Waters LC-MS system using a Waters Symmetry
column (C18, 5 µM, 19 mm diameter, 100 mm length) using a
mixture of water and acetonitrile with 1% TFA or acetic acid
(gradient from 5% to 95%) as solvent. The purities of compounds
for biological testing were assessed by analytical LC-MS on a
Waters 996 photodiode array detector system using standard
conditions such as the following: column, Symmetry C-18; solvent
3
A, water 0.1% formic acid; solvent B, CH CN; flow rate, 2.5 mL/
2
3
human tumor xenograft models, will be reported elsewhere.
min; run time, 4.5 min; gradient, from 0 to 100% solvent B; mass
detector, micro mass ZMD. TLCs were performed on precoated
silica gel plates (Merck Art. 5715), and the resulting chromatograms
Conclusions
1
We have described a new series of pyrazole-acetanilide-
substituted quinazoline Aurora kinase inhibitors that show
superior in ViVo activity over the other series that we have
previously reported. The compounds show significant selectivity
for Aurora B kinase, as measured in assays using recombinant
enzymes, and are highly potent in cellular assays. These
compounds have a novel mechanism of action (failure of cell
division) that is distinct from classical antimitotics, such as
Taxol, which cause cells to arrest in mitosis. The compounds
show low/modest protein binding, no significant inhibition of
the major P450 enzymes or the hERG ion-channel and good
were visualized under UV light at 254 nm. H NMR spectra were
recorded on an EX 400 JEOL FT or a Bruker DPX300 NMR
spectrometer. Chemical shifts are reported as δ values (ppm)
downfield from internal TMS in appropriate organic solutions. Peak
multiplicities are expressed as follows: s, singlet; d, doublet; dd,
doublet of doublet; t, triplet; br s, broad singlet; m, multiplet.
Accurate mass spectra were recorded on a Micromass LCT +ve-
ion electrospray spectrometer, and the results agreed with the
theoretical values to within 4 ppm. NMR and mass spectra were
run on isolated intermediates and final products and are consistent
with the proposed structures. The combustion analyses (C, H, N)
were performed with a Carlo Erba EA1108 analyzer, and the results

2
220 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 9
Mortlock et al.
·1.0H O)
+
agreed with the theoretical values to within (0.4%. Water was
measured by the Karl Fischer method using a Mettler DL 18; HCl
content was determined on a Metrohm 686 by titration using silver
nitrate solution.
HRMS-ESI m/z 568.2479 [MH] . Anal. (C28
C, H, N, H O.
H F
31 2
N
7
O
4
2
2
N-(2,3-Difluorophenyl)-2-{3-[(7-{3-[(2-hydroxyethyl)(propyl)-
amino]propoxy}-6-methoxyquinazolin-4-yl)amino]-1H-pyrazol-
5-yl}acetamide (29). An analogous reaction to that described for
compound 10, but starting with 2-(propylamino)ethanol (0.700 g,
68 mmol) and heating at 85 °C for 5 h, gave compound 29 (0.650
2
-{3-[(7-{3-[4-(Hydroxymethyl)piperidin-1-yl]propoxy}-6-
methoxyquinazolin-4-yl)amino]-1H-pyrazol-5-yl}-N-phenyl-
acetamide (6). {3-[(7-{3-[4-(Hydroxymethyl)piperidin-1-yl]pro-
poxy}-6-methoxyquinazolin-4-yl)amino]-1H-pyrazol-5-yl}acetic acid
1
g, 67% yield). H NMR (DMSO, TFA): δ 8.96 (s, 1H), 8.30 (s,
4
3 (0.118 g, 0.25 mmol) was added to a solution of aniline (0.035
1H), 7.75 (m, 1H), 7.33 (s, 1H), 7.18-7.22 (m, 2H), 6.84 (s, 1H),
4.30 (m, 2H), 4.00 (s, 3H), 3.94 (s, 2H), 3.78 (m, 2H), 3.30-3.45
(m, 2H), 3.28 (m, 2H), 3.15-3.20 (m, 2H), 2.28 (m, 2H), 1.73 (m,
g, 0.37 mmol) in DMA (1 mL). 1-(3-Dimethylaminopropyl)-3-ethyl-
carbodiimide hydrochloride (0.081 g, 0.43 mmol) and 2-hydroxy-
pyridine 1-oxide (0.042 g, 0.38 mmol) were then added, and the
suspension was heated at 55 °C for 2 h. The reaction mixture was
purified directly by LC/MS preparative chromatography, eluting
with a gradient of acetonitrile in water containing 1% acetic acid.
The fractions containing product were combined and evaporated
to dryness. The oily residue was dissolved in dichloromethane (with
a drop of methanol), and the product was precipitated by the
addition of diethyl ether. The mixture was filtered, and the residue
+
2H), 0.95 (t, 3H). HRMS-ESI m/z 570.2640 [MH] . Anal.
(C28
H
33
F N O
2 7 4
2 2
·0.4H O) C, H, N, H O.
N-(2,3-Difluorophenyl)-2-{3-[(7-{3-[(2R)-2-(hydroxymethyl)-
pyrrolidin-1-yl]propoxy}quinazolin-4-yl)amino]-1H-pyrazol-5-
yl}acetamide (33). An analogous reaction to that described for
compound 31, but starting with (2R)-pyrrolidin-2-ylmethanol (0.257
1
g, 2.54 mmol), gave compound 33 (0.206 g, 60% yield). H NMR
(DMSO): δ 11.60 (br s, 7H), 10.25 (s, 1H), 8.52 (m, 2H), 7.75
(m, 1H), 7.16 (m, 4H), 6.67 (s, 1H), 4.22 (t, 2H), 3.84 (s, 2H),
3.50 (d, 2H), 3.35 (m, 1H), 3.28 (m, 1H), 3.07 (m, 1H), 2.86 (m,
1H), 2.72 (m, 1H), 2.05 (m, 2H), 1.95 (m, 1H), 1.60-1.90 (m,
was dried at 50 °C under vacuum overnight to yield compound 6
1
(0.106 g, 75% yield). H NMR (DMSO, TFA): δ 8.94 (s, 1H),
8
4
2
2
.29 (s, 1H), 7.63 (d, 2H), 7.31 (t, 3H), 7.05 (t, 1H), 6.83 (s, 1H),
.27 (t, 2H), 3.99 (s, 3H), 3.82 (s, 2H), 3.60 (d, 2H), 3.30 (m, 4H),
+
4H). HRMS-ESI m/z 538.2378 [MH] . Anal. (C H F N O ·
2
7
29
2
7
3
.97 (t, 2H), 2.27 (m, 2H), 1.89 (d, 2H), 1.65 (m, 1H), 1.44 (m,
2 2
0.8H O) C, H, N, H O.
+
H). HRMS-ESI m/z 546.2829 [MH] . Anal. (C29
35 7 4
H N O ‚0.4H
2
O)
N-(3-Fluorophenyl)-2-{3-[(7-{3-[ethyl(2-hydroxyethyl)amino]-
propoxy}quinazolin-4-yl)amino]-1H-pyrazol-5-yl}acetamide (34).
An analogous reaction to that described for compound 31, but
starting with 2-(3-{[7-(3-chloropropoxy)quinazolin-4-yl]amino}-1H-
pyrazol-5-yl)-N-(3-fluorophenyl)acetamide 48 (1.0 g, 2.2 mmol) and
N-(ethylamino)ethanol (1.07 mL, 11.0 mmol), gave compound 34
2
C, H, N, H O
N-(3-Fluorophenyl)-2-{3-[(7-{3-[4-(hydroxymethyl)piperidin-
-yl]propoxy}-6-methoxyquinazolin-4-yl)amino]-1H-pyrazol-5-
1
yl}acetamide (8). A mixture of 2-(3-{[7-(3-chloropropoxy)-6-
methoxyquinazolin-4-yl]amino}-1H-pyrazol-5-yl)-N-(3-
fluorophenyl)acetamide 44 (0.136 g, 0.28 mmol) and piperidin-4-
ylmethanol (0.115 g, 1.0 mmol) in DMA (1 mL) was heated at 90
1
(0.66 g, 59% yield). H NMR (DMSO): δ 12.31 (m, 1H), 10.39
(s, 1H), 10.15 (m, 1H), 8.51 (s, 2H), 7.62 (d, 1H), 7.35 (m, 2H),
7.16 (m, 2H), 6.90 (t, 1H), 6.78 (m, 1H), 4.29 (m, 1H), 4.20 (t,
2H), 3.76 (s, 2H), 3.45 (m, 2H), 3.30 (m, 4H), 2.61 (t, 2H), 1.89
°
C for 8 h. The reaction mixture was purified directly by LC/MS
preparative chromatography, eluting with a gradient of acetonitrile
in water containing 1% acetic acid. The fractions containing product
were combined and evaporated to dryness. The oily residue was
dissolved in dichloromethane (with a drop of methanol), and the
product was precipitated by the addition of diethyl ether. The
mixture was filtered, and the residue was dried at 50 °C under
vacuum overnight to give compound 8 (0.080 g, 57% yield). H
NMR (DMSO, TFA): δ 8.96 (s, 1H), 8.30 (s, 1H), 7.63 (m, 1H),
7
3
2
HRMS-ESI m/z 564.27307 [MH] . Anal. (C29
2
C, H, N, H O.
N-(2,3-Difluorophenyl)-2-{3-[(7-{3-[4-(hydroxymethyl)piperi-
din-1-yl]propoxy}-6-methoxyquinazolin-4-yl)amino]-1H-pyr-
azol-5-yl}acetamide (10). A mixture of 2-(3-{[7-(3-chloropropoxy)-
+
(t, 2H), 0.95 (t, 3H). HRMS-ESI m/z 508.2472 [MH] . Anal.
(C H FN O ) C, H, N.
26
30
7
3
2-[[3-({4-[(5-{2-[(2,3-Difluorophenyl)amino]-2-oxoethyl}-1H-
pyrazol-3-yl)amino]-6-methoxyquinazolin-7-yl}oxy)propyl]-
(propyl)amino]ethyl Dihydrogen Phosphate (38). (a) Di-tert-
butyldiethylphosphoramidite (0.417 mL, 1.5 mmol) was slowly
added to a solution of compound 29 (0.569 g, 1.0 mmol) in DMF
(2.5 mL) in the presence of tetrazole (0.210 g, 3.0 mmol) at ambient
temperature under argon. The mixture was stirred at ambient
temperature for 1.5 h, and cooled to -10 °C, and hydrogen peroxide
(0.134 mL of a 9.0 N solution, 1.2 mmol) was slowly added. The
resulting mixture was stirred at ambient temperature for 2 h. The
mixture was cooled to 0 °C, and a solution of sodium metabisulphite
(0.570 g, 3.0 mmol) in water (2 mL) was added. The mixture was
allowed to warm to ambient temperature and stirred for 0.5 h. The
mixture was concentrated, and then a mixture of dichloromethane
and methanol (8:2) was added. The mixture was filtered and the
residue washed with a mixture of dichloromethane and methanol.
The filtrate was concentrated in Vacuo and the residue purified by
chromatography on silica gel, eluting with a mixture of dichloro-
methane and methanol (90:10) and then with dichloromethane/
methanol/ammonia (7.0 N) (90:10:1), to give di-tert-butyl 2-
[[3-({4-[(5-{2-[(2,3-difluorophenyl)amino]-2-oxoethyl}-
1H-pyrazol-3-yl)amino]-6-methoxyquinazolin-7-yl}oxy)propyl]-
(propyl)amino]ethyl phosphate as an off-white solid (0.319 g,
1
.36 (m, 3H), 6.90 (m, 1H), 6.84 (s, 1H), 4.30 (t, 2H), 4.01 (s,
H), 3.85 (s, 2H), 3.62 (d, 2H), 3.32 (d, 2H), 3.27 (m, 2H), 2.98 (t,
H), 2.29 (m, 2H), 1.90 (d, 2H), 1.67 (m, 1H), 1.42 (m, 2H).
+
H34FN
7
O
4
2
·0.32H O)
6
-methoxyquinazolin-4-yl]amino}-1H-pyrazol-5-yl)-N-(2,3-difluo-
rophenyl)acetamide 45 (0.151 g 0.3 mmol), potassium iodide (0.10
g, 0.6 mmol), and piperidin-4-ylmethanol (0.115 g, 1.0 mmol) in
NMP (1 mL) was heated at 60 °C for 8 h. The mixture was
evaporated and the residue purified by chromatography on silica
eluting with 5% methanol in dichloromethane followed by 5%
methanolic ammonia in dichloromethane to give compound 10
1
(0.138 g, 79% yield). H NMR (DMSO, TFA): δ 8.95 (s, 1H),
8
1
.29 (s, 1H), 7.75 (m, 1H), 7.32 (s, 1H), 7.17 (m, 2H), 6.83 (s,
H), 4.29 (m, 2H), 4.00 (s, 3H), 3.93 (s, 2H), 3.60 (d, 2H), 3.30
1
(m, 4H), 3.00 (t, 2H), 2.30 (m, 2H), 1.90 (m, 2H), 1.70 (m, 1H),
42% yield). H NMR (DMSO, TFA): δ 8.95 (s, 1H), 8.29 (s, 1H),
+
1
0
.40 (m, 2H). HRMS-ESI m/z 582.2640 [MH] . Anal. (C29
.51H O) C, H, N, H O.
N-(2,3-Difluorophenyl)-2-{3-[(7-{3-[(2R)-2-(hydroxymethyl)-
H
33
2
F N
7
O ·
4
7.72 (m, 1H), 7.33 (s, 1H), 7.18 (m, 2H), 6.84 (s, 1H), 4.20-4.35
(m, 4H), 4.00 (s, 3H), 3.94 (s, 2H), 3.53 (m, 2H), 3.39 (m, 2H),
3.20 (m, 2H), 2.30 (m, 2H), 1.73 (m, 2H), 1.44 (s, 18H), 0.95 (t,
2
2
+
pyrrolidin-1-yl]propoxy}-6-methoxyquinazolin-4-yl)amino]-1H-
pyrazol-5-yl}acetamide (18). An analogous reaction to that
described for compound 10, but starting with (2R)-pyrrolidin-2-
ylmethanol (0.101 g, 1.0 mmol), gave compound 18 (0.134 g, 79%
3H). MS-ESI m/z 762.5 [MH] . (b) A solution of hydrogen chloride
in dioxane (4 M, 0.5 mL, 2.05 mmol) was added to a stirred
suspension of compound di-tert-butyl 2-[[3-({4-[(5-{2-[(2,3-di-
fluorophenyl)amino]-2-oxoethyl}-1H-pyrazol-3-yl)amino]-6-meth-
oxyquinazolin-7-yl}oxy)propyl](propyl)amino]ethyl phosphate (0.316
g, 0.41 mmol) in dichloromethane (10 mL). The mixture was stirred
at room temperature for 15 h and then heated to 40-45 °C for 2 h.
The mixture was filtered, and the solid was dissolved in a mixture
1
yield). H NMR (DMSO, TFA): δ 8.95 (s, 1H), 8.29 (s, 1H), 7.75
(m, 1H), 7.32 (s, 1H), 7.16 (m, 2H), 6.84 (s, 1H), 4.30 (m, 2H),
4
3
.00 (s, 3H), 3.94 (s, 2H), 3.70-3.85 (m, 1H), 3.52-3.70 (m, 4H),
.15-3.30 (m, 2H), 2.25-2.35 (m, 2H), 1.75-2.20 (m, 4H).

SelectiVe Inhibitors of Aurora B Kinase
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 9 2221
of dichloromethane and methanol and then evaporated. The residue
was triturated with dichloromethane and then dried under vacuum
described for compound 38, but starting with di(tert-butyl) {(2R)-
1-[3-({4-[(5-{2-[(2,3-difluorophenyl)amino]-2-oxoethyl}-1H-pyra-
zol-3-yl)amino]-quinazolin-7-yl}oxy)propyl]pyrrolidin-2-yl}-
methyl phosphate (0.355 g, 0.49 mmol), gave compound 39 as the
dihydrochloride salt (0.355 g, 100% yield) as a pale yellow solid.
at 50 °C for 6 h to give compound 38 as the dihydrochloride salt
1
(
0.30 g, 100% yield). H NMR (DMSO, TFA): δ 8.96 (s, 1H),
8
1
.31 (s, 1H), 7.75 (m, 1H), 7.36 (s, 1H), 7.20 (m, 2H), 6.84 (s,
H), 4.31 (t, 2H), 4.24 (m, 2H), 4.01 (s, 3H), 3.94 (s, 2H), 3.50
1
H NMR (DMSO): δ 12.00 (br s, 1H), 10.40 (s, 1H), 8.93 (s, 1H),
(
0
m, 2H), 3.38 (m, 2H), 3.19 (m, 2H), 2.32 (m, 2H), 1.74 (m, 2H),
.95 (t, 3H). HRMS-ESI m/z 650.22998 [MH] . (c) Compound 38
8.82 (d, 1H), 7.68 (m, 1H), 7.40-7.50 (m, 2H), 7.15-7.25 (m,
2H), 6.75 (s, 1H), 4.35 (t, 2H), 4.10-4.30 (m, 1H), 3.92 (s, 2H),
3.81 (m, 1H), 3.55-3.70 (m, 2H), 3.27 (m, 1H), 3.18 (m, 1H),
2.10-2.35 (m, 4H), 1.85-2.10 (m, 3H), 1.75-1.85 (m, 1H).
+
was also prepared as the free base according to the following
method: compound 38·dihydrochloride (10 g, 13 mmol) was
dissolved in methanol (300 mL), and cyclohexene oxide (12.7 g,
+
HRMS-ESI m/z 618.20380 [MH] .
1
30 mmol) was added. The solution was stirred at ambient
2-[[3-({4-[(5-{2-[(3-Fluorophenyl)amino]-2-oxoethyl}-1H-pyra-
zol-3-yl)amino]-quinazolin-7-yl}oxy)propyl](ethyl)amino]ethyl
Dihydrogen Phosphate (5). (a) An analogous reaction to that
described for compound 38, but starting with compound 34 (0.62
g, 1.22 mmol), gave di(tert-butyl) 2-[[3-({4-[(5-{2-[(3-fluoro-
phenyl)amino]-2-oxoethyl}-1H-pyrazol-3-yl)amino]-quinazolin-7-yl}oxy)-
propyl](ethyl)amino]ethyl phosphate (0.539 g, 63% yield) as a pale
yellow solid which was used in the next step without further
characterization. (b) An analogous reaction to that described for
compound 38, but starting with di(tert-butyl) 2-[[3-({4-[(5-{2-[(3-
fluorophenyl)amino]-2-oxoethyl}-1H-pyrazol-3-yl)amino]-quinazo-
lin-7-yl}oxy)propyl](ethyl)amino]ethyl phosphate (0.539 g, 0.77
mmol), gave compound 5 as the dihydrochloride salt (0.504 g, 99%
temperature for 48 h, during which time a white solid precipitated.
The mixture was diluted with diethyl ether (100 mL), and the solid
was recovered by filtration, washed with ether, and dried in Vacuo
to give compound 38 as the free base (7.65 g, 88% yield) as a
1
light yellow powder. H NMR (DMSO, TFA): δ 8.96 (s, 1H), 8.30
(s, 1H), 7.74 (m, 1H), 7.34 (s, 1H), 7.19 (m, 2H), 6.84 (s, 1H),
4
2
.31 (m, 2H), 4.24 (m, 2H), 4.01 (s, 3H), 3.94 (s, 2H), 3.51 (m,
H), 3.38 (m, 2H), 3.18 (m, 2H), 2.29 (m, 2H), 1.73 (m, 2H), 0.96
+
(t, 3H). MS-ESI m/z 650 [MH] . Anal. (C28
H F
34 2
N O
7
7
P·1.04H O·
2
0
.03Et
(2R)-1-[3-({4-[(5-{2-[(2,3-Difluorophenyl)amino]-2-oxoethyl}-
H-pyrazol-3-yl)amino]-6-methoxyquinazolin-7-yl}oxy)propyl]-
2
O) C, H, N.
{
1
1
pyrrolidin-2-yl}methyl Dihydrogen Phosphate (37). (a) An
analogous reaction to that described for compound 38, but starting
with compound 18 (1.1 g, 1.9 mmol), gave di-tert-butyl {(2R)-1-
yield). H NMR (DMSO): δ 11.98 (s, 1H), 10.79 (s, 1H), 8.93 (s,
1H), 8.83 (d, 1H), 7.65 (d, 1H), 7.47 (d, 1H), 7.38 (m, 3H), 6.89
(t, 1H), 6.74 (s, 1H), 4.32 (t, 2H), 4.28 (m, 2H), 3.85 (s, 2H), 3.42
(m, 2H), 3.34 (m, 2H), 3.27 (q, 2H), 2.29 (m, 2H), 1.28 (t, 3H).
[3-({4-[(5-{2-[(2,3-difluorophenyl)amino]-2-oxoethyl}-1H-pyrazol-
+
3
-yl)amino]-6-methoxyquinazolin-7-yl}oxy)propyl]pyrrolidin-2-
MS-ESI m/z 587.8 [MH] . (c) An analogous reaction to that
1
yl}methyl phosphate (0.453 g, 31% yield). H NMR (DMSO,
TFA): δ 10.24 (s, 1H), 10.15 (s, 1H), 8.44 (s, 1H), 7.98 (s, 1H),
7
2
1
described for compound 38, but starting with the dihydrochloride
salt of compound 5 (10 g, 14 mmol), yielded the free base of
1
.72 (t, 1H), 7.19 (m, 2H), 7.13 (s, 1H), 6.83 (s, 1H), 4.17 (br s,
H), 3.93 (s, 3H), 3.85 (s, 1H), 3.77 (m, 1H), 3.56 (t, 1H), 3.54 (t,
H), 3.08 (t, 1H), 2.94 (m, 1H), 2.66 (m, 1H), 2.47 (m, 1H), 2.20
compound 5 as a pale yellow solid (6.54 g, 73% yield). H NMR
(DMSO): δ 10.53 (s, 1H), 8.57 (s, 1H), 8.54 (d, 1H), 7.62 (d, 1H),
7.37 (m, 2H), 7.27 (s, 1H), 7.21 (d, 1H), 6.88 (m, 1H), 6.65 (s,
1H), 4.27 (t, 2H), 4.05 (m, 2H), 3.75 (s, 2H), 3.24 (m, 2H), 3.21
(q, 1H), 1.94 (m, 2H), 1.86 (m, 1H), 1.69 (m, 2H), 1.60 (m, 1H),
+
1
.37 (s, 9H), 1.36 (s, 9H). MS-ESI m/z 758.5 [MH] . (b) An
(t, 2H), 3.13 (q, 2H), 2.18 (m, 2H), 1.24 (t, 3H). MS-ESI m/ z 588
+
analogous reaction to that described for compound 38 but starting
with di-tert-butyl {(2R)-1-[3-({4-[(5-{2-[(2,3-difluorophenyl)amino]-
2
[MH] . Anal. (C26
H31FN O
7 6
2 2
P·3.0H O) C, H, N, H O.
(5-((7-(3-Chloropropoxy)-6-methoxy-quinazolin-4-yl)amino)-
1H-pyrazol-3-yl)acetic Acid (41). A mixture of N′-(5-(3-chloro-
propoxy)-2-cyano-4-methoxyphenyl)-N,N-dimethylimidoform-
amide 4915 (14.78 g, 50 mmol) in acetic acid (40 mL) was heated
at reflux with (5-amino-1H-pyrazol-3-yl)acetic acid 50 (8.1 g, 57.5
mmol) for 1.5 h. The reaction mixture was cooled to ambient
temperature, water (250 mL) was added to the mixture, and the
solid was recovered by suction filtration. The solid was washed
first with water followed by ethyl acetate and finally with diethyl
ether and then dried in Vacuo at 50 °C to yield compound 41 as a
-oxoethyl}-1H-pyrazol-3-yl)amino]-6-methoxyquinazolin-7-yl}-
oxy)propyl]pyrrolidin-2-yl}methyl phosphate (0.445 g, 0.59 mmol)
gave compound 37 as the dihydrochloride salt (0.440 g, 94% yield).
1
H NMR (DMSO, TFA): δ 8.94 (s, 1H), 8.31 (s, 1H), 7.73 (m,
1
H), 7.40 (s, 1H), 7.19 (m, 2H), 6.83 (s, 1H), 4.31 (t, 2H), 4.20
(m, 2H), 4.01 (s, 3H), 3.94 (s, 2H), 3.82 (m, 1H), 3.70 (m, 1H),
3
1
6
.60 (m, 1H), 3.31 (m, 1H), 3.23 (m, 1H), 2.32 (m, 2H), 2.19 (m,
H), 2.04 (m, 1H), 1.95 (m, 1H), 1.85 (m, 1H). HRMS-ESI m/z
48.21417 [MH] . (c) An analogous reaction to that described for
+
1
compound 38, but starting with the dihydrochloride salt of
compound 37 (9.0 g, 0.012 mol) and recrystallizing the crude
product from 20% aqueous tetrahydrofuran gave the free base of
yellow solid (13.6 g, 69% yield). H NMR (DMSO, TFA): δ 8.95
(s, 1H), 8.28 (s, 1H), 7.32 (s, 1H), 6.80 (s, 1H), 4.33 (m, 2H), 4.00
(s, 3H), 3.83 (m, 2H), 3.74 (s, 2H), 2.40-2.50 (m, 2H). MS-ESI
1
+
compound 37 as a pale yellow solid (6.3 g, 81% yield). H NMR
m/z 392, 394 [MH] .
(
DMSO): δ 10.30 (s, 1H), 10.20 (s, 1H), 8.50 (s, 1H), 8.00 (s,
(3-{[7-(3-Chloropropoxy)quinazolin-4-yl]amino}-1H-pyrazol-
5-yl)acetic Acid (42). 4-Chloro-7-(3-chloropropoxy)quinazoline 53
(2.5 g, 9.72 mmol) and (5-amino-1H-pyrazol-3-yl)acetic acid 50
(1.37 g, 9.72 mmol) were combined in DMA (25 mL). A solution
of 4 M HCl in dioxane (1.25 mL, 4.8 mmol) was added and the
mixture heated to 90 °C for 40 min. The solution was cooled to
room temperature, diluted with water (250 mL), and filtered through
celite. The acidic solution was basified to pH 4.9 and the yellow
powder filtered. (At pH 3, a red solid precipitated which was
isolated, suspended in water, and basified to pH 12. Careful
adjustment back to pH 4.8 resulted in the precipitation of a yellow
powder, which was combined with the first crop). The solid was
washed with diethyl ether and dried over phosphorus pentoxide to
1
4
3
H), 7.70-7.80 (m, 1H), 7.20-7.30 (m, 3H), 6.70 (s, 1H), 4.30-
.40 (m, 2H), 4.10-4.20 (m, 1H), 3.90 (s, 3H), 3.80 (s, 2H), 3.70-
.75 (m, 1H), 3.40-3.50 (m, 1H), 3.30-3.35 (m, 1H), 3.20-3.25
(
2
m, 1H), 3.05-3.15 (m, 1H), 2.90-3.00 (m, 1H), 2.10-2.20 (m,
H), 1.90-2.00 (m, 1H), 1.70-1.80 (m, 3H). MS-ESI m/z 648
+
[MH] . Anal. (C28
H F
32 2
N
7
7 2 2
O P·2.3H O) C, H, N, H O.
{(2R)-1-[3-({4-[(5-{2-[(2,3-Difluorophenyl)amino]-2-oxoethyl}-
1
2
H-pyrazol-3-yl)amino]-quinazolin-7-yl}oxy)propyl]pyrrolidin-
-yl}methyl Dihydrogen Phosphate (39). (a) An analogous
reaction to that described for compound 38, but starting with
compound 33 (0.437 g, 0.81 mmol), gave di(tert-butyl) {(2R)-1-
[3-({4-[(5-{2-[(2,3-difluorophenyl)amino]-2-oxoethyl}-1H-pyrazol-
1
3
-yl)amino]-quinazolin-7-yl}oxy)propyl]pyrrolidin-2-yl}methyl phos-
give compound 42 as a pale orange solid (2.88 g, 82% yield). H
1
phate (0.355 g, 60% yield) as a pale yellow solid. H NMR
NMR (DMSO): δ 12.60 (br s, 2H), 10.78 (br s, 1H), 8.65 (s, 1H),
8.60 (d, 1H), 7.26 (d, 1H), 7.22 (s, 1H), 6.67 (s, 1H), 4.28 (t, 2H),
3.83 (t, 2H), 3.67 (s, 2H), 2.24 (m, 2H). MS-ESI m/z 362, 364
(
1
DMSO): δ 12.30 (br s, 1H), 10.20 (s, 2H), 8.50 (s, 2H), 7.68 (m,
H), 7.10-7.20 (m, 4H), 6.78 (br s, 1H), 4.15 (t, 2H), 3.80 (m,
H), 3.65 (m, 1H), 3.10 (m, 1H), 2.93 (m, 1H), 2.64 (m, 1H), 2.19
+
3
[MH] .
(
1
m, 1H), 1.80-1.95 (m, 3H), 1.68 (m, 2H), 1.60 (m, 1H), 1.33 (s,
{3-[(7-{3-[4-(Hydroxymethyl)piperidin-1-yl]propoxy}-6-meth-
oxyquinazolin-4-yl)amino]-1H-pyrazol-5-yl}acetic Acid (43). A
+
8H). MS-ESI m/z 730 [MH] . (b) An analogous reaction to that

2
222 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 9
Mortlock et al.
mixture of (3-{[7-(3-chloropropoxy)-6-methoxyquinazolin-4-yl]-
amino}-1H-pyrazol-5-yl)acetic acid 41 (7.83 g, 20 mmol) and
aniline (1.15 g, 8.95 mmol) was added to a suspension of (3-{[7-
(3-chloropropoxy)quinazolin-4-yl]amino}-1H-pyrazol-5-yl)acetic acid
42 (2.70 g, 7.46 mmol) in pyridine (30 mL) and the mixture cooled
to 0 °C. Phosphorus oxychloride (1.14 g, 7.46 mmol) was added
dropwise and the mixture stirred at 0 °C for 1 h. The mixture was
warmed to ambient temperature, more phosphorus oxychloride (0.5
mL) was added, and then the mixture was stirred for a further 4.5
h. The mixture was diluted with ethyl acetate:ether (100 mL: 37
mL) and stirred for 18 h. The mixture was filtered and the residue
suspended in water and neutralized with ammonium hydroxide (7%,
15 mL). The resultant yellow suspension was filtered, washed with
water, and dried over phosphorus pentoxide to give compound 47
4-(hydroxymethyl)piperidine (8.05 g, 70 mmol) in DMA (30 mL)
was heated at 100 °C for 2 h. The solvent was evaporated and the
residue triturated with a mixture of dichloromethane and ethyl
acetate (1:1). The residue was dissolved in a mixture of dichloro-
methane and methanol, ethanolic HCl (7.0 N) (10 mL, 70 mmol)
was added, and the mixture was evaporated. Methanol (200 mL)
was added to the residue, and the mixture was stirred for 0.5 h.
The mixture was reduced in volume and dichloromethane added.
The resultant solid was recovered by filtration and dried to yield
1
compound 43 (6.5 g, 60% yield) as a yellow solid. H NMR
1
(DMSO, TFA): δ 8.96 (s, 1H), 8.31 (s, 1H), 7.37 (s, 1H), 6.80 (s,
as an orange powder (3.15 g, 89% yield). H NMR (DMSO): δ
1
3
H), 4.31 (m, 2H), 4.00 (s, 3H), 3.75 (s, 2H), 3.59 (d, 2H), 3.24-
.30 (m, 4H), 2.97 (t, 2H), 2.35 (m, 2H), 1.86-1.91 (m, 2H), 1.68
10.64 (br s, 1H), 10.27 (s, 1H), 8.60 (s, 1H), 8.55 (d, 1H), 7.70 (m,
1H), 7.20 (m, 6H), 6.68 (s, 1H), 4.27 (t, 2H), 3.83 (m, 4H), 2.25
(m, 2H). MS-ESI m/z 473, 475 [MH] .
+
(m, 1H), 1.47 (m, 2H).
2
-(3-{[7-(3-Chloropropoxy)-6-methoxyquinazolin-4-yl]amino}-
2-(3-{[7-(3-Chloropropoxy)quinazolin-4-yl]amino}-1H-pyra-
zol-5-yl)-N-(3-fluorophenyl)acetamide (48). Pentafluorophenyl
trifluoroacetate (23.25 g, 83 mmol) was added dropwise to a
solution of (3-{[7-(3-chloropropoxy)quinazolin-4-yl]amino}-1H-
pyrazol-5-yl)acetic acid 42 (15.0 g, 41 mmol) and pyridine (6.7
mL, 83 mmol) in DMF (150 mL) with cooling to maintain the
solution temperature at <23 °C. The solution was stirred at ambient
temperature for 30 min before addition of 3-fluoroaniline (9.22 g,
83 mmol). The reaction was stirred for 2.5 h at ambient temperature,
a further portion of 3-fluoroaniline (2 mL) was added, and the
mixture was heated at 90 °C for 3 h. The reaction mixture was
poured into dilute hydrochloric acid (0.1 M) and ice (ca. 500 mL),
and the resultant solid was filtered, washed with water and then
diethyl ether, and then dried to give compound 48 (17.7 g, 94%
1
H-pyrazol-5-yl)-N-(3-fluorophenyl)acetamide (44). (5-((7-(3-
Chloropropoxy)-6-methoxyquinazolin-4-yl)amino)-1H-pyrazol-3-
yl) acetic acid 41 (7.83 g, 20 mmol) in DMA (78 mL) was reacted
with 3-fluoroaniline (2.44 g, 22 mmol) in the presence of EDC
(
4.2 g, 22 mmol), 2-hydroxypyridine 1-oxide (2.22 g, 20 mmol),
and diisopropylethylamine (2.8 g, 22 mmol) at 50 °C for 1.7 h.
The mixture was evaporated in Vacuo and the residue triturated
with water. The residue was purified by chromatography on silica
gel, eluting with dichloromethane:methanol (95:3 to 85:15) to give
compound 44 (4.5 g, 46% yield) as a beige solid. 1H NMR
(
(
2
DMSO): δ 8.47 (s, 1H), 8.02 (s, 1H), 7.60-7.68 (m, 1H), 7.40
m, 2H), 7.20-7.30 (s, 1H), 6.88 (m, 1H), 6.84 (s, 1H), 4.27 (m,
H), 3.96 (s, 3H), 3.84 (m, 2H), 3.78 (s, 2H), 2.26 (m, 2H). MS-
+
1
ESI m/z 485.6 [MH] .
-(3-{[7-(3-Chloropropoxy)-6-methoxyquinazolin-4-yl]amino}-
H-pyrazol-5-yl)-N-(2,3-difluorophenyl)acetamide (45). 3-{[7-
3-Chloropropoxy)-6-methoxyquinazolin-4-yl]amino}-1H-pyrazol-
-yl)aceti c acid 41 (3.91 g, 10 mmol) was suspended in pyridine
20 mL) in the presence of 2,3-difluoroaniline (1.55 g, 12 mmol)
yield) as a brown solid. H NMR (DMSO): δ 12.50 (br s, 1H),
2
10.42 (s, 1H), 8.59 (s, 1H), 8.54 (d, 1H), 7.62 (m, 1H), 7.35 (m,
2H), 7.24 (m, 1H), 7.19 (m, 1H), 6.90 (m, 1H), 6.67 (br s, 1H),
4.28 (t, 2H), 3.84 (t, 2H), 3.76 (s, 2H), 2.27 (quintet, 2H). MS-ESI
1
(
5
(
+
m/z 455 [MH] .
7-(3-Hydroxypropoxy)quinazolin-4(3H)-one (52). Sodium hy-
dride (14.6 g, 365 mmol) was added at 0 °C to a solution of 1,3-
propanediol (27.8 g, 365 mmol) in DMA (70 mL). 7-Fluoroquinazo-
lin-4(3H)-one 5121 (10 g, 60.9 mmol) was added portionwise and
the reaction mixture heated at 60 °C and then at 110 °C for 3 h.
The reaction was cooled to 0 °C, quenched with water (280 mL),
and adjusted to pH 5.9. The resulting suspension was filtered,
washed with water and then ether, and dried over phosphorus
pentoxide to give compound 52 as a white powder (12.41 g, 92%
under argon at 0 °C. Phosphorus oxychloride (1.53 g, 10 mmol) in
ethyl acetate (2 mL) was slowly added at 0 °C, and the resulting
mixture was allowed to warm to ambient temperature over 1.5 h.
The reaction mixture was diluted with ethyl acetate (150 mL) and
diethyl ether (50 mL), resulting in the precipitation of a red solid.
The solid was recovered by suction filtration, dried, and resuspended
in water (100 mL). The mixture was cooled to 0 °C and the pH
adjusted to ∼7 by addition of 1.5 N aqueous ammonium hydroxide
solution. After 15 min stirring, the solid was recovered, dried, and
purified by chromatography on silica gel. Elution with dichloro-
methane:methanol (95/5) and then with dichloromethane:methanolic
1
yield). H NMR (DMSO): δ 11.90 (br s, 1H), 8.04 (s, 1H), 8.00
(d, 1H), 7.10 (m, 2H), 4.17 (t, 2H), 3.58 (t, 2H), 1.92 (m, 2H).
+
MS-ESI m/z 221 [MH] .
1
ammonia (95:2) gave compound 45 (2.55 g, 50% yield). H NMR
4-Chloro-7-(3-chloropropoxy)quinazoline (53). 7-(3-Hydroxy-
propoxy)quinazolin-4(3H)-one 52 (10.5 g, 47.7 mmol) and thionyl
chloride (100 mL, 137 mmol) were combined. DMA (1 mL) was
added and the reaction mixture heated to 85 °C for 1 h. The mixture
was cooled to room temperature, diluted with toluene, and
evaporated to dryness. This was repeated until all the thionyl
chloride was removed. The residue was dissolved in dichloro-
methane and washed with a saturated sodium bicarbonate solution.
The aqueous layer was extracted with dichloromethane. The
organics were combined, dried (magnesium sulfate), and concen-
trated to leave a yellow solid. Trituration with diethyl ether removed
a less soluble impurity, and the ether filtrate was concentrated to
(
(
3
[
DMSO, TFA): δ 8.94 (s, 1H), 8.28 (s, 1H), 7.73 (m, 1H), 7.33
s, 1H), 7.15-7.22 (m, 1H), 6.84 (s, 1H), 4.30 (m, 2H), 4.00 (s,
H), 3.94 (s, 2H), 3.84 (m, 2H), 2.30 (m, 2H). MS-ESI m/z 503.9
+
MH] .
2-(3-{[7-(3-Chloropropoxy)-6-methoxyquinazolin-4-yl]amino}-
1
H-pyrazol-5-yl)-N-(3,5-difluorophenyl)acetamide (46). A sus-
pension of 3-{[7-(3-chloropropoxy)-6-methoxyquinazolin-4-yl]-
amino}-1H-pyrazol-5-yl)acetic acid 41 (3.91 g, 10 mmol) in DMF
20 mL) was reacted with 3,5-difluoroaniline (1.42 g, 11 mmol) in
the presence of 1-(3-dimethylaminopropyl)-3-ethylcarbodimide
hydrochloride (2.01 g, 10.5 mmol) and 2-hydroxypyridine 1-oxide
(
1
(
1.11 g, 10 mmol) at 60 °C for 1.75 h. The solvent was evaporated
leave compound 53 as an off-white solid (8.5 g, 70% yield). H
in Vacuo, and the residue was triturated twice with water. The
resulting paste was dissolved in a mixture of dichloromethane:
methanol (80:20), adsorbed onto silica gel and purified by chro-
matography on silica gel, eluting with dichloromethane:methanol
NMR (DMSO): δ 13.25 (br s, 1H), 8.34 (s, 1H), 8.06 (d, 1H),
7.17 (m, 2H), 4.21 (t, 2H), 3.83 (t, 2H), 2.23 (m, 2H). MS-ESI m/z
+
257, 259 [MH] .
3-Amino-5-(carboxymethyl)-1H-pyrazole-4-carboxylic Acid
(55). A suspension of 5-amino-3-(cyanomethyl)-1H-pyrazole-4-
carbonitrile 54 (12.05 g; 82 mmol) in an aqueous solution of sodium
hydroxide (10 M, 120 mL) was heated at 95f100 °C for 14 h.
The resultant solution was cooled to 5f10 °C in an ice/water bath
and then made acidic (pH ∼ 3) by the addition of concentrated
hydrochloric acid. The resultant gray solid was removed by
filtration, washed with water, and then dried in air with suction for
(
95:5 to 85:15) to yield compound 46 (2.45 g, 49% yield) as a
1
beige solid. H NMR (DMSO): δ 8.47 (s, 1H), 8.02 (s, 1H), 7.36
m, 2H), 7.20 (s, 1H), 6.94 (t, 1H), 6.84 (s, 1H), 4.27 (m, 2H),
.96 (s, 3H), 3.83 (m, 2H), 3.79 (s, 2H), 2.27 (m, 2H). MS-ESI
(
3
+
m/z 503.5, 505.5 [MH] .
-(3-{[7-(3-Chloropropoxy)quinazolin-4-yl]amino}-1H-pyr-
azol-5-yl)-N-(2,3-difluorophenyl)acetamide (47). 2,3-Difluoro-
2

SelectiVe Inhibitors of Aurora B Kinase
5 min. The residue was washed with ethyl acetate and then dried
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 9 2223
1
and compound 34 by solid-phase extraction followed by high-
performance liquid chromatography with tandem mass spectrometric
detection (HPLC-MS-MS). Pharmacokinetic analysis of the plasma
concentration data was performed using the pharmacokinetic data
analysis program WinNonLin Professional (Pharsight Corporation,
version 3.1).
at 50 °C under vacuum for 4 h to leave compound 55 (14 g, 84%
1
yield). H NMR (DMSO): δ 6.5-5.25 (m, 1H), 3.55 (s, 2H).
(5-Amino-1H-pyrazol-3-yl)acetic Acid (50). 3-Amino-5-(car-
boxymethyl)-1H-pyrazole-4-carboxylic acid 0.5HCl 55 (14 g, 69
mmol) was suspended in water (280 mL), and the mixture was
heated to reflux for 5 h. The mixture was allowed to cool to room
temperature and then filtered. The orange-brown filtrate was
evaporated to dryness, and the residue was triturated first with ethyl
acetate and then with diethyl ether and then dried over phosphorus
pentoxide under vacuum at 50 °C overnight to give compound 50
Acknowledgment. We would like to acknowledge the
excellent technical expertise of the scientists at AstraZeneca,
in particular, Chris Halsall for help with elemental analysis,
Madeleine Vickers for mass spectra analysis, Linette Ruston
for physicochemical assessment, and Chris Pollard for helpful
discussion concerning hERG.
1
(
2
10.5 g, 95% yield). H NMR (DMSO): δ 5.88 (s, 1H), 3.70 (s,
H).
Biological Evaluation. IC50 values reported are average figures
Supporting Information Available: Experimental details and
data for compounds 7, 9, 11-17, 19-28, 30-32, 35, 36; elemental
analyses for all target compounds of this study. This material is
available free of charge via the Internet at http://pubs.acs.org.
of at least two independent measurements.
In Vitro Kinase Assays. Drug compounds were incubated with
Aurora A (Upstate Cell Signaling Solutions cat. no. 14-511) or
Aurora B-INCENP and the substrate peptide. Twenty microliters
of reaction mix (25 mM Tris-HCl, 12.7 mM KCl, 2.5 mM NaF,
References
0
.6 mM dithiothreitol, 6.25 mM MnCl
2
, 7.5µM ATP for Aurora A
(1) Keen, N.; Taylor, S. Aurora-kinase inhibitors as anticancer agents.
assay and 15 µM for Aurora B-INCENP assay, 6.25 µM peptide
substrate (biotinyl-Ahx-tetra (LRRWSLG); Bachem) containing 0.2
Nature ReV. Cancer 2004, 4 (12), 927-936.
(
2) Andrews, P. D. Aurora kinases: shining lights on the therapeutic
horizon? Oncogene 2005, 24, 5005-5015.
33
µCi [γ P]ATP (Amersham Pharmacia, specific activity g 2500
Ci/mmol)) was then added to all test wells to start the reaction (to
give a final volume of 50 µL). Plates were incubated at room
temperature for 60 min, reactions were stopped by addition of 100
µL of 20% v/v orthophosphoric acid, and peptide substrate was
captured on positively charged nitrocellulose P30 filtermat (What-
man) using a 96-well plate harvester (TomTek) and then assayed
for incorporation of 33P with a Beta plate counter.
(
3) Expression of Aurora C seems to be restricted to the testes where it
may play an important role in meiosis. See: (a) Kimura, M.; Matsuda,
Y.; Yoshioka, T.; Okano, Y. Cell cycle-dependent expression and
centrosome localization of a third human aurora/Ipl1-related protein
kinase, AIK3. J. Biol. Chem. 1999, 274 (11), 7334-7340. (b) Li,
X.; Sakashita, G.; Matsuzaki, H.; Sugimoto, K.; Kimura, K.; Hanaoka,
F. et al. Direct association with inner centromere protein (INCENP)
activates the novel chromosomal passenger protein, Aurora-C. J. Biol.
Chem. 2004, 279, 47201-47211. (c) Sasai, K.; Katayama, H.;
Stenoien, D. L.; Fujii, S.; Honda, R.; Kimura, M.; et al. Aurora-C
kinase is a novel chromosomal passenger protein that can complement
Aurora-B kinase function in mitotic cells. Cell Motil. Cytoskeleton
Determination of Phospho Histone H3 Suppression in SW620
Tumor Cells. Carried out as described in reference 15.
In ViWo Studies. All animal experiments were conducted in full
accordance with the UK Home Office Animal (Scientific Proce-
dures) Act 1986. Male Swiss nu/nu mice (AstraZeneca, UK) were
housed in negative pressure isolators (PFI Systems Ltd, Oxon, UK).
SW620 tumor xenografts were established in 8-12-week-old mice
2
004, 59 (4), 249-263.
(
4) Matthews, N.; Visintin, C.; Hartzoulakis, B.; Jarvis, A.; Selwood,
D. L. Aurora A and B kinases as targets for cancer: will they be
selective for tumours? Expert ReV. Anticancer Ther. 2006, 6 (1),
109-120.
6
by injecting 1 × 10 tumor cells subcutaneously (200 µL volume;
(5) Ditchfield, C.; Johnson, V. L.; Tighe, A.; Ellston, R.; Haworth, C.;
Johnson, T.; Mortlock, A.; Keen, N.; Taylor, S. S. Aurora B couples
chromosome alignment with anaphase by targeting BubR1, Mad2,
and Cenp-E to kinetochores. J. Cell. Biol. 2003, 161, 267-280.
in serum-free media containing 50% Matrigel; Becton Dickinson,
UK) on the left dorsal flank. Animals were randomized into
treatment groups when tumors became palpable. Compounds (HCl
salts) were prepared in Tris buffer pH 9 and administered as a
continuous infusion for up to 48 h via Alzet osmotic minipumps
(
6) Hauf, S.; Cole, R. W.; LaTerra, S.; Zimmer, C.; Schnapp, G.; Walter,
R.; Heckel, A.; Van, Meel, J.; Rieder, C. L.; Peters, J.-M. The small
molecule Hesperadin reveals a role for Aurora B in connecting
kinetochore-microtubule attachment and in maintaining the spindle
assembly checkpoint. J. Cell Biol. 2003, 161, 281-294.
7) Yang, H.; Burke, T.; Dempsey, J.; Diaz, B.; Collins, E.; Toth, J.;
Beckmann, R. Ye, X. Mitotic requirement for aurora A kinase is
bypassed in the absence of aurora B kinase. FEBS Lett. 2005, 579,
3385-3391.
(Model 2001D, http://www.alzet.com/) implanted subcutaneously
on the right-hand side of the animal, distal to the tumor. Tumors
were measured up to three times per week with calipers, tumor
volumes were calculated, and the data were plotted using the
geometric mean for each group versus time. Tumor volume and
tumor growth inhibition were calculated as described in reference
(
(8) Girdler, F.; Gascoigne, K. E.; Eyers, P. A.; Hartmuth, S.; Crafter,
C.; Foote, K. M.; Keen, N. J.; Taylor, S. S. Validating Aurora B as
an anti-cancer drug target. J. Cell Sci. 2006, 119, 3664-3675.
23. Tumor tissue was excised and frozen at the end of dosing and
assessed for pharmacodynamic effects using flow cytometric
analysis. Cell suspensions were prepared from the snap-frozen
tumors by using an automated tissue disaggregation system (Medi-
machine; BD Biosystems) and fixed in 80% ethanol for a minimum
of 24 h. Once fixed, disaggregated tumors were prepared for DNA
content and PhH3 (ser10) assessment using a previously described
staining protocol and analyzed on a FACSCalibur flow cytometer
(
9) Harrington, E. A.; Bebbington, D.; Moore, J.; Rasmussen, R. K.;
Ajose-Adeogun, A. O.; Nakayama, T.; Graham, J. A.; Demur, C.;
Hercend, T.; Diu-Hercend, A.; Su, M.; Golec, J. M. C.; Miller, K.
M. VX-680, a potent and selective small-molecule inhibitor of the
Aurora kinases, suppresses tumour growth in ViVo. Nat. Med. 2004,
10, 262-267.
(
10) Mortlock, A. A.; Keen, N. J.; Jung, F. H.; Heron, N. M.; Foote, K.
M.; Wilkinson, R.; Green, S. Progress in the Development of
Selective Inhibitors of Aurora kinases. Curr. Top. Med. Chem. 2005,
5 (8), 807-821.
(
Becton Dickinson, Franklin Lakes, NJ).23
Pharmacokinetic Studies. Male and female Wistar-Hannover
rats (substrain crl:WI[Glx/BRL/Han]IGSBR; Charles River, UK)
were multiple-housed in standard conditions until dosed. Three
animals of each sex received compound 5 (68 µmol/kg) formulated
in 3.3% mannitol in phosphate buffer administered intravenously
(11) Fancelli, D.; Moll, J. Inhibitors of Aurora kinases for the treatment
of cancer. Expert Opin. Ther. Pat. 2005, 15 (9), 1169-1182.
(
12) Fancelli, D.; Berta, D.; Bindi, S.; Cameron, A.; Cappella, P.;
Carpinelli, P.; Catana, C.; Forte, B.; Giordano, P.; Giorgini, M. L.;
Mantegani, S.; Marsiglio, A.; Meroni, M.; Moll, J.; Pittala, V.;
Roletto, F.; Severino, D.; Soncini, C.; Storici, P.; Tonani, R.; Varasi,
M.; Vulpetti, A.; Vianello, P. Potent and Selective Aurora Inhibitors
Identified by the Expansion of a Novel Scaffold for Protein Kinase
Inhibition. J. Med. Chem. 2005, 48 (8), 3080-3084.
(bolus injection) via a tail vein at 10 mL/kg/min by an infusion
pump. Approximately 0.25 mL of blood was then collected into
tubes containing anticoagulant (EDTA) from the tail vein after 2,
15, and 30 min and 1, 4, 8, and 24 h in addition to a 24 h terminal
(
13) Pan, W.; Liu, H.; Xu, Y.-J.; Chen, X.; Kim, K. H.; Milligan, D. L.;
Columbus, J.; Hadari, Y. R.; Kussie, P.; Wong, W. C.; Labelle, M.
Pyrimido-oxazepine as a versatile template for the development of
inhibitors of specific kinases. Bioorg. Med. Chem. Lett. 2005, 15
(24), 5474-5477.
sample taken from the vena cava after administration of halothane.
Plasma samples where obtained by centrifugation and transferred
into separate neutral polypropylene tubes and stored frozen at or
below -20 °C. Plasma samples were analyzed for compound 5

2
224 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 9
Mortlock et al.
(
14) Heron, N. M.; Anderson, M.; Blowers, D. P.; Breed, J.; Eden, J. M.;
Green, S.; Hill, G. B.; Johnson, T.; Jung, F. H.; McMiken, H. H. J.;
Mortlock, A. A.; Pannifer, A. D.; Pauptit, R. A.; Pink, J.; Roberts,
N. J.; Rowsell, S. SAR and inhibitor complex structure determination
of a novel class of potent and specific Aurora kinase inhibitors.
Bioorg. Med. Chem. Lett. 2006, 16 (5), 1320-1323.
R.; Curwen, J. O.; Kendrew, J.; Lambert-van der Brempt, C. Design
and Structure-Activity Relationship of a New Class of Potent VEGF
Receptor Tyrosine Kinase Inhibitors. J. Med. Chem. 1999, 42 (26),
5369-5389.
(22) Foote, K. M.; Mortlock, A. A.; Heron, N. M.; Jung, F. H.; Keen, N.;
Wilkinson, R. W.; Wedge, S. R.; Brady, M. C.; McKillop, D.; Khatri,
L.; Green, S. The Discovery of AZD1152: A Novel and Highly
Potent Aurora Kinase Inhibitor. Proceedings of American Association
for Cancer ResearchsNational Cancer InstitutesEuropean Organi-
sation for Research and Treatment of Cancer, Molecular Targets and
Cancer Therapeutics, November 14-18, 2005, Philadelphia, PA,
Abstract C271.
23) Wilkinson, R. W.; Odedra, R.; Heaton, S.; Wedge, S. R.; Keen, N.
J.; Crafter, C.; Foster, J. R.; Brady, M. C.; Bigley, A.; Brown, E.;
Byth, K.; Barrass, N. C.; Mundt, K.; Foote, K. M.; Heron, N. M.;
Jung, F.; Mortlock, A. A.; Boyle, F. T.; Green, S. Clin. Cancer Res.
(
15) Jung, F. H.; Pasquet, G.; Lambert-van der Brempt, C.; Lohmann,
J.-J. M.; Warin, N.; Renaud, F.; Germain, H.; De Savi, C.; Roberts,
N.; Johnston, T.; Dousson, C.; Hill, G. B.; Mortlock, A. A.; Heron,
N.; Wilkinson, R. W.; Wedge, S. R.; Heaton, S. P.; Odedra, R.; Keen,
N. J.; Green, S.; Brown, E.; Thompson, K.; Brightwell, S. Discovery
of Novel and Potent Thiazoloquinazolines as Selective Aurora A and
B Kinase Inhibitors. J. Med. Chem. 2006, 49, 955-970.
(
(
16) Soncini, C.; Carpinelli, P.; Gianellini, L.; Fancelli, D.; Vianello, P.;
Rusconi, L.; Storici, P.; Zugnoni, P.; Pesenti, E.; Croci, V.; Ceruti,
R.; Giorgini, M. L.; Cappella, P.; Ballinari, D.; Sola, F.; Varasi, M.;
Bravo, R.; Moll, J. PHA-680632, a Novel Aurora Kinase Inhibitor
with Potent Antitumoural Activity. Clin. Cancer Res. 2006, 12 (13),
2007, in press.
(
24) Jung, F. H.; Pasquet, G. R. Preparation of substituted quinazoline
derivatives as inhibitors of aurora kinases. WO2003055491, 2002.
25) Heron, N. M.; Jung, F. H.; Pasquet, G. R.; Mortlock, A. A.
Preparation of phosphonooxy quinazoline derivatives and their
pharmaceutical use. WO2004058781, 2003.
26) Foote, K. M. Preparation of substituted quinazolinylaminopyrazolyl-
acetamides as anticancer agents. WO2006067391, 2005.
27) Cell activity data is given for a single cell line (SW620). Activity
for compounds in this pyrazole series in a range of human tumor
cell lines will be reported elsewhere.
28) While a number of different prodrug strategies were considered, only
the phosphate ester approach is described here.
29) For examples of â-oxygen effects on the pKa of organic bases, see:
Dissociation Constants of Organic Bases in Aqueous Solution:
Supplement 1972; Perrin, D. D., Ed.; Butterworths: London, 1972.
For example, the pKa of ethylamine is 10.72 and that of 2-methoxy-
ethyl ethylamine is 9.40.
30) Keen, N. J.; Crafter, C.; Brown, E.; Thompson, K.; Brightwell, S.;
Ahmed, A.; Mundt, K.; Wilkinson, R. W.; Odedra, R.; Heaton, S.;
Byth, K.; Foote, K.; Heron, N.; Jung, F.; Mortlock, A.; Green, S.
Manuscript in preparation.
31) Bridgland-Taylor, M. H.; Hargreaves, A. C.; Easter, A.; Orme, A.;
Henthorn, D. C.; Ding, M.; Davis, A. M.; Small, B. G.; Heapy, C.
G.; Abi-Gerges, N.; Persson, F.; Jacobson, I.; Sullivan, M.; Albertson,
N.; Hammond, T. G.; Sullivan, E.; Valentin, J. P.; Pollard, C. E.
Optimisation and validation of a medium-throughput electrophysi-
ology-based hERG assay using IonWorksTM HT. J. Pharmacol.
Toxicol. Methods 2006, 54 (2), 189-199.
4080-4089.
(
(
17) Fancelli, D.; Moll, J.; Varasi, M.; Bravo, R.; Artico, R.; Berta, D.;
Bindi, S.; Cameron, A.; Candiani, I.; Cappella, P.; Carpinelli, P.;
Croci, W.; Forte, B.; Giorgini, M. L.; Klapwijk, J.; Marsiglio, A.;
Pesenti, E.; Rocchetti, M.; Roletto, F.; Severino, D.; Soncini, C.;
Storici, P.; Tonani, R.; Zugnoni, P.; Vianello, P. 1,4,5,6-Tetra-
hydropyrrolo[3,4-c]pyrazoles: Identification of a Potent Aurora
Kinase Inhibitor with a Favorable Antitumour Kinase Inhibition
Profile. J. Med. Chem. 2006, 49, 7247-7251.
(
(
(
(
18) Moriarty, K. J.; Koblish, H. K.; Garrabrant, T.; Maisuria, J.; Khalil,
E.; Ali, F.; Petrounia, I. P.; Crysler, C. S.; Maroney, A. C.; Johnson,
D. L.; Galemmo, R. A. et al. The synthesis and SAR of 2-amino-
pyrrolo[2,3-d]pyrimidines: A new class of Aurora-A kinase inhibi-
tors. Bioorg. Med. Chem. Lett. 2006, 16, 5778-5783.
(
(
19) Young, M. A.; Shah, N. P.; Chao, L. H.; Seeliger, M.; Milanov, Z.
V.; Biggs, W. H.; Treiber, D. K.; Patel, H. K.; Zarrinkar, P. P.;
Lockhart, D. J.; Sawyers, C. L.; Kuriyan, J. Structure of the Kinase
Domain of an Imatinib-Resistant Abl Mutant in Complex with the
Aurora Kinase Inhibitor VX-680. Cancer Res. 2006, 66 (2), 1007-
(
1014.
(
(
20) Manfredi, M.; Ecsedy, J.; Meetze, K.; Balani, S.; Burenkova, O.;
Chen, W.; Hoar, K.; Huck, J.; Leroy, P.; Sells, T.; Stroud, S.; Vos,
T.; Weatherhead, G.; Wysong, D.; Zhang, M.; Claiborne, C.
MLN8054, an orally active Aurora A kinase small molecule inhibitor
in phase 1 clinical trials. Proc. Amer. Assoc. Cancer Res., 2006, 47,
1110 (Abstract #4724).
(21) Hennequin, L. F.; Thomas, A. P.; Johnstone, C.; Stokes, E. S. E.;
Ple, P. A.; Lohmann, J.-J. M.; Ogilvie, D. J.; Dukes, M.; Wedge, S.
JM061335F