A new series of neutral 5-substituted 4-anilinoquinazolines as potent, orally active inhibitors of erbB2 receptor tyrosine kinase

Article abstract of DOI:10.1016/j.bmcl.2007.11.052

Starting from initial lead 1 containing a basic 5-substituent, optimisation of the glycolamide-derived neutral 5-substituent led to potent inhibitors of erbB2 with good pharmacokinetics. Representative compounds 19 and 21 inhibited phosphorylation of erbB

Full text of DOI:10.1016/j.bmcl.2007.11.052

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry Letters 18 (2008) 674–678
A new series of neutral 5-substituted 4-anilinoquinazolines
as potent, orally active inhibitors of erbB2 receptor tyrosine kinase
Bernard Barlaam,a,* Peter Ballard,b Robert H. Bradbury,b Richard Ducray,a
a
b
b
b
´
Herve Germain, D. Mark Hickinson, Kevin Hudson, Jason G. Kettle,
Teresa Klinowska,b Franc¸oise Magnien,a Donald J. Ogilvie,b Annie Olivier,a
Stuart E. Pearson,b James S. Scott,b Abid Suleman,b Cath B. Trigwell,b
Michel Vautier,a Robin D. Whittakerb and Robin Woodb
aAstraZeneca, Centre de Recherches, Z.I.S.E. La Pompelle B.P. 1050, 51689 Reims Cedex 2, France
bCancer and Infection Research, AstraZeneca, Mereside, Alderley Park, Macclesfield, Cheshire SK10 4TG, UK
Received 30 October 2007; revised 14 November 2007; accepted 15 November 2007
Available online 21 November 2007
Abstract—Starting from initial lead 1 containing a basic 5-substituent, optimisation of the glycolamide-derived neutral 5-substituent
led to potent inhibitors of erbB2 with good pharmacokinetics. Representative compounds 19 and 21 inhibited phosphorylation of
erbB2 in a mouse BT474C xenograft model after oral administration.
ꢀ 2007 Elsevier Ltd. All rights reserved.
Aberrant function of the erbB family of receptor tyro-
sine kinases and their ligands has been described in
many human cancers.1 Within this family, erbB2 plays
a central role: mis-regulation of erbB2, for example by
over-expression/gene amplification, has been observed
to varying degrees in a range of tumours including those
of the prostate, cervix, ovary, endometrium, gastrointes-
tinal tract and, in particular, the breast.2
such as the erbB2 selective inhibitor CP-724714, and the
irreversible inhibitors HKI-272 and BIBW-2992 are
undergoing clinical trials (Fig. 1).
We previously reported work on anilinoquinazolines
substituted at the C-5 position, as inhibitors of Src,6
EGFR7 and erbB2.8 In the latter publication, compound
1 was described as a potent inhibitor of erbB2 that dis-
played good pharmacokinetics in mouse and inhibited
tumour growth in the mouse BT474C xenograft model.
With such a large body of evidence implicating erbB2 in
tumour progression to date,3 both antibody and small
molecule inhibitory approaches are being developed
and have proven to possess anti-tumour activity in the
clinic. The monoclonal antibody trastuzumab is now ap-
proved for use in erbB2-overexpressing metastatic breast
cancer, either as monotherapy or in combination with
chemotherapy.4 The mixed EGFR/erbB2 small molecule
inhibitor lapatinib has recently been approved in combi-
nation with capecitabine for women with metastatic
erbB2-positive breast cancer who have failed to respond
to trastuzumab.5 Additional small molecule inhibitors1b
O
Cl
F
H
N
HN
N
MeO2S
O
N
Lapatinib
O
N
O
HN
O
N
N
Keywords: Anilinoquinazoline; C-5 substitution; erbB2 kinase
inhibitor.
H
CP-724714
N
*
Corresponding author. Tel.: +33 3 26 61 68 97; fax: +33 3 26 61 68
42; e-mail: bernard.barlaam2@astrazeneca.com
Figure 1.
0960-894X/$ - see front matter ꢀ 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2007.11.052

B. Barlaam et al. / Bioorg. Med. Chem. Lett. 18 (2008) 674–678
675
However, this compound showed potent inhibition of
the hERG channel (IC50 = 5 lM), modest pharmacoki-
netics in other species (especially dog) and mild to severe
effects in liver and lung in rat toxicological studies
attributed to phospholipidosis, which precluded its pro-
gression beyond pre-clinical models. We reasoned that
the basicity of the N-methyl piperidine (pKa = 8.1) and
the high lipophilicity of 1 could be responsible for its
affinity to the hERG channel and phospholipidosis.
Work on neutral 5-substituted 4-anilinoquinazolines de-
rived from 1 has shown that good pharmacokinetics in
rat and dog and absence of phospholipidosis after
chronic dosing in the rat could be achieved in this
series.9
listed in Table 1, and synthetic routes are outlined in
the following Schemes.10
Compounds 2–14 and 17–22 were prepared from the key
fluoroanilinoquinazolines 23a8 and 23b (made from 5-
fluoro-4-chloroquinazoline8). The key sequence (Scheme
1) involves nucleophilic displacement of the fluoro
group with 2-acetamidoethoxide11 or with methoxide
followed by demethylation with pyridine hydrochloride
under mild conditions to give the corresponding phenols
24a and 24b, Mitsunobu reaction with the correspond-
ing lactate or glycolate and transformation of the ester
into the amide directly or via the carboxylic acid.12,13
Other analogues were synthesised using similar ap-
proaches, for example: Mitsunobu reaction of 24b with
(S)-N,N-dimethyl lactamide to give 21; alkylation of
the phenol 24a with the corresponding haloacetamide
(illustrated by 2 and 5); nucleophilic displacement of
the 5-fluoro group from 23a with ethyl glycolate to give
ester 27a which can be readily transformed into the cor-
responding amides. An efficient method (Scheme 2) for
variation of the aniline at the final step was also devel-
oped: 32 which was made from 286a in four steps was
activated with POCl3 and reacted with the correspond-
ing anilines10 to give 15–16.
In this publication, we describe a new series of neutral 5-
substituted anilinoquinazolines as potent and orally
active erbB2 inhibitors (Fig. 2). Representative com-
pounds prepared during the course of this work are
O
N
N
Cl
O
HN
R1
N
O
R3
R6
The compounds listed in Table 1 were evaluated in an
erbB2 autophosphorylation assay using a MCF7 breast
carcinoma cell line engineered to overexpress erbB2 as
described previously.8,10 Initial glycolamide-derived
compounds 2–5 showed moderate activity in this assay.
Introduction of a methyl on the position a to the amide
significantly improved potency as illustrated by the fol-
lowing pairs (8 vs 4, 6 and 7 vs 3).
N
R2
O
HN
R5
N
1
O
N
N
2-22
Figure 2.
Table 1. Inhibition data versus erbB2 cellular autophosphorylation and hERG for compounds 2–22
R5
R6
R3
NR1R2
Synthetic routea
(for Scheme 1)
p-erbB2 IC50 (lM)
hERG IC50 (lM)
b
c
Compound
2
Cl
Cl
CH2-(2-Pyridyl)
CH2-(2-Pyridyl)
CH2-(2-Pyridyl)
CH2-(2-Pyridyl)
CH2-(2-Pyridyl)
CH2-(2-Pyridyl)
CH2-(2-Pyridyl)
CH2-(2-Pyridyl)
CH2-(2-Pyridyl)
CH2-(2-Pyridyl)
3-Pyridyl-6-Me
3-Pyridyl-6-Me
3-Pyridyl-6-Me
3-Pyridyl-6-Me
3-Pyridyl-6-Me
3-Pyridyl-6-Me
3-Pyridyl-6-Me
3-Pyridyl-6-Me
3-Pyridyl-6-Me
3-Pyridyl-6-Me
3-Pyridyl-6-Me
H
NH2
a,e
c,f
0.180
0.120
0.140
0.350
0.031
0.079
0.019
0.034
0.043
0.066
0.130
0.022
0.035
0.027
0.023
0.200
0.120
0.069
0.078
0.023
0.030
3
H
N(Me)CH2CH2OH
4-Morpholine
4-Me-Piperazine
N(Me)CH2CH2OH
N(Me)CH2CH2OH
4-Morpholine
NMe2
4
Cl
Cl
H
c,f
a,e
5
H
6
Cl
Cl
(R)-Me
(S)-Me
(R)-Me
(R)-Me
(S)-Me
(R)-Me
H
a,b,g
a,b,g
a,b,f
a,b,f
a,b,f
a,b,g
a,d,f
a,b,f
a,b,f
12
11
10
14
10
32
7
8
Cl
Cl
9
10
11
12
13
14
15
16
17
18
19
20
21
22
Cl
Cl
NMe2
NHMe
Me
Me
Me
Cl
N(Me)CH2CH2OH
4-Morpholine
4-Morpholine
4-Morpholine
4-Morpholine
NHCH2CH2OH
N(Me)CH2CH2OMe
N(Me)CH2CH2OH
N(Me)CH2CH2OH
NMe2
(R)-Me
(S)-Me
(R)-Me
(R)-Me
(R)-Me
(R)-Me
(R)-Me
(S)-Me
(R)-Me
(S)-Me
OMe
Me
Me
Me
Me
Me
Me
a,b,f
a,b,f
a,b,f
a,b,f
12
a,b,f or a,h
a,b,f or a,h
8
11
NMe2
a Steps used for the synthesis (as referred to in Scheme 1).
b n P 2, standard error is typically 0.3 log unit.
c n P 2.

676
B. Barlaam et al. / Bioorg. Med. Chem. Lett. 18 (2008) 674–678
R3
R4
R4
N
MeO
R4
N
OH HN
F
HN
O
HN
N
O
(b)
(a)
N
series a: R4=
N
N
O
25a-b R3= (
26a-b R3= (
R
)-Me
23a-b
24a-b
N
S)-Me
(d)
(h)
Cl
(c)
(e)
(f) or (g)
R1
N
series b: R4=
O
R1
R3
R4
N
EtO
R4
N
R2
N
O
HN
N
R2
R4
O
HN
O
HN
O
O
(f)
O
N
N
N
N
27a-b
2-5, 12
6-11, 13-14, 17-22
Scheme 1. Synthesis of compounds 2–14 and 17–22. Reagents and conditions: (a) AcNHCH2CH2OH (2.5 eq), NaH (6 equiv), DMA, 120 ꢁC (series
a) or MeONa (3 equiv), MeOH, reflux, then pyridineÆHCl (4 equiv), pyridine, reflux (series b); (b) (t-BuO2C)2N2 or (EtO2C)2N2, PPh3, (R) or (S)-
methyl lactate, CH2Cl2, 20 ꢁC; (c) EtONa, ethyl glycolate, reflux (series a); (d) (EtO2C)2N2, PPh3, ethyl glycolate, CH2Cl2, 20 ꢁC (series b); (e)
BrCH2CONH2, K2CO3, DMA, 20 ꢁC (for 2); ClCH2CO-N-piperazine-N0-BOC, KI, K2CO3, DMA, 60 ꢁC, then TFA, 20 ꢁC, then 37% aq CH2O,
HCO2H, DMSO, 180 ꢁC lw heating (for 5); (f) NaOH, THF/MeOH or EtOH, 20 ꢁC; HATU, Ni-Pr2Et, NHR1R2, DMF, 20 ꢁC (for 3 and 4) or
HATU, Ni-Pr2Et, DMA, 50–70 ꢁC, then NHR1R2, 20 ꢁC (for 8–10, 12) or HOBT, EDCI, NHR1R2, DMF, 20 ꢁC (for 13–14, 17–22); (g) THF or
EtOH, NHR1R2, lw heating (for 6–7, 11); (h) (t-BuO2C)2N2, PPh3, (R) or (S)-N,N-dimethyl lactamide, CH2Cl2, 20 ꢁC (for 21–22).
O
O
N
N
Ar
N
O
O
N
OMe O
R'
O
N
O
HN
N
R''
O
POM
O
N
(a)
NH
(e)
N
(c)
N
28
31 R''= POM
32 R''= H
29 R'= OMe
30 R'= OH
15-16
(b)
(d)
Scheme 2. Synthesis of compounds 15–16. Reagents and conditions: (a) NaH, DMF, 25 ꢁC, then ClCH2OC(O)t-Bu; (b) MgBr2, pyridine, 120 ꢁC; (c)
(t-BuO2C)2N2, PPh3, 4-((S)-2-hydroxypropionyl)morpholine,14 CH2Cl2, 20 ꢁC; (d) 7 N NH3, MeOH, 20 ꢁC; (e) POCl3 (1.2 equiv), Ni-Pr2Et (2.5
equiv), ClCH2CH2Cl, 80 ꢁC; ArNH2 (1.05 equiv), CH3CN, 80 ꢁC.
A similar SAR for the C-5 substitution was observed
when using the aniline from series b at C-4 (compounds
12–14, 17–22): lactamides were preferred over glycola-
mides. The level of activity between the 2 anilines was
similar. Exploration of the SAR around the aniline from
series b was studied (15–16 vs 13). The 3-methyl on the
aniline could be replaced by a chlorine or a methoxy
without significant loss of activity.
Physical properties and pharmacokinetic parameters of
selected compounds were evaluated further (see Table
3). Although being more lipophilic than 9, 21 displayed
better solubility and higher levels of free drug. Excellent
pharmacokinetic parameters in rat and dog were seen
for both compounds, with low clearance and high
bioavailability.17
Compounds 19 and 21 were evaluated for inhibition of
phosphorylation of erbB2 in BT474C18 xenograft in
athymic mice. Compound 19, dosed at 100 mg/kg orally
with coadministration of ABT9,20 (an irreversible inhib-
itor19 of cytochrome P450 known to reduce P450 medi-
ated clearance in vivo), gave 78% and 36% inhibition,
respectively, at 1 and 8 h post dose, which correlates
with free exposure over the in vitro BT474C cell IC50
(2.3-fold at 1 h; 0.6-fold at 8 h). Compound 21, dosed
at 100 mg/kg orally without ABT, gave 37% and 55%
inhibition, respectively, at 1 and 8 h post dose.
Selectivity versus EGFR was evaluated against EGFR
and erbB2 isolated enzymes,8,10 and in BT474C and
KB cell lines, known to respond, respectively, to stimu-
lation of both erbB2 and EGFR, and EGFR alone.8
Excellent separation between erbB2 activity and EGFR
activity was seen with compounds bearing the aniline
from series b whereas a less marked difference was seen
with the aniline of series a (see 19 and 21 vs 9 in Table 2).
Compounds 9, 19 and 21 showed a clean selectivity pro-
file versus in-house and external kinase panels (data not
shown).15
In summary, we have discovered a new series of 5-
substituted anilinoquinazolines displaying good activity
and good pharmacokinetics compared to the initial lead
1 containing a basic 5-substituent. Representative com-
pounds 19 and 21 inhibited phosphorylation of erbB2
Inhibition of the hERG channel was evaluated10 on a
selection of compounds (see Table 1). Slightly reduced
hERG activity was seen compared to 1, as previously
observed with neutral compounds.16

B. Barlaam et al. / Bioorg. Med. Chem. Lett. 18 (2008) 674–678
677
Table 2. Inhibition data versus erbB2 and EGFR kinases, erbB2 cellular autophosphorylation, BT474C and KB cell proliferation assays for
compounds 9, 19 and 21
a
a
a
a
a,b
KB IC50 (lM)
Compound
erbB2 enz. IC50 (lM)
EGFR enz. IC50 (lM)
p-erbB2 IC50 (lM)
BT474C IC50 (lM)
9
<0.001
0.002
<0.001
0.022
2.0
1.1
0.034
0.069
0.023
0.100
0.270
0.320
1.0c
23c
2.7
19
21
a n P 2, standard error is typically 0.3 log unit.
b KB cell proliferation end point after EGF stimulation.
c n = 1.
Table 3. Pharmacokinetic parameters, plasma protein binding, solubility and logD for compounds 9, 19 and 21
Rat/dog Cl %hbfa
Rat/dog F%a
Rat/dog/mouse %freeb
Solubilityc (lM)
logDd
a
Rat/dog Vdss (L/kg)
Compound
9
4.3/24
23/34
12/48
0.5/1
1/0.9
75/89
34/33
67/47
0.4/1.1/—
2.9/3.8/2.5
1.7/2.5/1.1
6
3.7
3.8
19
21
66
28
0.9/1.8
>4.3
a Female Han Wistar rats dosed at 2 mg/kg iv and 5 mg/kg po except for 9 dosed at 0.4 mg/kg iv and 1 mg/kg po; mean values for male and female
beagle dogs dosed at 0.2 mg/kg iv and 0.4 mg/kg po except for 21 dosed at 1 mg/kg iv and 2 mg/kg po; Cl expressed as % of hepatic blood flow.
b Protein binding of compound in plasma, expressed as %free.
c Solubility in aqueous phosphate buffer, pH 7.4, at 24 h.
d Measured from octanol/water, pH 7.4.
7. Ballard, P.; Bradbury, R. H.; Harris, C. S.; Hennequin, L.
F. A.; Hickinson, M.; Johnson, P. D.; Kettle, J. G.;
Klinowska, T.; Leach, A. G.; Morgentin, R.; Pass, M.;
Ogilvie, D. J.; Olivier, A.; Warin, N.; Williams, E. J.
Bioorg. Med. Chem. Lett. 2006, 16, 1633.
in a mouse BT474C xenograft model after oral adminis-
tration. Further evolution of 5-substituted 4-anilinoqui-
nazolines will be reported in due course.
8. Ballard, P.; Bradbury, R. H.; Hennequin, L. F. A.;
Hickinson, D. M.; Johnson, P. D.; Kettle, J. G.; Kli-
nowska, T.; Morgentin, R.; Ogilvie, D. J.; Olivier, A.
Bioorg. Med. Chem. Lett. 2005, 15, 4226.
9. Ballard, P.; Barlaam, B. C.; Bradbury, R. H.; Dishington,
A.; Hennequin, L. F. A.; Hickinson, D. M.; Hollinsworth,
I. M.; Kettle, J. G.; Klinowska, T.; Ogilvie, D. J.; Pearson,
S. E.; Scott, J. S.; Suleman, A.; Whittaker, R.; Williams, E.
J.; Wood, R.; Wright, L. Bioorg. Med. Chem. Lett. 2007,
17, 6326.
10. For experimental procedures and hERG assay protocol,
see: Bradbury, R. H.; Kettle, J. G.; Scott, J. S.; Barlaam,
B. C. PCT Int. Appl. WO 2005118572. In particular, the
synthesis of the aniline precursor of compounds 15–16 is
described herein.
Acknowledgments
We warmly thank Elizabeth Pease for helpful discus-
sions and the following for technical assistance: Sarah
Beck, Dominique Boucherot, Rowena J. Callis, Sara
Davenport, Christian Delvare, Delphine Dorison-Du-
val, Paul M. Farrington, Nadine Gondouin, Janet A.
Jackson, Patrice Koza, Zaieda Parveen, Jacques Pell-
eter, Fabrice Renaud, Helen S. Ross, Robin Smith, John
P. Vincent and Lisa C. Woods.
References and notes
11. The phenol 24a resulted from the intramolecular nucleo-
philic attack of the anion of the acetamide (formed in the
reaction conditions because of excess of sodium hydride)
onto the ether adduct. This intermediate is formed by
displacement of the fluoro group of 23a with sodium 2-
acetamidoethoxide.
12. In the absence of amine, activation of the carboxylic acid
with EDCI or HATU gave the activated anilide resulting
from the intramolecular attack of the NH of the aniline
onto the activated carboxylic acid; this activated anilide
can be isolated and reacted with an amine to form the
expected amide. Alternatively in the presence of the amine,
activation of the carboxylic acid gave the expected amides
with minor amounts of the activated anilide.
13. Good enantiomeric purity was observed with these meth-
ods: 19 and 21 made as described in Scheme 1 gave,
respectively, 99.6% ee and >99.6% ee measured by chiral
HPLC; 21 made in a similar manner as compounds in
Scheme 2 gave 98.6% ee; however attempts to displace 23b
with the (R)-N,N dimethyl lactamide in the presence of
sodium hydride in THF at reflux gave partial racemisation.
1. (a) Rowinski, E. K. Horizons Cancer Ther. 2001, 2, 3, and
references herein; (b) Wallace, E. M.; Yeh, T. C.; Laird, E.
R.; Blake, J. F.; Lyssikatos, J. In Topics in Medicinal
Chemistry (Cancer); Bradbury, R. H., Ed.; Springer:
Heidelberg, 2007; p 83.
2. Gross, M. E.; Shazer, R. L.; Agus, D. B. Semin. Oncol.
2004, 31, 9, 1 Suppl 3.
3. Citri, A.; Yarden, Y. Nat. Rev. Mol. Cell Biol. 2006, 7,
505.
4. Emens, L. A.; Davidson, N. E. Oncology 2004, 18, 1117.
5. (a) Lackey, K. Curr. Top. Med. Chem. 2006, 6, 435; (b)
Script, 15 March 2006.
6. (a) Hennequin, L. F.; Allen, J.; Breed, J.; Curwen, J.;
Fennell, M.; Green, T. P.; Lambert van der Brempt, C.;
Morgentin, R.; Norman, R. A.; Olivier, A.; Otterbein, L.;
Ple, P. A.; Warin, N.; Costello, G. J. Med. Chem. 2006, 49,
6465–6488; (b) Barlaam, B.; Fennell, M.; Germain, H.;
´
Green, T.; Hennequin, L.; Morgentin, R.; Olivier, A.; Ple,
P.; Vautier, M.; Costello, G. Bioorg. Med. Chem. Lett.
2005, 15, 5446.

678
B. Barlaam et al. / Bioorg. Med. Chem. Lett. 18 (2008) 674–678
14. Tasaka, A.; Tamura, N.; Matsushita, Y.; Teranishi, K.;
Hayashi, R.; Okonogi, K.; Itoh, K. Chem. Pharm. Bull.
1993, 41, 1035.
15. The activity of compounds 9, 19 and 21 was assessed at a
concentration of 10 lM against a panel of recombinant
protein kinases (including KDR and Src) at Km ATP
concentration for each enzyme: 14 kinases for 9 and >40
for 19 and 21.
24 · 10ꢀ6 cm sꢀ1; about the MDCK-MDR permeability
assay, see: Kerns, E. H.; Di, L.; Petusky, S.; Farris, M.;
Ley, R.; Jupp, P. J. Pharm. Sci. 2004, 93, 1440.
18. Baselga, J.; Norton, L.; Albanell, J.; Kim, Y. M.;
Mendelsohn, J. Cancer Res. 1998, 58, 2825.
19. Balani, S. K.; Li, P.; Nguyen, J.; Cardoza, K.; Zeng, H.;
Mu, D.-X.; Wu, J.-T.; Gan, L.-S.; Lee, F. W. Drug Metab.
Disp. 2004, 32, 1092.
16. Jamieson, C.; Moir, E. M.; Rankovic, Z.; Wishart, G.
J. Med. Chem. 2006, 49, 5029.
17. Compound 21 showed very good permeability across
membranes (Papp in MDCK-MDR) cell line:
20. Increased blood levels of 19 dosed at 100 mg/kg orally
were seen after oral coadministration of 100 mg/kg of
ABT (AUC 670 lM h vs 130 lM h without ABT and
concentration at 8 h 23 lM vs 2 lM).