Discovery of a potent dual EGFR/HER-2 inhibitor L-2 (selatinib) for the treatment of cancer

Article abstract of DOI:10.1016/j.ejmech.2013.09.032

To develop potent dual EGFR/HER-2 inhibitors with improved druggability, a series of new lapatinib analogs were designed and synthesized. Compared with lapatinib, L-2, L-4 and M-6 were more active against BT-474 or NCI-N87 cells. In vivo efficacy studies indicated that L-2 significantly suppressed tumor growth in NCI-N87 (94.8% inhibition) or SK-OV-3 xenograft (85.7% inhibition) without causing significant loss of body weight. And the inhibition rates of lapatinib in the two xenograft models were 89.7% and 78.8%, respectively. Moreover, further studies revealed that the potent in vivo activities of L-2 may be mainly attributed to its superior aqueous solubility and oral bioavailability. In addition, a high-yielding one-pot procedure was developed for the synthesis of lapatinib and its analogs.

Full text of DOI:10.1016/j.ejmech.2013.09.032

European Journal of Medicinal Chemistry 69 (2013) 833e841
Contents lists available at ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Original article
Discovery of a potent dual EGFR/HER-2 inhibitor L-2 (selatinib) for the
treatment of cancer
,
b
c
a
a
Long Zhang a, Chuanwen Fan a , Zongru Guo , Ying Li , Shuyong Zhao , Shaobo Yang ,
*
Yingying Yang a, Jianrong Zhu c, Dong Lin a
a Qilu Institute of Pharmaceutical Research, Qilu Pharmaceutical Co. Ltd, 243 Gong Ye Bei Road, Jinan 250100, Shandong Province, PR China
b Department of Medicinal Chemistry, Institute of Materia Medica, Chinese Academy of Medical Sciences and Peking Union Medical College,
1 Xian Nong Tan Street, Beijing 100050, PR China
c State Key Laboratory of Applied Organic Chemistry, Lanzhou University, Lanzhou 730000, PR China
a r t i c l e i n f o
a b s t r a c t
Article history:
To develop potent dual EGFR/HER-2 inhibitors with improved druggability, a series of new lapatinib
analogs were designed and synthesized. Compared with lapatinib, L-2, L-4 and M-6 were more active
against BT-474 or NCI-N87 cells. In vivo efficacy studies indicated that L-2 significantly suppressed tumor
growth in NCI-N87 (94.8% inhibition) or SK-OV-3 xenograft (85.7% inhibition) without causing significant
loss of body weight. And the inhibition rates of lapatinib in the two xenograft models were 89.7% and
78.8%, respectively. Moreover, further studies revealed that the potent in vivo activities of L-2 may be
mainly attributed to its superior aqueous solubility and oral bioavailability. In addition, a high-yielding
one-pot procedure was developed for the synthesis of lapatinib and its analogs.
Received 30 March 2013
Received in revised form
30 August 2013
Accepted 15 September 2013
Available online 27 September 2013
Keywords:
EGFR
HER-2
Ó 2013 Elsevier Masson SAS. All rights reserved.
Lapatinib analogs
NCI-N87
SK-OV-3
Selatinib
1. Introduction
signal transduction of tumor proliferation and survival [11]. How-
ever, low aqueous solubility of lapatinib results in its incomplete
The signaling cascade of epidermal growth factor receptor
(EGFRs or ErbBs) family currently represents a major target of
anticancer agents [1,2]. Deregulated expression of EGFRs, in
particular EGFR and HER-2, has been implicated in the develop-
ment and malignancy of many human cancers [3,4]. Four EGFR
targeted tyrosine kinase inhibitors, gefitinib (IressaÒ) [5], erlotinib
(TarcevaÒ) [6], lapatinib (TykerbÒ) [7] and icotinib (ConmanaÒ),
have been marketed since 2002 (Fig. 1). Gefitinib and erlotinib
belong to the specific inhibitors of EGFR for treating non-small-cell
carcinoma (NSCLC), but later have been observed serious drug
resistance which is associated with the T790M mutation of EGFR
tyrosine kinase [8,9]. Lapatinib is a dual inhibitor of EGFR/HER-2
which is used to treat HER-2 positive breast cancer. Since the for-
mation of EGFR/HER-2 or HER-2/HER-3 heterodimers can enhance
the ligand binding, receptor tyrosine phosphorylation and cell
proliferation compared with EGFR homodimers [10], lapatinib has
superior potency comparing to single inhibitor of EGFR in inhibiting
and unstable absorption, great individual difference and low
bioavailability after oral administration [12]. The elimination half-
life of lapatinib between single-dose and repeated-doses showed
great variation, and long-term administration of lapatinib can cause
accumulation, which may intensify its toxicity [13]. All of these are
the burdens for its clinical use.
Suitable aqueous solubility is expected for oral potential drugs.
Water-soluble drugs usually show greater pharmacokinetic prop-
erties and efficacies compared with poorly water-soluble ones,
especially in low dose cases. Lipinski et al. found that 87% of oral
drugs in their study had a higher solubility than 65 mg/L, and only
7% of them showed a lower solubility than 20 mg/L [14]. The
measured solubility of lapatinib in water at 25 ꢀC is 7.0 mg/L and far
lower than the limit for oral potential drugs (20 mg/L) [15].
Therefore, there is still room for improvement in the druggability of
lapatinib, and a “Me Better” drug with improved aqueous solubility
and pharmacokinetic properties may be discovered by modifying
the structure of lapatinib.
In this paper, we described the work we performed around the
improvement in the druggability of lapatinib. This research allowed
* Corresponding author. Tel.: þ86 0531 8312 6937.
E-mail address: chuanwen.fan@qilu-pharma.com (C. Fan).
0223-5234/$ e see front matter Ó 2013 Elsevier Masson SAS. All rights reserved.
http://dx.doi.org/10.1016/j.ejmech.2013.09.032

834
L. Zhang et al. / European Journal of Medicinal Chemistry 69 (2013) 833e841
Fig. 1. EGFR family inhibitors currently on the market.
us to find a new dual EGFR/HER-2 inhibitor L-2 (selatinib) with
potent in vivo efficacy in NCI-N87 and SK-OV-3 xenografts as well as
improved druggablility and pharmokinetic profiles.
NaBH(OAc)3 produced the target product which was purified by
chromatography.
With the preparation of M-series compounds, the above syn-
thetic protocol could not be applied due to the difficult
2. Results and discussion
Table 1
The physicochemical properties of lapatinib and its analogs.
2.1. Analog design and calculation
O
Analysis of the four approved EGFR inhibitors revealed these
compounds shared a common 4-anilinoquinazoline core structure
(Fig.1). Erlotinib and gefitinib are type-1 ATP competitive inhibitors
and recognize the so-called active conformation of the kinase [16].
The crystal structure of active EGFR in complex with erlotinib re-
veals that the quinazoline structure of erlotinib simulates adenine
action mode, and the acetylene group recognizes the BP-I pocket
[16]. On the other hand, lapatinib is a type-2 inhibitor and recog-
nizes the inactive conformation of the kinase [17,18]. The crystal
structure of inactive form of EGFR in complex with lapatinib in-
dicates that the quinazoline part is hydrogen-bonded to the hinge
region between the eNH2 and eCOOH terminal, and the aniline
group is oriented deep in the back of the ATP binding site and
makes predominantly hydrophobic interactions with the protein
[19]. The BP-II pocket (a pocket formed by the side chains of
Met742, Leu753, Thr766, Thr830, Phe832, and Leu834) is well
recognized by the 3-fluorobenzyloxy group [19,20]. Further studies
have also confirmed that the des-benzyl metabolite of lapatinib was
inactive [21]. The methylsulfonylethylaminomethylfuryl group of
lapatinib is positioned at the solvent interface and does not make
any significant interactions with the protein [19]. Based on the
above rationale assessment, we assumed that the modification of
substituted R groups in the furan ring of lapatinib might be an ideal
method to develop a new candidate with improved drug-like
properties.
F
HN
N
Cl
R1
N
O
Comp.
R1
Average
log P
Average
log S
tPSAa
Solubilityb
(mg/L)
H
L-1
L-2
6.50
5.04
ꢁ6.67
ꢁ5.74
67.2
84.3
3.0
N
H3C
H3C
S
H
N
404.6
S
O
O
O
H3C
H
N
S
L-3
L-4
5.04
4.76
ꢁ6.36
ꢁ6.54
113.4
127.4
48.5
10.1
N
H
H
N
H2N
O
S
O
L-5
5.89
5.29
ꢁ6.18
ꢁ6.24
101.4
101.4
16.7
7.0
H
N
H C
lapatinib
S
O
O
Several compounds with different R groups in the furan ring of
lapatinib were designed, and the average Log P and Log S values of
these lapatinib analogs were calculated using software ALOGPS 2.1.
Calculation indicated these compounds covered a wide range of Log
P or Log S values (Table 1.). When compared with lapatinib, L-2, L-3
and L-4 have favorable Log P, and L-2, M-2, M-3, M-6 and K have
better Log S.
M-1
M-2
M-3
M-4
5.78
5.86
5.51
6.17
ꢁ5.94
ꢁ5.59
ꢁ5.79
ꢁ6.07
67.7
61.7
70.5
88.0
ND
ND
ND
ND
O
N
N
N
HN
N
N
O
O
N
2.2. Chemistry
S
O
N
N
M-5
M-6
5.23
5.53
ꢁ5.99
ꢁ5.75
95.8
78.8
ND
ND
O
O
A high-yielding synthetic procedure was developed for the
preparation of L-series compounds (Table 1). The two reactions
were carried out in one pot without isolating the intermediates
(Scheme 1). All starting materials A, B, Pd(dppf)2Cl2, triethylamine
and ReNH2$HCl were simply mixed in ethanol under N2 atmo-
sphere. After the first reaction took place at room temperature to
form the imine intermediate, raising the temperature to reflux, the
imine intermediate reacted with compound A to give coupling in-
termediate. Further reduction of the coupling intermediate with
N
N
N
N
K
5.47
ꢁ5.66
78.8
ND
O
a
tPSA values were calculated using software ChemDraw Ultra 10.0.
Solubility in unbuffered water was determined at 25 ꢀC by the shake-flask
method, and the solubility values were the average values of two experiments
from the same batch of material.
b

L. Zhang et al. / European Journal of Medicinal Chemistry 69 (2013) 833e841
835
O
O
F
F
HN
N
Cl
HN
N
Cl
H
I
N
a, b
N
N
O
R
NH2 HCl
R
+
+
OHC
B(OH)2
O
B
A
L-1 ~ L-5 and Lapatinib
Comp.
R
Comp.
R
H3C
H3C
H2N
L-1
L-4
S
S
O
O
O
L-2
L-5
S
O
F3C
S
O
O
L-3
Lapatinib
O
H3C
S
S
N
O
H3C
O
H
Scheme 1. Synthesis of L series compounds. Reagents and conditions: (a) Pd(dppf)2Cl2, CH3CH2OH, triethylamine; (b) NaHB(OAc)3.
condensation between the furan-aldehyde and the secondary
amine of piperazine-type nitrogen atom. The coupling intermediate
C had to be purified first, and then reacted with the appropriately
substituted heterocyclic using reductive amination to give the M-
series compounds (Scheme 2). Besides, the intermediate C was
oxidized by CrO3eH2SO4 to provide the acid D, followed by the
reaction with N-methyl piperazine to give the final compound K.
The tosylate salts of the target compounds were prepared by the
following protocol: to a solution of each compound (1.0 equivalent)
in THF was added a solution of p-toluenesulfonic acid (3.5 equiva-
lents for M-2 and M-3, and 2.5 equivalent for other compounds) in
ethanol at room temperature. The stirred mixture was refluxed for
2 h, and then allowed to cool to room temperature. The solid formed
was isolated by filtration and dried in vacuum at 45 ꢀC for 24 h.
in vitro anticancer activities of these lapatinib analogs. The cell-
based proliferation assay in HER-2-overexpressing cell lines BT-
474 (human breast cancer cells) and dual EGFR/HER-2 highly
expressing cell lines NCI-N87 (human gastric carcinoma cells) [22],
and the data were summarized in Table 2.
Similar to lapatinib, possibly because the significant interactions
with the target proteins were not changed, most compounds re-
ported here showed potent activity against EGFR or HER-2 kinase as
well as EGFR and HER-2, or HER-2 over-expressing cancer cell lines
except compound K (Table 2). When compared with lapatinib, the
lower EGFRs inhibitory activity and antiproliferative effect of
compound K suggested the amido bond in R groups might be not a
good linkage. In general, L-series compounds were superior to M-
series compounds against the two tested cancer cell lines. Com-
pounds with suitable log P and Log S values (such as L-2, L-3, L-4,
M-3 and M-6) displayed higher activities in comparison with others
of the same series compounds (Tables 1 and 2). When compared
with lapatinib, L-4 had higher antiproliferative effect against the
two tested cancer cell lines (IC50: 13.2 and 18.7 nM) or higher HER-2
inhibitory activity (P < 0.05), M-6 and L-2 were more active against
BT-474 cells (P < 0.05) with IC50 values of 19.2 and 22.5 nM,
respectively. Other L-series compounds had similar in vitro activity
to lapatinib.
2.3. Pharmacology
2.3.1. In vitro evaluation
Our principle behind the structural design of lapatinib analogs is
to improve the drug-like properties without significantly damaging
its EGFRs inhibitory or cancer cell proliferation inhibitory activities.
Therefore, EGFR and HER-2 tyrosine kinase inhibition assay and
cancer cell proliferation inhibition assay were used to assess the
O
O
O
Cl
F
F
F
HN
N
Cl
HN
N
Cl
HN
N
a
b
H
R1
N
I
N
O
O
N
+
OHC
O
B(OH)2
O
B
C
M-1 ~ M-6
A
c
O
O
F
F
N
HN
N
Cl
HN
N
Cl
d
N
N
O
HO
N
O
O
O
D
K
Scheme 2. Synthesis of M series compounds and K. Reagents and conditions: (a) Pd(OAc)2, K2CO3; (b) NaBH3CN or NaBH4; (c) CrO3, H2SO4; (d) SOCl2, N-Methyl piperazine.

836
L. Zhang et al. / European Journal of Medicinal Chemistry 69 (2013) 833e841
Table 2
the NCI-N87 xenograft model, L-2 (200 mg/kg/day) inhibited tumor
growth by 94.8% over the 28-day administration schedule, and
significant tumor regression was observed in all the six mice (the
corresponding values of lapatinib were 89.7% and 5). In the SK-OV-3
xenograft model, L-2 inhibited tumor growth by 85.7% during 21-
day administration cycle, and significant tumor regression was
observed in two animals (the corresponding values of lapatinib
were 78.8% and 1). The in vivo anticancer activity of L-1 (200 mg/kg/
day) was close to lapatinib, and its inhibition rates in two xenograft
models on the final day were 84.7% and 74.6%, respectively.
EGFR and HER-2 kinase inhibition as well as cancer cell proliferation inhibition by
lapatinib and its analogs.
Comp.
IC50 (nM)
EGFR
HER-2
NCI-N87 cells
BT-474 cells
L-1
L-2
L-3
L-4
25.1 ꢂ 9.9
13.0 ꢂ 9.4
18.7 ꢂ 6.1
10.8 ꢂ 2.6
21.1 ꢂ 5.2
26.0 ꢂ 12.4
12.5 ꢂ 3.2
10.8 ꢂ 4.0
28.0 ꢂ 7.8
59.9 ꢂ 26.4
11.4 ꢂ 3.5
232.4 ꢂ 97.5*
16.3 ꢂ 6.9
18.3 ꢂ 10.0
8.5 ꢂ 6.5
47.3 ꢂ 11.1
29.4 ꢂ 7.2
35.1 ꢂ 7.6
22.5 ꢂ 6.8*
27.4 ꢂ 7.7
18.7 ꢂ 3.6*
26.4 ꢂ 9.8
84.6 ꢂ 26.1*
67.5 ꢂ 20.5
41.9 ꢂ 13.0
88.1 ꢂ 26.9*
69.0 ꢂ 21.0
19.2 ꢂ 7.8*
61.8 ꢂ 17.3
37.4 ꢂ 6.4
15.9 ꢂ 4.3
3.9 ꢂ 1.2*
8.6 ꢂ 3.1
33.2 ꢂ 5.8
13.2 ꢂ 4.3*
51.3 ꢂ 12.0
62.8 ꢂ 14.6
42.8 ꢂ 8.0
L-5
M-1
M-2
M-3
M-4
M-5
M-6
K
13.1 ꢂ 4.3
23.7 ꢂ 6.3*
12.3 ꢂ 3.3
14.4 ꢂ 7.0
19.8 ꢂ 5.6
6.8 ꢂ 3.3
29.2 ꢂ 7.2
97.7 ꢂ 24.9*
71.8 ꢂ 15.8
24.8 ꢂ 9.4
2.4. Pharmacokinetics
53.3 ꢂ 19.9*
10.3 ꢂ 3.3
204.8 ꢂ 87.8*
46.2 ꢂ 14.3
The difference between in vitro and in vivo anticancer activities
of those compounds draws our attention: the structures of L-1 and
L-2 are similar to lapatinib, and they are possibly oxidized into
lapatinib in vivo during the absorption, distribution or metabolism.
It has been confirmed that lapatinib is mainly oxidatively metab-
olized by CYP3A4/5 and CYP2C19 in liver [23]. This may raise the
question whether the potent in vivo anticancer activity of L-2 comes
from L-2 itself or its metabolite lapatinib. Further pharmacokinetic
studies on the comparison of their ADME properties were done in
rats, and the exposures of L-1, L-2 and lapatinib after oral admin-
istration of their tosylates (100 mg/kg) were investigated (Table 3).
These compounds displayed linear elimination pharmacoki-
netics, and lapatinib could be detected as a metabolite in rats’
serum after treated with the tosylate of L-1 or L-2 (100 mg/kg by
oral route) (Fig. 4). Calculated by AUC0-t parameter, the metabolite
lapatinib in L-1 group was equivalent to 20% of the parent drug L-1,
and the metabolite lapatinib in L-2 group was equivalent to 7.8% of
the parent drug L-2 (Table 3). Obviously, lapatinib generated from
L-2 by metabolism was only a small portion, and was not the major
contributor to the potent in vivo anticancer activities of L-2. Besides,
L-2 had significantly better absorption and bioavailability after oral
administration, and the maximum serum concentration (Cmax) or
the area under the curve to infinity (AUC0eN) were approximately
twice as high for L-2 as for the positive control lapatinib (Table 3).
Compared with positive control lapatinib, the relative bioavail-
ability of L-2 calculated by AUC0-t was 187%. Moreover, L-2 couldn’t
be detected 24 h after oral administration (Fig. 4b) which suggested
that L-2 eliminated completely in 24 h, and long-term adminis-
tration wouldn’t cause drug accumulation. Because of the low
aqueous solubility, the oral absorption or exposure of L-1 was far
lower than that of the positive control lapatinib, and the relative
bioavailability calculated by AUC0-t was only 29.3% (Table 3).
Lapatinib
Data represent mean ꢂ SD, n ¼ 3, *P < 0.05 vs. lapatinib.
2.3.2. In vivo evaluation
After the promising in vitro results were observed, further in vivo
antitumor activities of L-1 w L-5, M-3 and M-6 were evaluated in
NCI-N87 or SK-OV-3 xenograft models (HER-2-overexpressing hu-
man ovarian tumor cell line) in athymic mice. After given the
tosylate of M-3 or M-6 at a 200 mg/kg/day dosage, the experi-
mental athymic mice became weak or lethargic, further developed
diarrhea or scruffy skin. Four days later, severe weight-loss or he-
maturia was observed, and mice began to die. The toxicity showed
by M-3 and M-6 suggested that the introduction of the piperazine
moiety in R group was less desirable, which terminated our further
studies on M-series compounds. However, all L-series compounds
were well tolerated, and no overt toxicity (eg, hematuria, bloody
stool) or weight-loss was observed (Figs. 2b and 3b).
When compared with the control, reductions of tumor size in
each group except L-4 were observed after treating with L series
compounds’ tosylates (200 mg/kg/day) (Figs. 2a and 3a). L-1, L-2
and lapatinib caused marked regression in NCI-N87 or SK-OV-3
xenograft model (P < 0.01 vs. control) (Table 4 in Supplementary
materials). L-3 and L-5 showed lower activities in comparison
with lapatinib or L-2, and couldn’t achieve effective inhibition in
both xenografts models (Figs. 2a and 3a). Presumably because of its
poor Log S and tPSA (Table 1), L-4, although had the best activity
in vitro, showed the lowest in vivo tumor inhibitory activity among
all compounds tested (Figs. 2a and 3a).
Comparing with lapatinib, the tumor inhibition rate of L-2 was
higher in both xenografts (Table 4 in supplementary materials). In
Fig. 2. Antitumor activities of L-1 w L-5 and lapatinib in NCI-N87 xenograft athymic mice at a dose of 200 mg/kg/day. (a) Tumor volume and (b) body weight were measured on the
indicated days, after treated with vehicle or tosylate salts of the test compounds once a day.

L. Zhang et al. / European Journal of Medicinal Chemistry 69 (2013) 833e841
837
Fig. 3. Antitumor activities of L-1 w L-5 and lapatinib in SK-OV-3 xenograft athymic mice at a dose of 200 mg/kg/day. (a) Tumor volume and (b) body weight were measured on the
indicated days, after treated with vehicle or tosylate salts of the test compounds once a day.
Table 3
Pharmacokinetic study of L-1, L-2 and lapatinib in rats treated with their tosylates.
Parameters
L-1
L-1
L-2
L-2
Lapatinib
Lapatiniba
Lapatinibb
Tmax (h)
3.0 ꢂ 0.0
3.0 ꢂ 0.0
242 ꢂ 104
1994 ꢂ 540
2058 ꢂ 580
8.34 ꢂ 1.04
4.18 ꢂ 0.77
3.5 ꢂ 1.0
4.5 ꢂ 1.0
645 ꢂ 214
4295 ꢂ 977
4309 ꢂ 970
5.88 ꢂ 0.41
2.64 ꢂ 0.58
2.5 ꢂ 0.6
Cmax (ng/ml)
AUC0-t (ng h/ml)
AUC0-N (ng$h/ml)
MRT (h)
T1/2 (h)
Fr
854ꢂ142**
7891ꢂ1305**
8134 ꢂ 1226**
8.19 ꢂ 0.66
4.43 ꢂ 0.69
29%
8881 ꢂ 2061**
50,299 ꢂ 12863*
53,236 ꢂ 12248*
5.38 ꢂ 0.59
2.32 ꢂ 0.5
3902 ꢂ 1208
26,908 ꢂ 9085
26,921 ꢂ 9092
5.52 ꢂ 0.90
1.91 ꢂ 0.45
100%
187%
Data represent mean ꢂ SD, n ¼ 4, *P < 0.05, **P < 0.01 vs. the positive control lapatinib.
a
The metabolite of L-1 in vivo.
The metabolite of L-2 in vivo.
b
2.5. Aqueous solubility
properties, and the modification of this group was an ideal method
to develop a drug candidate with improved aqueous solubility and
pharmacokinetic properties compared with lapatinib. Because M-6
was presumed to remove the acetyl in vivo metabolism, the great
toxicity of M-3 or M-6 might be attributed to the introduction of the
piperazine moiety in R group. For the optimization of drug-like
properties, a very favorable value of one factor doesn’t compensate
for a less favorable value of another [14]. That’s why when compared
with L-2, M-3 or lapatinib, M-2, although had the lowest Log S,
exhibitedloweractivitiesagainstcancercells,andtheunfavorableLog
P might affect its permeability (Tables 1 and 2). When compared with
lapatinib, L-2 had more favorable Log P value besides its improved
aqueous solubility (Table 1), which might result in its good absorption
and permeability, and made L-2 possess higher bioavailability. The
success of L-2 indicated that only other factors of drug-like properties
being comparable or improved, the improvement of aqueous solu-
bility could be a qualitative leap. These findings may provide a
Solubility studies revealed that the measured solubility for the
tosylate salts of L-1, L-2 and lapatinib in unbuffed water at 25 ꢀC
was 3.0 mg/L, 404.6 mg/L and 7.0 mg/L, respectively (Table 1),
which was consistent with their oral absorption, relative bioavail-
ability as well as in vivo anticancer activity. Taken together, we can
infer that the potent in vivo anticancer activity of L-2 may be mainly
attributed to its improved pharmacokinetic profiles and oral
bioavailability which are closely related to its suitable aqueous
solubility and improved Log P.
2.6. Discussion
Our
present
work
confirmed
that
the
methyl-
sulfonylethylaminomethylfuryl group of lapatinib, although did not
make any significant interactions with EGFR, affected its drug-like
Fig. 4. Plasma concentration versus time profiles of parent compound and its metabolite lapatinib in rats treated with the tosylate of L-1, L-2 or lapatinib (100 mg/kg): (a) the curves
of L-1 and its metabolite lapatinib; (b) the curves of L-2 and its metabolite lapatinib; (c) the curve of lapatinib.

838
L. Zhang et al. / European Journal of Medicinal Chemistry 69 (2013) 833e841
valuable clue for the optimization of other candidates or drugs with
similar problems.
4.1.3. 6-(5-((2-(Methylsulfonamido)ethylamino)methyl)furan-2-
yl)-N-(4-(3-fluorobenzyloxy)-3-chlorophenyl)quinazolin-4-amine
(L-3)
The title compound was prepared in a manner similar to that
3. Conclusions
described in L-1 (yield 45%). 1H NMR (600 MHz, DMSO-d6,
d ppm):
9.92 (s, 1H), 8.73 (d, 1H, J ¼ 1.8 Hz), 8.55 (s, 1H), 8.17 (dd, 1H,
J1 ¼ 1.8 Hz, J2 ¼ 8.4 Hz), 8.02 (d, 1H, J ¼ 2.4 Hz), 7.80 (d, 1H,
J ¼ 8.0 Hz), 7.75 (dd, 1H, J1 ¼ 2.4 Hz, J2 ¼ 8.0 Hz), 7.48 (m, 1H), 7.30e
7.35 (m, 2H), 7.29 (d, 1H, J ¼ 8.0 Hz), 7.19 (m, 1H), 7.05 (d, 1H,
J ¼ 3.0 Hz), 6.99 (m, 1H), 6.48 (d, 1H, J ¼ 3.0 Hz), 5.27 (s, 2H), 3.83 (s,
2H), 3.06 (m, 2H), 2.91 (s, 3H), 2.70 (t, 2H). ESI-MS (m/z): [M þ H]þ
596.4.
In summary, a series of new lapatinib analogs were designed and
synthesized, and most of them exhibited potent affinity for EGFR or
HER-2 kinase as well as excellent antiproliferative activities. A potent
EGFR/HER2 inhibitor N-(3-chloro-4-(3-fluorobenzyloxy)phenyl)-6-
(5-((2-(methylsulfinyl)ethylamino)methyl)furan-2-yl)quinazolin-4-
amine (L-2) has been discovered. On the basis of its favorable solu-
bility, pharmokinetic profiles and in vivo efficacies, L-2 (selatinib)
was progressed into full development. It is currently in phase I
clinical trials, and the phase I clinical data will be reported soon. In
addition, a high-yielding one-pot synthetic procedure was devel-
oped for the synthesis of lapatinib and its analogs.
4.1.4. 6-(5-((2-(Sulfamoyl)ethylamino)methyl)furan-2-yl)-N-(4-(3-
fluorobenzyloxy)-3-chlorophenyl) quinazolin-4-amine (L-4)
The title compound was prepared in a manner similar to that
described in L-1 (yield 74%). 1H NMR (600 MHz, DMSO-d6,
d ppm):
9.93 (s, 1H), 8.56 (s, 1H), 8.76 (s, 1H), 8.18 (dd, 1H, J1 ¼ 1.8 Hz,
J2 ¼ 8.0 Hz), 8.20 (d, 1H, J ¼ 2.4 Hz), 7.81 (d, 1H, J ¼ 8.4 Hz), 7.76 (dd,
1H, J1 ¼ 2.4 Hz, J2 ¼ 8.0 Hz), 7.48 (q, 1H), 7.30e7.35 (m, 2H), 7.29 (d,
1H, J ¼ 8.0 Hz), 7.20 (m,1H), 7.06 (d,1H, J ¼ 3.0 Hz), 6.89 (s, 2H), 5.27
(s, 2H), 3.87 (s, 2H), 3.19 (m, 2H), 3.00 (m, 2H), 3.13 (m, 2H). ESI-MS
(m/z): [M þ H]þ 582.3.
4. Experimental protocols
4.1. Chemistry
The reagents were all analytically grade. Yields referred to pu-
rified products and were not optimized. 1H NMR spectra were
recorded on a Varian 600 spectrometer in DMSO-d6 with TMS as
the internal reference. The values of chemical shifts were expressed
in ppm. MS spectra were determined using either an Agilent 1100
LC/MSD or a Thermo Fisher Scientific LCQFleet LC/MSn, and the
signals were given in m/z.
4.1.5. 6-(5-((2-(2,2,2-Trifluoroethylsulfonyl)ethylamino)methyl)
furan-2-yl)-N-(4-(3-fluorobenzyloxy)-3-chlorophenyl)quinazolin-
4-amine (L-5)
The title compound was prepared in a manner similar to that
described in L-1 (yield 53%). 1H NMR (600 MHz, DMSO-d6,
d ppm):
9.92 (s, 1H), 8.74 (s, 1H), 8.56 (s, 1H), 8.16 (dd, 1H, J1 ¼ 1.2 Hz,
J2 ¼ 8.4 Hz), 8.02 (d, 1H, J ¼ 2.4 Hz), 7.75 (dd, 1H, J1 ¼ 1.8 Hz,
J2 ¼ 9.0 Hz), 7.80 (d, 1H, J ¼ 8.4 Hz), 7.48 (q, 1H), 7.35 (d, 1H,
J ¼ 9.0 Hz), 7.32 (m, 1H), 7.28 (d,1H, J ¼ 9.6 Hz), 7.20 (m, 1H), 7.06 (d,
1H, J ¼ 3.0 Hz), 5.27 (s, 2H), 4.77 (m, 2H), 3.84 (s, 2H), 3.45 (m, 2H),
3.05 (t, 2H). ESI-MS (m/z): [M þ H]þ 649.4.
4.1.1. 6-(5-((2-(Methylthio)ethylamino)methyl)furan-2-yl)-N-(4-
(3-fluorobenzyloxy)-3-chlorophenyl)-4-quinazolinamine (L-1)
A mixture of 2-(methylthio)ethanamine hydrochloride (3.2 g,
25 mmol), 5-formylfuran-2-ylboronic acid (2.5 g, 18 mmol) and trie-
thylamine (28 ml, 0.2 mol) was stirred under nitrogen in ethanol
(200 ml), followed by the addition of N-(3-chloro-4-(3-
fluorobenzyloxy)phenyl)-6-iodoquinazolin-4-amine
(5.1
g,
4.1.6. N-(3-Chloro-4-(3-fluorobenzyloxy)phenyl)-6-(5-
(morpholinomethyl)furan-2-yl)quinazolin-4-amine (M-1)
10 mmol) and Pd(dppf)2Cl2 (0.6 g, 0.8 mmol). The mixture was stirred
at room temperature for 1 h, and then heated under reflux for 1.5 h.
After cooled to room temperature, NaHB(OAc)3 (2.5 g, 12 mmol) was
added, and the mixture was stirred at room temperature for 2 h. The
mixture was filtered carefully and washed with EtOH (3 ꢃ 30 ml). The
filtrate and washings were combined and concentrated, then diluted
with CH2Cl2 (500 ml). The organic phase was washed with H2O
(3 ꢃ 300 ml), dried (Na2SO4), filtered and concentrated to give the
crude product. The crude product was purified on a silica gel column
using Ethyl acetate/MeOH to provide the title compound (L-1) as a
A
mixture of N-(3-chloro-4-(3-fluorobenzyloxy)phenyl)-6-
iodoquinazolin-4-amine (10.2 g, 20 mmol), 5-formylfuran-2-
ylboronic acid (3.8 g, 27 mmol), Pd(OAc)2 (0.4 g, 1.8 mmol) and
K2CO3 (11.0 g, 80 mmol) in THF (100 ml) and ethanol (100 ml) was
stirred and heated to reflux for 1 h in a nitrogen atmosphere. The
reaction mixture was cooled to room temperature and diluted
with THF (200 ml) and ethanol (200 ml). Solid was filtered off and
washed with 100 ml of THF. The combined organic layers were
concentrated, diluted with H2O (500 ml) and stirred for 1 h. The
yellow solid was collected, washed with ethanol (100 ml) and
dried in a vacuum oven for two days at room temperature to give
intermediate C (7 g, purity 98%).
yellow solid (2.8 g, yield 50%). 1H NMR (600 MHz, DMSO-d6,
d ppm):
9.93 (s,1H), 8.73 (s,1H), 8.55 (s,1H), 8.16 (d,1H, J ¼ 2.4 Hz), 8.01 (d,1H,
J ¼ 2.4 Hz), 7.80 (d,1H, J ¼ 7.4Hz), 7.74 (dd,1H, J ¼ 2.4 Hz, J ¼ 9 Hz), 7.45
(m,1H), 7.34 (d,1H, J ¼ 7.8 Hz), 7.32 (s,1H), 7.29 (d,1H, J ¼ 8.4 Hz), 7.19
(t,1H, J ¼ 8.4 Hz), 7.05 (d,1H, J ¼ 3.0 Hz), 6.48 (d,1H, J ¼ 3.0 Hz), 5.25(s,
2H), 3.83 (s, 2H), 2.77 (t, 2H, J ¼ 7.2 Hz), 2.59 (t, 2H, J ¼ 7.2 Hz), 2.04 (s,
3H); ESI-MS (m/z): [M þ H]þ 548.5.
To a solution of intermediate C (0.47 g, 1 mmol) and morpho-
line (0.14 ml, 1.6 mmol) in CH2Cl2 (100 ml) and methanol (30 ml)
was added NaBH(OAc)3 (0.380 g, 1.80 mmol). After 25 h, the re-
action was diluted with H2O (300 mL) and extracted with CH2Cl2
(3 ꢃ 250 ml). The organic layers were combined, dried (Na2SO4)
and concentrated. Chromatography of the residue (in CH2Cl2/
MeOH) gave the title compound as an orange solid (0.21 g, 39%).
4.1.2. 6-(5-((2-(Methylsulfinyl)ethylamino)methyl)furan-2-yl)-N-
(4-(3-fluorobenzyloxy)3-chlorophenyl)-4-quinazolinamine (L-2)
The title compound was prepared in a manner similar to that
1H -NMR (600 MHz, DMSO-d6,
d ppm): 9.96 (s, 1H), 8.73 (s, 1H),
described in L-1 (yield 58%). 1H NMR (600 MHz, DMSO-d6,
d ppm):
8.56 (s, 1H), 8.15 (d, 1H, J ¼ 8.4 Hz), 8.01 (d, 1H, J ¼ 2.4 Hz), 7.80 (d,
1H, J ¼ 8.4 Hz), 7.74 (dd, 1H, J1 ¼ 9.0 Hz, J2 ¼ 2.4 Hz), 7.48 (m, 1H),
7.34 (d, 1H, J ¼ 7.8 Hz), 7.32 (s, 1H), 7.29 (d, 1H, J ¼ 9 Hz), 7.20 (m,
1H), 7.07 (d, 1H, J ¼ 3.0 Hz), 6.53 (d, 1H, J ¼ 3.0 Hz), 5.23 (s, 2H),
3.61 (broad s, 4H), 3.59 (s, 2H), 2.46 (broad s, 4H). ESI-MS (m/z):
[M þ H]þ 545.2.
10.13 (s, 1H), 9.13 (s, 1H), 8.57 (s, 1H), 8.21 (dd, 1H, J1 ¼ 1.2 Hz,
J2 ¼ 8.4 Hz), 8.15 (d, 1H, J ¼ 3.0 Hz), 7.89 (dd, 1H, J1 ¼ 1.8 Hz,
J2 ¼ 8.4 Hz), 7.81 (d,1H, J ¼ 8.0 Hz), 7.48 (q,1H), 7.35 (d,1H, J ¼ 9 Hz),
7.32 (m, 1H), 7.28 (d, 1H, J ¼ 9.6 Hz), 7.18e7.21 (m, 2H), 6.69 (d, 1H,
J ¼ 3.0 Hz), 5.27 (s, 2H), 4.17 (s, 2H), 3.22 (m, 4H), 3.0 (m,1H), 2.62 (s,
3H). ESI-MS (m/z): [M þ H]þ 565.1.

L. Zhang et al. / European Journal of Medicinal Chemistry 69 (2013) 833e841
839
4.1.7. N-(3-Chloro-4-(3-fluorobenzyloxy)phenyl)-6-(5-((4-
methylpiperazin-1-yl)methyl)furan-2-yl)quinazolin-4-amine (M-2)
The title compound was prepared in a manner similar to that
(preparation method: 11 ml of 98% H2SO4 was added into the
mixture of CrO3 (10 g) and H2O (50 ml) at 0 ꢀC). The reaction
mixture was stirred for 2 h at 0 ꢀC, then diluted with H2O (150 ml).
The solid was filtered, washed with saturated aqueous solution of
ammonium chloride and dried in a vacuum oven for 10 h at 60 ꢀC to
give the intermediate D (0.98 g).
A mixture of D (0.98 g) and SOCl2 (0.3 ml) in CH2Cl2 (50 ml) was
stirred and heated under reflux for 2 h. After the solvent was
removed, the residue was suspended in the solvent of CH2Cl2
(20 ml) and N-methyl piperazine (3 ml). The mixture was heated
under reflux for 4 h, then diluted with H2O (150 ml) and concen-
trated at 35 ꢀC to remove the CH2Cl2 and filtered. The residue was
dried, and purified on a silica gel column using ethyl acetate/MeOH
to provide the title compound (K) (0.35 g, yield 29%). 1H NMR
described in M-1 (yield 43%). 1H NMR (600 MHz, DMSO-d6,
d ppm):
9.96 (s, 1H), 8.72 (s, 1H), 8.55 (s, 1H), 8.15 (dd, 1H, J1 ¼ 9.0 Hz,
J2 ¼ 1.8 Hz), 8.00 (d, 1H, J ¼ 2.4 Hz), 7.80 (d, 1H, J ¼ 9.0 Hz), 7.73 (dd,
1H, J1 ¼ 9.0 Hz, J2 ¼ 3.6 Hz), 7.48 (m, 1H), 7.34 (d, 1H, J ¼ 7.8 Hz), 7.32
(s, 1H), 7.29 (d, 1H, J ¼ 9 Hz), 7.19 (m, 1H), 7.07 (d, 1H, J ¼ 3.6 Hz),
6.51 (d, 1H, J ¼ 3.6 Hz), 5.27 (s, 2H), 2.35e2.39 (m, 8H), 3.59 (s, 2H),
2.15 (s, 3H). ESI-MS (m/z): [M þ H]þ 558.3.
4.1.8. N-(3-Chloro-4-(3-fluorobenzyloxy)phenyl)-6-(5-(piperazin-
1-ylmethyl)furan-2-yl)quinazolin-4-amine (M-3)
The title compound was prepared in a manner similar to that
described in M-1 (yield 37%). 1H NMR (600 MHz, DMSO-d6,
d
ppm):
(600 MHz, DMSO-d6, d ppm): 10.04 (s, 1H), 8.89 (s, 1H), 8.23 (s, 1H),
10.16 (s, 1H), 8.92 (s, 1H), 8.56 (s, 1H), 8.15 (dd, 1H, J1 ¼ 9.0 Hz,
J2 ¼ 1.8 Hz), 8.07 (d, 1H, J ¼ 2.4 Hz), 7.80e7.82 (m, 2H), 7.47 (q, 1H),
7.32e7.35 (m, 2H), 7.28 (d, 1H, J ¼ 9.6 Hz), 7.18e7.20 (m, 2H), 6.56
(d,1H, J ¼ 3.6 Hz), 5.27 (s, 2H), 3.68 (s, 2H), 3.00 (m, 4H), 2.60 (broad
s, 4H). ESI-MS (m/z): [M þ H]þ 544.2.
8.22 (d, 1H, J ¼ 9.0 Hz), 8.03 (d, 1H, J ¼ 2.4 Hz), 7.85 (d, 1H,
J ¼ 9.0 Hz), 7.75 (dd, 1H, J1 ¼ 9.0 Hz, J2 ¼ 2.4 Hz), 7.48 (m, 1H), 7.33e
7.27 (m, 4H), 7.21e7.15 (m, 2H), 5.27 (s, 2H), 3.75 (s, 4H), 2.41 (s, 4H),
2.24 (s, 3H). ESI-MS (m/z): [M þ H]þ 572.3.
4.2. Biological evaluation
4.1.9. Ethyl 4-((5-(4-(3-chloro-4-(3-fluorobenzyloxy)phenylamino)
quinazolin-6-yl)furan-2-yl)methyl)piperazine-1-carboxylate (M-4)
To a stirring mixture of M-3 (544 mg, 1 mmol) and Na2CO3
(212 mg, 2 mmol) in CH2Cl2 (100 ml) at 0 ꢀC was added dropwise a
solution of ethyl chloroformate (119 mg, 1.1 mmol) in CH2Cl2
(25 ml). The mixture was stirred for 1 h at 0 ꢀC, then quenched with
H2O (1 ml). Solid was filtered off and washed with 20 ml of CH2Cl2.
The combined filtrate and washings were washed with H2O
(3 ꢃ 100 ml) and dried (Na2SO4). After the solvent was removed, the
residue was purified on a silica gel column using CH2Cl2/MeOH to
give the title compound (523.6 mg, 85% yield). 1H -NMR (600 MHz,
4.2.1. EGFR and HER-2 tyrosine kinase enzyme inhibition assay
EGFR or HER-2 tyrosine kinase activity was determined by an
enzyme-linked-immuno-sorbent assay (ELISA) [24] in 96-well
plates pre-coated with 2.5
the substrate. A mixture of serially diluted test compounds and
M ATP diluted in 100 l of reaction buffer (50 mM HEPES, pH 7.4,
mg/well of Poly (Glu, Tyr 4:1, Sigma) as
5
m
m
20 mM MgCl2, 0.2 mM MnCl2, 0.1 mM Na3VO4, 1 mM DTT) was
added to each well. The reaction was initiated by the addition of
0.5
Co.). After incubation for 1 h at 37 ꢀC, the plate was washed three
times with T-PBS (PBS containing 0.1% Tween20). 100 l of anti-
ml of EGFR (Upstate Co.) or HER-2 (Cell Signaling Technology
DMSO-d6,
d
ppm): 9.96 (s, 1H), 8.73 (s, 1H), 8.56 (s, 1H), 8.16 (dd, 1H,
m
J1 ¼ 9.0 Hz, J2 ¼ 1.8 Hz), 8.01 (d, 1H, J ¼ 2.4 Hz), 7.80 (d, 1H,
J ¼ 8.4 Hz), 7.73 (m, 1H), 7.48 (m, 1H), 7.35 (m, 1H), 7.29 (d, 1H,
J ¼ 9.0 Hz), 7.19 (m, 1H), 7.07 (d, 1H, J ¼ 3.0 Hz), 6.53 (d, 1H,
J ¼ 3.6 Hz), 5.27 (s, 2H), 4.02 (m, 2H), 3.64 (s, 2H), 3.38 (m, 4H), 2.44
(m, 4H), 1.16 (t, 3H). ESI-MS (m/z): [M þ H]þ 616.1.
phospho-tyrosine (PY99) (1:500) antibody was added. Incubation
for 1 h at room temperature, the plate was washed three times with
T-PBS. Next, the plate was incubated with HRP-conjugated goat
anti-mouse IgG secondary antibody for 1 h at room temperature
and washed as before. 100 ml of the reaction solution (0.03% H2O2,
2 mg/ml o-phenylenediamine in citrate buffer 0.1 M, pH 5.5) was
4.1.10. N-(3-Chloro-4-(3-fluorobenzyloxy)phenyl)-6-(5-((4-
(methylsulfonyl)piperazin-1-yl)methyl)furan-2-yl)quinazolin-4-
amine (M-5)
added, and the plate was incubated at room temperature until color
emerged. The reaction was terminated by the addition of 100 ml of
2 M H2SO4, and A492 was measured using a multi-well spectro-
photometer (SPECTRAMAX190Ô). IC50 calculations and analysis
were carried out using Microsoft Excel. Comparisons between all
groups were made using a two-tailed Student’s t test (SPSS version
11.5 software). Data represented as means ꢂ SD from three inde-
pendent experiments, and differences between groups were
considered statistically significant at p < 0.05.
The title compound was prepared in a manner similar to that
described in M-4 (yield 90%). 1H NMR (600 MHz, DMSO-d6,
d ppm):
9.95 (s,1H), 8.73 (s,1H), 8.56 (s,1H), 8.16 (d,1H, J ¼ 9.0 Hz), 8.01 (d,1H,
J ¼ 1.8 Hz), 7.80 (d,1H, J ¼ 8.4 Hz), 7.73 (dd,1H, J1 ¼9.0 Hz, J2 ¼ 2.4 Hz),
7.48 (m, 1H), 7.32e7.34 (m, 2H), 7.29 (d, 1H, J ¼ 8.4 Hz), 7.19 (m, 1H),
7.08 (d, 1H, J ¼ 3.0 Hz), 6.55 (d,1H, J ¼ 3.0 Hz), 5.27 (s, 2H), 3.68 (s, 2H),
3.14 (s, 4H), 2.57 (s, 4H), 2.87 (s, 3H). ESI-MS (m/z): [M þ H]þ 622.1.
4.2.2. Cancer cell proliferation inhibition assay
4.1.11. 1-(4-((5-(4-(3-Chloro-4-(3-fluorobenzyloxy)phenylamino)
quinazolin-6-yl)furan-2-yl)methyl)piperazin-1-yl)ethanone (M-6)
The title compound was prepared in a manner similar to that
Human gastric adenocarcinoma NCI-N87 and breast carcinoma
BT-474 cell lines were purchased from the American Type Culture
Collection (Manassas, USA) and cultured in RPMI 1640 medium
supplemented with 10% FBS. Cell cultures were maintained in a
humidified atmosphere of 5% CO2 at 37 ꢀC. Cells were seeded at
respective density (8000 cells per well for NCI-N87 and 10,000 cells
described in M-4 (yield 87%). 1H NMR (600 MHz, DMSO-d6,
d ppm):
9.96 (s, 1H), 8.73 (s, 1H), 8.56 (s, 1H), 8.16 (dd, 1H, J1 ¼ 8.4 Hz,
J2 ¼ 1.8 Hz), 8.01 (d, 1H, J ¼ 2.4 Hz), 7.80 (d, 1H, J ¼ 9.0 Hz), 7.73 (dd,
1H, J1 ¼ 9.0 Hz, J2 ¼ 2.4 Hz), 7.48 (m, 1H), 7.34 (d, 1H, J ¼ 7.2 Hz), 7.32
(s, 1H), 7.29 (d, 1H, J ¼ 8.4 Hz), 7.20 (m, 1H), 7.08 (d, 1H, J ¼ 3.0 Hz),
6.55 (d, 1H, J ¼ 3.0 Hz), 5.27 (s, 2H), 3.65 (s, 2H), 3.44 (m, 4H), 2.41
(m, 4H), 2.00 (s, 3H). ESI-MS (m/z): [M þ H]þ 586.4.
per well for BT-474) in 96-well plates in a volume of 200 ml per well.
The test compounds were dissolved in DMSO and diluted with
culture medium to different concentrations (the final concentration
of DMSO was 0.1%). 24 h after seeding, the medium was removed,
and 200 ml of the test compound solution was added in duplicates
4.1.12. (5-(4-(3-Chloro-4-(3-fluorobenzyloxy)phenylamino)
quinazolin-6-yl)furan-2-yl)(4-methylpiperazin-1-yl)methanone (K)
To a stirring mixture of intermediate C (1 g) in acetone (100 ml)
at 0 ꢀC was added dropwise the solution of CrO3eH2SO4 (10 ml)
and incubation continued for 72 h at 37 ꢀC in a humidified atmo-
sphere containing 5% CO2. After removing the medium from the
plate, cells were fixed with 10% trichloroacetic acid (50
ml) for 1 h at
37 ꢀC, and then washed 5 times with tap water and air-dried. Cells

840
L. Zhang et al. / European Journal of Medicinal Chemistry 69 (2013) 833e841
that survived were stained with 0.4% (w/v) SRB dye solution
(100 l) for 20 min at room temperature. After removing the SRB
dye solution, the plate was washed with water and dried. 150 l of
area of each compound at the concentration of 100 mg/L (prepa-
ration: 10 mg of compound was dissolved in the mixed solution of
water (35 ml), methanol (25 ml) and acetonitrile (40 ml)) was
m
m
Trizma base (10 mM) was added to each well, and the absorbance at
540 nm was measured using a microplate reader. IC50 calculations
and analysis were carried out using Microsoft Excel. Comparisons
between all groups were made using a two-tailed Student’s t test
(SPSS version 11.5 software). Data represented as means ꢂ SD from
three independent experiments, and differences between groups
were considered statistically significant at p < 0.05.
determined as Ar earlier using HPLC with a Xbridge, C18, 5 mm,
250 ꢃ 4.6 mm guard column. The chromatographic conditions
were as follows: mobile phase was a mixed solution of 30 mM
K2HPO4/Methanol/Acetonitrile (35/45/20), flow speed was 1.0 ml/
min, UV detection was at 262 nm, and temperature was at 35 ꢀC.
The examined compound was dissolved in solid excess in 10 ml of
water in a glass vial. The solution was shaken for 24 h on a Water-
bath Thermostatic Oscillator at 25 ꢀC until the solubility equilib-
rium. The solution was left to sediment for 24 h at 25 ꢀC, and then
taken the clear part. The peak area of each examined compound
was measured as As by HPLC using the above condition. The
aqueous solubility was calculated using the external standard
method by peak area as follows: solubility ¼ (As/Ar) ꢃ 100 mg/L.
4.2.3. In vivo efficacy study
Pathogen-free, 4e6 week-old, female BALB/c athymic mice (Vital
River Laboratory Animal Technology Co. Ltd) were housed under
sterile conditions. Human gastric adenocarcinoma NCI-N87 or ovary
cancer SK-OV-3 xenograft was established by inoculated 2 ꢃ 106 cells
subcutaneous in the right flanks of athymic mice. When the tumor
reached a volume of 100e300 mm3, the mice were randomly
assigned into control (n ¼ 10 per group) and treatment groups (n ¼ 6
per group). Control groups were given vehicle, and treatment groups
were orally administered with the tosylate salts of the tested com-
pounds (200 mg/kg, qd) by gavage. The size of tumor was measured
individually with microcalipers on the indicated days. Tumor volume
(V) was calculated as V ¼ (length ꢃ width2)/2. The individual relative
tumor volume (RTV) was calculated as follows: RTV ¼ Vt/V0, where
Vt represented the last tumor size measurement and V0 represented
the predosing tumor size measurement (in case where tumor
regression occurred, RTV < 1). Mean RTV of each group was calcu-
lated as follows: mean RTV ¼ Sum of RTV/n, where n represented the
number of animals per group. Results are presented as mean
RTV ꢂ SD for each group. Therapeutic effect of each compound was
expressed in terms of Tumor Inhibition Rate and the calculation
formula was: Tumor Inhibition Rate ¼ (1 ꢁ T/C) ꢃ 100%, where T was
mean RTV of the treated group and C was mean RTV of the control
group. Comparisons between all groups were made using a two-
tailed Student’s t test (SPSS version 11.5 software). Differences be-
tween groups were considered statistically significant at P < 0.05.
Acknowledgments
This work was supported by the Project of Innovative Antineo-
plastic Drug Incubation Base Construction from the Ministry of
Science and Technology of PR China (2011ZX09401-015).
Appendix A. Supplementary data
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.ejmech.2013.09.032.
References
[1] J. Schlessinger, Cell signaling by receptor tyrosine kinases, Cell 103 (2000)
211e225.
[2] Y. Yarden, M.X. Sliwkowski, Untangling the ErbB signalling network, Nat. Rev.
Mol. Cell Biol. 2 (2001) 127e137.
[3] M.A. Olayioye, R.M. Neve, H.A. Lane, et al., The ErbB signaling network: re-
ceptor heterodimerization in development and cancer, EMBO J. 19 (2000)
3159e3167.
[4] R.S. Herbst, C.J. Langer, Epidermal growth factor receptors as a target for
cancer treatment: the emerging role of IMC-C225 in the treatment of lung and
head and neck cancers, Semin. Oncol. 29 (2002) 27e36.
4.3. Pharmacokinetic study in rats
[5] M.G. Kris, R.B. Natale, R.S. Herbst, et al., Efficacy of gefitinib, an inhibitor of
the epidermal growth factor receptor tyrosine kinase, in symptomatic pa-
tients with non-small cell lung cancer: a randomized trial, JAMA 290 (2003)
2149e2158.
[6] M.H. Cohen, J.R. Johnson, Y.F. Chen, et al., FDA drug approval summary:
erlotinib (TarcevaÒ) tablets, The Oncologist 10 (2005) 461e466.
[7] S. Maheswaran, L.V. Sequist, S. Nagrath, et al., Detection of mutations in EGFR
in circulating lung-cancer cells, N. Engl. J. Med. 359 (2008) 366e377.
[8] M. Ranson, L.A. Hammond, D. Ferry, et al., ZD1839, a selective oral epidermal
growth factor receptoretyrosine kinase inhibitor, is well tolerated and active
in patients with solid, malignant tumors: results of a phase I trial, J. Clin.
Oncol. 20 (2002) 2240e2250.
[9] F. Ciardiello, G. Tortora, A novel approach in the treatment of cancer: targeting
the epidermal growth factor receptor, Clin. Cancer Res. 7 (2001) 2958e2970.
[10] S.R.D. Johnston, A. Leary, Lapatinib: a novel EGFR/HER-2 tyrosine kinase in-
hibitor for cancer, Drugs Today 42 (2006) 441.
[11] N.L. Spector, K.L. Blackwell, J. Hurley, et al., EGF103009, a phase II trial of
lapatinib monotherapy in patients with relapsed/refractory inflammatory
breast cancer (IBC): clinical activity and biologic predictors of response, J. Clin.
Oncol. 24 (2006) 502.
[12] H.A. Burris III, H.I. Hurwitz, E.C. Dees, et al., Phase I safety, pharmacokinetics,
and clinical activity study of lapatinib (GW572016), a reversible dual inhibitor
of epidermal growth factor receptor tyrosine kinases, in heavily pretreated
patients with metastatic carcinomas, J. Clin. Oncol. 23 (2005) 5305e5313.
[13] A.K. Bence, E.B. Anderson, M.A. Halepota, et al., Phase I pharmacokinetic
studies evaluating single and multiple doses of oral GW572016, a dual EGFR-
ErbB2 inhibitor, in healthy subjects, Invest New Drugs 23 (2005) 39e49.
[14] C.A. Lipinski, F. Lombardo, B.W. Dominy, et al., Experimental and computa-
tional approaches to estimate solubility and permeability in drug discovery
and development settings, Adv. Drug Deliv. Rev. 46 (2001) 3e26.
[15] N.R. Budha, A. Frymoyer, G.S. Smelick, et al., Drug absorption interactions
between oral targeted anticancer agents and PPIs: is pH-dependent solu-
bility the Achilles heel of targeted therapy? Clin. Pharmacol. Ther. 92 (2012)
203e213.
A Thermo-Finnigan LCQ ion trap mass spectrometer (San Jose, CA,
USA) equipped with an electrospray ionization (ESI) source was used
to identify L-1, L-2 and the metabolite lapatinib. Male Wistar rats
(220 ꢂ 10 g) were purchased from the Lab Animal Center of She-
nyang Pharmaceutical University (Shenyang, China). The formulation
for oral administration was prepared by suspending drugs in sodium
carboxymethylcellulose solution (0.5%). Male rats were randomly
assigned into three groups (n ¼ 4 per group), and were respectively
administered with the tosylate salts of L-1, L-2 and lapatinib orally at
100 mg/kg by gavage. Blood samples were separately collected at
0.25, 0.5, 1.0, 2.0, 3.0, 5.0, 7.0, 9.0, 12.0, 24.0 h. A solid-phase extrac-
tion method was used to extract L-1, L-2 and their metabolites as
well as lapatinib from the blood samples. Samples (100 ml) were
applied to a preconditioned Bond Elute C18 cartridge (Teda Fuji,
Tianjin, China) and analyzed by LC/MSn to identify L-1, L-2 and their
metabolites. When compared with lapatinib, relative bioavailability
of the test compound was calculated by AUC0-t parameter as follows:
Relative bioavailability (Fr) ¼ AUC0-t t/AUC0-t r ꢃ 100%, where AUC0-t
t represented the AUC0-t of the test compound, and AUC0-t r repre-
sented the AUC0-t of lapatinib.
4.4. Measurements of aqueous solubility
The aqueous solubility of the tosylate salts of L-1 w L-5 was
determined by the shake-flask method [25]. The reference peak

L. Zhang et al. / European Journal of Medicinal Chemistry 69 (2013) 833e841
841
[16] J. Stamos, M.X. Sliwkowski, C. Eigenbrot, Structure of the epidermal growth
factor receptor kinase domain alone and in complex with a 4-anilinoquina-
zoline inhibitor, J. Biol. Chem. 277 (2002) 46265e46272.
[17] C.H. Yun, T.J. Boggon, Y. Li, et al., Structures of lung cancer-derived EGFR
mutants and inhibitor complexes: mechanism of activation and insights into
differential inhibitor sensitivity, Cancer Cell 11 (2007) 217e227.
[18] M. Scaltriti, C. Verma, M. Guzman, et al., Lapatinib, a HER2 tyrosine kinase
inhibitor, induces stabilization and accumulation of HER2 and potentiates
trastuzumab-dependent cell cytotoxicity, Oncogene 28 (2009) 803e814.
[19] E.R. Wood, A.T. Truesdale, O.B. McDonald, et al., A unique structure for
epidermal growth factor receptor bound to GW572016 (Lapatinib) relation-
ships among protein conformation, inhibitor off-rate, and receptor activity in
tumor cells, Cancer Res. 64 (2004) 6652.
[21] P.J. Medina, S. Goodin, Lapatinib: a dual inhibitor of human epidermal growth
factor receptor tyrosine kinases, Clin. Ther. 30 (2008) 1426.
[22] D.W. Rusnak, K.J. Alligood, R.J. Mullin, et al., Assessment of epidermal
growth factor receptor (EGFR, ErbB1) and HER-2 (ErbB2) protein
expression levels and response to lapatinib (TykerbÒ, GW572016) in an
expanded panel of human normal and tumour cell lines, Cell Prolif. 40
(2007) 580e594.
[23] T.E. Kim, J.R. Murren, Lapatinib ditosylate GlaxoSmithKline, IDrugs 6 (2003)
886e893.
[24] L. Zhong, X.N. Guo, X.H. Zhang, et al., Expression and purification of
the catalytic domain of human vascular endothelial growth factor re-
ceptor
2 for inhibitor screening, Biochim. Biophys. Acta 1722 (2005)
254e261.
[20] J.M. Zhang, P.L. Yang, N.S. Gray, Targeting cancer with small molecule kinase
inhibitors, Nat. Rev. Cancer 9 (2009) 28e39.
[25] E. Baka, K. Takács-Novák, Study on standardization of shake-flask solubility
determination method, Eur. J. Pharm. Sci. 32 (2007) S23eS24.