Discovery of a Furanopyrimidine-Based Epidermal Growth Factor Receptor Inhibitor (DBPR112) as a Clinical Candidate for the Treatment of Non-Small Cell Lung Cancer

Article abstract of DOI:10.1021/acs.jmedchem.9b00722

Epidermal growth factor receptor (EGFR)-targeted therapy in non-small cell lung cancer represents a breakthrough in the field of precision medicine. Previously, we have identified a lead compound, furanopyrimidine 2, which contains a (S)-2-phenylglycinol

Full text of DOI:10.1021/acs.jmedchem.9b00722

Subscriber access provided by UNIV OF DURHAM
Article
Discovery of a Furanopyrimidine-Based Epidermal
Growth Factor Receptor Inhibitor (DBPR112) as a Clinical
Candidate for the Treatment of Non-Small Cell Lung Cancer
Shu-Yu Lin, Yung Chang Hsu, Yi-Hui Peng, Yi-Yu Ke, Wen-Hsing Lin, Hsu-Yi Sun, Fu-Ming Kuo, Pei-
Yi Chen, Tzu-Wen Lien, Chun-Hwa Chen, Chang-Ying Chu, Sing-Yi Wang, Kai-Chia Yeh, Ching-
Ping Chen, Tsu-An Hsu, Suying Wu, Teng-Kuang Yeh, Chiung-Tong Chen, and Hsing-Pang Hsieh
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.9b00722 • Publication Date (Web): 27 Sep 2019
Downloaded from pubs.acs.org on September 28, 2019
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 67
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 67
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Discovery of a Furanopyrimidine-Based Epidermal Growth Factor Receptor
Inhibitor (DBPR112) as a Clinical Candidate for the Treatment of Non-Small
Cell Lung Cancer
†
†
†
†
†
†
Shu-Yu Lin, Yung Chang Hsu, Yi-Hui Peng, Yi-Yu Ke, Wen-Hsing Lin, Hsu-Yi Sun, Fu-
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
†
†
†
†
†
Ming Kuo, Pei-Yi Chen, Tzu-Wen Lien, Chun-Hwa Chen, Chang-Ying Chu, Sing-Yi
†
†
†
†
†
†
Wang, Kai-Chia Yeh, Ching-Ping Chen, Tsu-An Hsu, Su-Ying Wu, Teng-Kuang Yeh,
†
,†,‡
Chiung-Tong Chen, and Hsing-Pang Hsieh*
†
Institute of Biotechnology and Pharmaceutical Research, National Health Research
Institutes, 35 Keyan Road, Zhunan, Miaoli County 35053, Taiwan, R.O.C.
‡
Department of Chemistry, National Tsing Hua University, Hsinchu 30013, Taiwan, R.O.C.
Abstract
EGFR-targeted therapy in NSCLC represents a breakthrough in the field of precision medicine.
Previously, we have identified a lead compound, furanopyrimidine 2, which contains a (S)-2-
phenylglycinol structure as a key fragment to inhibit EGFR. However, compound 2 showed
high clearance and poor oral bioavailability in its pharmacokinetics studies. In this work, we
optimized compound 2 by scaffold hopping and exploiting the potent inhibitory activity of
various warhead groups to obtain a clinical candidate, 78 (DBPR112), which not only displayed
a potent inhibitory activity against EGFRL858R/T790M double mutations, but also exhibited 10-
fold potency better than 3rd generation inhibitor, osimertinib, against EGFR and HER2 exon
2
0 insertion mutations. Overall, pharmacokinetic improvement through lead-to-candidate
optimization yielded 4-fold oral AUC better that afatinib along with F = 41.5%, an encouraging
safety profile, and significant antitumor efficacy in in vivo xenograft models. DBPR112 is
currently undergoing phase 1 clinical trial in Taiwan.
1
ACS Paragon Plus Environment

Page 3 of 67
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
INTRODUCTION
Lung cancer has the highest rate of morbidity and mortality among various cancer types, being
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1,2
responsible for over 1.69 million deaths per year. Approximately 80–85% of the lung cancer
3,4
patients diagnosed present non-small cell lung cancer (NSCLC). Mutations in the epidermal
growth factor receptor (EGFR) have been observed in ~50% of NSCLC patients, in particular,
in East Asian patients.4–6 The EGFR exon 19 deletions or exon 21 (L858R) substitution
mutations constitute 90% of all EGFR-activating mutations.3,7
The first generation EGFR-targeted inhibitors, gefitinib and erlotinib, were approved by US
8
9
FDA in 2003 and 2004 to treat mutation-positive patients. Even though the two drugs showed
remarkable objective response rates in EGFR-mutant NSCLC patients,10.11 resistance can be
acquired due to a secondary T790M mutation after treatment.12,13 More recently, the irreversible
second-generation EGFR tyrosine kinase inhibitor (TKI) afatinib was developed. It contains an
acrylamide moiety that can undergo covalent interaction with Cys797 at the active site to
14
enhance the binding affinity and prolong progression-free survival. FDA initially approved
15
16
afatinib in 2013 and broadened its indication in 2018 for patients with NSCLC. However,
afatinib showed potent activity against the wild-type (WT) EGFR, resulting in toxicity and
serious adverse effects.17,18 Very recently, AstraZeneca developed the selective covalent EGFR
inhibitor AZD9291 (osimertinib), which received regular approval in 2017 for EGFR-T790M-
mutation-positive NSCLC patients that showed cancer progression on EGFR TKI therapy.19,20
2
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 67
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Osimertinib also has been approved by FDA as a frontline treatment in 2018. 21 However, the
22
cost for a one-month treatment of osimertinib is $12,750 USD, and several studies have
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
revealed that patients acquire resistance after clinical use.23–25 On September 27, 2018, FDA
approved dacomitinib for the first-line treatment of metastatic NSCLC patients.26,27
Dacomitinib is orally irreversible second-generation EGFR-TKI same as afatinib.26
In a previous effort to develop EGFR inhibitors, we rapidly constructed a furanopyrimidine
kinase-targeted library and screened it for Aurora and EGFR kinase activities. And we
identified compound 1 with (S)-2-phenylglycinol moiety as a potent WT EGFR inhibitor, with
an IC50 value of 223 nM and no Aurora kinase inhibition. The docking of 1 into EGFR revealed
a possible enhancement of the EGFR kinase activity through the introduction of a Michael
28
acceptor group on the 3-position of the furan ring. Based on this, we introduced an acrylamide
28
group as a Michael acceptor, resulting in compound 2, which showed potent in vitro activity
in both wild and mutant EGFR kinases and potent anti-proliferative activity against HCC827
lung cancer cells. In addition, since the DFG (Asp-Phe-Gly) motif plays an important role in
regulating the kinase activity of EPGR,29–31 we prepared compound 3 by structure-based drug
design (SBDD), which is able to interact with the DFG motif by H-bonding and salt bridge
32
formation. The N,N-dimethylamino tail of 3 moves toward the DFG motif and forms a salt
bridge with the side chain of Asp855, which greatly contributes to its potency against EGFR.32
3
ACS Paragon Plus Environment

Page 5 of 67
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
In this work, we apply SBDD to guide the optimization of lead compound 2 and perform
intensive lead-to-candidate optimization to identify promising candidate into the next
development stage. As a result of this study, we hybridized the characteristic moieties of 2 and
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
to identify the EGFR-targeted drug candidate DBPR112 (Figure 1). DBPR112 has received
investigational new drug (IND) application approval from US FDA (NCT03246854) in 2016
and Taiwan FDA in 2017 and is currently ongoing phase I clinical trials in Taiwan.33
Figure 1. EGFR inhibitors design strategy from initial hit compound 1.
RESULTS AND DISCUSSION
Binding Mode Analysis of Compound 2
The crystalline structure of EGFR kinase complex with 1 (PDB ID 4JQ7) was used to initiate
the covalent docking study of 2. As shown in Figure 2, the binding mode of 2 in the ATP-
binding cleft was very similar to that of 1, except that the additional acryl group of 2 formed a
3
4
covalent bond with the SH group of Cys797. A hydrogen bond was also formed between the
4
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 67
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
pyrimidine N1 and Met793 in the kinase hinge region. The hydroxyl group of (S)-2-
phenylglycinol formed an additional H-bond with Asp855. Moreover, the furanopyrimidine
core and the three phenyl moieties made close contact as sigma-pi hydrophobic interactions
with the residues of Val726, Ala743, Lys745, Leu788, Leu718, and Thr790. The above analysis
indicated that the structure-activity relationship (SAR) studies should focus on the 5,6-fused
pyrimidine (hinge binding) and the perpendicular phenyl ring with the Michael acceptor group
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(covalent binding), and also explore possible modification sites to expand the structural
diversity and identify the most promising drug candidate.
Figure 2. The binding site analysis for compound 2 (orange) docking with EGFR X-ray protein
(PDB code: 4JQ7). Compound 2 forms a covalent bond with Cys797 (as shown in white color
residue), and hydrogen bond with Met793 (as shown in green color residue). The hydrophobic
effect was contributed by Leu718, Val726, Ala743, Lys745, Leu788, and Leu844 (as shown in
cyan color residues).
5
ACS Paragon Plus Environment

Page 7 of 67
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Chemistry
Compounds 42–48 were synthesized as illustrated in Schemes 1 and 2. We synthesized the
building blocks 7, 10, 13, and 17 from appropriate nitrobenzene derivatives (5, 8, 11, and 14)
through modification of published procedures or known methods.35–39 In short, the cyclization
of the 2-hydroxy-1-(3-nitrophenyl)-1-ethanone (6) or N-hydroxy-3-nitrobenzimidoyl chloride
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(16) with malononitrile under basic conditions yielded the furan 7 and the isoxazole 17,
respectively. The nitrobenzene derivatives 8 and 11 were individually condensed with ethyl
cyanoacetate and malononitrile followed by cyclization in the presence of octasulfur to obtain
40
the thiophenes 10 and 13. The required five-membered heteroaryl compounds 7, 10, 13, and
1
7 were cyclized and then chlorinated to synthesize the bicyclic [2,3-d]pyrimidines 22–26, as
shown in Scheme 2. The nucleophilic aromatic substitution with (S)-(+)-2-phenylglycinol
derivatives gave the desired compounds 29–35. Bromo-substituent 30 was cross-coupled with
methyl zinc chloride using Negishi coupling to prepare the methylated compound 31. The
reduction of 29–35 using 10% Pd/C in presence of hydrogen gas, iron powder under acidic
condition, or stannic chloride yielded the corresponding amino analogues 36–41. Amide bond
formation of compounds 36–41 with acrylic acid was accomplished in presence of 1-(3-
dimethylaminopropyl)-3-ethylcarbodiimide hydrochloride (EDCI·HCl) to obtain the
corresponding acrylamides 42, 43, and 45–47. The desilylation of OTBS-protected 47 was
carried out with tetrabutylammonium fluoride (TBAF) to give the desired isoxazolo[5,4-
6
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 67
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
d]pyrimidine 48. On the other hand, acrylation of the aniline 38 using acryloyl chloride led to
the acrylamide 44. The synthetic routes for the preparation of amide 70–85 are outlined in
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Scheme 3, considering various functional groups attached to the terminal end of the Michael
acceptor. Starting compounds 49–51 were synthesized by modifying reported procedures.36,41
Bromination of the furan and the thiophene rings at 3-position with N-bromosuccinimide (NBS)
followed by nucleophilic substitution by (R)-(+)-1-phenylethylamine, (R)-(−)- or (S)-(+)-2-
phenylglycinol led to analogues 55–59. Suzuki cross-coupling of analogues 55–59 with 3-
nitrophenylboronic acid yielded the nitro analogues 60–64. Similarly to the synthesis of 42–48,
aniline compounds 65–69 were obtained by reduction of the corresponding nitro compounds
6
0–64 and subsequent coupling with various carboxylic acids or 2-chloroethanesulfonyl
chloride to obtain the analogues 70–85.
7
ACS Paragon Plus Environment

Page 9 of 67
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Scheme 1. Synthesis of the building blocks 7, 10, 13, and 17.a
8
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 67
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Scheme 2. Synthetic route of the EGFR inhibitors for simplification and replacement of
furanopyrimidine through scaffold-hopping.a
9
ACS Paragon Plus Environment

Page 11 of 67
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Scheme 3. Synthesis of EGFR inhibitors with various Michael acceptors.a
Biological Evaluation of EGFR Activating Mutations and SAR Analysis
WT
L858R/T790M
All the novel synthesized compounds were evaluated by EGFR , EGFR
kinase
assays, as well as on two NSCLC cell lines: HCC827 (expressing EGFRdel19) and H1975
28
(
expressing EGFRL858/T790M). The biological data are shown in Tables 1 and 2. Compound 2
was used as the starting point for the structure-activity and lead optimization studies.
In order to understand how the EGFR inhibitory activity was affected by the compound
structure, a preliminary SAR study was performed. Removal of the hydroxyl group from the 2-
phenylglycinol fragment of 2 significantly decreased EGFRWT and EGFRL858R/T790M activity by
1
0
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 67
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
over 6-fold and 18-fold, respectively (cf. 2 versus 70). This result is in good agreement with
our SBDD analysis and indicates that the hydroxyl group (forming a H-bond with Asp855)
plays an important role in maintaining a potent activity against both wild-type and double
mutant EGFR kinases.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Replacement of the phenyl group in the 6-position of the furanopyrimidine scaffold with H and
Me led to a significantly reduced inhibition of EGFRWT and EGFRL858R/T790M (cf. 2 versus 42
and 43). These results indicate that the sigma-pi interactions formed by the phenyl group at the
6
-position are essential for EGFR inhibition, especially of EGFRL858R/T790M
.
Compounds 2 and
4
2 were then used by docking method to illustrate the binding interaction in EGFR double
mutations X-ray structure (PDBID: 5CAS). As shown in Figure 3A, the binding mode of
compound 42 was twisted and induced a steric clash with Met790 as well as influenced on the
Michael acceptor's binding, which is a crucial factor that compound 42 lost the activity of
EGFRL858R/T790M. Furthermore, the introduction of the ortho-methoxy group on the 6-phenyl
moiety caused a 6-fold loss of EGFRL858R/T790M inhibitory activity likely because of the steric
hindrance between two phenyl rings at 5- and 6-position. It can be concluded from
these observations that the 6-phenyl ring lies perpendicular to the 5-phenyl and that the right
positioning of the acrylamide relative to the thiol of Cys797 is crucial for covalent bond
formation. High EGFRL858R/T790M inhibition activity can be achieved when the acrylamide
moiety points to the Cys797 in the protein binding pocket.
1
1
ACS Paragon Plus Environment

Page 13 of 67
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Furthermore, we conducted a scaffold-hopping strategy by replacing the furanopyrimidine with
thieno[2,3-d]pyrimidine and isoxazolo[5,4-d]pyrimidine to obtain 44–46, 48, and 85. Although
these compounds maintained submicromolar IC50 and CC50 against EGFRWT kinase inhibition
and HCC827 cell proliferation, their inhibitory activity for EGFRL858R/T790M was greatly reduced
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(
44–46, 48, and 85; EGFRL858R/T790M IC50 = 650–4626 nM). It is thus concluded that
furanopyrimidine along with two phenyl rings at 5 and 6-positions is a promising scaffold on
which to conduct further SAR studies in order to identify potential EGFR clinical candidates.
Table 1. SAR investigation by scaffold modificationsa.
Compound
EGFRWT
IC50 (nM)
20±10
EGFRL858R/T790M
HCC827
CC50 (nM)
24±11
H1975
X
Y
R
b
b
c
c
No.
IC50 (nM)
CC50 (nM)
2
O
O
O
O
CPh
CPh
CH
OH
H
27±10
500
1,258±302
4300
7
4
4
0
2
3
117
58
OH
OH
159±65
62±22
>10,000
1,322±487
744±197
46±5
>20,000
>20,000
CCH
3
C-
82
O
OH
24±10
157±54
34±4
4,076±1,600
2
’OMePh
44
S
S
S
S
CH
OH
OH
OH
OH
82±44
81±36
2,316±819
1,080±365
1,259±201
650±209
56±21
51±11
35±13
146±32
>20,000
45
CCl
8,656±4,490
12,564±5,945
10,721±1,267
46
CCH
3
107±35
136±36
8
4
5
8
CPh
O
–
N
–
OH
–
416±244
29±13
14±6
4,626±2,359
6,746±1,205
22±7
105±16
26±6
>20,000
18,755±2,778
224±58
gefitinib
afatinib
–
–
–
14±4
1
2
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 67
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
aIC50 and CC50 data are expressed as the mean of at least three independent determinations,
b
presented as mean±SD; The IC50 value was defined as the amount of compound that induced
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
c
a 50% inhibition in enzyme activity in comparison with DMSO-treated controls; The
CC50 value was defined as the amount of compound that caused 50% reduction in cell viability
in comparison with DMSO-treated controls.
Table 2. SAR investigation of Michael acceptors and side chain chiralitya.
EGFRWT
IC50 (nM)
223
EGFRL858R/T790M
HCC827
CC50 (nM)
518
H1975
Compound
R1
R2 R3
b
b
c
c
IC50 (nM)
CC50 (nM)
ND
1
2
–
H
H
Ph
Ph
>10000
20±10
64±22
27±10
24±11
1258±302
6,740±4,429
7,243±2,813
7
7
1
2
H
H
Ph
6,370±2,043
2,459±987
182±16
Ph 3,675±1,196
2,707±127
7
7
7
3
4
5
H
H
H
Ph
Ph
239±49
>10,000
>10,000
>10,000
>10,000
307±130
679±28
13,211±2,450
>20,000
Ph 5,004±1,687
Ph 3,059±941
1,780±93
>20,000
76
77
78
H
H
H
>10,000
779±157
48±12
4,116±499
516±210
25±8
>20,000
7,007±3,715
620±104
Ph
Ph
79±33
15±5
1
3
ACS Paragon Plus Environment

Page 15 of 67
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
7
8
8
9
0
1
H
H
H
Ph
Ph
Ph
14±5
13±2
19±7
131±24
117±40
283±34
62±7
19±5
17±7
2,482±417
2,413±787
2,830±1,503
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
8
8
3
4
Ph
Ph
H
H
238±38
562±37
147±39
840±78
13,397±2.310
6,943±1.166
1,061±59
1,046±35
gefitinib
afatinib
-
-
-
-
-
-
29±13
14±6
6,746±1,205
22±7
26±6
14±4
18,755±2,778
224±58
aIC50 and CC50 data are expressed as the mean of at least three independent determinations,
b
presented as mean ± SD; The IC50 value was defined as the amount of compound that induced
c
a 50% inhibition in enzyme activity in comparison with DMSO-treated controls; The
CC50 value was defined as the amount of compound that caused 50% reduction in cell viability
in comparison with DMSO-treated controls.
Next, we explored the possibility of attaching a variety of Michael acceptors to the 5-phenyl
group of the furanopyrimidine core (Table 2). At first, we attempted to replace the acrylamide
group with butynamide and vinyl sulfonamide. The butynamide 71 inhibited EGFR in a
concentration range very similar to that of gefitinib, while the vinyl sulfonamide 72 drastically
reduced the kinases activities of wild-type and double mutant. Acrylamide group was thus
concluded to be an appropriate Michael acceptor to have in the optimized molecule.
1
4
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 67
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
A series of functional groups were introduced into C=C in order to improve cellular activity,
including hydrophobic groups (alkyl and phenyl groups) and hydrophilic groups (acid, ester,
amine, and alcohol). Molecule extension with various hydrophobic groups at the terminal
double bond resulted in a loss of potency against wild-type and double mutant kinases,
respectively (cf. 2 versus 73–75). To explain the poor activities of compounds 73–76 we applied
covalent docking strategy for compound 74 with the EGFR wild type X-ray (PDB code: 4JQ7)
and double mutation X-ray (PDB code: 5CAS) as representatives. In both of binding mode
analysis, the isopropyl group in terminal double bond of compound 74 has a steric clash with
the Asp800 of wild-type and double mutation proteins, that may give an explanation why
introduction of hydrophobic alkyl or phenyl group lost their inhibition activities (Figures 3B
and 3D). While attachment of carboxylic acid ethyl ester to Michael acceptor led to compound
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
7
7 displayed moderate inhibition in EGFR assays. According to its binding modes with
EGFRWT and EGFRL858R/T790M, the ethyl ester group of 77 evaded a steric clash from Asp800.
Figure 3C and 3E). In T790M and L858R mutation proteins, the modeling exhibited an extra
strong hydrogen bond interaction with the Cys797 backbone NH and two hydrogen bonds with
(
Asn842 and Arg855 (Figure 3E). By contrast, introduction of –CO H lost its bioactivities due
2
to electronic repulsion with –CO H of Asp800. (i.e., 76). Moreover, extension with di-
2
substituted amino groups at the terminal double bond (i.e., 78–81) maintained EGFR activities
i.e., 78–81). It is noteworthy that di-substituted amino functionalities are well tolerated for the
(
1
5
ACS Paragon Plus Environment

Page 17 of 67
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
activity against HCC827 cells (78–81, CC50 = 17–62 nM). When the size of the
alkyl substituent of amino group increased, inhibition of EGFRL858R/T790M slightly decreased (cf.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
8 versus 79–81), but inhibition of EGFRWT remained comparable to that by lead compound 2.
7
Based on the X-ray structure of EGFR proteins, Asp800 is adjacent to Cys797 in the binding
42
pocket. Incorporating a nitrogen containing a basic group onto the acrylamide β-position may
form a plausible ionic interaction between the protonated nitrogen of amino side chain and the
acid residue of Asp800, which improved the recognition of compound 78–81 in the active
site.42,43 It may also assist the nucleophilic attack at the Michael acceptor by Cys797.43 Among
all the new EGFR inhibitors synthesized, dimethylamino compound 78
showed the highest inhibitory activity against HCC827 (CC50 = 25 nM) and H1975 (CC50
=
6
20 nM) cell lines. In addition, (R)-enantiomers 83 and 84 led to over 5-fold lower kinase
activity and 10-fold anti-proliferative activity compared to the corresponding (S)-isomers (cf. 2
versus 83 and 78 versus 84), respectively (Table 2). These results indicated that the (S)
configuration of 2-phenylglycinol side chain is essential for the EGFR activities. Compound 78
with an acrylamide moiety possess irreversible kinase inhibitor character, which showed potent
activity against EGFRL858R/T790M similar to afatinib. However, further toxicity study and
antitumor efficacy of 78 in in vivo experiments are evaluated in following section.
1
6
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 67
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
(
(
(
A)
B)
D)
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(C)
(E)
Figure 3. Characterization of inhibitor binding modes: (A) the binding mode of compound 2
pink) overlaid with 42 (green) in EGFRL858R/T790M protein (PDB code: 5CAS); (B) EGFRWT
(
protein−inhibitor interactions for compound 74 (PDB code: 4JQ7); (C) EGFRWT
protein−inhibitor interactions for compound 77 (PDB code: 4JQ7); (D) EGFRL858R/T790M
1
7
ACS Paragon Plus Environment

Page 19 of 67
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
protein−inhibitor interactions for compound 74 (PDB code: 5CAS); (E) EGFRL858R/T790M
protein−inhibitor interactions for compound 77 (PDB code: 5CAS).
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Biological Evaluation of EGFR and HER2 Exon 20 Insertion Mutations.
Amino acids deletions of exon 19 and the exon 21 L858R point mutation are most common
44
EGFR mutations and considered as the classic activating mutations in NSCLC. Patients with
classic mutations are highly response to first and second-line standard EGFR TKIs therapy,
including erlotinib, gefitinib, afatinib and osimertinib.44 Highly heterogeneous exon 20
insertions are the third most prevalent non-classic EGFR activating mutations and accounted
45
for approximately 4–10% of all EGFR mutations. Response rates of patients with exon 20
insertions to standard EGFR targeted therapy are generally poor compared with classic
mutations.45−47 Oncogenic function of HER2 mutations and overexpression play an important
role in various cancer types. In particular, HER2 exon 20 insertions are found in 1–3% of
48
NSCLC patients. Currently no EGFR- or HER2-directed therapies are approved specifically
for the treatment of both mutations. Two investigational drugs, poziotinib and TAK-788, are
undergoing phase II (NCT03066206) and phase I/II clinical trial (NCT02716116),
respectively.49,50 Recent trial data displayed that poziotinib achieved 64% of objective response
rate (ORR) in EGFR and HER2 exon 20 mutant NSCLC (NCT03066206).51,52 Herein, we
selected four compounds (70, 78, 82, 85) to examine their inhibitory activities against three
1
8
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 67
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
types of EGFR and one type of HER2 exon 20 insertions enzymatic assays along with
pozitotinib and osimertinib as references. Those compounds were also evaluated for their
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
WT
inhibitory activity in A431 cell line (expressing EGFR ). As seen in Table 3, among of three
furanopyrimidines (70, 78, 82), 78 exhibited equal potency compared with poziotinib and 7−25
folds more potent than osimertinib in both EGFR and HER2 exon 20 insertion assays.53
Moreover, 78 showed 33-fold weaker inhibition than poziotinib against A431 cells and similar
inhibition to osimertinib that spare wild-type EGFR. The results indicated that 78 exhibited
potentially higher therapeutic window than poziotinib. Furanopyrimidines 70 and 82 possessed
moderate activities against three EGFR exon 20 insertion enzymes, but almost lost activity
against HER2 exon 20 insertion enzyme. More interestingly, thienopyrimidine 85 dramatically
decreased its inhibition in both exon 20 insertion assays, which indicated oxygen-containing
scaffold 78 is much tighter binding to exon 20 insertion enzymes than thiol-containing scaffold
8
5.
Table 3. Inhibition of EGFR and HER2 exon 20 insertions assays and A431 cellular assay.a,b
HER2 exon 20
a
EGFR exon 20 insertions IC50 (nM)
A431
a
Compound
insertion IC50 (nM)
HER2V777_G778insCG
8.13
b
CC50 (nM)
EGFRA763_Y764insFHEA
0.214
EGFRD770GY
0.641
EGFRD770_N771insNPG
7
0
1.48
0.133
0.636
449
1020
78
0.043
0.033
0.203
5.20
8
2
0.145
0.295
>8,010
1
9
ACS Paragon Plus Environment

Page 21 of 67
Journal of Medicinal Chemistry
1
2
3
4
5
6
85
1.15
0.078
0.421
4.82
0.082
0.858
17.8
0.218
0.950
369
0.206
3.31
>20,000
31
Poziotinib
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Osimertinib
967
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
aThe IC50 value was defined as the amount of compound that induced a 50% inhibition in
enzyme activity in comparison with DMSO-treated controls; Compounds were tested in 10-
dose IC50 mode with 4-fold serial dilution starting at 5 μM. And control compound,
b
staurosporine, was tested in 10-dose IC50 mode with 4-fold serial dilution starting at 20 μM. In
cell-based assay, compounds were tested in 9-dose CC50 mode with 4-fold serial dilution
starting at 20 μM.; The CC50 data are expressed as the mean of at least three independent
determinations.
X-ray Co-crystal Study of Wild-Type EGFR Kinase in Complex with 78
EGFR crystals were soaked in compound 78 and the resulting EGFR/78 complex crystal
structure was resolved to 2.86Å . The electron density map clearly showed that 78 formed a
covalent bond with Cys797 in the ATP-binding site through the Michael acceptor (enone)
(
Figure 4). The (S)-2-phenylglycinol fragment of 78 occupied the region around the gatekeeper
and formed hydrophobic interactions with Thr790, Lys745, and Met766 (Figure 4). The
hydroxyl group of 78 formed hydrogen bonds with Asp855 on the DFG motif. The pyrimidine
core structure aligned with the hinge region and formed hydrophobic interactions with Leu718
in the starting site of P-loop and also with Ala743, Thr790, and Leu844. Two separated phenyl
2
0
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 67
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
rings extended from the furanopyrimidine core perpendicular to each other. The 6-phenyl group
on the same plane as the pyrimidine core formed hydrophobic interactions with Leu844, while
the perpendicular 5-phenyl ring formed hydrophobic interactions with Val726 on the β-sheet.
The Michael acceptor linked to this perpendicular 5-phenyl ring formed the important covalent
bond with residue Cys797. This Michael addition was close to the acidic residue Asp800.
Moreover, the conserved DFG motif and His-Arg-Asp (HRD) triad motif of protein kinase play
a major role in the regulation of kinase activity.54−56 In EGFR/78 complex crystal structure, a
unique hydrogen bond network was observed between the hydroxyl group of the (S)-2-
phenylglycinol fragment of 78 and the DFG and HRD motifs in the activation loop of EGFR
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(Figure 5). The hydroxyl group formed hydrogen bonds with Asp855 of the DFG motif and to
the HRD motif via Asp855 and Asn842. While the hydrogen bond interactions are absent by
removed hydroxyl group from (S)-2-phenylglycinol moiety, the inhibitory activity decreased
significantly (cf. 70 versus 78 and 2). The hydrogen bond network interactions between 78 and
the DFG and HRD motif greatly contribute to its potency. Altogether, compound 78 occupied
the ATP-binding site and interacts with surrounding residues by covalent bonding, hydrogen
bonds, and hydrophobic interactions, which gave it a potent inhibitory activity against wild-
type EGFR.
2
1
ACS Paragon Plus Environment

Page 23 of 67
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 4. Crystal structure of EGFR in complex with 78 (yellow). For clarity, M766 (on αC
helix) and P-loop were omitted in the figure (PDB code: 6JZ0).
Figure 5. Compound 78 (yellow) forms a hydrogen bond network with the DFG (pink) and
HRD motifs (magenta) in the activation loop through the hydroxyl group of the (S)-2-
2
2
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 67
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
phenylglycinol fragment. The distances between 78 and EGFR kinase domain were labeled
(PDB code: 6JZ0).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Pharmacokinetic Study
We carried out pharmacokinetic (PK) evaluations of compounds 2 and 78 in rats. We found
that 78 showed a much better PK profile than compound 2 (Table 4). Moreover, when
administered by IV injection, 78 exhibited a lower clearance and volume distribution, as well
as 50% higher AUC than the approved irreversible-binding drug, afatinib. When administered
orally (P.O.), 78 displayed an approximately 4-fold higher AUC than afatinib and around 41.5%
oral bioavailability.
Table 4. Pharmacokinetics Profile of compound 2, 78, and afatinib in rat.
I.V. (5 mg/kg)
P.O. (20 mg/kg)
t
1/2
CL
V
ss
AUC(0-inf)
t
1/2
C
max
T
max
AUC(0-inf)
F
Compound
2
(
h)
(ml/min/kg) (l/kg) (ng/mL*h)
(h) (ng/mL)
(h)
(ng/mL*h) (%)
0.5
2.3
5.7
76.5
55.6
79.2
2.3
8.6
1096
1520
1063
0.9
3.4
6
193
508
75.2
1.2
3.3
4.2
303
2978
676
5.5
7
8a
41.5
15.9
afatinib
28.7
a5 mg/kg for I.V. injection and 23.6 mg/kg for P.O. administration
Western Blotting Analysis
In order to probe the full potential of 78, we evaluated its ability to interfere with EGFR
signaling inside the cell by western blotting. To this end, HCC827 and H1975 cells were treated
2
3
ACS Paragon Plus Environment

Page 25 of 67
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
with different doses of 78 and then were analyzed for EGFR autophosphorylation (pEGFR at
residue Tyr1068). As shown in Figure 6A, 78 showed a similar inhibitory level of EGFR
autophosphorylation compared to gefitinib in HCC827 cells. In Figure 6B, it is observed that
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
7
8 induced reduction of phosphorylated EGFR in a dose-dependent manner in H1975 cells. The
western blotting results were consistent with two cell-based growth inhibition data, which are
showed in Table 2.
(A)
(B)
Figure 6. Western blot analysis: (A) By using anti-phospho-EGFR Y1068 and anti-EGFR
2
4
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 67
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
immunoblots of HCC827 cell lysates treated with 78 and gefitinib; (B) By using anti-phospho-
EGFR Y1068 and anti-EGFR immunoblots of H1975 cell lysates treated with 78 and afatinib.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Kinase Profiling
To examine the specificity of the highly promising compound 78, we subjected it to kinase
profiling using the KINOMEscan technology against a panel of 456 kinases (containing 395
non-mutant kinases) at a concentration of 10,000 nM (Figure 7). EGFRWT and all EGFR
mutants were potently inhibited by 78 (0–7.6% of control value at 10,000 nM). Due to the
irreversible inhibition it causes, 78 also exhibited high affinity against kinases bearing a
cysteine structurally analogous to Cys797 in EGFR, such as BLK (0.25%), BMX (4.8%), BTK
(3%), ERBB2 (0%), ERBB4 (0.05%), JAK3 (0%), and TEC (5.6%). Moreover, other off-target
kinases showing apparent binding included LOK, SLK, and TXK (<10% of control value). The
selectivity score is a quantitative measure of compound selectivity and it is defined as S(10) =
(number of non-mutant kinases with <10% control value inhibition)/(number of non-mutant
kinases tested). The S(10) selectivity score, calculated using <10% of control as potency
threshold at a concentration of 10,000 nM, was determined to be 0.028 (11/395). These results
overall indicated that 78 has a high selectivity and low off-target effects, suggesting a safety
index suitable for further development.
2
5
ACS Paragon Plus Environment

Page 27 of 67
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 7. Kinase profiling of 78 using the KINOMEScan technology.
Toxicity and Efficacy In Vivo of Compound 78
First of all, we evaluated acute toxicity of 78 by orally treating 78 into ICR mice at a dose of
1
00 mg/kg once a day (QD) for two weeks. Afatinib (100 mg/kg) was also examined at the
same dosing schedule for comparison. Mice treated with afatinib were observed severe fur loss
in the nose and abdomen and skin rash, but no apparent changes were observed for the group
treated with 78. It is indicated that furanopyrimidine 78 possessed lower off-target effects and
higher dose tolerance than afatinib.
2
6
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 67
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
The compound profile of 78 encouraged us to assess its in vivo anticancer activity in EGFR
mutant NSCLC models. HCC827 and H1975 xenograft models were developed by inoculating
nude mice subcutaneously with the corresponding cancer cells. In the HCC827 tumor model,
mice were treated with oral administration of 78 for 5 days/week for 2 consecutive weeks (days
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
–5 and 8–12). At dosages of 20 and 50 mg/kg of 78, tumor growth was significantly reduced
(p<0.05), demonstrating the potent in vivo anticancer activity of 78 (Figure 8). No significant
differences in body weight change or other adverse effects were observed upon treatment with
compound 78. Administration of the reference compound gefitinib (20 mg/kg) and afatinib (20
mg/kg) by perioral route using the same schedule also shrank the tumor significantly (p<0.05).
Approximately 5% body weight loss was observed in afatinib-treated mice.
(A)
(B)
Figure 8. (A) In vivo antitumor efficacy of gefitinib, afatinib, and 78 in HCC827 xenograft
mouse model. (B) Body weight change (%) of the mice in HCC827 xenograft mouse model.
2
7
ACS Paragon Plus Environment

Page 29 of 67
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
In the H1975 tumor model, the in vivo efficacy was evaluated through oral administration of
afatinib (15 mg/kg) and 78 (50 mg/kg) QD for 15 days, respectively. The mean tumor size of
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
the vehicle control group reached 2044 mm on day 22 after tumor inoculation. Treatment with
afatinib at 15 mg/kg had a significant antitumor effect compared with the control group,
suppressing tumor growth by 65%. Treatment with 78 only at 50 mg/kg also had a significant
antitumor effect (mean tumor growth inhibition of 34%, range: 22–48%). The test compound
7
8 and afatinib were tolerated well by the tumor-bearing mice at 50 mg/kg and 15 mg/kg,
respectively, and no severe body weight loss was observed during the studies.
Table 5. Antitumor activity of 78 and afatinib as a single agent in the treatment of subcutaneous
H1975 human lung cancer xenograft model.
Tumor growth
Tumor Size (mm3)a
p-valueb
Treatment
inhibition (TGI, %)
Vehicle
2044±161
706±92
－
65
34
－
Afatinib (15 mg/kg)
<0.001
0.002
7
8 (50 mg/kg)
1354±115
a
b
Data are presented as the Mean ± SD; compared to vehicle control.
Conclusions
Over the past years, we developed a focused library of furanopyrimidines which was rapidly
synthesized using a variety of primary and secondary amines as nucleophiles in parallel reactors
and in situ screened for EGFR kinase inhibitory activity to identify an initial hit (1). Very
2
8
ACS Paragon Plus Environment