Discovery of a Noncovalent, Mutant-Selective Epidermal Growth Factor Receptor Inhibitor

Article abstract of DOI:10.1021/acs.jmedchem.6b00995

Inhibitors targeting the activating mutants of the epidermal growth factor receptor (EGFR) have found success in the treatment of EGFR mutant positive non-small-cell lung cancer. A secondary point mutation (T790M) in the inhibitor binding site has been li

Full text of DOI:10.1021/acs.jmedchem.6b00995

Subscriber access provided by Northern Illinois University
Article
Discovery of a Non-Covalent, Mutant-Selective
Epidermal Growth Factor Receptor Inhibitor
Bryan K Chan, Emily J. Hanan, Krista Bowman, Marian C. Bryan, Daniel J. Burdick, Emily W. Chan,
Yuan Chen, Saundra Clausen, Trisha Dela Vega, Jennafer Dotson, Charles Eigenbrot, Richard L
Elliott, Robert Andrew Heald, Philip S Jackson, Jamie D Knight, Hank La, Michael David Lainchbury,
Shiva Malek, Hans E. Purkey, Gabriele Schaefer, Stephen Schmidt, Eileen M Seward, Steve
Sideris, Lily Shao, Shumei Wang, Kuen S Yeap, Ivana Yen, Christine Yu, and Timothy P Heffron
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.6b00995 • Publication Date (Web): 26 Aug 2016
Downloaded from http://pubs.acs.org on August 26, 2016
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 free 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 accessible to all
readers and 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.
Journal of Medicinal Chemistry 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 46
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Discovery of a Non-Covalent, Mutant-Selective
Epidermal Growth Factor Receptor Inhibitor
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
*
, †
†
†
†
⊥
Bryan K. Chan, Emily J. Hanan, Krista K. Bowman, Marian C. Bryan, Daniel Burdick,

§
◊
†
⊥
Emily Chan, Yuan Chen, Saundra Clausen, Trisha Dela Vega, Jennafer Dotson, Charles
‡
∆
∆
∆
∆
Eigenbrot, Richard L. Elliott, Robert A. Heald, Philip S. Jackson, Jamie D. Knight, Hank
§
∆
◊
†
#
La, Michael D. Lainchbury, Shiva Malek, Hans Purkey, Gabriele Schaefer, Stephen
◊
∆
◊

†
∆
Schmidt, Eileen M. Seward, Steve Sideris, Lily Shao, Shumei Wang, Kuen S. Yeap, Ivana
◊
‡
†
Yen, Christine Yu, Timothy P. Heffron
†
‡
§
Departments of Discovery Chemistry, Structural Biology, Drug Metabolism and
◊
#

Pharmacokinetics, Biochemical and Cellular Pharmacology, Molecular Oncology, Research
⊥
Oncology and Protein Expression Genentech Inc., 1 DNA Way, South San Francisco,
∆
California, 94080, United States. Discovery, Charles River, 7/9 Spire Green Centre, Flex
Meadow, Harlow, Essex CM19 5TR, United Kingdom.
KEYWORDS
Kinase, kinase inhibitor, EGFR, T790M, resistance mutant, structure-based drug design
ACS Paragon Plus Environment
1

Journal of Medicinal Chemistry
Page 2 of 46
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
Abstract
Inhibitors targeting the activating mutants of the epidermal growth factor receptor (EGFR) have
found success in the treatment of EGFR-mutant positive non-small cell lung cancer.
A
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
secondary point mutation (T790M) in the inhibitor binding site has been linked to the acquired
resistance against those first generation therapeutics. Herein, we describe the lead optimization
of a series of reversible, pan-mutant (L858R, del746-750, T790M/L858R and T790M/del746-750
)
)
EGFR inhibitors. Using a non-covalent double mutant (T790M/L858R and T790M/del746-750
selective EGFR inhibitor (2) as a starting point, activities against the single mutants (L858R and
del746-750) were introduced through a series of structure-guided modifications. The in vitro
ADME-PK properties of the lead molecules were further optimized through a number of rational
structural changes. The resulting inhibitor (21) exhibited excellent cellular activity against both
the single and double mutants of EGFR, demonstrated target engagement in vivo and ADME-PK
properties that are suitable for further evaluation. The reversible, non-covalent inhibitors
described complement the covalent pan-mutant EGFR inhibitors that have shown encouraging
results in recent clinical trials.
Introduction
The role of activating mutants of the epidermal growth factor receptor (EGFR) in non-small cell
1
-5
lung cancer (NSCLC) is well-documented. The most common activating mutants, the point
mutation L858R or deletions within exon 19 (del746-750), increase EGFR-driven cell proliferation
and survival. The first generation EGFR inhibitors (erlotinib and gefitinib) that target those
activating mutants have demonstrated a remarkable initial rate of clinical response in the
6
-10
treatment of EGFR-mutated NSCLC.
Additionally, a number of second-generation EGFR
ACS Paragon Plus Environment
2

Page 3 of 46
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
inhibitors that covalently target the Cys797 residue within the EGFR ATP binding site have been
1
1-12
reported.
While both first and second generation EGFR inhibitors demonstrated clinical
efficacy, this efficacy is accompanied by skin and gastrointestinal toxicities that are believed to
be associated with their inhibitory activity towards the wild type form of EGFR (wtEGFR).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
2-16
In addition, acquired resistance against these therapeutic agents follows.
Approximately
6
0% of the acquired resistance arises from a secondary point mutation of the gatekeeper residue
1
7-19
Thr790 to methionine (T790M) within the EGFR kinase domain.
This mutation, while
maintaining catalytic function of the enzyme, reduces the inhibitory effect of erlotinib and
gefitinib by occluding part of the binding site for these molecules while simultaneously
2
0-24
increasing the kinase’s affinity for ATP.
Recently, promising clinical results have been reported for the third generation inhibitors
osimertinib and rociletinib) that were designed to covalently inhibit both the activating
(
1
6,25-30
mutations and the T790M resistance mutation while sparing wtEGFR.
These results have
led to the FDA approval of osimertinib for the treatment of patients with metastatic EGFR
T790M-positive NSCLC who have progressed on or after other EGFR-blocking therapies.
Despite this remarkable advance, clinical cases of the Cys797 to serine point mutation (C797S)
conferring resistance to covalent inhibitors by eliminating the point of covalent modification
1
6,31-33
have recently been reported.
Anticipating such a resistance mechanism, it was our goal to
identify a non-covalent, mutant selective EGFR inhibitor.
We previously reported a series of potent and selective inhibitors of the resistance mutants of
3
4,35
EGFR (1 and 2, Table 1).
Aminopyrimidine inhibitors 1 and 2 demonstrated excellent
biochemical potency against the T790M-L858R (TMLR) and T790M-del746-750 (TMdel)
resistance double mutants of EGFR. These molecules inhibit the T790M mutants with high
ACS Paragon Plus Environment
3

Journal of Medicinal Chemistry
Page 4 of 46
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
selectivity over wtEGFR, which was appealing for safety considerations. However, 1 and 2 each
lack biochemical inhibitory effects on the initial T790-containing single activating mutants of
EGFR (L858R and del746-750). A similar T790M selectivity profile of 1 and 2 was also observed
in cellular EGFR autophosphorylation (pEGFR) assays, comparing activity in the TMLR
expressing (H1975) and TMdel expressing (PC9-ER) cell lines against activity in the del746-750
expressing (PC9) and wtEGFR (H292) cell lines.
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
As reports indicate the importance of sustained inhibition of EGFR activating mutants beyond
initial progression on first generation EGFR inhibitor therapy, we sought to improve upon 1 and
2
by incorporating inhibition of the activating single mutants of EGFR along with the T790M
36
double mutants. Given their attractive overall physicochemical profiles, high selectivity against
wtEGFR and our knowledge of their structure-activity relationships, the aminopyrimidine
inhibitors (1 and 2) were chosen as the starting points towards a non-covalent, wtEGFR-sparing,
pan-mutant EFGR inhibitor.
Table 1. Reversible T790M resistance mutant-selective inhibitors.
Cmpd TMLR TMdel L858R del746-750 wtEGFR pEGFR
pEGFR
PC9-ER
IC50d
pEGFR pEGFR
Kiappc
Kiappc
(nM)
Kiappc
(nM)
Kiappc
(nM)
Kiappc
(nM)
H1975
IC50d
PC9
IC50d
(M)
H292
IC50
(M)
(
nM)
d
(
M)
(M)
a
1
1.1
2.5
7.0
6.0
77
490
572
122
313
0.021
0.261
0.027
0.081
2.6
5.4
>10
>10
b
2
223
ACS Paragon Plus Environment
4

Page 5 of 46
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
a
Compound 1 was a 1:1 mixture of (R)-1-(6-((2-((3R,4S)-3-fluoro-4-methoxypiperidin-1-
yl)pyrimidin-4-yl)amino)-1-((S)-1,1,1-trifluoropropan-2-yl)-1H-imidazo[4,5-c]pyridin-2-
yl)ethan-1-ol and (R)-1-(6-((2-((3S,4R)-3-fluoro-4-methoxypiperidin-1-yl)pyrimidin-4-
yl)amino)-1-((S)-1,1,1-trifluoropropan-2-yl)-1H-imidazo[4,5-c]pyridin-2-yl)ethan-1-ol.
Compound 2 was a 1:1 mixture of (3R,4S)-3-fluoro-1-(4-((2-((R)-1-hydroxyethyl)-1-((S)-1,1,1-
trifluoropropan-2-yl)-1H-imidazo[4,5-c]pyridin-6-yl)amino)pyrimidin-2-yl)piperidin-4-ol and
3S,4R)-3-fluoro-1-(4-((2-((R)-1-hydroxyethyl)-1-((S)-1,1,1-trifluoropropan-2-yl)-1H-
imidazo[4,5-c]pyridin-6-yl)amino)pyrimidin-2-yl)piperidin-4-ol. Biochemical assays. All Ki
data measurements reported are the geometric means of at least two independent experiments.
Inhibition of EGFR auto-phosphorylation. All IC50 data measurements reported are the
3
4
b
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(
3
4 c
d
geometric means of at least two independent experiments. H1975: Double mutant TMLR cell
line. PC9-ER: Double mutant TMdel cell line. PC9: Activating mutant del746-750 EGFR cell line.
H292: wild type EGFR cell line.
Results and Discussions
As previously described, the selectivity of 1 and 2 for the T790M mutants over wtEGFR and
the activating mutants was attributed to the favorable hydrophobic interactions between the 4-
3
4
alkoxypiperidine group and the side chain of the methionine gatekeeper (Figure 1B). Fluorine
substitution at the 3-position of the piperidine further enhanced those favorable interactions and
resulted in enhanced inhibition of the T790M mutants, while sparing the activating single
mutants. The 4-hydroxyl or 4-alkyoxy substitutions of the piperidine group were critical to
potency as they form a hydrogen bonding interaction with the conserved, catalytic Lys745. In
order to generate an inhibitor that exhibits activities against the T790M resistance double
mutants as well as the activating single mutants of EGFR, we hypothesized that it would be
necessary to identify a replacement for the T790M-selective piperidine motif.
Through a library approach, a number of aliphatic amines were examined and found to be
3
4
inadequate replacements for the piperidine group. Significant loss of potency was typically
observed. In parallel, we embarked on a structure-based design approach with the aim of
identifying a replacement for the 4-alkoxypiperidine group that could engage Lys745 and the P-
loop residue Phe723. Since those two residues are conserved among all EGFR forms, we
ACS Paragon Plus Environment
5

Journal of Medicinal Chemistry
Page 6 of 46
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
hypothesized that engaging those residues could result in favorable binding interactions toward
both T790- and T790M-carrying mutants.
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 1. (A) T790M selective inhibitor 3. Absolute stereochemistry of all asymmetric centers is
as shown. (B) Overlay of a model of N-alkylsulfonylpyrazole design (green) with x-ray structure
of resistance mutant-selective inhibitor 3 complexed with EGFR TMLR double mutant (ligand:
orange; protein: cyan; PDB code 5CAS). Predicted hydrogen bond interactions are shown as
yellow dashed lines.
A
B
ACS Paragon Plus Environment
6

Page 7 of 46
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
An examination of the x-ray structure of a representative T790M selective inhibitor (3, Figure
1
A) in the TMLR double mutant form of EGFR suggested that it would be desirable to position a
hydrogen bond acceptor approximately five bond lengths distal from the pyrimidine ring in order
to engage Lys745 (Figure 1B). A number of aliphatic systems that satisfied this requirement
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3
4
were tested and yielded disappointing results. Further modeling results suggested that an N-
alkylsulfonylpyrazole would position a hydrogen bond acceptor, namely the sulfonyl oxygen,
within hydrogen bonding distance from the ammonium group of Lys745 (Figure 1B).
Furthermore, the terminal alkyl group would also be within the appropriate distance to engage
the side chain of Phe723 via hydrophobic interactions.
To test the hypothesis, compound 6 was synthesized using the simplified achiral
azabenzimidazole moiety as a model system for synthetic ease (Table 2). Compound 6
demonstrated either comparable or improved activity against the TMLR double mutant and
TMdel double mutants of EFGR when compared to the previously optimized piperidine-based
inhibitors 4 and 5 (Table 2). Gratifyingly, compound 6 also showed improved potency against
the L858R and del746-750 single activating mutants. Expansion of the methylsulfonyl group to a
cyclopropylsulfonyl group (7) further improved potency for all forms of EGFR. The increase in
activity for the bulkier cyclopropyl analog could possibly be due to the increase in hydrophobic
2
interactions or, more specifically, -interactions between the partially sp -hybridized cyclopropyl
group and the -face of the aromatic side chain of Phe723. The cyclopropylmethyl analog 8
showed reduced inhibitory activity against the TMLR double mutant, which we attribute to its
inability to engage Lys745 in a hydrogen bonding interaction (the other mutant forms were not
tested).
Table 2. Piperidine Replacements.
ACS Paragon Plus Environment
7

Journal of Medicinal Chemistry
Page 8 of 46
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
Cmpd
R
TMLR
Kiapp (nM)
TMdel
L858R
del746-750
Kiapp (nM)
wtEGFR
Kiapp (nM)
a
a
a
Kiapp
Kiapp
a
a
(nM)
(nM)
b
4
13
43
41
1100
608
1300
1800
905
896
5
3.6
6
7
18
3.5
1.8
126
43
38
12
324
55
4.1
8
34
-
-
-
1200
a
Biochemical assays. All Ki data measurements reported are the geometric means of at least
two independent experiments. Racemic; Relative stereochemistry as shown.
b
Figure 2. X-ray structure of 7 complexed with EGFR TMLR double mutant (PDB code
5
HCX). Predicted hydrogen bond interactions are shown as yellow dashed lines. Interatom
distance is indicated by dashed purple lines with values in Å.
ACS Paragon Plus Environment
8

Page 9 of 46
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
An x-ray structure of 7 in the TMLR double mutant was obtained to confirm the binding mode
of the inhibitor (Figure 2). In line with the previously disclosed structures of the inhibitors
within this series, the amino-azabenzimidazole forms two hydrogen bonds with the backbone of
34,35
Gln791 and Met793 in the hinge region of the binding site.
The pyrimidyl group of the
inhibitor is positioned under the Met790 gatekeeper side chain and engages Thr854 via hydrogen
bonding. The N-cyclopropylsulfonyl-pyrazole appears to engage the ammonium group of
Lys745 via both the sulfonyl oxygen (3.0 Å) and the pyrazole nitrogen (3.0 Å). The
cyclopropylsulfonyl motif is also situated to engage the phenyl side chain of Phe723 via
hydrophobic interactions.
Table 3. In vitro ADME data for 4 and 7.
Cmpd
Liver microsome Clhep
Hepatocyte Clhep
MDCK Papp A-B
-
1
-1 a
-1 -1 b
-6
-1 c
H/R/M (ml min kg )
H/R/M (ml min kg )
9.0/11/69
(x 10 cm s )
4
7
5.2/12/76
19
21
7.7/20/70
4.0/17/49
ACS Paragon Plus Environment
9

Journal of Medicinal Chemistry
Page 10 of 46
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
c
H/R/M = human/rat/mouse. in vitro stability in cryopreserved hepatocytes. A-B, apical-to-
basolateral. See experimental section for details of assay errors and controls.
Compound 7 demonstrated comparable in vitro metabolic stability and permeability properties
to the corresponding hydroxypiperidine analog 4 (Table 3). Most importantly, both 4 and 7 had
acceptable predicted human clearance as extrapolated from in vitro liver microsome and
hepatocyte data. Because of its promising overall profile, the N-cyclopropylsulfonyl-pyrazole
group was used in subsequent optimization of the other parts of the molecule with the goal of
further improving potency while maintaining acceptable ADME attributes.
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
The vector extending from the 3-position of the azabenzimidazole system had been
underexplored to this point (Table 4). To access this vector, replacement of the azabenzimidazole
group with bioisosteric systems such as an azaindazole or azaindole would be required.
Optimization on an azaindole system was first explored as it was previously found to have
3
5
potency comparable to that of the azabenzimidazole. Due to the high lipophilicity of the
scaffold in general (cLogP of 7 = 3.6) and a desire to reduce the metabolic liability of an
electron-rich azaindole heterocycle, we first focused on polar electron-withdrawing substituents
such as carboxamides.
Table 4. Azaindole-3-carboxamide and azaindazole-3-carboxamide SAR.
TMLR
TMdel
Kiappa
(nM)
L858R
Kiappa
(nM)
del746-750 wtEGFR MDCK A-B
-6
K. Sol
(M)
1
2
Cmpd
X
R
R
Kiappa
Kiappa
Kiappa
Papp (10
cms )
c
-1 b
(
nM)
(nM)
(nM)
ACS Paragon Plus Environment
1
0

Page 11 of 46
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
9
CH
CH
A
B
1.1
5.3
0.35
27
4.1
1.2
7.8
4.5
26
<1
3
10
363
323
167
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
1
1
1
2
3
N
B
A
A
29
1.1
1.2
-
-
-
>1000
4.6
-
-
7
CH
CH
2.0
1.6
6.4
3.8
2.0
1.4
2.3
4.6
3.4
11
a
Biochemical assays. All Ki data measurements reported are the geometric means of at least
two independent experiments. A-B, apical-to-basolateral. Aqueous kinetic solubility at pH 7.4.
b
c
Gratifyingly, the 3-carboxamide azaindole 9 resulted in improved potency compared to
azabenzimidazole 7 for both the T790- and T790M-carrying mutants of EGFR (Table 4). The
gain in potency could possibly be attributed to water-mediated hydrogen bonding interactions
between the carboxamide and the backbone carbonyl of Pro794 (Figure 3A). In the context of the
4
-methoxypiperidine group (10), we again observed reduced potency for the L858R and del746-750
single mutants relative to the T790M double mutants, suggesting that the 3-carboxamide group
did not alter the mutant-selectivity of the inhibitor.
2
Modest permeability was observed for 9, possibly due to its high TPSA (150 Å ). Reducing the
number of hydrogen bond donors in a molecule is a well-established strategy to improve
3
7
permeability. We envisioned that conversion of the azaindole to an azaindazole could mask the
NH functionality of the amide via an intramolecular hydrogen bond, and therefore improve
permeability. The 3-amido-azaindazole 11, however, showed a significant loss of potency. With
the increase in polarity, the azaindazole may have reduced favorable, hydrophobic contacts with
the surrounding lipophilic residues (vide supra).
ACS Paragon Plus Environment
1
1

Journal of Medicinal Chemistry
Page 12 of 46
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
Alternatively, the primary amide was replaced with a variety of the secondary amides to reduce
the hydrogen bond donor count and improve aqueous solubility by deterring intermolecular
hydrogen bonding (e.g. dimerization). Indeed, the oxetanyl (12) and tetrahydropyranyl (13)
carboxamides demonstrated improved aqueous solubility. However, permeability remained
modest. Possibly due to the combination of modest solubility and permeability, after a single oral
dose of 100 mg/kg (MCT suspension) in mouse, plasma concentration of 13 was below the
lowest level of quantification. The 3-carboxamide series was thus deprioritized. Nonetheless, the
3-amido-azaindole analogs demonstrated that significant potency gain can be achieved by
substituting at the 3-position of the [6,5]-bicyclic scaffold.
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 3. (A) X-ray structure of 13 in EGFR TMLR double mutant (PDB code 5HCY).
Predicted hydrogen bond interactions are shown as yellow dashed lines (B) X-ray structure of 13
in EGFR TMLR double mutant with interaction surface shown. Color of surface indicates
element (pink = oxygen; cyan = nitrogen; gray = carbon).
A
ACS Paragon Plus Environment
1
2

Page 13 of 46
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
B
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Close examination of the x-ray structure of 13 suggested that the C3-vector of the azaindole is
situated under the lipophilic chain side of Leu718 in a narrow channel and 3-dimensional steric
bulk should only be tolerated beyond approximately two bond lengths from the azaindole core
structure (Figure 3B). To optimally occupy this space, a substitution would need to be lipophilic
and structurally planar proximal to the [6,5]-bicyclic scaffold. Despite their apparent fit to this
pocket, aryl motifs were considered to be undesirable due to their propensity to negatively
3
8
impact aqueous solubility. We hypothesized that a dialkylamino motif should satisfy the
requirements. To prevent further elevation of lipophilicity, heteroatoms were incorporated into
the C3-substituents.
Table 5. 3-Amino-azaindazole SAR.
ACS Paragon Plus Environment
1
3

Journal of Medicinal Chemistry
Page 14 of 46
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
c
Cmpd
R
TMLR
wtEGFR
p-EGFR
p-EGFR
p-EGFR
PC9 IC50
(M)
p-EGFR
CL HLM /
MDCK
Kin
Sol.
(M)
Kiapp
Kiapp
H1975 IC50
PC9-ER IC50
H292 IC50
A-B Papp
a
a
b
b
b
b
d
-6
f
(nM)
(nM)
(M)
(M)
(M)
CL HHep
(x 10 cm
-
1 e
s )
-
1
(
ml min
kg )
-1
1
1
4
5
0.79
2.1
8.2
10
0.064
0.074
0.014
0.019
0.080
0.076
3.9
13/8.4
16
12
2
1
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
0.85
9.3/6.7
7.0/7.7
1
6
1.2
5.5
2.0
0.092
0.019
0.015
0.046
0.028
1.5
12
1
3
17
<0.6
0.0081
0.096
10/13
8.1
1
8
9
<0.6
24
7.0
0.032
2.4
0.010
0.91
0.13
3.0
0.21
>10
8.9/7.1
8.5/15
20
<1
<1
1
>531
3.9
a
Biochemical assays. All Ki data measurements reported are the geometric means of at least
b
two independent experiments. Inhibition of EGFR auto-phosphorylation. All IC50 data
measurements reported are the geometric means of at least two independent experiments. H1975:
Double mutant TMLR EGFR cell line. PC9-ER: Double mutant TMdel EGFR cell line. PC9:
Activating mutant del746-750 EGFR cell line. H292: Wild type EGFR cell line. Metabolic
clearance from human liver microsomes. Metabolic clearance from cryopreserved human
c
d
e
f
hepatocytes. A-B, apical-to-basolateral. Kinetic solubility at pH 7.4.
Although the azaindole system had demonstrated excellent potency, we opted to focus on
azaindazoles to study the C3-amine substitution to avoid the presence of a highly electron-rich 3-
amino-azaindole system that could be a metabolic liability. In addition, given the excellent
biochemical potency of the C3-substituted azaindole or azaindazole inhibitors, we opted to
evaluate the potency of the compounds through the use of the cellular autophosphorylation
assays (Table 5).
We previously found that 3-amino-azaindazoles had good activity against the TMLR double
mutant, and that finding was recapitulated in the context of the N-cyclopropylsulfonyl-pyrazole
35
(
14). Expansion of the amino group to the morpholine (15) maintained cellular potency for
ACS Paragon Plus Environment
1
4

Page 15 of 46
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
both the activating del746-750 mutant (PC9) and T790M double mutant (H1975 and PC9-ER) cell
lines. The 4-hydroxypiperidine analog 16 was also tolerated by all mutant forms, while the
hydroxymethylazetidine analog 17 provided improved potency. Introduction of the gem-dimethyl
substitution (18) to sterically mask the hydrogen bond donor improved permeability. The
tertiary alcohol 18 also showed improved hepatocyte stability relative to primary alcohol 17. As
predicted, replacement of the dialkylamino motif with a 3-dimensional alkyl group (19) resulted
in considerable loss in activity. At this point, compound 18 was chosen for further optimization
due to its balanced pharmacological and ADME profiles.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
From our previous work on the azabenzimidazole series, we learned that further potency gains
could be realized by expanding the N1-alkyl substitution of the azabenzimidazole system, given
that it is a vector that extends towards the lipophilic ribose pocket in the EGFR ATP binding site
34
that is lined with multiple hydrophobic residues (Leu844, Val726, Cys797). However, as
shown in Table 6, expansion of the isopropyl group to a sec-butyl group within the current series
did not improve potency (20 and 21 vs 18). Gratifyingly, superior metabolic stability for the
more lipophilic 21 was observed in both in vitro human liver microsome and hepatocyte assays.
Improved metabolic stability was only observed for the S-enantiomer when compared to the
isopropyl analog, which suggests that the N-alkyl substitution may have affected the binding
and/or activity of cytochrome P450 in a specific manner. The fact that 21 demonstrated
appealing predicted human metabolic stability and excellent potency led us to compare the
potential of this molecule to the reported clinical inhibitors.
Table 6. Optimization of N-alkyl group.
ACS Paragon Plus Environment
1
5

Journal of Medicinal Chemistry
Page 16 of 46
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
b
MDCK
A-B P
p-EGFR
PC9-ER
p-EGFR
H292
HLM /
Kin
Sol.
p-EGFR
p-EGFR
PC9 ICa50
(M)
c
app
HHep
(ml min
kg )
-
6
Cmpd
R
H1975 IC50
-1
(
x 10
IC50
IC50
a
-1 d
(M)
a
a
-1
cm s )
e
(
M)
(M)
(M)
1
2
2
8
0
CH
3
0.032
0.040
0.027
0.010
0.015
0.13
0.057
0.033
0.21
0.58
0.22
8.9/7.1
9.9/7.8
3.8/2.7
20
19
<1
2
CH
CH
CH
(R)
(S)
2
3
1
2
CH
3
0.0080
8.9
2
a
Inhibition of EGFR phosphorylation. All IC data measurements reported are the geometric
means of at least two independent experiments. H1975: Double mutant TMLR EGFR cell line.
PC9-ER: Double mutant TMdel EGFR cell line. PC9: Activating mutant del EGFR cell
line. H292: Wild type EGFR cell line. In vitro stability in human liver microsomes. In vitro
stability in cryopreserved human hepatocytes. A-B, apical-to-basolateral. Kinetic solubility at
pH 7.4.
5
0
746-750
b
c
d
e
Compared to afatinib, an irreversible second generation EGFR inhibitor, 21 showed
significantly reduced activity against wtEGFR (H292), which we believe to be important for
safety considerations (Table 7). We were very pleased to see that, like osimertinib and in
contrast to the first generation EGFR inhibitors (erlotinib and getifinib), 21 demonstrated
excellent potency against both the activating and resistant EGFR mutants in the cellular
mechanistic assays, yet without requiring a covalent interaction. In addition, 21 demonstrated
strong anti-proliferative effect on the T790M mutant carrying H1975 cell line (IC = 0.361 M)
5
0
and the single activating mutant PC9 cell line (IC50 = 0.151 M). Furthermore, compound 21
also showed good selectivity against other kinases when evaluated in a 225-kinase panel (12/225
ACS Paragon Plus Environment
1
6

Page 17 of 46
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
kinases inhibited at >70% when tested at 0.1 M, 61-fold over the TMLR Ki and 63-fold over
the TMdel Ki; see supporting information).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table 7. Comparison of cellular activity profiles.
a
pEGFR IC (M)
5
0
Cmpd
H1975
0.027
2.6
PC9-ER
0.009
1.8
PC9
H292
0.218
0.006
0.008
0.004
0.483
2
1
0.033
0.003
0.005
0.001
0.031
erlotinib
gefitinib
1.1
1.9
afatinib
osimertinib
0.010
0.043
0.035
0.017
a
Inhibition of EGFR phosphorylation. All IC data measurements reported are the geometric
5
0
means of at least two independent experiments. H1975: Double mutant TMLR cell line. PC9-
ER: Double mutant TMdel cell line. PC9: Activating mutant del746-750 cell line. H292: Wild type
cell line.
With the encouraging in vitro pharmacological profile, we elected to further evaluate 21 in
vivo. The pharmacokinetic profile of 21 is outlined in Table 8. In mouse, post intravenous and
oral administration, the plasma clearance of 21 was determined to be 104 mL/min/kg with a
bioavailability of 19%. In dogs, the plasma clearance was 13 mL/min/kg with an oral
bioavailability of 30%. While high clearance at approximately liver blood flow was observed in
mouse, the good correlation between in vivo clearance and in vitro clearance from liver
microsomes in dog increased our confidence that 21 would have low clearance in human as the
in vitro clearance measured from human liver microsomes and hepatocytes were both low.
Table 8. Pharmacokinetic parameters of 21.
-
1
-1
Species
CL Liver
CL Hepatocytes iv Clp (ml min kg ) /
F%
-1 c
microsomes
Vss (L kg )
-1
-1 b
(
ml min kg )
ACS Paragon Plus Environment
1
7

Journal of Medicinal Chemistry
Page 18 of 46
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
-1
-1 a
(
ml min kg )
d
e
Mouse
66.5
12.6
52.6
6.5
104 / 3.4
13 / 6.9
19%
Dog
30%
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
a
b
Clearance determined from liver microsomes. Clearance determined from cryopreserved
hepatocytes. Clp = plasma clearance; Compound dosed iv (mouse = 5 mg kg ; dog = 1 mg kg )
as a solution in 10/35/55 ethanol/PEG/water solution. Compound dosed po (25 mg kg ) as a
solution in corn oil. Compound dosed po (2 mg kg ) as a suspension in MCT.
c
-1
-1
d
-1
e
-1
Figure 5. Compound 21 was administered orally to H1975 tumor-bearing mice with either a 50
or 100 mg/kg single dose. Tumors were collected at indicated times and phosphorylation of
EGFR, ERK, AKT was determined using commercially available MSD assays. Bars represent
individual animals (four animals per group). Dotted line on the free plasma concentration plot
shows the in vitro H1975 p-EGFR IC . Error bars represent the mean ± SD. Plasma
5
0
concentrations of 21 were measured using liquid chromatography-tandem mass spectrometry
assay, and free concentration was calculated using F of 0.009.
u
ACS Paragon Plus Environment
1
8

Page 19 of 46
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
To examine its inhibitory effect on pEGFR levels in vivo, inhibitor 21 was studied in a mouse
H1975 (TMLR) xenograft model (Figure 5). After single oral dose of 21 at 50 mg/kg, free
plasma concentrations of 21 at or exceeding the in vitro p-EGFR IC50 of 0.027 M were
sustained over 8 h. When administered at 100 mg/kg, the coverage of p-EGFR IC50 was
extended to the last measured time point of 16 h post dose. Corresponding knockdown of p-
EGFR and the downstream effectors pERK1/2 and AKT levels were observed at those time
points, suggesting target engagement in vivo.
Chemistry
Scheme 1.
ACS Paragon Plus Environment
1
9

Journal of Medicinal Chemistry
Page 20 of 46
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
(a) 4-Amino-2-chloropyrimidine, Pd (dba) , XantPhos, Cs CO , 1,4-dioxane, 100 C, 4 h,
2
3
2
3
(
26%); (b) 25, 26 or 27, Na CO , Pd(Amphos) Cl , CH CN, water, 100 C, 3 h.
2 3 2 2 3
Scheme 2.
(
a) For 25 and 26, R-SO Cl, NaH, DMF, rt, 5 h; (b) For 27, bromomethylcyclopropane,
2
Cs CO , DMF, 60 C, 12 h (45%); (c) For 26, 2-chloropyrimidin-4-amine, Na CO ,
2
3
2
3
Pd(Amphos) Cl , CH CN, water, 100 C, 2 h (85%).
2
2
3
3
4
Compounds 1-5 were prepared using procedures previously described. Compounds 6-8 were
synthesized as outlined in Scheme 1 and 2. Palladium-catalyzed coupling between chloro-
3
4
azabenzimidazole 22 and 2-chloropyrimidin-4-amine afforded intermediate 23. Suzuki
coupling between intermediate 23 and the corresponding 4-pyrazole pinacol boronic esters 25-27
yielded 6-8. The 4-pyrazole pinacol boronic esters 25-27 were prepared via the alkylation or
sulfonylation of the commercially available 4-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)-1H-
pyrazole (24).
ACS Paragon Plus Environment
2
0

Page 21 of 46
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Scheme 3.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(
a) 2-Bromopropane, NaH, DMF, 0 C to rt, 2 h (quant.); (b) POCl , DMF, rt, 2 h (80%); (c)
3
NaOCl , NaH PO , 2-methyl-2-butene, water, t-BuOH, rt, 18 h (94%); (d) NH R, HBTU, DIEA,
DMF, rt, 40 min; (e) 4-amino-2-chloropyrimidine, Xantphos, Pd (dba) , Cs CO , dioxane, 120
C, 6 h; (f) 26, Na CO , Pd(Amphos) Cl , CH CN, water, 100 C, 1 h.
2
2
4
2
2
3
2
3

2
3
2
2
3
Compounds 9, 12 and 13 were prepared according to Scheme 3. A four-step sequence
involving N-alkylation, Vilsmeier-Haack formylation, Pinnick oxidation, and amide bond
formation efficiently advanced bromoazaindole 29 to intermediates 30-32. Buchwald-Hartwig
reactions between 30-32 and 2-chloropyrimidin-4-amine, followed by a Suzuki coupling with the
pyrazole pinacol boronic ester 26, afforded 9, 12 and 13.
Scheme 4.
ACS Paragon Plus Environment
2
1

Journal of Medicinal Chemistry
Page 22 of 46
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) 2-bromopropane, NaH, DMF, 0 C to rt, 2 h (quant.); (b) POCl , DMF, rt, 2 h (80%); (c)
3
NaOCl , NaH PO , 2-methyl-2-butene, water, t-BuOH, rt, 18 h (94%); (d) NH R, HBTU, DIEA,
DMF, rt, 40 min; (e) I , K CO , DMF, rt, 1 h (quant.); (f) 2-bromopropane, NaH, DMF, 0 C to
rt, 2 h (81%); (g) Pd(OAc) , Et N, CH OH, CO, 65 C, 3 h (40%); (h) i. (CH ) SnOH, DCM, 60
C, 4 h; ii. NH OH, HBTU, DIEA, DMF, rt, 20 min (25% over 2 steps); (i) 2-(4-
methoxypiperidin-1-yl)pyrimidin-4-amine, chloro{[BrettPhos][2-(2-
aminoethylphenyl]palladium(II)]}/[Brettphos] admixture, t-BuONa, t-BuOH, 120 C, 3 h.
2
2
4
2
2
2
3
2
3
3
3 3
º
4
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
Compounds 10 and 11 were synthesized according to Scheme 4. Azaindole 38 was prepared
via the four step sequence described above starting from 6-chloro-5-azaindole (36). Iodination
3
4
and alkylation of 6-chloro-5-azaindazole (37), followed by methoxycarbonylation and amide
bond synthesis, afforded azaindazole 39. Palladium-catalyzed coupling of 38 or 39 with 2-(4-
methoxypiperidin-1-yl)pyrimidin-4-amine yielded 10 and 11.
Scheme 5.
(
a) I , K CO , DMF, rt (quant.); (b) 2-bromopropane, NaH, DMF, 0 C to RT (44%); (c)
2 2 3
Benzophenone imine, Cs CO , Pd (dba) , Xantphos, 1,4-dioxane, 90 C (58%); (d) 28,
2
3
2
3
.
Pd (dba) , Xantphos, Cs CO , 1,4-dioxane, 100 C (50%); (e) NH OH HCl, NaOAc, rt (14%).
2
3
2
3
2
Scheme 6.
ACS Paragon Plus Environment
2
2

Page 23 of 46
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(
a) I , K CO , DMF, rt; (b) 2-bromopropane or 2-bromobutane, NaH, DMF, 0 C to rt (quant.);
2
2
3
(c) HNR , CuI, L-proline, K CO , 90 C, 3.5 h; (d) 28, Pd (dba) , XPhos, Cs CO , 1,4-
2
2
3
2
3
2
3
º
dioxane, 100 C.
Compound 14 was prepared according to Scheme 5. Iodination and N-alkylation of 6-bromo-5-
azaindazole (40) afforded 41. Installation of the protected 3-amino group was accomplished by a
palladium-catalyzed coupling using benzophenone imine. Subsequent coupling with 28,
followed by imine cleavage gave inhibitor 14. Similarly, inhibitors 15-18 and 20-21 were
synthesized according to Scheme 6. Iodination and N-alkylation of the commercially available
3
7 or 40 followed by copper-mediated amination yielded intermediates 43-44. Subsequent
palladium-catalyzed amination using the previously described aminopyrimidines 28 afforded the
inhibitors 15-18 and 20-21. Chiral separation of 20 and 21 was achieved by using chiral
supercritical fluid chromatography.
Absolute stereochemistry of 21 was assigned by
crystallography (see supporting information).
Scheme 7.
ACS Paragon Plus Environment
2
3

Journal of Medicinal Chemistry
Page 24 of 46
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) 4,4,5,5-tetramethyl-2-(prop-1-en-2-yl)-1,3,2-dioxaborolane, Pd(dppf)Cl , Na CO , KOAc,
2
2
3
water, CH CN, 100 C, W (45%); (b) 28, Pd (dba) , XPhos, Cs CO , 1,4-dioxane, 100 C
(65%); (c) H , Pd(OH) /C, rt (56%).
3
2
3
2
3
2
2
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
The synthesis of 19 began with a Suzuki coupling to yield isopropene 46 (Scheme 7).
Buchwald-Hartwig coupling between 46 and 28, followed by hydrogenation afforded 19.
Conclusions
By using a structure-based design approach, we identified a novel, non-covalent EGFR
inhibitor that exhibits excellent activity against both the activating and resistance mutants of the
receptor. In sparing the wild type form of EGFR, a wider therapeutic window can be anticipated
relative to first and second generation inhibitors. Without the reliance on a covalent interaction
with Cys797, inhibitors of this type should be able to achieve the efficacy of the covalent
inhibitors without being subject to the emerging C797S resistance mutation. With strong
pathway knock-down in vivo and low predicted human clearance, compound 21 shows excellent
potential for the treatment of EGFR-mutant NSCLC. Further in vivo evaluation of compound 21
will be disclosed in due course.
EXPERIMENTAL SECTION
1
Synthesis. General methods: H NMR spectra were recorded at ambient temperature using either
a Varian Unity Inova (400MHz) spectrometer with a triple resonance 5mm probe, an Avance III
(
300 MHz) spectrometer or a Bruker Ultrashield (400 MHz or 500 MHz) spectrometer. Chemical
shifts are expressed in ppm relative to tetramethylsilane. The following abbreviations have been
used: br = broad signal, s = singlet, d = doublet, dd = double doublet, t = triplet, q = quartet, m =
ACS Paragon Plus Environment
2
4

Page 25 of 46
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
multiplet. Microwave experiments were carried out using a CEM Discover, Smith Synthesiser
or a Biotage Initiator 60™, which uses a single-mode resonator and dynamic field tuning, both of
which give reproducibility and control. Temperatures from 40-250 °C can be achieved and
pressures of up to 30 bar can be reached. HPLC/UPLC - Mass Spectrometry experiments to
determine retention times and associated mass ions were performed using various methods which
are fully described in the Supporting Information. All final compounds were assessed for purity
by UPLC and found to be ≥95% purity. Diastereomeric and enantiomeric purity of final
compounds was assessed by UPLC and SFC respectively and found to be >95% de and >95% ee.
Experimental procedures for selected inhibitors are described below. Experimental procedures
for all inhibitors are provided in the supporting information. Compounds 1-5 were prepared as
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3
4, 35
previously described.
N-(2-(1-(cyclopropylsulfonyl)-1H-pyrazol-4-yl)pyrimidin-4-yl)-1-isopropyl-2-methyl-1H-
imidazo[4,5-c]pyridin-6-amine(7)
1
-(Cyclopropylsulfonyl)-4-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)-1H-pyrazole (26). To
a solution of 4-pyrazoleboronic acid pinacol ester (1.02 g, 5.26 mmol) in DMF (15 mL) was
added sodium hydride (60%, 0.32 g, 8 mmol) and cyclopropylsulfonyl chloride (0.8 g, 5.78
mmol). The reaction was stirred at room temperature for 5 h. The mixture was then diluted with
brine and extracted with EtOAc (3x). The combined extracts were washed with brine, dried over
sodium sulfate, filtered and concentrated. The crude material was purified by flash
1
chromatography (0-100% EtOAc in heptanes) to give the title compound (0.8 g, 50%). H NMR
(
400 MHz, CDCl ) δ 8.29 (s, 1H), 8.01 (s, 1H), 2.77 (m, 1H), 1.48 (m, 2H), 1.33 (s, 12H), 1.16
3
(m, 2H).
ACS Paragon Plus Environment
2
5

Journal of Medicinal Chemistry
Page 26 of 46
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
N-(2-chloropyrimidin-4-yl)-1-isopropyl-2-methyl-1H-imidazo[4,5-c]pyridin-6-amine (23). Into
a 250 mL 3-necked round-bottom flask purged and maintained with nitrogen, was added a
3
4
solution of 6-bromo-1-isopropyl-2-methyl-1H-imidazo[4,5-c]pyridine (3.50 g, 13.0 mmol) 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
1
,4-dioxane (70.0 mL), 2-chloropyrimidin-4-amine (1.70 g, 13.1 mmol), XantPhos (1.70 g, 2.94
.
mmol), Pd (dba) CHCl (1.70 g, 1.64 mmol) and Cs CO (11.0 g, 33.7 mmol). The resulting
2
3
3
2
3
mixture was stirred at 100 C for 4 h. The mixture was cooled to room temperature, the solids
were removed by filtration and the filtrate was concentrated under vacuum. The residue was
purified by silica gel flash chromatography (DCM/MeOH, 50:1) to afford the title compound
+
1
(
1.10 g, 26.1%) as a yellow solid. LCMS (ESI): [M+H] = 303. HNMR (300Hz, DMSO-d ): δ
6
1
0.55 (s, 1H), 8.57 (s, 1H), 8.26 (d, J = 5.7 Hz, 1H), 8.13 (s, 1H), 7.48 (s, 1H), 4.72-4.68 (m,
1H), 2.58 (s, 3H), 1.57 (d, J = 6.9, 6H).
N-(2-(1-(cyclopropylsulfonyl)-1H-pyrazol-4-yl)pyrimidin-4-yl)-1-isopropyl-2-methyl-1H-
imidazo[4,5-c]pyridin-6-amine (7). To a reaction vial was added N-cyclopropylsulfonyl-4-
pyrazoleboronic acid pinacol ester (0.56 g, 1.8 mmol), N-(2-chloropyrimidin-4-yl)-1-isopropyl-
2
-methyl-1H-imidazo[4,5-c]pyridin-6-amine (0.56 mg, 1.8 mmol), Pd(Amphos) Cl (65 mg,
2 2
0.092 mmol), a 2 M aqueous solution of Na CO (1.4 mL, 2.8 mmol) and acetonitrile (2.5 mL).
2
3
The mixture was degassed by nitrogen bubbling for 20 min. The reaction vial was sealed and
stirred at 100 C in an oil bath for 3 h. The reaction mixture was cooled to room temperature,
filtered and concentrated. The crude product was purified by reverse phase HPLC to give the
+
1
title compound (41 mg, 5 %). LCMS (ESI): [M+H] 439.2. H NMR (400 MHz, DMSO-d ) δ
6
1
0.16 (s, 1H), 8.66 (s, 1H), 8.56 (d, J = 0.8 Hz, 1H), 8.47 (d, J = 0.5 Hz, 1H), 8.39 (m, 2H), 7.24
(
s, 1H), 4.78 (m, 1H), 2.58 (s, 3H), 1.63 (d, J = 6.9 Hz, 6H), 1.38 – 1.21 (m, 4H).
ACS Paragon Plus Environment
2
6

Page 27 of 46
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
6
-((2-(1-(Cyclopropylsulfonyl)-1H-pyrazol-4-yl)pyrimidin-4-yl)amino)-1-isopropyl-1H-
pyrrolo[3,2-c]pyridine-3-carboxamide (9).
-Bromo-1-isopropyl-1H-pyrrolo[3,2-c]pyridine. To a solution of 6-bromo-5-azaindole (1.0 g,
.9 mmol) in DMF (8 mL) was added 60% sodium hydride (0.23 g, 5.8 mmol) at 0 ºC. The
6
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
4
mixture was stirred 0 C for 5 min prior to the addition of 2-bromopropane (0.7 mL, 7 mmol).
The reaction was stirred at room temperature for 2 h. The reaction was then quenched by pouring
onto a saturated aqueous solution of NH Cl. The product was extracted with EtOAc (3x). The
4
combined extracts were washed with brine, dried over sodium sulfate, filtered and concentrated.
The crude product was purified by silica gel flash chromatography (0-100 % EtOAc in heptane)
+
to give the title compound (1.5 g, quant.). LCMS (ESI): [M+H] = 239.2.
6-Bromo-1-isopropyl-1H-pyrrolo[3,2-c]pyridine-3-carbaldehyde. To cooled (0 C) DMF (40
mL) was added POCl (3.0 mL, 31 mmol). The reaction was stirred at 0 ºC for 20 min. A
3
solution of 6-bromo-1-isopropyl-1H-pyrrolo[3,2-c]pyridine (4.9 mmol) in DMF (1 mL) was
added drop wise at room temperature. The reaction was stirred at 70 C for 2 h. The mixture was
then slowly neutralized by pouring onto a saturated aqueous NaHCO solution. The product was
3
extracted with EtOAc (3x). The combined organic extracts were washed with saturated aqueous
NaHCO and brine, dried over sodium sulfate, filtered and concentrated. The crude product was
3
purified by silica gel flash chromatography (0-100 % EtOAc in heptane) to give the title
+
compound (1.34 g, 80%). LCMS (ESI): [M+H] = 267.3.
6
-Bromo-1-isopropyl-1H-pyrrolo[3,2-c]pyridine-3-carboxamide (30). To a solution of 6-
bromo-1-isopropyl-1H-pyrrolo[3,2-c]pyridine-3-carbaldehyde (1.3 g, 4.9 mmol) in t-BuOH (10
mL) was added 2-methyl-2-butene (2 M in THF, 4.1 mL, 49 mmol). A solution of sodium
chlorite (1.1 g, 9.7 mmol) and monosodium phosphate (5.8 g, 49 mmol) in water (15 mL) was
ACS Paragon Plus Environment
2
7

Journal of Medicinal Chemistry
Page 28 of 46
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
then added. The reaction was stirred at room temperature for 18 h. The mixture was then diluted
with brine and extracted with EtOAc (3x). The combined extracts were washed with brine, dried
over sodium sulfate, filtered and concentrated to give the title compound (1.3 g, 94%). The crude
material was used without further purification.
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
To a solution of the crude 6-bromo-1-isopropyl-1H-pyrrolo[3,2-c]pyridine-3-carboxylic acid
(1.25 g, 4.4 mmol) in DMF (10 mL) was added DIPEA (1.2 mL, 6.6 mmol) and HBTU (2.64 g,
6
.62 mmol). The mixture was stirred at room temperature for 15 min before the addition of
ammonium hydroxide (10 mL). The resulting mixture was stirred vigorously for 1 h. The
mixture was then diluted with brine and extracted with EtOAc (3x). The combined extracts were
washed with brine, dried over sodium sulfate, filtered and concentrated. The crude material was
purified by silica gel flash chromatography (0-10 % MeOH in DCM) to give the title compound
+
(
0.72 g, 58%). LCMS (ESI): [M+H] = 282.3.
6
-((2-Chloropyrimidin-4-yl)amino)-1-isopropyl-1H-pyrrolo[3,2-c]pyridine-3-carboxamide
(33). To a pressure reaction vessel was added 6-bromo-1-isopropyl-1H-pyrrolo[3,2-c]pyridine-3-
carboxamide (0.15 g, 0.53 mmol), 4-amino-2-chloropyrimidine (70 mg, 0.53 mmol), Xantphos
(32 mg, 0.053 mmol), Pd (dba) (25 mg, 0.027 mmol), Cs CO (0.35 g, 1.1 mmol) and 1,4-
2
3
2
3
dioxane (2 mL). The mixture was degassed by nitrogen bubbling for 20 min. The reaction vessel
was then sealed and stirred at 120 C for 6 h. The reaction was then filtered and concentrated.
The crude product was purified by silica gel flash chromatography (0-10 % MeOH in DCM) to
+
give the title compound (0.70 g, 40%). LCMS (ESI): [M+H] = 331.
6
-((2-(1-(Cyclopropylsulfonyl)-1H-pyrazol-4-yl)pyrimidin-4-yl)amino)-1-isopropyl-1H-
pyrrolo[3,2-c]pyridine-3-carboxamide (9). To a glass vial was added 1-(cyclopropylsulfonyl)-4-
(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)-1H-pyrazole (95 mg, 0.30 mmol), 6-((2-
ACS Paragon Plus Environment
2
8

Page 29 of 46
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
chloropyrimidin-4-yl)amino)-1-isopropyl-1H-pyrrolo[3,2-c]pyridine-3-carboxamide (50 mg,
0
0
.15 mmol), Pd(Amphos) Cl (5 mg, 0.008 mmol), a 2 M aqueous solution of Na CO (0.11 mL,
2 2 2 3
.23 mmol) and acetonitrile (2.5 mL). The mixture was degassed by nitrogen bubbling for 20
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
min. The reaction vial was sealed and stirred at 100 C in an oil bath for 1 h. The reaction
mixture was cooled to room temperature, filtered and concentrated. The crude product was
purified by reverse phase HPLC and lyophilized to give the title compound (20 mg, 30%).
+
1
LCMS (ESI): [M+H] = 467.2. H NMR (400 MHz, DMSO-d ) δ 10.27 (s, 1H), 9.08 (t, J = 2.4
6
Hz, 1H), 8.70 (d, J = 0.5 Hz, 1H), 8.50 (s, 1H), 8.47 (d, J = 0.5 Hz, 1H), 8.37 (d, J = 5.9 Hz, 1H),
8
.23 (s, 1H), 7.49 (s, 1H), 7.11 (s, 1H), 6.97 (d, J = 8.7 Hz, 1H), 6.63 (d, J = 8.8 Hz, 1H), 4.80
(m, 1H), 1.56 (d, J = 6.7 Hz, 6H), 1.40 – 1.31 (m, 2H), 1.30 – 1.23 (m, 2H).
N-(2-(1-(cyclopropylsulfonyl)-1H-pyrazol-4-yl)pyrimidin-4-yl)-1-isopropyl-3-morpholino-
1
H-pyrazolo[4,3-c]pyridin-6-amine (15).
-(6-bromo-1-isopropyl-1H-pyrazolo[4,3-c]pyridin-3-yl)morpholine. A mixture of 6-bromo-3-
4
iodo-1-isopropyl-1H-pyrazolo[4,3-c]pyridine (1 g, 2.73 mmol), CuI (103.8 mg, 0.550 mmol), L-
proline (94.3 mg, 0.820 mmol), K CO (943 mg, 6.82 mmol) and morpholine (713 mg, 8.19
2
3
mmol) in DMF (10 mL) was irradiated with microwave radiation for 2 h at 100 C under an
atmosphere of nitrogen. The resulting solution was diluted with water and extracted with ethyl
acetate. The organic extract was washed with brine then dried over anhydrous sodium sulfate and
concentrated in vacuo. The crude product was purified by silica gel flash chromatography (10%
EtOAc in petroleum ether) to afford the title compound (140 mg, 16%) as a white solid. LCMS
(
ESI): [M+H]+ = 325/327.
-(1-(Cyclopropylsulfonyl)-1H-pyrazol-4-yl)pyrimidin-4-amine (28). To a reaction vessel was
added Pd(Amphos) Cl (370 mg, 0.50 mmol), 1-(cyclopropylsulfonyl)-4-(4,4,5,5-tetramethyl-
2
2
2
ACS Paragon Plus Environment
2
9