Discovery of N-(4-(2,4-Difluorophenoxy)-3-(6-methyl-7-oxo-6,7-dihydro-1H-pyrrolo[2,3-c]pyridin-4-yl)phenyl)ethanesulfonamide (ABBV-075/Mivebresib), a Potent and Orally Available Bromodomain and Extraterminal Domain (BET) Family Bromodomain Inhibitor

Article abstract of DOI:10.1021/acs.jmedchem.7b00746

The development of bromodomain and extraterminal domain (BET) bromodomain inhibitors and their examination in clinical studies, particularly in oncology settings, has garnered substantial recent interest. An effort to generate novel BET bromodomain inhibi

Full text of DOI:10.1021/acs.jmedchem.7b00746

Subscriber access provided by Purdue University Libraries
Article
Discovery of N-(4-(2,4-difluorophenoxy)-3-(6-methyl-7-oxo-6,7-
dihydro-1H-pyrrolo[2,3-c]pyridin-4-yl)phenyl)ethanesulfonamide
(ABBV-075/mivebresib), a Potent and Orally Available Bromodomain
and Extraterminal domain (BET) Family Bromodomain Inhibitor
Keith F. McDaniel, Le Wang, Todd Soltwedel, Steven D. Fidanze, Lisa A. Hasvold, Dachun Liu, Robert
A. Mantei, John K. Pratt, George S. Sheppard, Mai H Bui, Emily J Faivre, Xiaoli Huang, Leiming Li,
Xiaoyu Lin, Rongqi Wang, Scott E. Warder, Denise Wilcox, Daniel H Albert, Terrance J. Magoc, Ganesh
Rajaraman, Chang H. Park, Charles W. Hutchins, Jianwei J Shen, Rohinton P. Edalji, Chaohong C. Sun,
Ruth Martin, Wenqing Gao, Shekman Wong, Guowei Fang, Steven W. Elmore, Yu Shen, and Warren M Kati
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.7b00746 • Publication Date (Web): 26 Sep 2017
Downloaded from http://pubs.acs.org on September 26, 2017
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 51
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
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 51
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 N-(4-(2,4-difluorophenoxy)-3-(6-methyl-7-oxo-6,7-dihydro-1H-pyrrolo[2,3-c]pyridin-
-yl)phenyl)ethanesulfonamide (ABBV-075/mivebresib), a Potent and Orally Available
Bromodomain and Extraterminal domain (BET) Family Bromodomain Inhibitor
4
a,
a
b
a
a
a
a
a
a
a
Keith F. McDaniel, * Le Wang, Todd Soltwedel, Steven D. Fidanze, Lisa A. Hasvold, Dachun Liu,
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
a
a
a
a
Robert A. Mantei, John K. Pratt, George S. Sheppard, Mai H. Bui, Emily J. Faivre, Xiaoli Huang,
a
a
a
a
a
Leiming Li, Xiaoyu Lin, Rongqi Wang, Scott E. Warder, Denise Wilcox, Daniel H. Albert,
b
b
b
a
Terrance J. Magoc, Ganesh Rajaraman, Chang H. Park, Charles W. Hutchins, Jianwei J. Shen,
a
a
a
a
a
Rohinton P. Edalji, Chaohong C. Sun, Ruth Martin, Wenqing Gao, Shekman Wong, Guowei Fang,
a
a
a
Steven W. Elmore, Yu Shen, Warren M. Kati
a
AbbVie Inc., Oncology Discovery, 1 North Waukegan Rd., North Chicago, IL 60064, USA
Former AbbVie employee
b
ꢀ ABSTRACT
The development of bromodomain and extraterminal domain (BET) bromodomain inhibitors and
their examination in clinical studies, particularly in oncology settings, has garnered substantial recent
interest. An effort to generate novel BET bromodomain inhibitors with excellent potency and DMPK
properties was initiated based upon elaboration of a simple pyridone core. Efforts to develop a bidentate
interaction with a critical asparagine residue resulted in the incorporation of a pyrrolopyridone core,
which improved potency by 9ꢀ to 19ꢀfold. Additional structureꢀactivity relationship (SAR) efforts aimed
both at increasing potency and improving pharmacokinetic properties led to the discovery of the clinical
candidate 63 (ABBVꢀ075/mivebresib), which demonstrates excellent potency in biochemical and
cellular assays, advantageous exposures and halfꢀlife both in animal models and in humans, as well as in
vivo efficacy in mouse models of cancer progression and inflammation.
1
ACS Paragon Plus Environment

Page 3 of 51
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
ꢀ
INTRODUCTION
The transcription of different genes is turned on or off in different cell types, explaining why, for
instance, a liver cell functions differently from a nerve cell or skin cell despite all the cells within an
1
individual having a common genetic sequence. This epigenetic regulation of gene transcription is
mediated, in part, by the acetylation of specific lysine residues on histone and other proteins. Proteins
containing a conserved structural fold known as a bromodomain specifically bind to the acetylꢀlysine
2
marks, thereby facilitating gene transcription and other downstream events. However, in some cancer
and inflammatory disease states the epigenetic regulation of gene transcription is dysfunctional,
3
resulting in the aberrant expression of growth promoting genes and proꢀinflammatory cytokines.
Consequently, small molecules which block the acetylꢀlysine/bromodomain interaction could have
4
therapeutic utility by modulating disease specific dysfunctional gene transcription.
There are 46 human proteins known to contain bromodomains and these proteins can be
segregated into 8 groups based on phylogenetic/structural modeling. The Bromodomain and Extraꢀ
terminal (BET) family represents one entire group, consisting of four family members (BRD2, BRD3,
BRD4 and BRDT). Each BET family member contains two Nꢀterminal bromodomains (BDI and BDII)
5
along with a Cꢀterminal extraterminal domain. The first BET bromodomain inhibitors to be reported,
6
initially from the Mitsubishi Tanabe Pharmaceutical Corporation (1a, MS417) and subsequently by the
7
Dana Farber Cancer Institute (1b, JQ1), incorporated the thieno[3,2ꢀf][1,2,4]triazolo[4,3ꢀ
a][1,4]diazepine core (thienotriazolodiazepine) to generate potent and selective BET bromodomain
2
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 51
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
inhibitors via occupation of the acetylꢀlysine (KAc) binding site. For this core, the methyltriazole
moiety mimics the acetylꢀlysine group of the native peptide ligand. Additional methyltriazolodiazepines
8
9
have also been developed, including clinical candidates 1c (OTXꢀ015) and 2 (IꢀBET762). As an
alternative to the methyltriazole binding moiety, a wide variety of 3,5ꢀdimethylisoxazoles have also
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
10
been discovered, as exemplified by 3 (IꢀBET151). Constellation Pharmaceuticals combined the 3,5ꢀ
dimethylisoxazole moiety with the benzodiazepine core to generate their clinical candidate, 4 (CPIꢀ
1
1
0610).
Another alternative to the triazolodiazepine core utilizes the acetylated 2ꢀ
12
methyltetrahydroquinoline core as the acetylꢀlysine mimic, as exemplified by compound 5 (THQ).
Preclinical studies with these and other compounds have established that BET bromodomain inhibitors
exhibit significant efficacy in xenograft models representing a variety of hematological and solid tumors
as well as in a diverse group of inflammatory disease models. Each of these distinct BET bromodomain
inhibitor cores presents novel vectors which allow access to additional binding sites of the BET
bromodomain protein, with a particular focus on the incorporation of aryl groups reaching into the
9
,13
hydrophobic binding site known as the WPF pocket to gain potency and selectivity. Although several
1
4
compounds utilizing these cores have progressed into the clinic, there remains substantial interest in
the development of alternative BET bromodomain inhibitors, based upon different core structures, to
provide more potent molecules along with superior pharmacokinetic properties.
3
ACS Paragon Plus Environment

Page 5 of 51
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
Figure 1. Representative BET bromodomain inhibitors.
1
5
We recently reported the discovery of pyridazinone fragment 6 as a novel fragment core to
utilize for the design of BET bromodomain inhibitors. Xꢀray crystallographic studies of this core
demonstrated the valuable interaction of the Nꢀmethyl moiety of these pyridazinones in the amphipathic
water pocket of the bromodomain protein, as well as the binding of the pyridazinone carbonyl with the
15
NH of Asn433.
SAR efforts led to the replacement of the pyridazinone of 6 by pyridone and
2
incorporation of a crucial biphenyl ether moiety to generate Nꢀmethylpyridone BET bromodomain
inhibitors based upon compound 7. Utilization of this core provided inhibitors demonstrating subꢀµM
activity in TRꢀFRET binding assays against BRD4, and substantial efficacy in several in vivo mouse
1
5
xenograft models. Herein we describe the SAR development leading from this pyridone core to the
discovery of pyrrolopyridoneꢀbased inhibitor 63 (ABBVꢀ075/mivebresib), an extremely potent BET
4
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 51
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
bromodomain inhibitor demonstrating excellent pharmacokinetic properties which currently is
1
6
undergoing Phase I clinical trials (ClinicalTrials.gov identifier: NTC02391480).
ꢀ SYNTHESIS
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
Synthesis of the inhibitors described and characterized herein relied mainly upon a Suzukiꢀ
17
Miyaura cross coupling reaction to form the bond between the pyridoneꢀbased core and the biaryl
ether, as is depicted in its simplest form for the generation of pyridones 9, 11, and 13 in Scheme 1.
Thus, reaction of 2ꢀphenoxyphenylboronic acid with bromo pyridones 8, 10, or 12 under standard
SuzukiꢀMiyaura coupling conditions provided compounds 9, 11, and 13 in good yield.
a
Scheme 1
a
Reagents and conditions: (i) Pd(PPh ) , CsF, DME/MeOH, 120 °C,
3
4
7
2
9% (9), 82% (13); (ii) Pd(PPh ) Cl , Na CO , DME/MeOH, 120 °C,
3 2 2 2 3
3% (11).
An analogous approach provided access to the unadorned pyrrolopyridone 19. Pyrrolopyridone
bromide 18 was generated in excellent overall yield utilizing a fiveꢀstep sequence beginning with 5ꢀ
bromoꢀ2ꢀmethoxyꢀ4ꢀmethylꢀ3ꢀnitropyridine (14), and was utilized in the crucial SuzukiꢀMiyauri
coupling reaction with 2ꢀphenoxyphenylboronic acid followed by deprotection to form 19 (Scheme 2).
5
ACS Paragon Plus Environment

Page 7 of 51
Journal of Medicinal Chemistry
1
a
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
Scheme 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
a
Reagents and conditions: (i) LiOMe, DMF, 100 °C, 76%; (ii) RaꢀNi
800, ethyl acetate, rt, 72%; (iii) NaH, pTsCl, DMF, rt, quant. yield;
iv) 4 M HCl, 1,4ꢀdioxane, 40 °C, 94%; (v) NaH, MeI, DMF, rt,
6%; (vi) 2ꢀphenoxyphenylboronoic acid, Pd(PPh ) , CsF, DME/
2
(
9
3
4
MeOH, 120 °C, then add K CO , water, 120 °C, 59%.
2
3
Three complimentary approaches detailed in Schemes 3ꢀ5 were used to prepare more highly
functionalized pyrrolopyridone analogs, incorporating substitution on both aryl groups of the crucial
biaryl ether. Two crucial transformations, a Suzuki coupling and a nucleophilic aromatic substitution,
were carried out in each approach, although the order of these two transformations was altered based
upon ease of synthesis and the availability of starting materials. The first of these three approaches is
outlined in Scheme 3 for the generation of compounds 24-27, which were prepared in order to explore
SAR on the central aryl ring. This approach began with a SuzukiꢀMiyaura coupling reaction of bromide
1
8, which was reacted with aryl boronic acid 20 to form the biaryl intermediate 21. With this biaryl core
in place, generation of biaryl ether 22 was accomplished readily via an S Ar displacement reaction of
N
the activated aryl fluoride of 21 by phenol in the presence of cesium carbonate. The tosyl protecting
group was also removed during this transformation. Ironꢀcatalyzed reduction of the nitro group of 22 in
6
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 51
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 presence of ammonium chloride followed by functionalization of the resulting aniline 23 provided
facile access to sulfonamides 24-26 and sulfamide 27.
a
Scheme 3
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
Reagents and conditions: (i) Pd(PPh ) , Na CO , DME/water, 120
3
4
2
3
°C, 52%; (ii) Phenol, Cs CO , DMSO, 100 °C, 72ꢀ84%; (iii) Fe,
2
3
NH Cl, THF/EtOH/H O, 95 °C 82% (23); (iv) RSO Cl, NEt ,
4
2
2
3
CH Cl , rt, then 1N NaOH, 90 °C, 45ꢀ77% (24-26) or Me NSO Cl,
2
2
2
2
Cs CO , DMF, 80 °C, 11% (27).
2
3
The second general approach to the formation of inhibitors containing functionalized biaryl
ethers is outlined in Scheme 4. The central premise of this approach relied upon formation of the
functionalized biaryl ether as the first step in the sequence, followed by SuzukiꢀMiyaura coupling with
pyrrolopyridone boronate 28 to generate the complete inhibitor framework. To facilitate this approach,
pyrrolopyridone boronate 28 was generated in good yield by reaction of bromide 18 with
4
,4,4',4',5,5,5',5'ꢀoctamethylꢀ2,2'ꢀbi(1,3,2ꢀdioxaborolane)
in
the
presence
of
tris(dibenzylideneacetone)dipalladium(0)
,
XꢀPHOS, and potassium acetate. The availability of boronate
7
ACS Paragon Plus Environment

Page 9 of 51
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
28 allowed for the use of aryl bromides and iodides as biaryl ether coupling partners, which simplified
the incorporation of a wide variety of more highly functionalized analogs into the SAR evaluation and
also provided a more convergent synthetic approach. For example, S Ar reaction of 3ꢀbromoꢀ2ꢀchloroꢀ
N
5ꢀnitropyridine (29) with phenol provided brominated biaryl ether 30 in excellent yield (Scheme 4).
Coupling of pyrrolopyridone boronate 28 with biaryl ether bromide 30 produced protected
pyrrolopyridone 31 in excellent yield. Reduction of the nitro group of 31 followed by sulfonylation and
removal of the tosyl protecting group of 32 gave pyridine sulfonamide 33 in excellent overall yield.
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
Scheme 4
a
Reagents and conditions: (i) 4,4,4',4',5,5,5',5'ꢀoctamethylꢀ2,2'ꢀ
bi(1,3,2ꢀdioxaborolane, KOAc, Pd (dba) , XꢀPHOS, 1,4ꢀdioxane, 80
2
3
°
C, 73%; (ii) Phenol, Cs CO , DMSO, 80 °C, 91%; (iii) KOAc,
2 3
Pd (dba) ,
(1,3,5,7ꢀtetramethylꢀ6ꢀphenylꢀ2,4,8ꢀtrioxaꢀ6ꢀ
2
3
phosphaadamantane), 1,4ꢀdioxane/H O, 60 °C, quant. yield; (iv) Fe,
2
NH Cl, THF/EtOH/H O, 100 °C, quant. yield; (v) MeSO Cl, NEt ,
4
2
2
3
CH Cl , rt, then 1 M NaOH, 90 °C, 77%.
2
2
8
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 51
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 third approach to the generation of pyrrolopyridone inhibitors employed an even more
convergent synthetic sequence (Scheme 5), and utilizes the crucial SuzukiꢀMiyaura coupling as the final
step in the sequence. For example, S Ar displacement of 34 by phenol, subsequent reduction of the
N
nitro group of 35, and functionalization of the resulting aniline 36 provided fully functionalized
sulfonamide 37, suitable for coupling with boronate 28 to establish an alternate route to inhibitor 24. As
exemplified by the routes highlighted in Schemes 3ꢀ5, the availability of multiple potential approaches
to each compound ensured that synthetic availability was not a limiting factor for the SAR evaluation of
these inhibitors. All of the additional inhibitors examined in this study were generated by the straightꢀ
forward application of one of the approaches outlined in Schemes 3, 4, or 5, including compound 63
from compound 21 via the approach outlined in Scheme 3 as described in the Experimental Section.
Detailed experimental procedures and characterization of compounds 38ꢀ62 and 64ꢀ120 can be found in
the Supporting Information (Schemes S1ꢀ7).
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
Scheme 5
a
Reagents and conditions: (i) Phenol, Cs CO , DMSO, 80 °C,
2
3
quant. yield; (ii) Fe, NH Cl, THF/EtOH/H O, 100 °C, quant. yield;
4
2
(iii) NEt , MeSO Cl, CH Cl , rt, then NaOH, 1,4ꢀdioxane, 70 °C,
3
2
2
2
75% (iv) 28, KOAc, Pd (dba) , (1,3,5,7ꢀtetramethylꢀ6ꢀphenylꢀ
2
3
2
,4,8ꢀtrioxaꢀ6ꢀphosphaadamantane), 1,4ꢀdioxane/H O, 60 °C, 79%.
2
9
ACS Paragon Plus Environment

Page 11 of 51
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
ꢀ RESULTS AND DISCUSSION
To further improve the potency of pyridoneꢀbased BET bromodomain inhibitors such as 7,
exploration of compounds that might provide an even more productive interaction between the inhibitor
core and the conserved Asn433 residue of the BET protein was undertaken. Compounds were evaluated
using a timeꢀresolved fluorescence resonance energy transfer (TRꢀFRET) binding assay and two
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
complementary cellular assays. The TRꢀFRET binding assay was used to determine the affinities (K ) of
i
compounds for a construct containing the two bromodomains of BRD4. Target engagement in cells was
measured using a luciferase reporter assay based on the contribution of BRD4 to human papilloma virus
(HPC) E2ꢀmediated transcriptional repression, where BRD4 is part of the HPV long control region
18
(LCR) promoter repression complex with E2 and EP400. In this assay, engagement of BRD4 with a
BET bromodomain inhibitor deꢀrepresses the HPV promoter engineered to drive luciferase transcription,
resulting in an increase of luciferase signal (see Supporting Information, Figure S3). Cancer cell lines
1
9
are dependent on BET proteins for growth, and so, as an orthogonal cellular assay, the impact of
compounds on cancer cell proliferation was measured using the triple negative breast cancer cell line
MXꢀ1 (ATCC) in a 3ꢀday proliferation assay. Good correlation was observed between the two cellular
assays, an indication that cell killing is the result of engagement of the BRD4 target (Supporting
Information, Figure S5). Examination of the proteinꢀinhibitor coꢀcrystal Xꢀray structures of
pyridazinone 6 and related pyridone analogs indicated that although the carbonyl moieties of these cores
15
are situated at an ideal distance away from the Asn433 NH group (2.9 Ǻ for 6), there does not appear
2
to be a productive contact between the Asn433 amide carbonyl and the pyridazinone or pyridone. The
first structural motif to be examined in order to provide a bidentate interaction with Asn433 incorporated
the 3ꢀmethyl NHꢀpyridone 11 in place of the Nꢀmethyl pyridone core of 9. It was proposed that a
bidentate interaction between the NH of the pyridone and the Asn433 carbonyl would provide an
improved binding interaction while maintaining the previous positive interactions of both the methyl
10
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 51
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
group with the amphipathic water pocket and the carbonyl group of the pyridone with the Asn433 amide
NH moiety. Examination of NHꢀpyridone 11 revealed, however, no improvement in biochemical or
2
cellular activity compared to Nꢀmethyl pyridone 9 (Table 1), and in fact revealed a slight decline in
2
0
LipE. An Xꢀray structure of pyridone 11 bound in BRD4 BDII (PDB code: SUVZ) indicated that
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
while the pyridone
Table 1. Biochemical and cellular potency of compounds 9, 11, 13, and 19.
O
O
H
N
NH
N
O
O
Compound ID
9
11
13
19
BRD4 TRꢀFRET Ki
0
.89 ± 0.037
1.7 ± 0.06
2.4
3.0 ± 0.86
0.048 ± 0.001
0.55
a
(
µM)
MXꢀ1 Proliferation
4
4
.8
.1
> 5
b
EC (µM)
5
0
BRD4 Engagement
2.3
1.5
8.5
2.0
0.48
b
EC (µM)
5
0
LipE
1.8
2.8
a
b
TRꢀFRET BRD4 K values are reported as the geometric mean derived from 3 or more independent measurements. EC50 values are
i
reported as the mean derived from two measurements.
carbonyl/Asn433ꢀNH distance remained optimal (2.9 Ǻ), there is not a direct interaction of the pyridone
2
NH with the Asn433 carbonyl moiety but instead a waterꢀmediated association (Figure 2) which does
not provide additional binding compared to pyridone 9. The slightly weaker K of NHꢀpyridone 11
i
compared to Nꢀmethylpyridone 9 may be a reflection of the presence of a fixed water molecule between
2
1
the NH of 9 and the Asn433 carbonyl. The entropy cost of rigidifying the water may affect the Ki
11
ACS Paragon Plus Environment

Page 13 of 51
Journal of Medicinal Chemistry
1
2
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
negatively. SZMAP calculation of the ∆G of this water gives a value of ꢀ6.64 kcal/mol. While one
cannot compare this freeꢀenergy value directly with energy values derived from binding data, the low
value indicates the water may have high occupancy.
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 2. Compound 11 bound to BRD4 BDII (Resolution =
1
.6 Ǻ, PDB code: SUVZ). Leu387 and Gly386 removed for
clarity.
In a second attempt to generate a more productive bidentate interaction between the crucial
Asn433 moiety and the pyridone inhibitor, it was next hypothesized that incorporation of an amine NH
group one atom closer to the Asn433 carbonyl would position the requisite NH group in a more suitable
location. Binding of this core would also be likely to displace a water molecule in the structure of
pyridone 11. Thus, 3ꢀaminoꢀ1ꢀmethylpyridone 13 was generated and examined (Table 1). Binding and
cellular data for 13, however, indicated that this exocyclic amine moiety also failed to provide any
improvement in potency. A slight improvement in LipE was realized, mainly a result of a lower cLogP.
A substantial improvement in biochemical binding potency, cellular activity, and lipophilic efficiency
was realized, however, with the incorporation of a pyrrole, in the form of pyrrolopyridone 19 (Table 1).
Compared to pyridone 9, addition of the pyrrole of 19 improved both biochemical and cellular activity
2
2
by 9ꢀ to 19ꢀfold (Table 1). Examination of the Xꢀray coꢀcrystal structure of pyrrolopyridone 19 (PDB
12
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 51
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
code: SUVY) in BRD4 BDII revealed the basis for this boost in activity (Figure 3). In addition to
maintaining the valuable interactions of the phenyl ether moiety in the WPF pocket, the Nꢀmethyl group
in the amphipathic water pocket, and the pyridone carbonyl with the NH of Asn433, the Xꢀray structure
2
confirms that a valuable interaction between the pyrrole NH and the Asn433 carbonyl has been
introduced, with an ideal 2.8 Ǻ distance measured between these two moieties. In addition, the pyrrole
ring of the heterocycle displaces a water molecule which is in a hydrophobic environment; the freeꢀ
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
2
1
energy of this water is +1.2 kcal/mol. The combination of replacing the water molecule bound to the
Asn433 carbonyl with a good hydrogen bond donor and replacing a water molecule in a hydrophobic
environment with the hydrophobic pyrrole ring would be expected to improve affinity, as demonstrated
by an improvement in K from 890 nM to 48 nM.
i
Figure 3. Compound 19 (yellow) bound to BRD4 BDII
(Resolution = 2.3 Ǻ, PDB code: SUVY). Tyr432 removed
for clarity.
With an effective and potent pyrrolopyridone BET bromodomain inhibitor core in place,
examination of SAR at additional sites on the molecule was undertaken to further improve potency and
to assess DMPK properties. The first region of the protein to be investigated was the area accessible
13
ACS Paragon Plus Environment

Page 15 of 51
Journal of Medicinal Chemistry
1
2
3
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
from the central aryl ring of the core, reaching into the pocket bounded by the ZA loop of the protein
and the adjacent helix. In fact, access to this pocket was facilitated by the chemistry developed to
generate biaryl ethers, as described above. In general, incorporation of an electronꢀwithdrawing group,
such as a nitro group, an ester, or a sulfone, promoted nucleophilic aromatic substitution on this central
ring, and allowed for rapid synthesis of the biaryl ether moiety with a functional group attached para to
the ether oxygen. Examination of the SAR available directly from this functional group on the central
ring, or after additional manipulation of this functional group, revealed a wide range of substituents with
the potential for substantial positive protein backbone interactions (Table 2). For example, nucleophilic
substitution on the central aryl moiety containing a nitro group, followed by reduction of the nitro group
and formation of sulfonamides provided compounds 24ꢀ26, which all demonstrated improved potency in
the TRꢀFRET binding assay and in cellular assays compared to the unsubstituted analog 19, along with a
considerable improvement in lipophilic efficiency (e.g., LipE for 19 = 2.8, LipE for 24 = 5.3). An Xꢀray
coꢀcrystal structure of sulfonamide 24 (PDB code: SUVX) in BRD4 BDII (Figure 4) confirmed the
presence of a productive hydrogen bond between the NH of Asp381 and one of the sulfonamide
oxygens, with a measured distance of 2.8 Ǻ. Notably, the positioning of the sulfonamide moiety
allowed for the formation of this hydrogen bond without disturbing the valuable interactions between the
pyrrolopyridone carbonyl (2.9 Ǻ) and the pyrrole NH (2.9Ǻ) with Asn433 and also maintained the
critical position of the phenyl ether in the WPF pocket. A similar interaction and boost in potency was
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
15
noted with the analogous pyridone BET bromodomain inhibitor core and also with a related 2ꢀ
2
4
methylpyrroleꢀ3ꢀcarboxaminde core. Similar increases in potency were demonstrated for sulfamides
27) and amides (39), although not to the extent shown for sulfonamides. Analogs were also prepared
(
that incorporated a hydrogen bond accepting group directly attached to the central aryl group, including
reversed amides (45 and 46), sulfones (51), and reversed sulfonamides (56). Although these analogs
also demonstrated improved potency compared to the unsubstituted analog 19, this increased potency
was not as substantial as for sulfonamides 24ꢀ26. We hypothesize that for these moieties to be
14
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 51
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
positioned in the proper alignment and distance to take advantage of an interaction with Asp381,
torqueing of the biaryl bond between the central aryl group and the pyrrolopyridone is necessary, thus
requiring a repositioning of the phenyl ether into a less favorable arrangement in the WPF pocket.
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. Compound 24 (yellow) bound to BRD4 BDII
(
Resolution = 1.5Ǻ, PDB code: SUVX). Leu385 removed for
clarity.
In addition to the enhanced potency achieved by incorporation of hydrogen bond acceptors on
the central aryl group, a substantial improvement in the in vitro stability of these compounds in liver
microsomes was also evident compared to unsubstituted analog 19. For example, whereas the stability
of compound 19 was limited in human liver microsomes and particularly poor in mouse and rat liver
microsomes, sulfonamides 24ꢀ26 and 33 demonstrated excellent stability in human liver microsomes
and moderate stability in mouse microsomes (Table 2). The MDCK cell permeability of these
ꢀ6
compounds was high (> 10 x 10 cm/s), although compounds containing a central pyridine ring (e.g.,
3) demonstrated only moderate permeability.
3
15
ACS Paragon Plus Environment

Page 17 of 51
Journal of Medicinal Chemistry
1
Table 2. Activity and ADME data for aryl substitutions.
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
c
TRꢀFRET
MXꢀ1
BRD4
Liver Microsome Clint.u
MDCK
d
BRD4
Proliferation Engagement
Human
Mouse
L/hr/kg
Rat
Permeability
a
b
b
ꢀ6
ID
X
R
H
K (nM)
i
EC50 (nM)
550
EC50 (nM)
480
10 cm/s
1
9
CH
48 ± 1.4
2.4 ± 1.1
1.5 ± 0.2
4.4 ± 0.4
2.9 ± 0.1
4.1 ± 0.3
9.8 ± 1.5
14 ± 1.4
32 ± 3.9
5.8 ± 0.2
3.6 ± 0.1
100
5.0
5600
18
2400
120
200
76
31
12
22
16
19
6
24
CH
CH
CH
CH
N
ꢀNHSO
2
Me
Et
16
14
11
8.1
22
43
88
26
71
70
50
39
2
2
2
5
6
7
ꢀNHSO
2
< 2.7
2.8
55
ꢀNHSO
CH
CF
31
39
2
2
3
ꢀNHSO
NMe
65
14
140
< 5.6
210
27
240
36
2
2
33
ꢀNHSO
ꢀNHCOMe
ꢀCONH
ꢀCONHEt
ꢀSO Me
ꢀSO NH
2
Me
46
3.8
3
4
9
5
CH
CH
CH
CH
CH
46
6.8
46
28
14
30
24
22
120
130
52
2.9
33
2
46
4.1
43
240
65
5
5
1
6
2
5.5
23
56
< 2.6
31
79
2
2
a
b
TRꢀFRET BRD4 K values are reported as the geometric mean derived from 3 or more independent measurements; EC values are
i
50
c
reported as the mean derived from two measurements; For reference, average inꢀhouse positive control clearance values for
d
dextromethorphan and verapamil, respectively, are Clint,u (human) = 7.4 L/hr/kg (low) and Clint,u (human) = 30 L/hr/kg (high). For
ꢀ
6
reference, average inꢀhouse positive control permeability values for atenolol and metoprolol, respectively, are 0.5 x 10 cm/s (low)
ꢀ
6
and 36 x 10 cm/s (high).
With analogs in hand that demonstrated excellent potency and promising in vitro microsomal
stability and permeability properties, the pharmacokinetic properties of lead compounds were examined.
16
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 51
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 general trend observed for these compounds is exemplified by the mouse and rat PK data collected
for sulfonamide 24. In agreement with the mouse liver microsome stability data (Clint,u = 18 L/hr/kg),
sulfonamide 24 exhibited moderately low unbound IV clearance (39 L/hr/kg) and good oral absorption
(FaFg = 0.77) leading to reasonable oral exposures (AUC = 0.34 µM*hr, F = 55%) in mouse PK studies.
However, rat PK studies for 24 exhibited substantially higher clearance (IV Clp,u = 74 L/hr/kg), with
limited oral absorption (FaFg = 0.32) and exposure (AUC = 0.016 µM*hr) leading to poor
bioavailability (F = 10%). A metabolic pathways analysis of 24 in rat and human liver microsomes
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 S2 and Table S1, Supporting Information) revealed that extensive oxidative metabolism in rat
liver microsomes occurs at the unsubstituted phenyl ether moiety, as might be expected for this electronꢀ
rich substrate. This study also indicated that cleavage of the phenyl ether and Nꢀdemethylation at the Nꢀ
methyl substituent on the pyrrolopyridone provide additional metabolites, but these metabolic liabilities
were much less significant compared to oxidation of the phenyl ether. Oxidation of the unsubstituted
phenyl ether of 24 was also determined to be the main metabolic pathway in human liver microsomes,
although this metabolism was less extensive in human liver microsomes than in rat liver microsomes.
To overcome the metabolic liabilities posed by the unsubstituted phenyl ether of analogs such as
2
4 and 25, a set of analogs was generated to examine the scope and limitations of replacements for this
phenyl ether, examining these analogs on core structures that contained either a methyl or ethyl
sulfonamide on the central phenyl ring (Table 3). It was hoped that the incorporation of electronꢀ
withdrawing substituents on the phenyl ether ortho or para to the ether oxygen would substantially
reduce the metabolic instability of this moiety, and thus most of the initial analogs generated contained
halogenated phenyl ethers or other electronꢀpoor substituents (Table 3, compounds 57ꢀ64). Initial
results from this evaluation indicated that the incorporation of more electronꢀpoor phenyl ethers did in
general provide compounds with substantially improved stability in rat liver microsomes, a trend also
1
5
24
noted for similarly substituted Nꢀmethylpyridone
and 2ꢀmethylpyrroleꢀ3ꢀcarboxamide
BET
bromodomain inhibitors. Incorporation of one halogen (57) or multiple halogens (61ꢀ64) gave
17
ACS Paragon Plus Environment

Page 19 of 51
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
compounds with 6ꢀ to 34ꢀfold improvement in rat liver microsome stability, while generally maintaining
potency in both the biochemical binding assay and in cellular assays. Phenyl ethers substituted with
nitriles (58) or trifluoromethyl groups (59, 60) also provided improved stability in rat liver microsomes,
although in the case of the trifluoromethyl substituents this improvement came at the expense of stability
in mouse liver microsomes and also overall potency. Examination of heteroaryl replacements (65, 66)
failed to provide inhibitors that improved liver microsome stability while maintaining potency. Of these
aryl analogs, 2,4ꢀdifluorophenyl ethers 62 and 63 provided the best combination of potency and
stability.
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
In addition to aryl replacements for the unsubstituted phenyl ethers of 24 or 25, compounds
containing alkyl replacements were also examined. Benzyl ether 67, which incorporates a methylene
spacer in the ether moiety, lost substantial potency compared to the phenyl ethers. Cyclohexyl ether 68
maintained the potency of the phenyl ethers, but failed to provide an improvement in metabolic stability.
Attempts to improve the ADME properties of this cyclohexyl moiety by replacement with
tetrahydropyran (69) or with the 4,4ꢀdifluorinated cyclohexyl analog 70 also gave compounds with
reduced potency or limited metabolic stability, and also inadequate permeability. Acyclic analogs such
as 71 lost substantial potency compared to aryl ethers. Incorporation of hydrophilic moieties,
represented by dimethylamine 72 or piperidine 73, led to improved liver microsome stability, but both
the potency and permeability of these compounds was limited. Since none of these alkyl analogs
provided potency or ADME properties comparable to 2,4ꢀdifluorophenylether analogs 62 and 63, all
additional efforts focused on the examination of compounds containing this moiety.
18
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 51
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
Table 3. Activity and ADME data for phenyl ether replacements
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
Liver Microsome Clint.u
MDCK
BRD4
MXꢀ1
BRD4
d
TRꢀFRET Proliferation Engagement Human Mouse Rat
Permeability
1
2
a
b
b
ꢀ6
1
0 cm/s
ID
R
R
K (nM)
i
EC (nM)
EC (nM)
L/hr/kg
18
50
50
24
25
57
Me
Et
Ph
Ph
2.4 ± 1.1
1.5 ± 0.2
3.4 ± 0.3
16
14
22
11
8.1
17
5.0
120
200
12
12
22
21
< 2.7
< 5.8
55
Et
pꢀClꢀphenyl
54
58
Et
pꢀCNꢀphenyl
9.4 ± 2.5
17
48
2.5
< 6.5
4.9
11
5
6
6
9
0
1
Et
pꢀCF
3
3
ꢀphenyl
ꢀphenyl
9.0 ± 0.4
6.4 ± 0.9
3.1 ± 0.7
4.5 ± 0.6
30
83
27
29
65
76
35
21
7.7
32
170
81
35
15
11
7.8
11
17
Me
Me
Me
oꢀCF
2ꢀClꢀ4ꢀFꢀphenyl
2,4ꢀdiꢀFꢀphenyl
14
51
19
62
9.3
17
< 3.9
6
3
Et
Et
2,4ꢀdiꢀFꢀphenyl
1.5 ± 0.2
12 ± 4.8
13
20
19
< 2.9
6.1
33
54
< 4.9
14
29
21
64
2,4,6ꢀtriꢀFꢀphenyl
7.9
6
6
5
6
Me
Et
3ꢀpyridyl
6.3 ± 2.8
32 ± 1.5
34
67
5.2
< 2
32
14
21
3.0
1.0
5ꢀpyrimidine
130
170
2.9
6
6
7
8
Me
Me
benzyl
68 ± 1.7
3.7 ± 0.3
110
14
120
18
7.3
27
54
34
96
24
23
cyclohexyl
220
4
ꢀtetrahydroꢀ
pyran
6
7
9
0
Me
Me
68 ± 15
19 ± 1.0
72
19
90
20
< 1.8
< 2.3
< 5.4
43
14
12
1.7
12
4
,4ꢀdiꢀFꢀ
cyclohexane
19
ACS Paragon Plus Environment

Page 21 of 51
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
71
Et
Et
ꢀCH
ꢀ(dimethylꢀ
amino)cyclohexyl
ꢀmethylꢀ
piperidinꢀ4ꢀyl
CMe
38 ± 20
26 ± 2.8
77
86
79
17
62
26
43
2
3
4
7
7
2
3
270
< 1.6
7.5
< 2.7
2.7
1
Et
430 ± 45
210
880
< 1.7
< 5.1
4.8
0.55
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
TRꢀFRET BRD4 K values are reported as the geometric mean derived from 3 or more independent measurements; EC values are
i
50
c
reported as the mean derived from two measurements; For reference, average inꢀhouse positive control clearance values for
d
dextromethorphan and verapamil, respectively, are Clint,u (human) = 7.4 L/hr/kg (low) and Clint,u (human) = 30 L/hr/kg (high). For
ꢀ
6
reference, average inꢀhouse positive control permeability values for atenolol and metoprolol, respectively, are 0.5 x 10 cm/s (low) and 36
ꢀ6
x 10 cm/s (high).
PK studies were next undertaken to assess whether or not the improvements in rat liver
microsome stability for 2,4ꢀdifluorophenyl ethers 62 and 63 compared to unsubstituted phenyl ether 24
would translate into improved rat PK properties. Examination of the rat PK of 62 and 63 revealed a
substantial improvement in IV clearance and absorption (FaFg), leading to higher bioavailability and
improved oral exposures (Table 4). In addition, modest improvement in mouse PK in both clearance
and bioavailability for 2ꢀ4ꢀdifluorophenyl analogs 62 and 63 was noted compared to phenyl analog 24.
Furthermore, ethyl sulfonamide 63 was preferred over methyl sulfonamide 62 based upon its modestly
improved biochemical and cellular activity together with its low clearance in human liver microsomes.
Compound 63 also exhibited physiochemical properties typically targeted for the generation of drugꢀlike
20,25
molecules,
including molecular weight (459 g/mol), cLogP (3.8), TPSA (92 Ǻ), number of Hꢀbond
acceptors (4) and donors (2), and LipE (4.9).
20
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 51
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
Table 4. DMPK data for compounds 24, 62 and 63.
Protein
Binding
Mouse
Protein
Binding
Rat
Mouse PK
Rat PK
(
IV Clp,u
39
3 mg/kg iv and 10 mg/kg po)
(1 mg/kg iv and po)
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
c
d
a
b
c
d
ID
PO AUCu
0.34
FaFg
0.77
0.70
0.90
F (%)
55
(fu,plasma
0.039
0.033
0.020
)
IV Clp,u
74
PO AUCu
0.016
0.19
FaFg
0.32
0.52
0.77
F (%)
10
(fu,plasma
0.034
0.036
0.023
)
2
4
62
10
1.3
65
25
40
6
3
19
0.63
83
27
0.27
65
a
d
b
c
Clint,u [L/hr/kg]; AUCu [µM*hr]; FaFg = F/(1ꢀClp/Qh). Qh = liver blood flow for mouse: 5.2 L/hr/kg; for rat: 3.8 L/hr/kg;
fu,plasma (unbound fraction in plasma)
Although difluorophenyl ether sulfonamide 63 displayed the desired potency and DMPK
properties to support moving this compound forward toward clinical candidate consideration, a final
round of SAR was carried out holding the 2,4ꢀdifluorophenyl ether moiety in place and reꢀexamining a
wider range of substituents on the central aromatic ring. This effort is described in the Supporting
Information (Table S3). It was found that incorporation of certain substituents on this central aromatic
ring also provided potent and stable compounds. However, compound 63 demonstrated a superior
combination of attributes including excellent potency in cellular assays and an advantageous
pharmacokinetic profile. Compound 63 also compared favorably in terms of biochemical binding
7
8
potency and cellular activity with known BET inhibitors such as 1b and 1c (see Table S1 in the
Supporting Information). Thus, ethyl sulfonamide 63 was chosen as the lead compound and was
examined in a variety of in vitro and in vivo efficacy experiments.
21
ACS Paragon Plus Environment

Page 23 of 51
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
Figure 5. Structure of compound 63
Compound 63 displayed cellular activities characteristic of BET family inhibitors, such as
2
6
inhibition of cꢀMyc expression and displacing BRD4 from the Myc promoter. Onꢀtarget inhibition
with compound 63 was also confirmed in vivo by assessing the effect on cytokine release in a murine
model of LPSꢀinduced endotoxic shock. The acute upregulation of inflammatory cytokine genes such as
ILꢀ6 observed in the shock model has been shown to be BETꢀmediated and consequently cytokine
9,27
inhibitory responses can be considered as a biomarker of BET bromodomain inhibition. As shown in
Figure 6a, oral administration of compound 63 resulted in a doseꢀdependent inhibition of ILꢀ6
concentration measured in plasma two hours post LPS challenge. The magnitude of inhibition correlated
with the concentration of compound 63 in plasma at the time of cytokine measurement (Figure 6b).
These results indicate that robust target inhibition (>50%) was achieved with plasma concentrations
ranging from 0.03 to 0.1 µM.
22
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 51
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
00
1
00
80
60
40
20
0
8
0
0
6
40
20
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
//
//
0
.0 00 1
0.01
0.1
1.
10.
100.
0.
0.001
0.01
0.1
1.
10.
dose (mg/kg)
plasma concentration (µM)
(a)
(b)
Figure 6. (a) Dose–dependent inhibition of LPSꢀinduced ILꢀ6 by compound 63. (b)
Relationship between plasma concentration of compound 63 and response. Plasma
concentration of ILꢀ6 and compound 63 were determined 2 hours post LPS challenge.
Values represent mean ± SE (n=5/group).
In vivo antitumor efficacy of compound 63 is exemplified by activity in a Kasumiꢀ1 AML mouse
xenograft model shown in Figure 7. Compound 63 was dosed orally QD at 1 mg/kg for 25 days and
achieved 99% tumor growth inhibition (TGI) with acceptable tolerability (weight loss ≤10%). For
comparison a representative therapy for AML, 5ꢀazacitidine, achieved 76% TGI when administered at
its MTD (IV Q7d, 8 mg/kg). An estimate of an efficacious exposure target for compound 63 can be
determined by adjusting the cellular antiproliferative potency (EC of 0.0021 µM) for protein binding
5
0
(98%), resulting in an exposure target of 0.10 µM. Based on plasma exposure values obtained in a
pharmacokinetic study in nonꢀtumor bearing animals, the plasma concentration target was achieved for
approximately 12 hours/day in the efficacy study (Supporting Information, Figure S7). The robust
antitumor activity achieved with compound 63 at the targeted plasma concentration is in agreement with
the activity in the LPSꢀinduced cytokine biomarker study and strongly supports onꢀtarget inhibition. The
breadth of activity of 63 dosed orally QD at doses ranging from 1ꢀ2 mg/kg over 14ꢀ28 days against both
23
ACS Paragon Plus Environment

Page 25 of 51
Journal of Medicinal Chemistry
1
2
6,28
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
heme and solid tumors has been reported elsewhere,
indicating that BET bromodomain inhibitor 63
has the potential for utility in a wide range of oncology indications.
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. Kasumiꢀ1 mouse xenograft study with compound 63; Values represent mean ± SE
(
n=8/group), WL: maximum mean weight loss. REM: removed from study due to morbidity.
In addition to the evaluation of efficacy, further characterization of 63 across species also
demonstrated desirable PK profiles (Table 6) supporting the choice of 63 as a clinical candidate. Indeed,
24
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 51
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
6a
in an ongoing clinical trial where safety, tolerability, and human PK for 63 were evaluated in subjects
with cancer, a prolonged halfꢀlife of 25 hours following oral administration of 1 mg (n = 3) was
observed, reflecting the excellent metabolic stability observed both in vitro (microsome and hepatocyte)
and in vivo (IV CLp). The observed human PK profile (1 mg oral dose) supports once daily dosing for
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
63 as a therapeutic agent.
Table 6. Mean PK parameters for 63 across species.
Parameter
Mouse
8.3
3.5
0.5
68
Rat
4.0
4.0
0.6
65
Dog
3.1
2.1
0.19
41
Monkey
2.3
Human
< 1.5
0.87
a
Microsome CLint (L/hr/kg)
a
Hepatocyte CLint (L/hr/kg)
3.1
a
b
IV CL (L/hr/kg)
0.33
33
0.14
p
c
F (%)
50
PO CLp/F (L/hr)
t (hr)
ꢀ
ꢀ
ꢀ
ꢀ
5.03
25.1
2.7
2.6
5.8
4.4
1
/2
a
Not corrected for protein binding as no species difference in protein binding was observed
b
Estimated IV CLp based on observed PO CLp/F and estimated F (%)
c
Estimated F (%) based on observed F (%) from preclinical species
Direct binding of compound 63 to BRD4 bromodomains I and II was tested by isothermal
titration calorimetry (ITC). Compound 63 binds to both domains tightly with Kd < 10 nM (Supporting
Information, Figure S8). The selectivity of compound 63 across both the BET family and a set of
bromodomainꢀcontaining proteins was also examined. Compound 63 binds to the tandem domains of
BRD2, BRD4, and BRDT with similar affinities (Table 7), although binding of compound 63 to the
tandem domain of BRD3 is approximately 10ꢀfold weaker. In addition, examination of the activity of
compound 63 against a set of bromodomainꢀcontaining proteins revealed moderate activity against
EP300 (K = 87 nM, 54ꢀfold selectivity vs. BRD4) and potential weak activity against SMARCA4 (70%
d
25
ACS Paragon Plus Environment

Page 27 of 51
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
inhibition at 1 µM), but displayed K > 1 µM for 18 other bromodomain proteins that were examined
d
(
Table S4, Supporting Information).
Table 7. Comparative binding data for 63 within BET family (K in nM)
i
BRD2
BDIꢀBDII
G73ꢀA560
1.0
BRD3
BDIꢀBDII
P24ꢀP416
12.2
BRD4
BDIꢀBDII
K57ꢀK550
1.5
BRDT
BDIꢀBDII
N21ꢀP380
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
ID
6
3
In order to evaluate potential offꢀtarget interactions, compound 63 was evaluated in a Cerep
panel screen (Cerep, http://www.cerep.fr) against 79 molecular targets, including receptors and
enzymes, measured at 10 µM. Compound 63 displayed excellent selectivity, as only three targets
displayed > 80% displacement of controlꢀspecific binding (A1 @ 97%, A2A @ 87%, and peripheral
benzodiazepine @ 86%, see Table S5 in the Supporting Information). Followꢀup studies were
conducted to establish whether the binding of compound 63 to these receptors modulated their function.
Compound 63 only modulated the functional activity of the A2B receptor (IC = 2.6 µM, antagonist
5
0
activity) and the peripheral benzodiazepine receptor (IC = 0.80 µM), thus providing at least a 500ꢀfold
5
0
window vs. BRD4 tandem domain binding.
The metabolism profile of compound 63 was also examined. A rat mass balance study using
radiolabeledꢀ63 demonstrated that 63 is primarily eliminated via metabolism (76% of the dose
administered). In vitro studies further identified CYP3A4/5 as the major metabolic enzymes that
contribute to > 60% of the metabolism of 63 in human hepatocytes, which suggest it may carry victim
liability upon coꢀadministration with strong CYP3A inhibitors or inducers in humans. From a
perpetrator perspective, 63 does not appear to carry significant liabilities at its efficacious concentration
based on in vitro characterization.
ꢀ
CONCLUSIONS
26
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 51
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 structureꢀbased design of the pyrrolopyridone BET family bromodomain inhibitor 63 has
been described. Incorporation of a pyrrole moiety with the original pyridone core enabled a
substantially more productive bidentate interaction with the crucial asparagine moiety of the BET
protein and provided a concomitant boost in activity (9ꢀ to 19ꢀfold). Examination of a range of
substituents reaching off of the central aromatic ring of the core pyrrolopyridone inhibitor 19 led to the
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
incorporation of a valuable sulfonamide moiety (compounds 24ꢀ26) and an improvement in activity
resulting from a useful interaction with a backbone protein aspartic acid residue. Finally, a metabolic
pathways analysis identified the pendant unsubstituted phenylether moiety as a hotspot for metabolism,
particularly in rat. Incorporation of electronꢀwithdrawing substituents on this aromatic ring substantially
reduced this problematic metabolism, and led directly to the discovery of 63, a structurally novel BET
bromodomain inhibitor with superior potency in TRꢀFRET binding assays and excellent activity in a
wide range of cellular phenotypes. Compound 63 also displays an advantageous DMPK profile across
multiple species, and demonstrates excellent exposures and a 25 h halfꢀlife in humans. Compound 63
demonstrates valuable activity in a wide range of in vivo efficacy models, and currently is under
examination in Phase I clinical trials.
ꢀ
EXPERIMENTAL SECTION
Protein expression and purification. Human BRD4 BDII (residues 352ꢀ457) and BDR4 BDIꢀ
BDII (residues 57ꢀ550) were cloned into the pET28b vector to make Nꢀterminal His6 with thrombin
cleavage site constructs. All the proteins were expressed in E. Coli BL21(DE3) cells and purified from
the soluble fraction using a NiꢀNTA column. For Xꢀray studies, the His6ꢀtag was cleaved with thrombin
protease and the protein was further purified using size exclusion chromatography.
BRD protein crystallization method. Human BRD4 BDII protein was concentrated to ~4
mg/mL in 10 mM BisꢀTris, pH 6.8, 100 mM NaCl, 5 mM DTT buffer. Protein was incubated with
compounds at a 3:1 mM ratio of compound to protein at 4 °C for 2 h. The proteinꢀcompound complexes
27
ACS Paragon Plus Environment

Page 29 of 51
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
were screened against SGCꢀ1 and SGC Redwing custom screens (prepared by Rigaku) at 17 °C. Some
proteinꢀcompound complexes were also screened against commercially available screens PEGRx and
SaltRx (Hampton Research) at 17 °C. Vaporꢀdiffusion sitting drops were prepared using a Mosquito
liquid dispenser (TTP Labtech) in MRC 2 Well Crystallization plates (Hampton Research.) The drops
contained 0.3 µL of protein and 0.3 µL of reservoir solution over wells of 40 µL of reservoir solution.
TR-FRET bromodomain binding assay. A timeꢀresolved fluorescence resonance energy
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
transfer (TRꢀFRET) assay was used to determine the affinities (K ) of compounds for the tandem
i
bromodomain of BRD4. Compound dilution series were prepared in DMSO via an approximately 3ꢀ
fold serial dilution. Compound dilutions were added directly into white, lowꢀvolume assay plates
(Perkin Elmer Proxiplate 384 Plus# 6008280) using a Labcyte Echo in conjunction with Labcyte Access
and Thermo Multidrop CombinL robotics. Compounds were then suspended in eight microliters (µL) of
assay buffer (20 mM sodium phosphate, pH 6.0, 50 mM NaCl, 1 mM ethylenediaminetetraacetic acid
disodium salt dihydrate, 0.01% Triton Xꢀ100, 1 mM DLꢀdithiothreitol) containing Hisꢀtagged
bromodomain, europiumꢀconjugated antiꢀHis antibody (Invitrogen PV5596) and Alexaꢀ647ꢀconjugated
probe. The final concentration of 1X assay mixture contained 0.5% DMSO, 5 nM His tagged BRD4
(BDIꢀII_K57ꢀ K550) and 30 nM probe, and 1 nM europiumꢀconjugated antiꢀHisꢀtag antibody, and
compound concentrations in the range of: 49.75 µMꢀ0.18 nM. After a 1 h equilibration at room
temperature, TRꢀFRET ratios were determined using an Envision multilabel plate reader (Ex 340, Em
495/520). TRꢀFRET data were normalized to the means of 24 noꢀcompound controls (“high”) and 8
controls containing 1 µM unꢀlabeled probe (“low”). Percent inhibition was plotted as a function of
compound concentration and the data were fit with the 4 parameter logistic equation to obtain IC s.
50
Inhibition constants (K ) were calculated from the IC s, probe K (0.021 µM) and probe concentration.
i
50
d
The TRꢀFRET binding assay had a running MSR = 1.2 and Testꢀretest MSR = 2.1. Compound 1b was
tested as a positive control for this assay, with a determined K value of 77 ± 17 nM. A representative
i
curve used for the determination of TRꢀFRET K for compound 63 is shown in the Supporting
i
28
ACS Paragon Plus Environment