Discovery of BIIB068: A Selective, Potent, Reversible Bruton's Tyrosine Kinase Inhibitor as an Orally Efficacious Agent for Autoimmune Diseases

Article abstract of DOI:10.1021/acs.jmedchem.0c00702

Autoreactive B cell-derived antibodies form immune complexes that likely play a pathogenic role in autoimmune diseases. In systemic lupus erythematosus (SLE), these antibodies bind Fc receptors on myeloid cells and induce proinflammatory cytokine producti

Full text of DOI:10.1021/acs.jmedchem.0c00702

Subscriber access provided by University of Birmingham
Drug Annotation
Discovery of BIIB068: A Selective, Potent, Reversible Bruton’s Tyrosine
Kinase Inhibitor as an Orally Efficacious Agent for Autoimmune Diseases
Bin Ma, Tonika Bohnert, Kevin L Otipoby, Eric Tien, Million Arefayene, Judy Bai, Bekim Bajrami,
Eris Bame, Timothy R Chan, Michael Humora, J Michael MacPhee, Douglas Marcotte, Devangi
Mehta, Claire M Metrick, George A. Moniz, Evelyne Polack, Urjana Poreci, Annick Prefontaine, Sarah
Sheikh, Patricia Schroeder, Karen Smirnakis, Lei Zhang, Fengmei Zheng, and Brian T Hopkins
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.0c00702 • Publication Date (Web): 22 Jul 2020
Downloaded from pubs.acs.org on July 22, 2020
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 53
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Discovery of BIIB068: a selective, potent, reversible
Bruton’s tyrosine kinase inhibitor as an orally
efficacious agent for autoimmune diseases
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
*
#
Bin Ma , Tonika Bohnert, Kevin L. Otipoby , Eric Tien, Million Arefayene, Judy Bai, Bekim
§
†
¶
Bajrami, Eris Bame, Timothy R. Chan , Michael Humora , J. Michael MacPhee , Douglas
ƒ
ɟ
Marcotte , Devangi Mehta , Claire M. Metrick, George Moniz, Evelyne Polack, Urjana Poreci,
ǂ
ƒ
Annick Prefontaine, Sarah Sheikh , Patricia Schroeder , Karen Smirnakis, Lei Zhang, Fengmei
Zheng, Brian T. Hopkins *.
Research & Development, Biogen, 225 Binney Street, Cambridge, MA, 02142, USA
ABSTRACT: Autoreactive B cell-derived antibodies form immune-complexes that likely play a
pathogenic role in autoimmune diseases. In Systemic Lupus Erythematosus (SLE), these
antibodies bind Fc receptors on myeloid cells and induce proinflammatory cytokine production
by monocytes and NETosis by neutrophils. Bruton’s tyrosine kinase (BTK) is a non-receptor
tyrosine kinase that signals downstream of Fc receptors and plays a transduction role in antibody
expression following B cell activation. Given the roles of BTK in both the production and
sensing of autoreactive antibodies, inhibitors of BTK kinase activity may provide therapeutic
value to patients suffering from autoantibody-driven immune disorders. Starting from an in-
house proprietary screening hit followed by structure-based rational design, we have identified a
ACS Paragon Plus Environment
1

Journal of Medicinal Chemistry
Page 2 of 53
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
potent, reversible BTK inhibitor, BIIB068 (1), which demonstrated good kinome selectivity with
good overall drug like properties for oral dosing, was well tolerated across preclinical species at
pharmacologically relevant doses with good ADME properties, and achieved >90% inhibition of
BTK phosphorylation (pBTK) in humans.
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
.
INTRODUCTION
Bruton’s Tyrosine Kinase (BTK) is a member of the Tec family of protein tyrosine kinases
(TFK) that functions downstream of the B cell antigen receptor (BCR) in B cells and Fc
1
receptors (FcRs) in myeloid cells such as neutrophils and basophils. BTK activation drives the
activation of key BCR-signaling pathways, phospholipase C-γ (PLCγ), phosphatidalyinositol-3-
kinase/Akt and NF-κB, which eventually regulates B cell activation and expression of
immunoglobulins (Ig). BTK plays a critical role in early B cell differentiation and mature B cell
activation, and is expressed throughout B cell development but is lost upon terminal
2
3
differentiation to plasma cells. In 1952, Ogden Bruton first reported a condition called X-linked
4
agammaglobulinemia (XLA), manifested by lack of Igs and B cells, that affects approximately 1
5
6
in 200,000 males. Mutations in the X-linked BTK gene are responsible for the XLA phenotype.
Similarly, crippling mutations in the mouse BTK gene disrupt BTK expression and give rise to
the X-linked immunodeficiency (xid) phenotype in mice, and those xid mice are resistant to
7
acquiring lupus and collagen induced arthritis. While BTK’s role in B cell biology has been best
8
9
characterized, inhibition of kinase activity with either covalent or reversible BTK inhibitors
additionally exhibits functional effects in myeloid cell populations,10 such as NETosis by
11
12
13
neutrophils, inflammation by basophils and cytokine production by monocytes.
ACS Paragon Plus Environment
2

Page 3 of 53
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
For the past two decades, the pharmaceutical community has directed significant resources to
1
4
identify BTK inhibitors for potential therapeutic use in various oncology and auto-immune
indications. Ibrutinib (2, Figure 1) is a first generation covalent BTK inhibitor that was approved
by the FDA in 2013 to treat patients with mantle cell lymphoma (MCL) and chronic lymphocytic
leukemia (CLL).15 Acalabrutinib (3), a second generation, more selective covalent BTK
inhibitor, exhibited a considerably improved safety profile (reduced adverse events of skin rash,
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
6
and bleeding risk ) and was approved in 2017, followed in 2019 by the approval of Zanubrutinib
(
4),17 another covalent BTK inhibitor from the same generation. Despite clinical success in
oncology, some characteristics of these covalent inhibitors preclude them from use in
autoimmune diseases that are often chronic. For example, elevations in liver aminotransferase
18
levels were reported with these covalent BTK inhibitors, both inherent and acquired resistance
1
9
to drug activity has been reported in various lymphomas, and an increased risk of idiosyncratic
toxicity20 is generally associated with covalent inhibitors. Therefore, the development of a
reversible, selective BTK inhibitor may provide autoimmune patients a treatment with equivalent
efficacy and a decreased risk profile.
Although numerous clinical trials have been reported targeting BTK inhibition as a potential
2
1
22
treatment for autoimmune diseases (Spebrutinib for rheumatoid arthritis (RA), Poseltinib for
2
3
24
25
RA and lupus, BMS-986195 for RA, Evobrutinib for multiple sclerosis (MS), PRN1008 for
2
6
pemphigus vulgaris, TAS5315 for RA , and Remibrutinib for chronic spontaneous urticaria and
2
7
28
Sjogren’s syndrome ), only three reversible BTK inhibitors (Fenebrutinib, BMS-986142 and
AS 1763) have been reported to be in clinic to date. The most advanced molecule is
Fenebrutinib, which was recently reported by Roche/Genentech to have achieved the primary
2
9
end point of ACR50 response in a phase II RA study , thus validating that inhibition of BTK
ACS Paragon Plus Environment
3

Journal of Medicinal Chemistry
Page 4 of 53
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
with a reversible small molecule can provide clinical efficacy for treatment of autoimmune
indications such as RA when compared to covalent inhibitors.30
Here we report the identification of a highly selective reversible BTK inhibitor compound 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
(BIIB068) in phase I clinical trial for Lupus, from an in-house proprietary screening hit followed
by structure-based rational design.
N
O
O
NH
H
N
O
O
N
O
H N
2
H N
2
H N
2
N
N
N
N
N
O
N
N
N
N
O
N
N
N
N
HN
N
N
N
N
H
O
O
1 (BIIB068)
2 (Ibrutinib)
3 (Acalabrutinib)
4 (Zanubrutinib)
Figure 1. Chemical structures of BTK inhibitors BIIB068 (1), Ibrutinib (2), Acalabrutinib (3) and
Zanubrutinib (4).
.
RESULTS AND DISCUSSION
Hit optimization. The initial hit pyrrolopyrimidine 5 (BTK IC = 3700 nM, Figure 2) was
5
0
31
identified from Biogen’s internal proprietary screening collection. It was co-crystalized with
the BTK protein to obtain a structure to 1.4 Å resolution (Figure 3, Supplementary Table S-1).
Compound 5 is an orthosteric ATP-competitive inhibitor which binds to the BTK in a DFG-in
conformation. In contrast to Ibrutinib, which occupies the kinase back pocket behind the
32
gatekeeper Tyr476 and interact with Cys481, compound 5 binds to the hinge region within the
ACS Paragon Plus Environment
4

Page 5 of 53
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
active site with the benzamide moiety reaching toward the pocket formed by Tyr551, Leu542
and Ser543 (termed as H3 pocket). Targeting the H3 pocket has been demonstrated to be
essential for obtaining selectivity for BTK versus the other Tec family members.33 Since this
structure demonstrated ample room to grow into the H3 pocket toward Tyr551, the phenyl amide
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
3
was augmented with a para tert-butyl group similar to CGI-1746. This modification led to
compound 6 (BTK IC = 142 ± 51 nM, N = 2), which exhibits more than 20-fold improvement
50
in potency over compound 5, though the ligand-lipophilicity efficiency (LLE = 1.8) remains low
and comparable to that of compound 5 (LLE = 2.2). The kinase selectivity of compound 6 was
profiled using the DiscoverX kinome panel and it showed modest selectivity (15/403 kinase >
8
0% inhibition @ 10 M, Supplementary Table S-2), consistent with the tert-butyl group
occupying the H3 pocket. An X-ray co-crystal structure of 6 was obtained to 1.6 Å resolution
which confirmed this placement of the tert-butyl group (Figure 3, Supplementary Table S-1).
H
N
H
H
N
N
O
O
O
N
N
N
N
H
N
H
N
H
N
N
N
6
5
(hit)
7
BTK IC50 = 142 nM
LLE 1.8
BTK IC50 = 3700 nM
LLE 2.2
BTK IC50 = 5 nM
LLE 2.8
Figure 2. Chemical structures of novel pyrrolopyrimidine BTK hit 5 and analogs 6 and 7.
ACS Paragon Plus Environment
5

Journal of Medicinal Chemistry
Page 6 of 53
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 3. An overlay of BTK co-crystal structures with compound 5 (blue, PDB 6VXQ) and
compound 6 (green, PDB 6W06) bound in the ATP active site. H3 residues Y551, S543 and
L542 are marked in orange and structural features, binding interactions, and residues are labeled.
The co-crystal structure of 6 with BTK suggests that 2-substitution on the central phenyl core
could reach towards the glycine rich P-loop and gain additional hydrophobic interactions
(Supplementary Figure S-1). Thus, the next effort was focused on modifying the 2-position of
the central phenyl ring to improve potency and LLE. As summarized in Table 1, adding a methyl
group yielded compound 7 (BTK IC = 5 ± 0.1 n   ) with a 30-fold increase in potency
5
0
and improved LLE (2.8 for 7 vs. 1.8 for 6). Introducing larger hydrophobic substituents such as
the ethyl or trifluoromethyl (8 and 9, BTK IC50 = 15 n and 94 n, respectively) was
detrimental to the biochemical potency. Introducing polar moieties such as the primary amide in
1
0 (BTK IC = 3200 n) led to a loss of activity. This is consistent with the in-silico docking
5
0
ACS Paragon Plus Environment
6

Page 7 of 53
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
studies which suggested that small hydrophobic moieties were optimal for sitting in this pocket
formed by lipophilic amino acid side chains.
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
Table 1. Characteristics of analogs of 6 featuring 2-substitution on the central phenyl core.
H
N
R
O
N
N
H
N
a
Log D b
4.95
Compound
R
BTK IC (n)
LLE
1.8
5
0
6
7
8
9
H
142
5
Me
Et
5.18
2.8
15
2.35
1.8
CF
3
94
2.43
1.1
1
0
CONH
2
3200
1.99
2.1
a IC50 values were determined using non-phosphorylated BTK protein in a FRET assay in duplicates.
b Measured using Analiza ELogD assay.
Compound 7 showed higher BTK potency and improved off target kinome selectivity (4/403
kinase > 80% @ 10 µM, FLT3, KIT, PDGFRB and YSK4, with K = 82, 150, 53 and 160 n
d
respectively) as compared to compound 6. Despite the high lipophilicity with LogD = 5.18,
compound 7 exhibited cellular activity (in vitro whole blood (WB) pBTK IC = 7.8 ± 0.2 µM, N
5
0
ACS Paragon Plus Environment
7

Journal of Medicinal Chemistry
Page 8 of 53
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
=
2). Thus, compound 7 was selected to advance to the next stage to optimize drug-like
properties.
Hit to lead optimization. With the achievement of good biochemical potency and kinase
selectivity by compound 7, the next challenge for this series was to improve the poor drug-like
properties represented by high LogD (5.18), poor aqueous solubility (kinetic solubility < 1 µg/ml
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
at pH 6.8), weak WB cellular potency (IC = 7.8 µM), and high in vivo rat clearance (CL %Q >
50
h
9
0). To address these issues, our primary goals were to reduce the lipophilicity and improve the
LLE of this chemical series. A series of compounds were synthesized in parallel to replace the
distal 4-tert-butyl phenyl amide with various substituted 5- and 6-membered heteroaryl moieties.
This effort led to the discovery of the isothiazole 11 which demonstrated comparable
biochemical potency (BTK IC = 3 n) and improved LLE (3.9) when compared to compound
5
0
7
(Table 2). Furthermore, compound 11 showed improved solubility (10 µg/mL at pH 6.8) and
moderate in vivo clearance in rat IV PK (CL %Q = 43). Although an initial disconnect between
h
in vivo and in vitro clearance rates (IVIVC) of compound 11 (in vitro RLM CL %Q > 90) was
h
3
4
observed, correcting the in vitro clearance with plasma unbound fraction (fu) (RLM CL %Q =
h
33), brought these values more in line. To investigate the contribution of the hinge binding
moiety on the potency and ADME properties, the pyrrolopyrimidine hinge binder was replaced
with either the 4-aminopyrimidine 12 or the 2-aminopyrimidine 13. Both changes brought a
modest improvement in the in vitro rat metabolic stability (RLM CL %Q = 72 and 65,
h
respectively) and exhibited low clearance when dosed in the rat IV PK study (12 CL %Q = 6
h
and 13 CL %Q = 11). The in vivo clearance values correlated with the fu-corrected in vitro rat
h
clearance rates (12 RLM CL %Q = 10 and 13 RLM CL %Q = 4).
h
h
ACS Paragon Plus Environment
8

Page 9 of 53
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Table 2. Characteristics of BTK inhibitors 11, 12 and 13 featuring different hinge binders.
N
H
N
S
O
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
R
Compound
11
12
13
R
N
N
N
N
N
NH2
N
NH2
N
H
BTK IC50 (nM) a
LLE
3
9
6
3.9
4.1
4.4
4.04
13
Log D a
4.35
10
3.87
Solubility pH 6.8 (µg/mL) b
Human PPB fu % c
6
<1
3
<1
HLM/RLM (CL %Q
h
) d
64/91
43
53/72
6
56/65
11
Rat PK (CL %Q
h
) e
a See notes from Table 1.
b Measured using Analiza miniaturized shake flask.
c Unbound fraction was measured at a concentration of 2 μM for human plasma via rapid equilibrium dialysis (RED).
d Metabolic stability in human and rat liver microsome (%Qh hepatic blood flow (portal vein plus hepatic artery)).
e IV dosing in male SD rats, measured in triplicates.
ACS Paragon Plus Environment
9

Journal of Medicinal Chemistry
Page 10 of 53
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
Lead optimization to candidate. Having demonstrated the ability to attenuate the in vivo
clearance while maintaining good biochemical potency and LLE, our efforts switched to further
optimizing the biochemical potency and physicochemical properties of compound 13 in order to
improve the WB cell activity (WB pBTK IC = 1.8 µM). The crystal structure of compound 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
5
0
(Supplementary Figure S-2) indicated that substituents adjacent to the hinge and extending
toward the solvent front could interact with the surrounding residues (Leu408, Tyr476, Ala478
and Gly480) and potentially improve the biochemical potency. Introducing various alkyl
substitutions on the amino group including Et, Me, i-Pr, t-Bu or cyclic alkyls decreased
biochemical potency, but switching to aromatic substituents led to phenyl analog 14 which
exhibited good biochemical potency (BTK IC = 1.8 n) but did not show cellular activity
5
0
likely due to the high lipophilicity (LogD = 3.44) and poor solubility (kinetic solubility < 1
µg/ml at pH 6.8) (Table 3). Introducing the 2-pyridine moiety 15 resulted in a loss in potency
(
BTK IC = 32 n). Including the 4-pyridine compound 16 increased biochemical potency
5
0
(
BTK IC = 0.4 n), however, like compound 14, this improvement in biochemical potency
5
0
was not paralleled by an improvement in the WB cell potency (WB pBTK IC = 5.3 µM). To
5
0
further probe the SAR within this region of the protein, a series of 5-membered heteroaryl
analogs were synthesized. Modulating the lipophilicity (LogD) by introducing substituted
thiadiazole (17), isoxazole (18), and triazole (19) moieties resulted in compounds that were
active in the biochemical assay. Unfortunately, even the most potent analog 18 (BTK IC = 1.7
5
0
n) did not exhibit WB cellular activity. The most interesting SAR emerged with the synthesis
of the 1,3 disubstituted pyrazole (20, BTK IC = 2.1 n), which exhibited comparable activity
5
0
in the WB cellular assay (WB pBTK IC = 4.7 µM) to pyridyl analog 16, but with increased
5
0
LogD (3.47). Appreciating the impact of the nitrogen atom within pyridyl-containing analogs
ACS Paragon Plus Environment
1
0

Page 11 of 53
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
had on binding affinity, the 1,4-disubstituted pyrazole 21 was synthesized to mimic the
placement of the pyridyl nitrogen. This adjustment provided a 10-fold improvement on both the
biochemical potency and cellular activity (BTK IC = 0.2 ± 0.12 nM (N = 5), human WB pBTK
5
0
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
IC = 0.43 ± 0.25 µM (N = 17)) and improved LLE (4.9) with higher LogD (4.76) (Table 3).
5
0
Compound 21 exhibited low in vivo clearance and modest oral bioavailability (CL %Q = 6, IV 1
h
mg/kg, and %F = 24 in CMC/Tween) when dosed in rat. However, to further evaluate this
chemical series in preclinical efficacy models, it was necessary to address the physicochemical
properties including high LogD (4.76), low solubility (kinetic pH 6.8 <10 µg/mL) and high
plasma protein binding (>99.5%).
Table 3. Characteristics of analogs of 13 featuring amino substitution off the pyrimidine hinge
binder.
N
H
N
S
O
N
R
N
N
H
Compound
R
BTK IC50 Log D a
(nM)
Solubility
pH 6.8
µg/mL
LLE
a
a
1
4
1.8
3.44
<0.1
2.2
ACS Paragon Plus Environment
1
1

Journal of Medicinal Chemistry
Page 12 of 53
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
5
32
0.4
40
1.7
10
2.17
2.94
2.07
2.97
2.79
0.23
<0.1
ND
2.4
4.2
2.8
3.7
4.1
N
16
N
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
1
1
7
8
9
N
N
S
N
0.12
ND
O
N
N
N
N
20
2.1
3.47
4.76
<0.1
2.7
4.0
4.9
N
2
1
N
0. 2
N
a See notes from Table 2.
ND - not determined.
With the goal of reducing the lipophilicity of the lead series while maintaining the desirable
biological and ADME properties of compound 21, our attention moved to exploring the SAR
around the amide moiety that occupies the H3 pocket. The thiadiazole analog 22 showed
improved solubility but a notable loss in biochemical potency (0.2 n vs 2 n, Table 4), and a
similar trend was observed with a few other types of heterocycles (data not shown). In-silico
3
modelling studies suggested that sp moieties such as 5-membered aliphatic groups would be
ACS Paragon Plus Environment
1
2

Page 13 of 53
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
tolerated in place of the heteroaryl moiety, and that might improve the physicochemical
properties of the series, including increasing solubility. To probe this hypothesis, the cyclopentyl
analog 23 was synthesized. Compound 23 was less potent (BTK IC = 2 n tested as a mixture
5
0
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
of 4 isomers) than 21 in the biochemical assay, however, its activity nevertheless validated the
rationale that non-planar moieties might be tolerated in this position. Furthermore, the
corresponding urea compound 24 showed significant improvement in both biochemical and
cellular potency, as well as LLE (BTK IC = 0.2 n, human WB pBTK IC = 0.35 µM, LLE =
5
0
50
5
), though no improvement in solubility was observed. Additional attempts to enhance the
solubility by reducing the lipophilicity led to the replacement of the pyrrolidine with the 3-
substituted azetidine (25). This modification maintained good biological activity (BTK IC =
5
0
0
.4 n, human WB pBTK IC = 0.65 ± 0.79 µM (N = 3)) with lower LogD (3.34) but still no
50
improvement in solubility was observed. Introducing an oxygen linkage at the 3-position of
azetidine led to the discovery of compound 1 (BTK IC = 1 ± 1.1 n N = 5). This simple
5
0
modification reduced lipophilicity (LogD = 2.32) substantially, improved the solubility, and
reduced the plasma protein binding which all contributed to improving the WB cell potency
(
human WB pBTK IC = 0.12 ± 0.15 µM, N = 51).
50
Table 4. Characteristics of analogs of 21 featuring modifications to the distal H3 pocket region.
ACS Paragon Plus Environment
1
3

Journal of Medicinal Chemistry
Page 14 of 53
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
H
N
N
R
O
N
N
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
N
N
H
Comp
ound
R
BTK
IC50
(nM) a
LLE Log
WB
Unbound Hum Corr.
WB
IV
Rat
% CL 6.8
Q
h
Solubility
pH 1.5,
a
D
pBTK
IC (µM) pBTK
an
PPB
RLM,
HLM CL
50
a
a
a
(µg/mL) a
b
IC (nM) fu%
%Q
h
5
0
c
N
21
22
23
0. 2
4.9 4.76
4.8 3.01
3.5 4.58
5.0 3.72
5.0 3.34
0.43
1.4
1.3
9.8
7
0.3
0.7
0.07
0.2
0.9
1, 2
6
<10, <10
S
N
N
2
1, 3
18
31
4
>50, <10
<10, <10
<10, <10
<10, <10
S
2
>10
0.35
0.65
3, 14
1, 19
1, 8
2
4
N
0. 2
0.4
0.7
5.9
25
1
17
N
N
O
1
5.0 2.32
0.12
7.1
5.9
3, 9
1
>50, >10
a See notes from Table 2.
b Measured in duplicates except those stated in the text.
c Unbound WB pBTK IC50 = WB pBTK IC50 x Human PPB fu.
ACS Paragon Plus Environment
1
4

Page 15 of 53
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
To sum up, focusing the medicinal chemistry strategy on optimizing the physicochemical
properties of the lead series resulted in the discovery of compound 1. Compound 1 exhibited
improvement in in vitro biological potency, as well as good drug-like properties (LLE = 5) which
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
resulted in low in vivo clearance (CL %Q = 6) and moderate oral bioavailability (%F = 48)
h
when dosed in rats.
Figure 4. X-ray co-crystal structure of compound 1 in BTK (PDB 6W07). The active site surface
is shown semi-transparent, and H3 residues Y551, S543 and L542 are marked in orange.
A co-crystal structure (Figure 4) of compound 1 bound in the active site of BTK confirmed that
the 2-aminopyridimine moiety was acting as a dual hydrogen-bond acceptor/donor with residues
in the hinge region of the protein. The methyl substituent on the phenyl core was positioned
under the P-loop, and the benzyl urea linker oriented the azetidine into the H3 pocket which
ACS Paragon Plus Environment
1
5

Journal of Medicinal Chemistry
Page 16 of 53
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
sequestered Tyr551 in an inactive conformation. Similar to the published literature whereby
binding in the H3 pocket of the active site conferred excellent kinome selectivity,28a, 33 1 hit only
10 kinases with >70% inhibition at 1 µM when screened against a panel of 395 kinases at
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
DiscoverX (Supplementary Table S-3). Affinity measurement (K ) were subsequently
d
3
5
determined for all 10 kinases (Table 5), which led to the conclusion that compound 1 yields >
00-fold selectivity for BTK versus all the other kinases tested including the structurally similar
Tec family members (BMX, ITK, TEC, TXK).
4
Table 5. K values for kinases exhibiting > 70% affinity toward compound 1 relative to control at
d
1
M.
Kinase name
K
d
(nM)
Selectivity fold
over BTK
Adaptor-associated protein Kinase
1
430
1433
(AAK1)
BMP2-Inducible Kinase (BMP2K)
BTK
1100
0.3
3666
-
Cyclin-dependent kinase-like 2 (CDKL2)
Janus Kinase 2 (JAK2)
420
460
1400
1533
>10000
Phosphatidylinositol-4-phosphare 5-Kinase >10000
type 1 Alpha (PIP5K1A)
RIO Kinase 1 (RIOK1)
1200
2400
880
4000
8000
2933
433
RIO Kinase 2 (RIOK2)
Serine-threonine kinase (STK1)
Mitogen-activated protein kinase kinase 130
kinase 19 (YSK4)
ACS Paragon Plus Environment
1
6

Page 17 of 53
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 5. Phenotypic BioSeek study data of compound 1 at 1.1, 3.3, 10 and 30 µM. The gray
area shows the 95% confidence area of each biomarker response in the absence of drug. Data
points that fall outside of the gray area signify inhibition (below) or augmentation (above) of the
biomarker response at the indicated concentration of compound 1.
To further evaluate the selectivity of compound 1, it was tested in vitro at 4 concentrations (30,
1
0, 3.3, and 1.1 µM) in the BioMAP system developed by BioSeek/DiscoverX. This platform
comprises of 12 human primary cell-based assays that enable the assessment of both on- and off-
target activity of a compound in a cellular system using pathway- and disease-relevant
3
6
biomarkers. Compound 1 exhibited dose-dependent activity which was restricted to the culture
systems that included hematopoietic cells that express BTK protein (LPS, SAg, BT and Mphg).
The predominant activity was observed in primary B cells and PBMCs (BT culture) and
impacted biomarkers such as secreted IgG and sTNF-alpha (Figure 5). Furthermore, at 1.1 M,
the only culture system impacted by compound 1 was BT culture stimulated with anti-IgM and a
3
7
sub-threshold amount of anti-TCR. Compared to the BioSeek literature data of a nonselective
BTK inhibitor Ibrutinib (Supplementary Table S-3) at the same concentration, which showed
ACS Paragon Plus Environment
1
7

Journal of Medicinal Chemistry
Page 18 of 53
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
activity in every cellular system, the BioSeek profile of compound 1 at 1.1 M is consistent with
its greater kinase selectivity.
In addition, compound 1 inhibited BCR mediated PLCγ2 phosphorylation in Ramos B cells (IC50
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
=
0.4 µM), anti-IgD induced and anti-IgM BCR-induced B cell activation in human PBMCs
(
IC = 0.11 µM and 0.21 µM, respectively) consistent with the activity on the BT culture system
5
0
in the BioMAP study.
Drug metabolism & pharmacokinetics. Compound 1 exhibited low volume of distribution, low
to moderate clearance in rat, dog and cynomolgus monkey, moderate oral bioavailability (26-
32% in dogs and cynomolgus monkeys and 48% in rats), and a short half-life across species
(
Table 6). In in vitro ADME screening studies (Table 7), compound 1 exhibited high
permeability and moderate efflux in the Caco-2 cell system. Tests for protein binding revealed a
range across all preclinical species and humans, with fu (1.28%) < fu
rat
cyno
(3.62%) < fuhuman
(5.95%) < fudog (11.3%). When evaluated for potential major human CYP-mediated drug-drug
interactions, compound 1 did not inhibit any of the major CYP isoforms (1A2, 2C9, 2C19, 2D6,
and 3A4) in human liver microsomes (IC > 25 M) and did not result in CYP (1A2, 2B6, and
50
3
A4) induction in the sandwich cultured human hepatocyte assay at concentrations up to 50 M.
Compound 1 was stable in the plasma of mouse, rat, beagle dog, cynomolgus monkey, and
human (>95% parent compound remaining after 6 hours of incubation) and was not sequestered
into the red blood cells of those species.
Compound 1 was extensively metabolized in vitro in liver microsomes and hepatocytes across all
preclinical species (rat, dog, monkey) and humans. The primary biotransformation pathways of
compound 1 in human liver microsomes were due to O-deisopropylation, oxidation at the
ACS Paragon Plus Environment
1
8

Page 19 of 53
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
pyrazole moiety, and N-demethylation. Oxidation of the isopropyl group appeared to be a minor
pathway, with additional minor metabolites formed by combinations of these primary pathways.
Multiple CYPs were responsible for the oxidative metabolism with major (66%) contribution by
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
-1
3
A4 (K = 50 µM, k
= 0.08 h ). The metabolic pathways in liver microsomes and hepatocytes
I
inact
were similar across all four species with only quantitative differences detected. All metabolites
detected in human liver microsomes and hepatocytes were also detected in more than one
preclinical species, with no formation of unique human metabolites or any GSH or CN adducts.
Overall, compound 1 showed acceptable ADME properties (Table 7) to warrant further
development.
Table 6. Pharmacokinetics characteristics of compound 1 in preclinical species.
PK
PO, 5 mpk a
AUC (ng*hr/mL)
11907
IV, 1 mpk b
C
max (ng/mL) %F CL (mL/min/kg)
t1/2 (h)
Vss(L/kg) %Q
h
Rat
2920
652
48
26
32
3
1.2
2.1
0.9
0.2
1.2
0.7
6
Dog
Cyno
1938
11
11
37
26
2346
282
a Vehicle: 0.1% CMC/ 0.2% Tween-80.
b Vehicle: 40% NMP solution for rat, and 70:30 PEG-400:H
2
O solution for dog and cyno.
Table 7. Detailed in vitro profile of compound 1.
Assay
Result
1 ± 1.1 (N = 5)
BTK IC50 (nM) a
d
BTK K (nM)
0
0
.3 ± 0.2 (N = 2)
.12 ± 0.15 (N = 51)
Human WB pBTK IC50 (µM) a
KINOMEscan @ 1 µM (>90% Control) b
1/394
d
YSK4 (K = 130 nM)
ACS Paragon Plus Environment
1
9

Journal of Medicinal Chemistry
Page 20 of 53
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
Panlabs @ 10 µM (> 50% inhibition) b
2/68
2
(IC50 = 4.2 µM), H (IC50 = 26.5 µM)
A
3
CYP inhibition (1A2, 2C9, 2C19, 2D6, and 3A4) IC50 (µM) b
all isoforms >25
K
I
= 50 µM, kinact = 0.08 h-1
CYP 3A4 HLM time-dependent inhibition b
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
LM (rat/dog/cyno/human) CL %Q
hERG inhibition IC50 (µM) c
HepG2 Cytotoxicity IC50 (µM) b
Caco2 (PA-B/Efflux ratio) b
PPB fu% (human) a
h
2/8/26/7
>50
>20
15.4/4.1
5.95
Aq. solubility (g/mL, pH7/2) d
1/>100
Measured LogD7.4 (shake flask method)
a See notes from Table 4.
2.92 ± 0.17 (N = 2)
b. See supporting information.
c. In the automated patch-clamp assay.
d. In simulated intestinal fluid (pH7) and simulated gastric fluid (pH2).
Chemistry. Compound 1 was synthesized in seven steps starting with the commercially
available 4-bromo-2-methylbenzonitrile 26 as shown in Scheme 1. Reduction of the nitrile
moiety followed by Boc protection gave arylbromide 28. To facilitate the aryl coupling to the
pyrimidine moiety, it was necessary to first form the boronic ester 29, followed by Suzuki
3
8
coupling to yield compound 30. The 3-amino pyrazole moiety was installed via a palladium
39
facilitated amination and the Boc group was then removed upon treatment with HCl to produce
the benzyl amine hydrochloride salt 32. Finally, a CDI-mediated urea formation provided
compound 1. With a viable synthetic route in hand to support the synthesis for preclinical
toxicology studies, it was deemed necessary to identify a crystalline form for further
development. After an extensive salt screening, the hemi-adipate salt was identified as a suitable
crystalline form with the necessary physicochemical properties for use in clinic.
ACS Paragon Plus Environment
2
0

Page 21 of 53
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
NHBoc
NH2
NHBoc
CN
Br
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
b
c
B
O
O
Br
Br
28
26
27
29
NHBoc
NHBoc
Cl
N
NH2
N
N
N
(
Cl
1.2 eq)
(1.3 eq)
N
N
e
N
d
N
N
Cl
N
N
H
31
30
NH2
O
f
.HCl
HClHN
BIIB068 (1)
N
N
g
N
N
N
H
3
2
o
Scheme 1. Synthesis of BIIB068 (1). Reagents and conditions: (a) 3eq BH , THF, 0-80 C, 16hr,
3
9
3
0%; (b) 1.1 eq Boc O, 3 eq NEt , THF, rt, 1hr, 95%; (c) 1.2 eq PinB-BPin, 0.1 eq Pd(dppf)Cl ,
2 3 2
o
eq KOAc, 1,4-dioxane, 100 C, 2hr, 69%; (d) 0.1 eq Pd(dppf)Cl , 2 eq K CO , 1,4-dioxane/H O
2
2
3
2
o
o
(4:1), 90 C, 2hr, 80%; (e) 0.1 eq Pd (dba) , 0.2 eq S-Phos, 2 eq Cs CO , 1,4-dioxane, 120 C, 2hr,
2
3
2
3
6
3%; (f) HCl/MeOH, rt, 6hr, 90%; (g) CDI, DIPEA, DMF, 30%.
ACS Paragon Plus Environment
2
1

Journal of Medicinal Chemistry
Page 22 of 53
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
Efficacy and pharmacodynamic measures. BIIB068 (compound 1) inhibits B cell, neutrophil
and monocyte activation in vitro, but not T cell responses (Table 8). In an in vitro assay
measuring total phosphorylated BTK (pBTK) in mouse WB, BIIB068 shows IC = 0.17 µM.
5
0
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 further biological profiling, BIIB068 demonstrated dampening of BCR-mediated B cell
activation, as measured by CD69 expression after PBMC simulation with anti-IgD (CD69 IC =
5
0
0
.111 µM). After factoring protein binding, this value correlates well with the compound’s
ability to inhibit phosphorylation of BTK (IC = 1 nM) in the biochemical assay. Similarly,
5
0
BIIB068 was shown to inhibit FcγR-mediated ROS production when added to human neutrophils
from healthy donors that were activated with immune complexes (IC) from SLE patients (SLE
IC) (IC = 0.054 µM). Blocking receptor access with antibodies to FcγRII (CD32) prevented
5
0
SLE IC-mediated ROS production by neutrophils, demonstrating that the BIIB068 functional
mechanism is dependent on FcγR signaling. Furthermore, to show that inhibition of BTK with
BIIB068 selectively blocks B cell activation via BCR signaling, treatment of PBMCs with other
receptor ligands including anti-CD40 and IL-4 in the presence of BIIB068 exhibited an IC =
5
0
+
>
10 µM for induction of CD69 expression on CD19 B cells. Collectively, the biological profile
observed for BIIB068 was consistent with a molecule that selectively targets BTK.
Table 8. In vitro biological profile of BIIB068 (1).
a
Assay (measured output, stimulus, cell type)
B cell activation (CD69, anti-IgD/BCR, PBMC)
ROS generation (ROS, SLE IC/FcR, neutrophils)
Monocyte (PB IgG, IL-6, IL-1b, TNF/FcR, PBMC)
IC50 ± SD (µM)
0.111 ± 0.087 (N = 10)
0.054 ± 0.057 (N = 4)
0.063 ± 0.034 (N = 4)
ACS Paragon Plus Environment
2
2

Page 23 of 53
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
pBTK inhibition (pTyr, n/a, human WB)
pBTK inhibition (pTyr, n/a, mouse WB)
0.12 ± 0.15 (N = 51)
0.17 ± 0.100 (N = 5)
B cell activation (CD69, IL-4/anti-CD40/CD40, PBMC)
>10
>10
T cell activation (CD69, anti-CD3/CD28/TCR, PBMC)
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 See supporting information for assay conditions.
To understand the translatability from in vitro biological potency to in vivo efficacy, BIIB068
was tested in a thymus-independent type 2 (TI-2) antigen mouse model which had previously
been reported to be BTK dependent.40
In contrast to thymus dependent (TD) antigens, thymus independent (TI) antigens induce a
productive antibody response in neonatally thymectomized mice or nude mice, which also lack a
thymus. TI antigens can be further differentiated based on whether they can (TI-1) or cannot (TI-
2) generate an antibody response in xid mice. These three classes of model antigens differ in how
they mechanistically induce an antibody response in vivo, with TI-2 antigens relying heavily on
BCR signaling. BTK mutation was later shown to be the genetic cause responsible for the xid
4
1
phenotype . In retrospect, TI-2 antigens were defined as those antigens that require BTK
expression to generate a productive antibody response in vivo, thus making it a well-suited in
vivo efficacy model to test compounds that target BTK function.42
To directly evaluate the effects of BIIB068 on BTK-dependent B cell activation in vivo, the
antigen-specific antibody response was measured from mice dosed orally with BIIB068 and
immunized with a model TI-2 antigen (4-hydroxy-3-nitrophenyl) acetyl (NP) -Ficoll. Twice
daily (BID) dosing of BIIB068 continued for 10 consecutive days. Serum, plasma, and WB were
collected on day 10 to evaluate the antigen-specific antibody responses, plasma concentration of
ACS Paragon Plus Environment
2
3

Journal of Medicinal Chemistry
Page 24 of 53
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
BIIB068 and target engagement using the WB PD assay. NP-specific IgM antibody titers were
compared to vehicle control animals. Substantial reductions of these antibody titers were
observed in mice treated BID with 10-50 mg/kg/dose of BIIB068 (Figure 7). Specifically, anti-
NP IgM antibody titers were reduced by 93%, 90%, 63%, 22% and -23% compared to vehicle
control treated animals for BID BIIB068 doses of 50, 30, 10, 3 and 1 mg/kg/dose, respectively.
The calculated ED50 of BIIB068 for IgM antibody titer was 4.4 mg/kg/dose. To further
interrogate the pharmacokinetics of BIIB068 and degree of target engagement achieved in the
TI-2 efficacy model, plasma and WB samples from each of the dose or vehicle arms were
collected to determine exposures and degree of target engagement using pBTK assay concurrent
with serum collection.
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 pre-last dose group was used to determine the steady-state trough concentration of BIIB068
and the corresponding degree of target engagement in each dose arm. The samples taken at 1 and
4 hours after the final dose were used to determine the concentration and target engagement of
BIIB068 at or near the tmax in each dose arm. The effects on the NP-specific antibody response
are negligible during the maximum of 4-hour time difference between samples. Figure 6 shows
the average inhibition of target engagement (% inhibition of pBTK figure 6B) or total exposure
(nM, Figure 6C) in each dose group. BTK target engagement remains high, but dose-responsive,
for at least 4 hours after PO dosing in all dose groups. In contrast, at steady-state trough levels of
BIIB068 seen in the pre-last dose group, target-engagement is low or absent in all but in the
highest dose arm, 50 mg/kg/dose. The duration of target engagement was dependent on the dose.
The in vivo pBTK IC for BIIB068 is 0.13 µM (Supplementary Figure S-3), which is similar to
5
0
the in vitro pBTK IC of 0.17 µM in mouse whole blood. The mice dosed with 30 mg/kg/dose
5
0
show ~90% efficacy, indicating that full target coverage throughout the entire dosing period is
ACS Paragon Plus Environment
2
4

Page 25 of 53
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
not necessary to drive high level of efficacy, at least in the TI-2 antibody response model.
BIIB068 demonstrated good PK/PD correlation with in vivo EC = 0.16 µM based on antibody
5
0
response, similar to in vivo WB pBTK potency in mice.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 6. PK and PD analysis of BIIB068 (1) in the TI-2 efficacy model. A). workflow of the TI-
2 mice efficacy model. B). pBTK % inhibition at 1 hr, 4 hr and pre-last dose. C). Total drug
ACS Paragon Plus Environment
2
5

Journal of Medicinal Chemistry
Page 26 of 53
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
concentration at 1 hr, 4 hr and pre-last dose. Figure shows mean and standard deviation of
triplicate determination in one experiment.
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. Dose response trial of BIIB068 (1)-mediated inhibition of the antigen-specific antibody
response to a TI-2 antigen. Figure shows mean and standard deviation of triplicate determination
in one experiment.
Preclinical safety. Toxicity profiling for BIIB068 was conducted in rodent (Sprague-Dawley
rat) and non-rodent (cynomologus monkey) species for up to 13 weeks in GLP toxicity studies.
The cynomologus monkey was chosen as the non-rodent species based on the similarities in
metabolic profiles for BIIB068 in both human and monkey hepatocytes and microsomes. In both
ACS Paragon Plus Environment
2
6

Page 27 of 53
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
the rat and monkey 13 weeks studies, BIIB068 was demonstrated to be safe and well tolerated.
No compound related adverse effects in clinical signs, body weight, food consumption, ECG,
clinical pathology, organ weight or gross pathology/histopathology were observed. BIIB068-
related effects in rats were limited to non-adverse microscopic changes in the pancreas at ≥ 15
mg/kg/day.43 Thus, the No Observed Adverse Effect Level (NOAEL) was determined at 50
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
mg/kg/day for rat and 100 mg/kg/day for monkey which corresponded to a daily rat AUC 376
inf

g*hr/mL and monkey AUC 129 g*hr/mL.
inf
BIIB068 exhibited no adverse effects in the single dose CNS and respiratory studies in rats nor in
a single dose cardiovascular study in monkeys. BIIB068 demonstrates low genotoxic potential.
BIIB068 was negative for clastogenic and mutagenic responses in vitro with and without
metabolic activation (Aroclor-induced rat liver S9). BIIB068 was negative in the in vivo
micronucleus assay in Sprague Dawley rats, where it showed no clastogenic activity and/or
disruption of the mitotic apparatus in the bone marrow, and negative for micronuclei in rats.
Clinical study. Key human PK parameters were simulated utilizing both standard scaling IVIVE
approaches extrapolated from the preclinical species’ observed PK data and Simcyp® simulator
(version 13) PBPK modeling. Clinical doses required to maintain a target level of EC for pBTK
7
5
44
inhibition (based on TI-2 efficacy study) in human WB for approximately > 10 hours coverage
in a 12 hour dosing interval, was in the range of 50 mg/dose BID (100 mg/day) - 125 mg/dose
BID (250 mg/day) daily dose. The phase I SAD study45 was designed with goal of establishing
the safety and tolerability for BIIB068. BIIB068 was dosed orally as a tablet in healthy
volunteers at a starting dose of 10 mg which was chosen to afford > 10-fold margin to the
NOAEL determined from the GLP preclinical toxicity studies. Dose escalation continued
through the 120 mg cohort which based on the human PK had achieved exposures equivalent to
ACS Paragon Plus Environment
2
7

Journal of Medicinal Chemistry
Page 28 of 53
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 NOAEL exposures (plasma total AUC) identified from the cyno GLP toxicity studies.
BIIB068 was well tolerated with peak plasma concentration (C ) occurring at ~0.9-2 hours
max
after dosing and median t1/2 ranges from 6-9 hours across all cohorts (Figure 8). BIIB068 was
effective at inhibiting BTK phosphorylation and B-cell activation dose-dependently. At single
doses ≥30 mg BIIB068, BTK phosphorylation was inhibited ≥70% for up to 12 hours post-
dosing. Inhibition of BTK phosphorylation was ≥90% from approximately 1 hour through 8
hours post-dosing at 60 and 120 mg BIIB068. Peak inhibition occurred at approximately the
same time as peak BIIB068 plasma concentration. Peak inhibition of B-cell activation (60%)
occurred at 1 to 2 hours post-dosing (same time as peak BIIB068 plasma concentration) at 120
mg BIIB068 then gradually decreased by 8 hours post-dosing. Overall, BIIB068 was well
tolerated and effective at inhibiting BTK phosphorylation in healthy subjects. BIIB068 also
inhibited B-cell activation, although the inhibition was not as robust as inhibition of BTK
phosphorylation. The reason why inhibition of B-cell activation was less robust than inhibition of
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
32
BTK phosphorylation is unknown and may be a future area of investigation. More details will
be disclosed in due course.
ACS Paragon Plus Environment
2
8

Page 29 of 53
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
A.
ACS Paragon Plus Environment
2
9