Discovery of Potent, Selective, and Peripherally Restricted Pan-Trk Kinase Inhibitors for the Treatment of Pain

Article abstract of DOI:10.1021/acs.jmedchem.8b00633

Hormones of the neurotrophin family, nerve growth factor (NGF), brain derived neurotrophic factor (BDNF), neurotrophin 3 (NT3), and neurotrophin 4 (NT4), are known to activate the family of Tropomyosin receptor kinases (TrkA, TrkB, and TrkC). Moreover, in

Full text of DOI:10.1021/acs.jmedchem.8b00633

Subscriber access provided by University of Sussex Library
Article
Discovery of Potent, Selective and Peripherally Restricted
Pan-Trk Kinase Inhibitors for the Treatment of Pain
Sharan K Bagal, Mark Andrews, Bruce M. Bechle, Jianwei Bian, James Bilsland, David C Blakemore, John
Braganza, Peter J. Bungay, Matthew S. Corbett, Ciarán N Cronin, Jingrong Jean Cui, Rebecca Dias, Neil
J Flanagan, Samantha E Greasley, Rachel Grimley, Kim James, Eric Johnson, Linda Kitching, Michelle
L Kraus, Indrawan McAlpine, Asako Nagata, Sacha Ninkovic, Kiyoyuki Omoto, Stephanie Scales, Sarah
E. Skerratt, Jianmin Sun, Michelle Tran-Dubé, Gareth J. Waldron, Fen Wang, and Joseph S Warmus
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.8b00633 • Publication Date (Web): 26 Jun 2018
Downloaded from http://pubs.acs.org on June 26, 2018
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 74
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 74
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 Potent, Selective and Peripherally
Restricted Pan-Trk Kinase Inhibitors for the
Treatment of Pain
0
1
2
3
4
5
6
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*
d
e
e
c
Sharan K. Bagal, Mark Andrews, Bruce M. Bechle Jianwei Bian, James Bilsland, David C.
a
f
b
e
f
Blakemore, John F. Braganza, Peter J. Bungay, Matthew S. Corbett, Ciaran N. Cronin,
f
c
c
f
c
Jingrong Jean Cui, Rebecca Dias, Neil J. Flanagan, Samantha E. Greasley, Rachel Grimley,
g
f
c
f
f
Kim James, Eric Johnson, Linda Kitching, Michelle L. Kraus, Indrawan McAlpine, Asako
f
f
a
f
a
Nagata, Sacha Ninkovic, Kiyoyuki Omoto, Stephanie Scales , Sarah E. Skerratt, Jianmin
e
f
c
f
e
Sun, Michelle Tran-Dubé, Gareth J. Waldron, Fen Wang, Joseph S. Warmus.
aWorldwide Medicinal Chemistry, Pfizer Worldwide R&D UK, The Portway Building, Granta
Park, Cambridge, CB21 6GS, UK
bPharmacokinetics, Dynamics & Metabolism, Pfizer Worldwide R&D UK, The Portway
Building, Granta Park, Cambridge CB21 6GS, UK
cPfizer Worldwide R&D UK, The Portway Building, Granta Park, Cambridge CB21 6GS, UK
dPfizer Worldwide R&D UK, Sandwich, Kent, CT13 9NJ, UK
ePfizer Worldwide R&D, Groton Laboratories, Eastern Point Road, Groton, Connecticut 06340,
USA
1
ACS Paragon Plus Environment

Page 3 of 74
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
fPfizer Worldwide R&D, La Jolla Laboratories, 10770 Science Center Drive, San Diego,
California, 92121, USA
gPeakdale Molecular, Discovery Park House, Ramsgate Road, Sandwich CT13 9ND, UK
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
KEYWORDS
Nerve growth factor receptor, NTRK1, NTRK2, NTRK3, NGF, TrkA, TrkB, TrkC, pan-Trk,
Tropomyosin Receptor Kinase, Kinase inhibitor
ABSTRACT
Hormones of the neurotrophin family: nerve growth factor (NGF), brain derived neurotrophic
factor (BDNF), neurotrophin 3 (NT3) and neurotrophin 4 (NT4) are known to activate the family
of Tropomyosin receptor kinases (TrkA, TrkB, TrkC). Moreover, inhibition of the TrkA kinase
pathway in pain has been clinically validated by the NGF antibody tanezumab leading to
significant interest in the development of small molecule inhibitors of TrkA. Furthermore, Trk
inhibitors having an acceptable safety profile will require minimal brain availability. Herein we
discuss the discovery of two potent, selective, peripherally restricted, efficacious and well-
tolerated series of pan-Trk inhibitors which successfully delivered three candidate quality
compounds 10b, 13b and 19. All three compounds are predicted to possess low metabolic
clearance in human that does not proceed via aldehyde oxidase-catalyzed reactions, thus
addressing the potential clearance prediction liability associated with our current pan-Trk
development candidate PF-06273340.
2
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 74
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
INTRODUCTION
Hormones of the neurotrophin family: nerve growth factor (NGF), brain derived neurotrophic
factor (BDNF), neurotrophin 3 (NT3) and neurotrophin 4 (NT4) are known to activate the family
0
1
2
3
4
5
6
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
of Tropomyosin receptor kinases (TrkA, TrkB, TrkC). In particular, NGF has been shown to have
2
a key role in the pathogenesis of inflammatory pain in preclinical studies. Moreover, clinical
studies in osteoarthritis, chronic low back pain and interstitial cystitis have demonstrated that NGF
neutralizing antibodies such as tanezumab demonstrate robust efficacy. 3 4 5 6 Since the NGF
pathway involves NGF binding to the TrkA kinase receptor, inhibition of TrkA has been clinically
validated as a target for pain leading to increased interest in the development of small molecule
inhibitors of TrkA.
The majority of small molecule kinase inhibitors tend to be ATP competitive Type I or Type II
binders, based on the conformation of the highly conserved aspartate-phenylalanine-glycine
(DFG) residues in the beginning of the activation loop in the kinase domain.7-8 Type I and Type II
binders of TrkA tend to exhibit pan-Trk (TrkA/B/C) inhibition rather than subtype selectivity
because TrkA, TrkB and TrkC kinases have no residue differences in the ATP binding site.9
Furthermore, Pan-Trk inhibitors require restriction to the peripheral compartment to avoid
undesirable side effects associated with Trk inhibition in the central nervous system (CNS). TrkA
is highly expressed in cholinergic neurons of the CNS with ablation of the TrkA gene in preclinical
species leading to dysfunction in cholinergic circuitry.10-12 TrkB is expressed throughout the body
and is involved in excitatory signaling, long-term potentiation and feeding behavior.13,14,15-16
17
Moreover, human genetic data has associated TrkB with obesity and development. TrkC is
widely expressed in neural and non-neural tissues and plays a role in the development and survival
of the sympathetic nervous system.18-19 Finally, clinical studies on the CNS penetrant pan-Trk/Tie2
3
ACS Paragon Plus Environment

Page 5 of 74
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
kinase inhibitor CE-245677 (advanced by Pfizer as an oncology agent), produced CNS side-effects
leading to the termination of Phase I multiple dose trials. The reported adverse events fully
resolved upon cessation of dosing and included cognitive problems, personality changes and sleep
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
20
disturbances. Given that the safety and toleration risks are associated with CNS Trk receptor
occupancy and efficacy of Trk kinase inhibitors is expected to be driven by target engagement in
peripheral neurons we can address the safety risks by restricting Trk inhibitors to the peripheral
compartment. This profile would require small molecule TrkA inhibitors that possess acceptable
oral bioavailability but which are also peripherally restricted.21-25 These properties can be designed
by operating within physicochemical space appropriate for absorption across the gastrointestinal
(GI) epithelium (e.g. molecular weight (MW) <500, polar surface area (PSA) <140, <10 rotatable
26
bonds), whilst combining activity as substrates for the efflux transporters P-glycoprotein (P-gp)
and Breast Cancer Resistance Protein (BCRP) that are expressed at the blood-brain barrier
(BBB).21-23, 27 However, targeting P-gp and BCRP, which are also present in the apical membrane
of intestinal epithelial cells, introduces the risk of restricting transport across the intestinal
epithelium, resulting in low and variable oral bioavailability. This risk, however, is low because
adequately soluble oral drugs at commonly prescribed doses are expected to saturate efflux
transporters, given that substrates of P-gp typically possess Km values in the range 1−100 μM. In
addition, the high concentration gradient across the intestinal epithelium will lead to a significant
driving force for flux of compound, particularly for drugs with high intrinsic permeability.28
We have previously reported the discovery of our pan-Trk development candidate PF-06273340
(
Figure 1).9, 29 The profile of metabolism of this compound involved a component that was
mediated by aldehyde oxidase (AO) reducing the confidence with which human metabolic
clearance (CL) could be predicted.30 Hence, contingency pan-Trk candidates were sought,
4
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 74
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
whereby AO-mediated metabolism was avoided in order to mitigate human pharmacokinetic (PK)
risk associated with PF-06273340. This work describes the discovery of three candidate quality
pan-Trk compounds that have high confidence in human PK prediction.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 1. Pfizer Pan-Trk development candidate PF-06273340 and hit molecule 1
RESULTS AND DISCUSSION
Hit identification and optimization. In order to find novel pan-Trk chemical matter the Pfizer
kinase selectivity dataset was analyzed. This dataset is a collection of compounds that have been
assessed for off target kinase selectivity at Invitrogen by screening in a kinase selectivity panel.
The Pfizer kinase selectivity panel included a TrkA assay in a Z’-LYTE biochemical assay format,
whereby benzamide hit 1 was identified as having 98% inhibition of TrkA at 1µM (Figure 1).
Furthermore, 1 was found to be a potent pan-Trk inhibitor in cell based assays with TrkA cell IC50
31
2
3nM and TrkB/C cell IC50 <30nM. Benzamide hit 1 exhibited a ligand efficiency (LE) 0.3 and
lipophilic efficiency (LipE) 3.9 (LipE= -Log(TrkA cell IC50)- LogD) both of which were
considered acceptable start points for optimization. Moreover, 1 had low metabolic turnover in
5
ACS Paragon Plus Environment

Page 7 of 74
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
human liver microsomes (HLM) and human hepatocytes (hHep) (intrinsic clearance (CLint) in
HLM 14L/min/mg protein, hHep <6µL/min/million cells) suggesting that this benzamide series
was capable of delivering compounds with low predicted human CL. In order to achieve restriction
from the CNS to the periphery and avoid unwanted CNS side effects P-gp and BCRP efflux
transporter substrates are required. P-gp and BCRP substrates are determined as having an efflux
ratio (ER) >2.5 in the current assay format to assess P-gp and BCRP efflux. Hit molecule 1 was a
P-gp substrate with an efflux ratio (ER) of 4 (BCRP not determined) indicating this benzamide
series had potential for the desired restriction to the peripheral compartment. The combination of
potency, LE, LipE, metabolic stability and P-gp efflux suggested that 1 was an exciting hit that
warranted further investment.
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
The human Ether-à-go-go-Related Gene (hERG) fluorescence polarization (FP) competition assay
and patch express assay (PX) have been shown to be predictive of compound QT prolongation
3
2
effects via hERG blockade. The hERG FP assay can also be run in a high throughput 384-well
plate manner.32 Hit molecule 1 was shown to inhibit hERG in the hERG FP assay (hERG FP IC50
3
.8µM) indicating there was some cardiovascular risk, which may have been associated with the
3
3
relatively high LogD of 1 (LogD 3.7). Modification of urea 1 to an amide 2 improved potency
and lowered LogD moderately (TrkA cell IC50 8.7nM, LogD 3.5) such that LipE improved by 0.7
units to 4.6, although hERG FP IC50 did not improve significantly (Figure 2). Additional
improvements in LipE were made by changing the N-methyl pyrazole into the isomeric N-methyl
imidazole 3 (LipE 5.3) through decreasing LogD further (LogD 2.9) whilst maintaining potency
(Figure 2). Furthermore, the reduction in LogD led to a concomitant decline in hERG liability
(hERG FP IC50 18µM). Ligand 3 was also assessed in the hERG PX assay which suggested 3 was
>1300-fold selective for TrkA in cell based assays over hERG inhibition (TrkA cell IC50 7.0nM,
6
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 74
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
hERG PX IC50 9500nM), making it an excellent lead molecule. Ligand 3 exhibited low turnover
in HLM and hHep (CLint HLM <8 L/min/mg protein; hHep 8 µL/min/million cells). Moreover,
metabolite identification studies suggested the major metabolism pathways involved only phase I
oxidative pathways catalyzed by cytochrome P450 (CYP) (data not shown). Thus, the absence in
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
of metabolism catalyzed by AO, would lend higher confidence to predictions made from in vitro
metabolic CL when compared with PF-06273340 (Figure 1). The passive permeability of 3 was
-6
-1
measured in the RRCK transcellular flux assay (RRCK Papp10x10 cms ) and was consistent with
good absorption.31, 34-35
Figure 2. Discovery of benzamide lead molecule 3
Lead molecule 3 was progressed to oral and i.v. pharmacokinetic (PK) studies in rats and
demonstrated moderate systemic CL (plasma CL 38 mL/min/Kg), and moderate oral
bioavailability (45%), consistent with complete absorption. The ability of 3 to act as a substrate
for P-gp and BCRP was also assessed and 3 was found to be a substrate for both these transporters
(P-gp ER 18, BCRP ER 2.5). Ligand 3 was progressed to CNS penetration studies in rat in order
to understand the level of restriction from the CNS achievable, and was found to be peripherally
restricted with Cb,u/Cp,u of 0.0065. These data provided evidence that restriction from the CNS
combined with acceptable oral absorption was indeed achievable in this chemical series. The
7
ACS Paragon Plus Environment

Page 9 of 74
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
binding mode of 3 was determined via X-ray crystallography with a 2.91Å co-crystal structure of
3
with TrkA protein (Figure 3). Ligand 3 was a Type II inhibitor which binds to the DFG-out
conformation of TrkA formed by the movement of the activation loop to an inactive conformation.
The movement of the activation loop involves relocation of the DFG motif which creates an
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
unoccupied back pocket.7-8 In the case of ligand 3 the back pocket is occupied by the p-OCF
aryl
3
group. The primary carboxamide of 3 interacts with the kinase hinge through hydrogen bonds
between the ligand C=O and backbone NH of Met592, and ligand NH with the backbone C=O of
Glu590. Ligand 3 extends into TrkA passing the gatekeeper Phe589 and DFG Phe669 and these
phenylalanine residues sandwich the piperidine linker, with the central amide carbonyl of 3
forming a hydrogen bond with the backbone NH of Asp668. The p-OCF aryl group occupied the
3
back pocket and there was a CH-π interaction between this aryl unit and the backbone CH of
Gly667. The N-methyl imidazole had a vector towards the solvent exposed kinase exit which
suggested that the structure activity relationship (SAR) in this region will be relatively tolerant,
which for example, explained the similar TrkA cellular potencies of inhibitors 2 and 3 (Figure 2).
Variation of the p-OCF aryl group revealed a number of potential replacements although since
3
the back pocket was formed by hydrophobic residues such as Leu564, Leu567, Phe646, Ile666 the
active back pocket substituents identified were relatively lipophilic (Table 1). Substitution of the
benzylic carbon was tolerated by TrkA as exemplified by 4 and 5 although one enantiomer was
over 60-fold more potent than the other (TrkA cell IC50 4: 0.011µM and 5: 0.72µM), presumably
due to a clash between the benzylic methyl substituent and Phe589 in 5. Substitution of the p-OCF
aryl group by 3-F in 6 was tolerated by TrkA but did not offer any advantages when compared
with the unsubstituted analogue 3. The p-OCF substituent was successfully replaced by a
3
3
cyclopropyl ether as demonstrated by ligand 7 and this modification improved hERG liability
ACS Paragon Plus Environment
8

Journal of Medicinal Chemistry
Page 10 of 74
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
considerably (hERG FP IC50 > 40µM). However, the cyclopropyl ether unit underwent greater
metabolic CL in HLM (HLM 23L/min/mg protein) when compared with lead molecule 3 (HLM
<
8L/min/mg protein). At this stage in the program the p-OCF
3
phenyl group in 3 was favored as
0
1
2
3
4
5
6
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 back pocket substituent based on TrkA cell potency, LipE and metabolic CL. The benzamide in
3
8
could be successfully replaced by a nicotinamide as demonstrated by 8 (Table 1). Nicotinamide
exhibited a similar TrkA potency to 3 but had a lower LogD (8: LogD 2.4, 3: LogD 2.9) and
hence a greater LipE than 3.31
A
B
Figure 3. A) Co-crystal structure of lead molecule 3 bound to TrkA protein in a DFG-out
conformation highlighting key protein-ligand interactions with black dashed lines. B) Co-crystal
structure of lead molecule 3 bound to TrkA protein highlighting binding site surface with brown
representing hydrophobic regions, green neutral polarity regions, blue polar regions. Some protein
residues have been omitted for clarity. PDB code is 6DKW for compound 3.
ACS Paragon Plus Environment
9

Page 11 of 74
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Table 1. Back pocket SAR and associated in vitro data
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
TrkA cell
IC50 (M)
RRCK P-gp
Herg FP
ER IC50 (M)
ID
4
A
Ar
LogDb LipEc HLMd hHepe
a
f
g
P
app
C
0.011
3.5
3.4
3.0
2.3
2.4
4.5
2.7
5.0
5.1
5.6
31
<8
<8
23
<8
ND
ND
13
6.2
7.7
6.4
10
11
11
34
30
10
6.5
4.2
5
6
7
C
C
C
N
0.72
0.011
0.040
0.010
ND
>40
4.5
ND
13
8
13
a
TrkA data were generated in PathHunter cells from DiscoveRx with at least 3 tests on 3 different
assay runs.3 LogD measured at pH 7.4. LipE= -Log(TrkA cell IC50) – LogD. Human liver
microsome metabolic CLint in mL/min/mg protein. Human hepatocyte CLint in µL/min/million
cells. Apparent permeability (x 10 cms ) measured with low efflux MDCKII cell line (RRCK).
Data for LogD, HLM, HLMUGT, hHep, RRCK and P-gp were results from at least 2 replicate
determinations. hERG FP assay data with at least 2 tests on 2 different assay runs. ND denotes
not determined.
1 b
c
d
e
f
-6
-1
35
g
32
ACS Paragon Plus Environment
1
0

Journal of Medicinal Chemistry
Page 12 of 74
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
Figure 3B shows the binding site surface of ligand 3 when bound to TrkA protein. The binding
site surface suggests that the piperidine linker can be further substituted to occupy more of the
binding site. Moreover, piperidine substitution may enhance interactions with Phe589 and Phe669
0
1
2
3
4
5
6
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
through increased van de Waals interactions and potential CH-π interactions if the piperidine is
substituted by an electron withdrawing group. To this end, 3-fluoropiperidine derivatives of the
most LipE efficient compound 8 in Table 1, were prepared (Table 2). Both enantiomers of the cis
and trans diastereomers were synthesized and gave up to a 3-fold enhancement in TrkA potency
and improved LipE by up to 0.4 units when compared with lead molecule 3 (ligand 10b, LipE 5.7).
All compounds were pan-Trk inhibitors with similar potencies at TrkA/B/C (within 3-fold), P-gp
and BCRP efflux substrates and exhibited moderate apparent permeability. Ligand 10b offered the
best balance of TrkA potency (TrkA cell IC50 1.9nM, TrkB cell IC50 2.6nM, TrkC cell IC50 1.1nM),
LipE, metabolic stability as measured by hHep CLint, and hERG liability as measured by hERG
FP (Ligand 10b hERG FP IC50 14µM). The binding mode of 10b was established via X-ray
crystallography with a 2.68Å co-crystal structure with TrkA protein (Figure 4). Ligand 10b made
similar interactions with TrkA protein to ligand 3 (Figure 3), but had additional van de Waals
interactions between F and Phe589. There were also CH-π interactions introduced between the
piperidine and Phe589/Phe669. Compound 10b was screened in 41 biochemical kinase assays
(including TrkA) at a concentration of 1µM at Invitrogen for kinase selectivity (see supporting
information). Gratifyingly 10b exhibited superb Trk selectivity with >95% inhibition of TrkA and
>40% inhibition of only 1 other kinase VEGFR2, (cell based assay follow-up at Caliper
Lifesciences indicated VEGFR2 IC50 >5µM). Furthermore, wide ligand profiling of 10b at a
concentration of 10µM in 84 targets to assess off target based liabilities indicated a clean profile
ACS Paragon Plus Environment
1
1

Page 13 of 74
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
(
1 hit with % inhibition >50% at 10µM which gave follow-up alpha1a antagonist IC50 8µM). The
exciting in vitro profile of 10b led to the progression of this compound to in vivo studies.31
Table 2. Piperidine linker SAR and associated in vitro data
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
TrkA cell
RRCK P-gp BCRP
Herg FP
IC50 (M)
ID
9a
A
R1
LogDb LipEc HLMd hHepe
IC50 (nM)a
P
f
ER
ER
g
app
N
74
5.4
3.0
3.0
3.0
3.0
4.1
5.3
5.5
<8
<8
<8
14
12
12
11
9
13
2.8
7.0
24
9
b
N
N
14
10
3.3
2.5
1
0a
13
9.0
1
0b
N
C
1.9
2.8
3.0
3.1
5.7
5.5
<8
24
7
17
10
10
13
4.6
14
1
1
11
ND
ND
a
TrkA data were generated in PathHunter cells from DiscoveRx with at least 3 tests on 3 different
assay runs.3 LogD measured at pH 7.4. LipE= -Log(TrkA cell IC50) – LogD. Human liver
microsome metabolic CLint in mL/min/mg protein. Human hepatocyte CLint in µL/min/million
cells. Apparent permeability (x 10 cms ) measured with low efflux MDCKII cell line (RRCK).
Data for LogD, HLM, hHep, RRCK, P-gp and BCRP were results from at least 2 replicate
determinations. hERG FP assay data with at least 2 tests on 2 different assay runs. ND denotes
not determined.
1 b
c
d
e
f
-6
-1
35
g
32
ACS Paragon Plus Environment
1
2

Journal of Medicinal Chemistry
Page 14 of 74
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 4. Co-crystal structure of ligand 10b bound to TrkA protein in a DFG-out conformation,
highlighting key protein-ligand interactions with black dashed lines. Ligand 10b shown in magenta
and TrkA protein shown in orange. Some protein residues have been omitted for clarity. PDB code
is 6DKB for compound 10b.
Design of benzamide alcohols as pan-Trk inhibitors
The co-crystal structure of PF-06273340 with TrkA protein has been published recently by our
group.9, 29 PF-06273340 and 10b have different hinge binders with the aminopyrimidine of PF-
0
6273340 interacting with the backbone NH and C=O of Met592 whilst the amide in 10b interacts
with the backbone NH of Met592 and backbone C=O of Glu590. Despite this difference an overlay
of the ligand bound conformation with TrkA of PF-06273340 and compound 10b revealed an
opportunity to remove the imidazole group in 10b and append an alcohol moiety heading into the
ribose pocket (Figure 5). Keeping both the imidazole and alcohol groups would result in high strain
ACS Paragon Plus Environment
1
3

Page 15 of 74
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
and high molecular weight. To this end the hybrid ether linked compounds shown in Table 3 were
prepared.36 Gratifyingly, the alcohols 12-14 were active in TrkA cell based assays and also
demonstrated a lower hERG liability when compared with the benzamides in Table 2. The cis
diastereomers 13a/13b were 5-50 fold more active than the corresponding trans diastereomers
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
2a/12b. The primary alcohol isomer 14 was also active in TrkA cell based assays but showed
significantly higher metabolic CL in hHeps possibly due to enhanced metabolism involving the
primary alcohol when compared with a tertiary alcohol.37
Figure 5. Overlay of the bound ligand conformation in TrkA protein of DFG-out binders PF-
0
6273340 (grey) and compound 10b (magenta).
ACS Paragon Plus Environment
1
4

Journal of Medicinal Chemistry
Page 16 of 74
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. Benzamide alcohols and associated in vitro data
0
1
2
3
4
5
6
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
TrkA cell
RRCK P-gp BCRP
Herg FP
IC50 (M)
ID
R1
R2
LogDb LipEc HLMd hHepe
IC50 (nM)a
P
f
ER
ER
g
app
1
2a
410
3.1
3.0
3.2
3.1
3.0
3.3
4.2
4.7
5.0
4.6
<8
<8
<8
<8
11
ND
5.9
6.1
2.0
25
5.7
11
48
ND
29
1
2b
6
1
22
64
72
22
ND
6.5
7.0
ND
38
38
42
35
13a
13
7.7
28
5.4
5.2
11
1
3b
1
4
a
TrkA data were generated in PathHunter cells from DiscoveRx with at least 3 tests on 3 different
36 b
c
d
assay runs. LogD measured at pH 7.4. LipE= -Log(TrkA cell IC50) – LogD. Human liver
microsome metabolic CLint in mL/min/mg protein. Human hepatocyte CLint in µL/min/million
cells. Apparent permeability (x 10 cms ) measured with low efflux MDCKII cell line (RRCK).
Data for LogD, HLM, hHep, RRCK, P-gp and BCRP were results from at least 2 replicate
determinations. hERG FP assay data with at least 2 tests on 2 different assay runs. ND denotes
not determined.
e
f
-6
-1
35
g
32
ACS Paragon Plus Environment
1
5

Page 17 of 74
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Benzamide alcohol 13b offered the best balance of TrkA potency (TrkA cell IC50 7.7nM, TrkB
cell IC50 15nM, TrkC cell IC50 3.9nM), LipE, and metabolic stability as determined in hHep (CLint
6
= 2L/min/10 cells). The binding mode of 13b was established via X-ray crystallography with a
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
2
.53Å co-crystal structure with TrkA protein (Figure 6). Ligand 13b made similar interactions
with TrkA to ligand 10b (Figure 4), however 13b had additional van de Waals interactions between
the gem dimethyl groups and Phe669 and interactions between the alcohol and water molecules
towards the kinase exit. Compound 13b was screened in 42 biochemical kinase assays (including
TrkA) at a concentration of 1µM at Invitrogen for kinase selectivity (see supporting information).
Gratifyingly 13b exhibited superb Trk selectivity with >95% inhibition of TrkA with no other
kinase being inhibited by >40% at 1µM. Furthermore, wide ligand profiling at a concentration of
1
0µM at 82 targets to assess off target based liabilities showed a clean profile with no hits having
>50% inhibition at 10µM. The exciting in vitro profile of 13b led to the progression of this
compound to in vivo studies.36
ACS Paragon Plus Environment
1
6

Journal of Medicinal Chemistry
Page 18 of 74
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 6. Co-crystal structure of ligand 13b bound to TrkA protein in a DFG-out conformation,
highlighting key protein-ligand interactions with black dashed lines. Ligand 13b shown in orange
and TrkA protein shown in green. Some protein residues have been omitted for clarity. PDB code
is 6DKG for compound 13b.
Design of aminopyridine pan-Trk inhibitors
Aminopyridines have been described as efficient kinase hinge binding motifs and this concept led
to an additional hybrid design wherein the benzamide hinge binder in molecules 2-3 (Figure 2)
38
was changed to an aminopyridine. This hypothesis led to the synthesis of the aminopyridines
3
9
shown in Figure 7 and Table 4. The aminopyridine in 15 replaces the benzamide hinge binder in
ligands 2-3 (Figure 2) and Figure 8 displays an overlay of benzamide 3 when bound to TrkA
protein (Figure 3) with aminopyridine 15 docked into TrkA protein. The methyl pyrazole
substitution in 15 was thought to be optimal based on the overlay in Figure 8. Aminopyridine 15
ACS Paragon Plus Environment
1
7

Page 19 of 74
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
was a potent Trk inhibitor in cell based assays, however replacement of the piperidine ring linker
with a pyrrolidine ring during SAR exploration improved TrkA cellular potency by 2-3-fold and
LipE by 1 unit in the case of the more active pyrrolidine enantiomer 16b (Figure 7, TrkA cell IC50
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
7
.7nM, LipE 5.5). Aminopyridines having a pyrrolidine linker were an exciting prospect and
further optimisation was conducted on this motif. Based on the learnings with ligand 3 thus far
(Figure 2, Table 2) the N-methyl pyrazole unit was replaced by an N-methyl imidazole and
fluorinated pyrrolidine derivatives were prepared (Table 4).39
Figure 7. Generation of aminopyridine pan-Trk inhibitors
ACS Paragon Plus Environment
1
8

Journal of Medicinal Chemistry
Page 20 of 74
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. Aminopyridine amides and associated in vitro data
0
1
2
3
4
5
6
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
TrkA
cell
Herg
FP
P-
gp
ER
RRCK
BCRP
ER
ID
R1
LogDb LipEc HLMd hHepe
f
IC50
M)
P
app
IC50
(M)
a
g
(
1
7a
1.1
2.6
2.6
2.8
2.8
3.4
5.5
2.9
6.2
15
<8
ND
17
17
22
22
22
10
12
12
13
ND
ND
ND
5.0
16
17b
0.008
1.8
15
21
11
1
8a
8b
9.1
9.7
ND
9.1
1
0.001
a
TrkA data were generated in PathHunter cells from DiscoveRx with at least 3 tests on 3 different
assay runs.3 LogD measured at pH 7.4. LipE= -Log(TrkA cell IC50) – LogD. Human liver
microsome metabolic CLint in mL/min/mg protein. Human hepatocyte CLint in µL/min/million
cells. Apparent permeability (x 10 cms ) measured with low efflux MDCKII cell line (RRCK).
Data for LogD, HLM, hHep, RRCK, P-gp and BCRP were results from at least 2 replicate
determinations. hERG FP assay data with at least 2 tests on 2 different assay runs. ND denotes
not determined.
9 b
c
d
e
f
-6
-1
35
g
32
ACS Paragon Plus Environment
1
9

Page 21 of 74
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 8. Overlay of the bound ligand conformation in TrkA protein of lead molecule 3 (blue,
Figure 3) and aminopyridine 15 docked into TrkA protein.
Mono and difluorinated pyrrolidines (Table 4) were potent TrkA inhibitors with 100-1000-fold
preference of TrkA for the more potent enantiomers (17b, 18b) over the weaker enantiomers (17a,
1
8a). Cis monofluorinated derivatives proved more challenging to synthesize than the
corresponding trans monofluorinated isomers. Difluorinated pyrrolidine 18b was 8-fold more
active when compared with the monofluorinated derivative 17b possibly due to enhanced van der
Waals interactions. Moreover 18b exhibited greater metabolic stability than 17b as measured by
6
hHep Clint (17b and 18b hHep CLint = 17 and 9.1L/min/10 cells respectively) conceivably due
to blocking of oxidative metabolism by the additional F in 18b. Metabolite identification studies
in hHeps with 16b (Figure 7, data not shown) indicated the major metabolism pathways involved
N-demethylation, pyrrolidine unit oxidation and glucuronidation on the amino moiety. Ligand 18b
was active in the hERG FP and PX assay (hERG FP IC50 14µM, hERG PX IC50 7.7µM) although
1
8b had a good therapeutic index (TI) for TrkA over hERG due to its nanomolar potency at TrkA.
Based on the SAR developed to date the imidazole substituent in 18b could be replaced as
demonstrated by the benzamide alcohols in Table 3. Moreover, amides are well established
ACS Paragon Plus Environment
2
0

Journal of Medicinal Chemistry
Page 22 of 74
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
bioisosteres for heterocycles and hence an amide substituent was introduced in place of the
imidazole moiety in 18b to give amides 19 and 20 (Figure 9). Methyl amide 19 was a potent and
LipE efficient TrkA inhibitor (TrkA cell IC50 8.4nM, LipE 5.6) with low metabolic turnover in
human in vitro assays (HLM <8L/min/mg protein, hHep 1.4µL/min/million cells). Moreover 19
demonstrated excellent selectivity over hERG in the hERG FP assay (IC50 79µM) and hERG PX
assay (IC50 88µM). Ethyl amide 20 was also a potent TrkA inhibitor (TrkA cell IC50 10nM, LipE
0
1
2
3
4
5
6
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
.2) although it offered no advantage over 19 and had a LipE that was ~0.4 units lower than 19.
This relatively flat SAR was consistent with the terminal amide occupying a region of TrkA that
exits the kinase towards solvent, much like the imidazole group in Figure 3. Ligand 19 seemed to
offer the best balance of TrkA potency, LipE, metabolic stability and hERG liability when
compared with 18b (Table 4) and was chosen for further progression. The binding mode of 19 was
determined via X-ray crystallography with a 2.11Å co-crystal structure with TrkA protein (Figure
10). The aminopyridine group did indeed act as a hinge binder with the amino unit NH interacting
with the main chain C=O of Glu590 and the pyridyl N interacting with the main chain Met592
NH. The terminal amide had a vector towards the solvent exposed kinase exit and appeared to
interact with water. The pyrrolidine linker exhibited van der Waals interactions with the
surrounding residues such as Phe669, Phe589, Val524 along with potential CH-π interactions with
Phe589 and Phe669. Consistent with the previous TrkA co-crystal structures discussed (Figures 3-
4
, and 6) the central amide carbonyl of 19 formed a hydrogen bond with the main chain NH of
Asp668 and the p-OCF
3
phenyl group occupied the back pocket.
ACS Paragon Plus Environment
2
1

Page 23 of 74
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 9. Discovery of aminopyridine amide 19
Figure 10. Co-crystal structure of aminopyridine 19 bound to TrkA protein in a DFG-out
conformation, highlighting key protein-ligand interactions with black dashed lines. Ligand 19
shown in yellow and TrkA protein shown in green. Some protein residues have been omitted for
clarity. PDB code is 6DKI for compound 19.
ACS Paragon Plus Environment
2
2

Journal of Medicinal Chemistry
Page 24 of 74
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
Compound 19 was screened at 44 biochemical kinase assays (including TrkA) at a concentration
of 1µM at Invitrogen for kinase selectivity (see supporting information). Ligand 19 exhibited
outstanding Trk selectivity with >95% inhibition of TrkA with no other kinase being inhibited by
0
1
2
3
4
5
6
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
>40% at 1µM. Furthermore, wide ligand profiling to assess off target based liabilities at a
concentration of 10µM in 82 targets indicated a clean profile with no hits having >50% inhibition
at 10µM. The exciting in vitro profile of 19 led to the progression of this compound to in vivo
studies.39
Surface plasmon resonance data for pan-Trk ligands 10b, 13b and 19
Ligands 10b, 13b and 19 were evaluated using surface plasmon resonance with immobilised TrkA
protein (residues 441-796). With non-activated unphosphorylated TrkA protein, ligands 10b, 13b
4
0
and 19 demonstrated relatively slow association and dissociation kinetics, with measured K
D
5
-1 -1
-4 -1
values consistent with TrkA cellular potencies 10b: K
a
6.9x10 M s , K
d
4.7 x10 s , K
D
0.7
4
-1 -1
-4 -1
4
-1 -1,
-4 -1
nM; 13b: K
a
4.4 x10 M s , K
d
1.1 x10 s , K
D
2.6 nM; 19: K
a
5.7 x10 M s , K
d
1.3 x10 s ,
K
D
2.3 nM.
In vivo rat PK data for pan-Trk ligands 10b, 13b and 19
The optimized pan-Trk ligands 10b, 13b and 19 were examined in oral and i.v. PK studies in rats
where they demonstrated low systemic CL and moderate oral bioavailability that was consistent
with good absorption (Table 5). These compounds were also found to be peripherally restricted
with Cb,u/Cp,u of <0.05 in all cases in CNS penetration studies in rat. These in vivo data confirm
that in these optimized pan-Trk ligands, acceptable oral absorption and restriction from the CNS
can be combined. These properties that are demonstrated in rats are desirable in pan-Trk inhibitors
ACS Paragon Plus Environment
2
3

Page 25 of 74
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
intended for use as oral medicines for chronic pain, and can be reasonably expected to translate to
human.
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 5. Compounds 10b, 13b and 19: pharmacokinetic properties in rat
i.v. administration
plasma CL
p.o. administration
dose
T½
Vss
dose
Tmax
(hr)
Oral F
ID
Cb,u/Cp,u
(
mg/Kg) (hr) (mL/min/Kg) (L/Kg)
(mg/Kg)
(%)
1
0b
1
1
1
2.6
1.9
3.9
13.4
8.9
0.65
0.6
0.043
0.026
0.018
3
0.5
0.5
0.6
52
13b
3
3
35
56
1
9
18.5
1.4
Compounds were administered via intravenous (i.v.) and oral (p.o.) routes to separate groups of
two male rats. Cb,u/Cp,u was determined separately following i.v. infusion in 4 male rats. Further
details of PK and CNS penetration experiments are given in the supporting information.
Predicted human clearance of pan-Trk ligands 10b, 13b and 19
Metabolic pathways of 10b, 13b and 19 were examined in HLM and in hHep in vitro (see
supporting information). In HLM, metabolism of all three analogues was identified as occurring
through pathways consistent with CYP-mediated oxidation. In separate incubations of compounds
in human liver cytosol fraction, disappearance of compounds over time was undetectable. Thus,
the pattern of metabolite formation and absence of detectable turnover in human liver cytosol
suggested that AO was not involved in metabolism of these compounds. In hHep, the benzamides
1
0b and 13b were found to form some of the oxidized metabolites detected in HLM, without
evidence of phase 2 conjugation pathways. The aminopyridine 19 also formed oxidized
ACS Paragon Plus Environment
2
4

Journal of Medicinal Chemistry
Page 26 of 74
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
metabolites in hHep but also formed small amounts of glucuronidated products, consistent with
conjugation at the aminopyridine moiety. In the absence of metabolism by AO, prediction of
systemic CL in human by scaling in vitro metabolic turnover data can be performed with higher
confidence than previously possible for the development candidate pan-Trk inhibitor PF-
0
1
2
3
4
5
6
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
6273340. Turnover of analogues 10b, 13b and 19 in HLM was too low to be quantified
(<8L/min/mg), whereas in hHep, CLint was measurable using the ‘relay’ assay technique
41
designed for compounds exhibiting high metabolic stability. CLint values were 7.8, 2.0 and 1.4
6

L/min/10 cells, for 10b, 13b and 19, respectively, which scaled to systemic plasma CL values
in human of 1.5, 0.3 and 0.6 mL/min/Kg, respectively, consistent with minimal hepatic extraction.
Solubility of pan-Trk ligands 10b, 13b and 19
Solubility is a drug property that plays a key role in the development potential and PK profile of
clinical candidates.34, 42 The solubility of our previous development candidate PF-06273340 was
low for the crystalline form at approximately 2µg/mL in a pH 6.5 aqueous phosphate buffer. In
comparison the solubility of 10b, 13b and 19 was significantly higher 15, 76, and 55µg/mL
respectively. This improvement in solubility can be attributed to, in part, the higher fraction of sp3
hybridised carbons of 10b, 13b and 19 when compared with PF-06273340, and fewer HBDs in
1
0b, 13b and 19 when compared with PF-06273340 that can result in high lattice energies (PF-
06273340.43-44 Ligands 10b and 19 gave highest measured pKa values (between 4.2-4.7). The
improved solubility of 10b, 13b and 19 gave further confidence in the clinical development
potential of these analogues.
ACS Paragon Plus Environment
2
5

Page 27 of 74
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
In vivo efficacy in UV burn hyperalgesia models
Compounds 10b, 13b and 19 were assessed in the UV irradiation-induced thermal hyperalgesia
(UVIH) model of inflammatory pain in rats whereby their effects on thermal hyperalgesia resulting
from UV irradiation were determined by measuring paw withdrawal latency (PWL) (Figure 11).
Single oral doses of 10b (0.03, 0.32 and 3 mg/Kg), 13b (0.15, 1.5 and 15mg/Kg) and 19 (0.03,
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
0
6
.32 and 3 mg/Kg) were administered 48 hours after UV treatment and PWL measured at 1, 3 and
hours post dose. The doses of compounds were chosen in order to achieve multiples of TrkA cell
IC50 of 0.1x, 1.0x and 10x at some time between 1 and 6 hours. The effect on hyperalgesia of all
three compounds at the highest dose studied is shown in Figure 11 and their effects at all doses
administered are summarized in Table 6. All three compounds significantly reversed hyperalgesia
at the highest two doses employed at each of the time-points studied (p<0.05, 2-way ANOVA vs.
vehicle). The positive control ibuprofen (100mg/Kg, p.o.) also reversed thermal hyperalgesia at
the 1 and 3 hour time-points (individual compound dose response data compared with vehicle and
positive control ibuprofen for 10b, 13b and 19 is provided in the supporting information). Whilst
no effect on PWL was observed at any dose of 10b or 19 for the contralateral hindpaw, the highest
dose of 13b displayed a significant effect on contralateral PWL (P<0.05, data not shown).
Unbound plasma concentrations of compounds were determined in animals at 6 hours post dose.
This demonstrated that statistically significant efficacy was observed at unbound plasma
concentrations equivalent to the following multiples of TrkA cell IC50: 10b: 0.5x (0.03mg/Kg), 4x
(3mg/Kg); 13b: 0.8x (0.15mg/Kg), 13x (15mg/Kg); 19: 0.15x (0.3mg/Kg), 2x (3mg/Kg). Overall,
it is possible to say that in these experiments, efficacy was achieved at doses of compounds that
corresponded with unbound exposure at multiples of TrkA cell IC50 of 2x – 13x.
ACS Paragon Plus Environment
2
6

Journal of Medicinal Chemistry
Page 28 of 74
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 6. Effects of 10b, 13b and 19 on UV irradiation-induced hyperalgesia in rats
ID
dose (mg/Kg)
significance of difference from vehicle
1
h
3h
6h
0
1
2
3
4
5
6
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
0b
0.03
N.S.
p<0.05
p<0.001
p<0.001
p<0.05
p<0.001
p<0.001
N.S.
N.S.
0
.3
p<0.001
p<0.001
N.S.
p<0.001
p<0.001
N.S.
p<0.001
p<0.001
N.S.
p<0.001
p<0.001
N.S.
3
13b
0.15
.5
1
1
5
1
9
0.03
.3
0
p<0.05
p<0.001
p<0.001
p<0.001
p<0.05
p<0.001
3
The statistical significance of effects on PWL (ipsilateral paw, n = 8) were determined according
to 2-way ANOVA vs. vehicle data (N.S = not significant (p>0.05)). Dose response data compared
with vehicle and positive control ibuprofen for 10b, 13b and 19 are provided in the supporting
information.
ACS Paragon Plus Environment
2
7

Page 29 of 74
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 11. Effects of compounds 10b, 13b and 19 in the UVIH model in rats at the highest dose
studied. Ipsilateral paw data are shown, n=8 rats per dose group. Statistical significance according
to 2-way ANOVA vs. vehicle data is given in Table 6. Dose response data compared with vehicle
and positive control ibuprofen for 10b, 13b and 19 is provided in the supporting information.
Ligands 10b, 13b and 19 were progressed to either 2 week (ligands 10b and 19) or 4 week (13b)
exploratory toxicity studies in rat. No adverse effects (including behavioral) were observed at free
average concentrations (Cavg) of 725x, 882x and 133x TrkA cell IC50 for 10b, 19 and 13b
respectively. Moreover 10b, 19 and 13b exhibited a favorable genetoxicity profile suitable for
advancement.
ACS Paragon Plus Environment
2
8