Discovery of TRPM8 Antagonist (S)-6-(((3-Fluoro-4-(trifluoromethoxy

Subscriber access provided by UNIV OF DURHAM

Article

Discovery of TRPM8 Antagonist (S)-6-(((3-fluoro-4-(trifluoromethoxy)phenyl) (3-fluoropyridin-2-yl)methyl)carbamoyl)nicotinic acid (AMG 333), a Clinical Candidate for the Treatment of Migraine Daniel B. Horne, Kaustav Biswas, James Brown, Michael D. Bartberger, Jeffrey Clarine, Carl D Davis, Vijay K Gore, Scott Harried, Michelle Horner, Matthew R. Kaller, Sonya G. Lehto, Qingyian Liu, Vu V Ma, Holger Monenschein, Thomas T Nguyen, Chester Yuan, Beth D. Youngblood, Maosheng Zhang, Wenge Zhong, Jennifer R. Allen, Jian Jeffrey Chen, and Narender Gavva J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.8b00518 • Publication Date (Web): 27 Aug 2018 Downloaded from http://pubs.acs.org on August 28, 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 71 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of Medicinal Chemistry

Discovery of TRPM8 Antagonist (S)-6-(((3-fluoro-4-(trifluoromethoxy)phenyl)(3fluoropyridin-2-yl)methyl)carbamoyl)nicotinic acid (AMG 333), a Clinical Candidate for the Treatment of Migraine Daniel B. Horne,†‡* Kaustav Biswas,† James Brown,† Michael D. Bartberger,† Jeffrey Clarine,Ω Carl D. Davis,Ω Vijay K. Gore,† Scott Harried,† Michelle Horner,∆ Matthew R. Kaller,† Sonya G. Lehto,† Qingyian Liu,† Vu V. Ma,† Holger Monenschein,† Thomas T. Nguyen,† Chester C. Yuan,† Beth D. Youngblood,† Maosheng Zhang,† Wenge Zhong,† Jennifer R. Allen,† Jian Jeffrey Chen,† and Narender R. Gavva† Departments of †Amgen Discovery Research, ΩPharmacokinetics and Drug Metabolism, and ∆Comparative Biology and Safety Sciences, Amgen Inc., One Amgen Center Drive, Thousand Oaks, CA 91320 & Amgen Inc., 360 Binney Street, Cambridge, MA 02142

Transient receptor potential melastatin 8 (TRPM8), the predominant mammalian cold temperature thermosensor, is a non-selective cation channel expressed in a subpopulation of sensory neurons in the peripheral nervous system, including nerve circuitry implicated in migraine pathogenesis: the trigeminal and pterygopalatine ganglia. Genome-wide association studies have identified an association between TRPM8 and reduced risk of migraine. This disclosure focuses on medicinal chemistry efforts to improve drug-like properties of initial leads, particularly removal of CYP3A4 induction liability and improvement of PK properties. A novel series of biarylmethanamide TRPM8 antagonists was developed, and a subset of leads were evaluated in preclinical toxicology studies to ACS Paragon Plus Environment

Journal of Medicinal Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 71

identify a clinical candidate with acceptable preclinical safety profile leading to clinical candidate AMG 333, a potent and highly selective antagonist of TRPM8 which was evaluated in human clinical trials.

INTRODUCTION The transient receptor potential (TRP) channels are non-selective, Ca2+ permeable ionchannels, some of which function as molecular sensors. The TRP channel family is comprised of 28 distinct members which are activated by a variety of stimuli including voltage, pH, temperature (cold and heat), mechanical pressure, osmotic pressure, and endogenous and exogenous ligands. Amongst the TRP channels, a subset are implicated in temperature sensation and thermoregulation (TRPV1-4 activated by heat/warmth; TRPM8 activated by innocuous cold; and TRPA1 activated by noxious cold). TRPM8 ion-channels are activated by cool temperatures (18-23 °C) as well as exogenous ligands such as menthol and icilin.1 TRPM8 is expressed in the lung, bladder, artery, prostate, and in key components of the peripheral nervous system such as the dorsal root ganglia (DRG), pterygopalatine (SPG), and trigeminal ganglia (TG) that are involved in pain sensation and migraine. Potential applications of modulating TRPM8 for treatment of neuropathic2 and chemotherapy-induced cold pain,3 prostate cancer,4 and overactive and painful bladder syndromes5 have been described. Previous investigations from various groups have resulted in identification of several small-molecule scaffolds that modulate

ACS Paragon Plus Environment

Page 3 of 71 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


TRPM8 activity.6-11 Herein we communicate our efforts targeting TRPM8 inhibition as a potential treatment for migraine. We previously described SAR studies leading to the identification of TRPM8 inhibitors 112a and 212b (Figure 1) and results in rat in vivo pharmacodynamic (PD) models of TRPM8 target engagement, the icilin-induced wet-dog shake (WDS) and cold pressor test (CPT). Compound 2 was a potent inhibitor of TRPM8 (h / rTRPM8 IC50 = 25 / 115 nM), had reasonable pharmacokinetic (PK) properties in rats, and was effective in vivo in both the rat WDS and CPT models of TRPM8 inhibition. Initially, we evaluated potent TRPM8 antagonists as potential therapeutics for neuropathic pain, but in our hands, 2 and other TRPM8 antagonists were ineffective in animal models of neuropathic pain.13 However, data later emerged from genome-wide association studies (GWAS) that identified an association between single nucleotide polymorphs (SNPs) that are intergenic to TRPM8 (one SNP located about 1000 bp 5’ to the start codon and a second SNP in the intron), with a reduced risk for migraine.14 Based on this new genetic data, we pivoted our existing TRPM8 preclinical program with the intent of seeking a therapeutic for migraine pain. Our lead optimization efforts described herein, seeking to improve in vitro and in vivo properties of early leads, ultimately led to the nomination of development candidate (S)6-(((3-fluoro-4-(trifluoromethoxy)phenyl)(3-fluoropyridin-2yl)methyl)carbamoyl)nicotinic acid (35) that was subsequently advanced into human clinical studies.



Page 4 of 71

Figure 1. Previously reported tetrahydroisoquinoline and tetrahydronaphthyridine TRPM8 antagonists. RESULTS AND DISCUSSION As previously reported,12b optimization of a tetrahydroisoquinoline series of TRPM8 antagonists resulted in tetrahydronaphthyridine 2 which showed improved PK properties and demonstrated potent activity in two TRPM8 engagement PD models: (i) the icilininduced rat WDS model with complete inhibition at 10 mg/kg, Cu = 0.19 µM; (ii) the rat CPT, with complete inhibition of blood pressure increase at 10 mg/kg po, Cu = 0.49 µM. However, despite achieving plasma exposures >10-fold over the exposure levels resulting in complete inhibition of rat WDS or CPT, 2 failed to show efficacy in rodent models of neuropathic pain. In addition, we discovered that potent induction of CYP3A4 in human hepatocytes, regulated through the pregnane X receptor (PXR), was endemic throughout this series. Since CYP3A4 plays an important role in human drug metabolism, upon reinitiating our program towards a treatment for migraine, we undertook efforts to eliminate the CYP induction potential of this series of compounds. In addition, we desired to restart our efforts with a novel chemical scaffold that might provide potentially improved properties, since fairly extensive SAR investigations


Page 5 of 71 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 2. Identification of potent biarylmethanamine based TRPM8 antagonists as a starting scaffold. had already been explored in the tetrahydroisoquinoline and tetrahydronaphthyridine series. In an earlier investigation into alternative syntheses of the tetrahydronaphthyridine core of compound 2, we discovered that one of the uncyclized intermediates in that synthesis, urea analog 3, suffered minimal loss of potency compared to the cyclized analog 2 (Figure 2). We viewed analog 3 as a novel starting scaffold we could use to potentially address the suboptimal properties of the series. At this point we also discovered that the preferred stereochemistry of the uncyclized analogs was opposite that of the tetrahydronaphthyridine series. Confirmation of stereochemistry by X-ray crystal and VCD or optical rotation confirmed our stereochemical assignment of both series (Figure 3 and SI). We postulate that the uncyclized series may be binding in an alternative binding conformation than the cyclized series since there is some divergence in the SAR between



Page 6 of 71

the series, although there is some common SAR that suggests the series share the same binding pocket. Initial SAR in the biarylmethanamide series began with substitution of the 3-Br substituent of analog 3 seeking improvements in potency. The 3-position was tolerant of several lipophilic groups, e.g. F (120 nM), allyl (28 nM), CF3 (4, 51 nM), and propynyl (48 nM), and the 3-CF3 analog was selected for further SAR exploration. A small library of amides (52 analogs) and ureas (203 analogs) was made based on the biarylmethanamine of analog 4. A variety of aryl amides from the library were well tolerated and retained potency on TRPM8 (34 of 52, 20-fold loss of activity, but that TRPM8 inhibition could be


Page 7 of 71 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


increased with incorporation of increasingly lipophilic groups at the 3- and 4-positions (9-11, 13-14) of the pendant phenyl group. In contrast, substitution at the 2-position of the phenyl (12), incorporation of hetereocycles (19-20), or benzylic groups (21) were generally not well tolerated, leading to less potent analogs. Investigation of the A-ring demonstrated that removal or transposition of the pyridine nitrogen (22-23) led to losses in potency. In addition, the changes above that led to more potent compounds, had minimal effects on CYP induction and in vitro stability, likely due to the inability to incorporate polarity into this sector of the molecule. It became apparent at this stage that a handle to address CYP3A4 induction and in vitro stability liabilities would need to come from modifications made elsewhere in the molecule. Since it has been demonstrated that PXR transactivation and the resulting CYP3A4 induction potential can be decreased by lowering lipophilicity and/or introducing polarity into the periphery of the molecule,15 we turned our attention to the C-ring of the molecules based on the previous SAR showing that it was unlikely that incorporation of polarity was going to be tolerated in the A- or B-rings with respect to maintaining TRPM8 potency.

Table 1. SAR Summary for Biarylmethanamides



Page 8 of 71

CLintb IC50 (nM)a Compound

Aring

ring

M8

101±

4 7

--

--

--

mTRP

HL

RL

M8

M

M

D

14

51

--

145

18±8

34±6

53±18

7.4)d

(%) @

1 µM

10 µM

16

--

--

3.9

20,0

>20,00 --

12 00

146

335

--

--

4.8

0

95±2 13

--

224

57

251

--

--

4.3

--

67±31

20

15

--

--

5.3

--

218

23

51

--

--

4.9

--

49±40

50

48

59

48

5.0

--

119

51

71

--

--

4.2

--

--

4.7

--

--

2.8

1 22±1 14 6

15

157

19±0. 16 1 17

75

>39 18

301

--

373

706 9

>10,0 19

>10,00 --

00

10,0 20

>10,00 -0

1140

2050± --

±109

a

15

00

21

Page 10 of 71

160

--

--

3.5

--

--

4.2

>39 78

670

9

22

311

--

663

20

25

--

--

4.5

23

109

--

212

61

31

37

39

6.1

IC50 values based on inhibition of icilin (1 µM) induced influx of Ca2+ into human, rat,

or mouse TRPM8-expressing CHO cells. Each IC50 value reported represents mean IC50 calculated from three 22-point concentration-response plots; when replicant experiments were performed, the data is reported as the mean value and standard deviation. b In vitro microsomal stability (intrinsic clearance, CLint), measured in a high-throughput automated format in human liver microsomes (HLM) and rat liver microsomes (RLM). c Expressed as a percentage of the activation response with rifampicin when both compounds are used at a concentration of 1 or 10 µM d Calculated LogD (cLogD) calculated with the ACD/LogD Suite. www.acdlabs.com/logdsuite/ Advanced Chemistry Development, Inc., Toronto, ON, Canada. 24 April 2007.

Initial incorporation of polar groups into the periphery of the C-ring such as hydroxy (25), nitrile (26), pyridyl (27), or methoxy pyridyl (28) functionalities were not success-


Page 11 of 71 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


ful in reducing PXR transactivation as compared to the unsubstituted benzamide (Table 2), since these analogs all still showed potent in vitro PXR transactivation, >90% POC @ 2 µM and was confirmed for analog 28 in the human hepatocyte CYP3A4 induction assay. These changes also had only a minor effect on calculated LogD (cLogD) of the analogs. Our first breakthrough in reducing CYP induction was a change to the C-ring of the analogs that drastically lowered the cLogD of the molecule, pyridone 29, which for the first time showed lower CYP induction (17% @ 1 µM) while still maintaining potency on TRPM8 (hTRPM8 IC50 = 6 nM). We next attempted to reduce lipophilicity further by replacement of the A-ring trifluoromethyl group (at position X). We demonstrated that substitution of the trifluoromethyl group with a fluorine was tolerated, with analog 31 reducing CYP induction significantly (6% @ 1 µM) while still maintaining potency on TRPM8 (hTRPM8 = 10 nM). In contrast, attempts to reduce CYP induction by replacing the trifluoromethyl group in the early urea analog 4 with fluoro, gave fluoro urea analog 30, which resulted in a significant loss of potency and was also confirmed to still have potent activation of hPXR. Further SAR studies around the pyridone 31 scaffold identified several additional analogs, including pyridones 32 and 33. These pyridones were potent on human and rodent TRPM8, had low turnover in liver microsomes, and reduced CYP induction activity. Pyridones 31-33 were differentiated in subsequent in vivo PK and toxicological studies. Concurrent with the discovery of the pyridones as a C-ring subunit with desirable in vitro properties, carboxylic acid 34 emerged as a functional group that was simultaneous-



Page 12 of 71

ly able to reduce CYP induction and maintain potency. Additional analogs of carboxylic acid 34 identified nicotinic acid 35, which had a slight improvement in potency while maintaining other properties in the appropriate range. Substitution of the 3-fluoro of 35 with a trifluoromethyl group gave analog 36, in which CYP induction was significantly increased.

It should also be noted that the reduction in CYP induction potential, as

achieved by lowering lipophilicity with pyridone and carboxylic analogs 29-35, was also accompanied by a significantly decreased intrinsic clearance in liver microsomes.

Table 2. Optimization of CYP Induction Liability

Human CLintb

Co

CYP3A4 InIC50 (nM)

m po

Hepatocyte

X

R

R

1

R

a

[µL/(min

2

•mg)]

un

duction (mRNA POC)c

d hTRP rTRP mTRP HL M8

M8

M8


M

RL

POC

POC

cLog

M

(%) @

(%)

D

Page 13 of 71 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1 µM

@ 10

(pH

µM

7.4)

CF 24

CF3

F

27

--

43

29

57

62

45

5.1

--

--

--

4.6

3

CF 25

13 CF3

H

8±7

21±1

6±4

CF3

H

27

--

40

19

21

--

--

4.7

CF3

H

35

--

54

19

50

--

--

4.2

CF3

H

15±5

--

24±11

15

18

56

94

5.0

Discovery of TRPM8 Antagonist (S)-6-(((3-Fluoro-4-(trifluoromethoxy

Recommend Documents