Orally bioavailable imidazoazepanes as calcitonin gene-related peptide (CGRP) receptor antagonists: Discovery of MK-2918

Article abstract of DOI:10.1016/j.bmcl.2010.12.054

In our ongoing efforts to develop CGRP receptor antagonists for the treatment of migraine, we aimed to improve upon telecagepant by targeting a compound with a lower projected clinical dose. Imidazoazepanes were identified as potent caprolactam replacemen

Full text of DOI:10.1016/j.bmcl.2010.12.054

Bioorganic & Medicinal Chemistry Letters 21 (2011) 2683–2686
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Orally bioavailable imidazoazepanes as calcitonin gene-related peptide
(CGRP) receptor antagonists: Discovery of MK-2918
Daniel V. Paone a, , Diem N. Nguyen a, Anthony W. Shaw a, Christopher S. Burgey a, Craig M. Potteiger a,
⇑
James Z. Deng a, Scott D. Mosser b, Christopher A. Salvatore b, Sean Yu c, Shane Roller c, Stefanie A. Kane b,
Harold G. Selnick a, Joseph P. Vacca a, Theresa M. Williams a
a Department of Medicinal Chemistry, Merck Research Laboratories, West Point, PA 19486, USA
b Department of Pain and Migraine Research, Merck Research Laboratories, West Point, PA 19486, USA
c Department of Drug Metabolism, Merck Research Laboratories, West Point, PA 19486, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
In our ongoing efforts to develop CGRP receptor antagonists for the treatment of migraine, we aimed to
improve upon telecagepant by targeting a compound with a lower projected clinical dose. Imidazoaze-
panes were identified as potent caprolactam replacements and SAR of the imidazole yielded the tertiary
methyl ether as an optimal substituent for potency and hERG selectivity. Combination with the aza-
benzoxazinone spiropiperidine ultimately led to preclinical candidate 30 (MK-2918).
Ó 2010 Elsevier Ltd. All rights reserved.
Received 19 November 2010
Revised 8 December 2010
Accepted 13 December 2010
Available online 30 December 2010
Keywords:
CGRP
Migraine
Calcitonin gene-related peptide
Telcagepant
MK-2918
Migraine is a debilitating disorder characterized by severe, epi-
sodic headaches which can lead to significant loss of productivity.1
The current standard of care, 5-HT1B/1D agonists known as triptans,
is contraindicated for patients with cardiovascular disease due to
inherent vasoconstrictive activity.2 Thus a novel migraine therapy
with an improved cardiovascular safety profile constitutes an un-
met medical need. Calcitonin gene-related peptide (CGRP) is a 37
amino acid neuropeptide present in both the central and periphe-
ral nervous systems, with a variety of physiological functions that
include nociception, vasodilation, and neurogenic inflammation.3
Growing evidence suggests that CGRP plays a key role during mi-
graine attacks,4 and indeed clinical proof of concept has been
established with two small-molecule CGRP receptor antagonists.
The first was the intravenously-administered olcegepant,5 which
displayed triptan-like efficacy in a phase II migraine clinical trial
with no reported cardiovascular effects.6 More recently the oral-
ly-administered telcagepant demonstated comparable efficacy to
a triptan positive control in a phase III clinical trial with lower
overall adverse event rates.7
we sought to design a compound that would result in a lower
anticipated clinical dose. To this end, targeted areas for improve-
ment were potency (in vitro and in vivo) and pharmacokinetic pro-
file (aqueous solubility, metabolism). We reasoned that
incorporation of heterocyclic rings to the caprolactam core of tel-
cagepant would improve solubility and reduce plasma protein
binding, resulting in lower serum shifts. Initially surveyed was
fused triazole derivative 1 (Fig. 1), and we were gratified to see
comparable potencies in the CGRP binding assay11 (Ki = 1 nM) and
the cell-based functional assays (cAMP IC50 = 1.6 nM, cAMP + 50%
human serum IC50 = 2.4 nM). Additionally, 1 shows improved aque-
ous solubility over telcagepant (fivefold).12 However this does not
translate to improved pharmacokinetics (rat F = 2%), as low perme-
ability may be limiting oral absorption (Papp = 4 Â 10À6 cm/s). Sub-
stitution to a fused imidazole substructure gives 2, which maintains
good potency and shows greatly increased permeability
(Papp = 24 Â 10À6 cm/s), perhaps aiding in the improved oral bio-
availability observed in rats (F = 35%).13
Given the promising attributes of 2, extensive SAR was con-
ducted at both C5 and C4 of the imidazole ring (designated R1
and R2, respectively, in Table 1). Derivatives bearing substituents
at R2 (4–6) were routinely less potent than those with compara-
ble substitution at R1 (2, 7 and 8) as well as the unsubstituted
analog 3. Analogs with hydrogen bond donor/acceptor (8–10) or
aryl/heteroaryl (11 and 12) substituents all proved less potent
than trifluoroethyl analog 2. Only branched alkyl (13) or ether
A series of reports from these laboratories detailed the evolution
of a benzodiazepinone lead8 to an azepanone scaffold9 which ulti-
mately led to the discovery of telcagepant (MK-0974).10 Within
the context of improving upon this development candidate,
⇑
Corresponding author. Tel.: +1 215 993 2269; fax: +1 215 652 7310.
E-mail address: daniel_paone@merck.com (D.V. Paone).
0960-894X/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2010.12.054

2684
D. V. Paone et al. / Bioorg. Med. Chem. Lett. 21 (2011) 2683–2686
Table 1
F3C
O
Imidazole and triazole SAR
N
O
R2
NH
O
NH
R1
N
N
N
N
N
O
F
NH
O
F
NH
N
N
telcagepant (MK-0974)
N
F
Ki = 0.8 nM
cAMP IC50 = 2 nM
+ 50% HS IC50 = 11 nM
rat PK:
F = 20%
F
P
app = 21 x 10-6 cm/s
b
c
Compd
R1
R2
Kia
(nM)
cAMP IC50
(nM)
+HS IC50
(nM)
2
3
H
H
0.14
3.1
0.37
6.7
2.6
19
F3C
N
F3C
N
N
H
N
N
4
H
3.9
7.2
33
F3C
NH
O
NH
5
6
7
H
H
Me
3.2
4
2.2
12
8.7
4.1
26
28
12
HO
Me
H
O
F
F
F
F
2
1
8
9
H
H
2.5
5.7
7.9
12
12
15
HO
Ki = 0.14 nM
cAMP IC50 = 0.37 nM
Ki = 1 nM
cAMP IC50 = 1.6 nM
H
N
+ 50% HS IC50 = 2.6 nM
+ 50% HS IC50 = 2.4 nM
O
rat PK:
F = 2%
rat PK:
F = 35%
10
H
2.8
4.1
7.4
N
Papp = 4 x 10-6 cm/s
Papp = 24 x 10-6 cm/s
11
12
H
H
1.1
1.1
2.6
3.6
34
64
Figure 1. Fused heterocyclic analogs of telcagepant.
N
S
substitution (14 and 15) afforded compounds with comparable or
improved potencies to 2. However, these analogs typically dis-
played poor rat oral bioavailability (F = 4% for 15).
13
14
15
H
H
H
0.69
0.38
0.04
0.62
0.63
0.31
3.2
1.9
O
O
We previously outlined a synthetic approach to polysubstituted
imidazoles developed for acyclic systems.14 Application of this
methodology with slight modifications allowed access to the de-
sired fused imidazoazepanes (Scheme 1). Caprolactam 16 was
available in large quantities as a common intermediate in our var-
ious telcagepant syntheses.15 Treatment with Lawesson’s reagent16
gave the corresponding thioamide 17 in high yield. Mercury-
mediated addition of amino alcohol 18 provided amidine 19, which
after subsequent oxidation with pyridinium dichromate under-
went spontaneous cyclization to imidazole 20.17 Deprotection
and urea coupling of the resulting amine with substituted piperi-
dines gave the final targets 21.18
Poor rhesus PK continued to plague most of the compounds
in this series, including 2, with oral bioavailabilities in the range
of telcagepant (F <2%). Concurrent efforts toward reducing
metabolism of the azabenzimidazolone substructure yielded
two promising spiropiperidine alternatives,20 azaoxindole A and
0.61
Values are a means of at least two experiments; inhibition of [125I]-CGRP
binding to the CGRP receptor (recombinant human CLR/RAMP1).
Inhibition of CGRP-stimulated cAMP production in cells.
Inhibition of CGRP-stimulated cAMP production in the presence of 50% human
serum.
a
b
c
H
H
O
S
N
N
Lawesson's
NHBoc
NHBoc
90 °C
(95%)
F
F
F
F
16
17
R1
R1
HO
NH2
azabenzoxazinone
B (Table 2). These novel CGRP privileged
H
N
N
HO
PDC
CH3CN
(~ 70% 2 steps)
18
structures were coupled with several of the more potent fused
imidazoazepane scaffolds with the aim of improving PK profiles
while maintaining potency. Azaoxindole analogs 22–25 showed
decreased potencies versus the corresponding azabenzimidazol-
ones and a range of rat oral bioavailabilities depending on the
nature of the R1 substituent. The azabenzoxazinone derivatives
were consistently more potent than the azaoxindoles and gener-
ally showed improved PK profiles versus the corresponding aza-
benzimidazolones. Trifluoroethyl analog 26 maintained good
potency and was orally bioavailable in three species, including
rat (F = 33%), dog (F = 28%) and rhesus (F = 17%). However more
thorough characterization of 26 revealed lower than expected
efficacy in our primary in vivo model, the rhesus capsaicin in-
duced dermal vasodilation (CIDV) assay21 (EC90 = 1000 nM).
Additionally, undesirable levels of hERG activity were observed
NHBoc
HgCl2, 55 °C, EtOH
F
F
19
R1
R1
N
N
N
N
1. TFA, CH 2Cl2
NHBoc
NH
R2
2. 4-NO2PhCO2Cl
Et3N, R2 amines
O
F
F
(~ 75% 2 steps)
F
F
20
21
Scheme 1. Synthesis of imidazoazepanes.
(IC50 = 3.3
l
M).
to decrease hERG affinity. Though selectivities were indeed
improved, the spiropiperidines were unable to rescue the poor
PK imparted by the triazole scaffolds. We then focused on attenu-
In an attempt to increase hERG selectivity fused triazoles were
revisited (27 and 28), as these heterocycles had previously shown

D. V. Paone et al. / Bioorg. Med. Chem. Lett. 21 (2011) 2683–2686
2685
Table 2
Imidazole and triazole spiropiperidine SAR
R1
X
O
O
N
O
N
NH
R2
=
N
N
NH
N
NH
O
R2
N
F
F
A
B
R1
X
R2
Kia (nM)
cAMP IC50 (nM)
+HS IC50 (nM)
Rat %F d
hERG IC50/cAMP IC50
b
c
f
Compd
22
CH
A
0.61
0.94
4
47
9300
F3C
23
24
25
CH
CH
CH
A
A
A
0.63
0.62
0.4
1.4
8.6
4.3
1.4
880
9500
9700
F3C
O
1.7
21e
5e
0.82
O
26
27
28
29
CH
N
B
A
B
B
0.07
0.48
0.19
0.14
0.41
2.3
2
33
6e
8700
>13,000
>36,000
8900
F3C
2.7
1.3
1.7
F3C
F3C
O
N
0.83
0.61
<1
15
CH
30
31
32
33
34
35
CH
CH
CH
CH
CH
CH
B
B
B
B
B
B
0.05
0.24
0.86
0.32
0.65
0.25
0.21
0.63
1.5
16
3
32,000
23,000
23,000
6100
O
0.29
0.05
HO
2.9
O
0.16
0.039
0.036
13
F3C
O
O
0.42
0.76
3200
8
17,000
O
a
b
c
d
e
f
Values are a means of at least two experiments; inhibition of [125I]-CGRP binding to the CGRP receptor (recombinant human CLR/RAMP1).
Inhibition of CGRP-stimulated cAMP production in cells.
Inhibition of CGRP-stimulated cAMP production in the presence of 50% human serum.
10 mpk po dosed as a suspension in 1% aqueous methylcellulose.
10 mpk po dosed as a solution in 1:1 imwitor:tween.
Ratio of inhibition of [35S]-MK-499 binding19 to the hERG channel/cAMP IC50
.
ation of hERG selectivity of the fused imidazoles by variation of R1,
specifically targeting oxygenated substituents. A wide array of
ethers was prepared and most were well tolerated. Tertiary methyl
ether 30 showed exceptional in vitro potency (cAMP + 50% HS
IC50 = 0.63 nM) and favorable hERG selectivity (32,000-fold). Subtle
modifications to this particular motif resulted in either potency
loss (alkyl chain extension analogs 32 and 33), decrease in hERG
selectivity (diethyl analog 34), or decrease in oral bioavailability
(THF analog 35).
During larger scale (>10 g) campaigns toward 30, the workup of
the PDC reaction became cumbersome. Therefore an alternate
sequence was developed in which the oxidation step could be cir-
cumvented by use of amino ketone 37 (Scheme 2). This synthon is
1. KOt-Bu, Boc2NH, THF
HO
NH3Cl
37
Br
2. KMnO4, acetone/H2O/AcOH
3. HCl, MeOH
O
36
Further evaluation of 30 in our in vivo model indicated im-
proved efficacy (rhesus CIDV EC90 = 300 nM). Though permeability
(Papp = 24 Â 10À6 cm/s) and aqueous solubility (sol pH
3.0 = 1.1 mg/mL) are respectable, oral bioavailability is only mod-
erate in rat (F = 16%), and low in dog (F = 10%) and rhesus
(F = 5%). However, 30 shows increased microsomal stability in hu-
man liver microsomes vs. preclinical species. The major metabolite
is tertiary hydroxyl derivative 31, which is itself a highly potent
antagonist (cAMP + 50% HS IC50 = 1.5 nM) with similar hERG selec-
tivity (23,000-fold) and in vivo efficacy (rhesus CIDV
EC90 = 330 nM) but is not orally bioavailable. Because substantial
levels of alcohol 31 are observed upon dosing of ether 30 (rat
F–OH = 26%, dog F–OH = 14%, rhesus F–OH = 29%),22 31 would be
expected to contribute significantly to clinical efficacy. Taken
together 30 has a lower anticipated clinical dose than telcagepant.
In addition, 30 shows >6000-fold selectivity in a panel of assays
representing over 160 receptors, transporters, and enzymes, and
was selected as a preclinical candidate (MK-2918).
(80% 3 steps)
HO
H
N
N
S
N
37
MsOH
NHBoc
NHBoc
HgCl2, 60 °C
Et3N, EtOH
MeOH
60 °C
(95%)
F
F
(75%)
F
F
17
38
O
O
HN
NH
O
O
N
N
N
N
N
40
O
O
NH2
NH
N
O
4-NO2PhCO2Cl
Et3N, THF
NH
F
F
N
(80%)
F
F
39
30 (MK-2918)
Scheme 2. Synthesis of MK-2918.

2686
D. V. Paone et al. / Bioorg. Med. Chem. Lett. 21 (2011) 2683–2686
6. (a) Olesen, J.; Diener, H.-C.; Husstedt, I. W.; Goadsby, P. J.; Hall, D.; Meier, U.;
Pollentier, S.; Lesko, L. N. Engl. J. Med. 2004, 350, 1104; (b) Doods, H.;
Hallermayer, G.; Wu, D.; Entzeroth, M.; Rudolf, K.; Engel, W.; Eberlein, W. Br. J.
Pharm. 2000, 129, 420.
7. (a) Ho, T. W.; Ferrari, M. D.; Dodick, D. W.; Galet, V.; Kost, J.; Fan, X.;
Leibensperger, H.; Froman, S.; Assaid, C.; Lines, C.; Koppen, H.; Winner, P. K.
Lancet 2008, 372, 2115; (b) Ho, T. W.; Mannix, L.; Fan, X.; Assaid, C.; Furtek, C.;
Jones, C.; Lines, C.; Rapoport, A. Neurology 2008, 70, 1305.
8. Williams, T. M.; Stump, C. A.; Nguyen, D. N.; Quigley, A. G.; Bell, I. M.;
Gallicchio, S. N.; Zartman, C. B.; Wan, B.-L.; Della Penna, K.; Kunapuli, P.; Kane,
S. A.; Koblan, K. S.; Mosser, S. D.; Rutledge, R. Z.; Salvatore, C.; Fay, J. F.; Vacca, J.
P.; Graham, S. L. Bioorg. Med. Chem. Lett. 2006, 16, 2595.
readily accessed in 3 steps from bromide 36 after alkylation with
Boc-protected ammonia, oxidation with KMnO4 and acidic depro-
tection. Steric hindrance of the adjacent quaternary center
undoubtedly aids in minimizing self condensation of 37 during
the addition to thioamide 17, allowing for clean amidine forma-
tion. Continued heating then provides imidazole 38 directly.23 Sol-
volysis under strongly acidic conditions in methanol yields
deprotected methyl ether 39. Finally urea coupling with aza-
benzoxazinone spiropiperidine 40 affords MK-2918.
In summary, fused heterocyclic derivatives of telcagepant were
explored with the overall goal of achieving a lower projected clin-
ical dose. The imidazoazepane scaffold was identified as a potent
replacement for the caprolactam, with permeabilities sufficient
for enabling good oral exposures. Utilization of the azabenzoxazi-
none spiropiperidine decreased metabolism and resulted in im-
proved rhesus PK profiles, and the tertiary methyl ether was
discovered as a potency-enhancing substituent which mitigated
hERG activity. These investigations culminated in the selection of
30 (MK-2918) as a preclinical candidate for the treatment of acute
migraine.
9. Shaw, A. W.; Paone, D. V.; Nguyen, D. N.; Stump, C. A.; Burgey, C. S.; Mosser, S.
D.; Salvatore, C. A.; Rutledge, R. Z.; Kane, S. A.; Koblan, K. S.; Graham, S. L.;
Vacca, J. V.; Williams, T. M. Bioorg. Med. Chem. Lett. 2007, 17, 4795.
10. (a) Paone, D. V.; Shaw, A. W.; Nguyen, D. N.; Burgey, C. S.; Deng, Z. J.; Kane,
S. A.; Koblan, K. S.; Salvatore, C. A.; Hershey, J. C.; Wong, B.; Roller, S. G.;
Miller-Stein, C.; Graham, S. L.; Vacca, J. P.; Williams, T. M. J. Med. Chem.
2007, 50, 5564; (b) Salvatore, C. A.; Hershey, J. C.; Corcoran, H. A.; Fay, J.
F.; Johnston, V. K.; Moore, E. L.; Mosser, S. D.; Burgey, C. S.; Paone, D. V.;
Shaw, A. W.; Graham, S. L.; Vacca, J. P.; Williams, T. M.; Koblan, K. S.; Kane,
S. A. JPET 2008, 324, 416.
11. For the details of these assays see: Bell, I. M.; Graham, S. L.; Williams, T. M.;
Stump, C. A. WO 2004/087649.
12. Aqueous solubilities at pH 7.4: telcagepant = 30 lg/mL; triazole 1 = 158 lg/mL.
13. As a general trend, fused heterocycles in this series with low permeability
resulted in low oral exposures.
14. Paone, D. V.; Shaw, A. W. Tetrahedron Lett. 2008, 49, 6155.
15. Burgey, C. S.; Paone, D. V.; Shaw, A. W.; Deng, J. Z.; Nguyen, D. N.; Potteiger, C.
M.; Graham, S. L.; Vacca, J. P.; Williams, T. M. Org. Lett. 2008, 10, 3235.
16. Scheibye, S.; Pederson, B. S.; Lawesson, S. O. Bull. Soc. Chim. Belg. 1978, 87,
229.
Acknowledgments
We thank the NMR, MS, PK support and analytical groups for
their valuable assistance. We also thank Jim Hershey’s group for
performing the rhesus CIDV studies.
17. For additional synthetic details, see: Burgey, C. S.; Paone, D. V.; Shaw, A. W.;
Nguyen, D. N.; Deng, Z. J.; Vacca, J. P.; Selnick, H. G.; Williams, T. M.; Potteiger,
C. WO 2006/044504 A2.
18. All final compounds were characterized by 1H NMR, HRMS, and HPLC.
19. Raab, C. E.; Butcher, J. W.; Connolly, T. M.; Karczewski, J.; Yu, N. X.; Staskiewicz,
S. J.; Liverton, N.; Dean, D. C.; Melillo, D. G. Bioorg. Med. Chem. Lett. 2006, 16,
1692.
References and notes
1. Goadsby, P. J.; Lipton, R. B.; Ferrari, M. D. N. Engl. J. Med. 2002, 346, 257.
2. (a) Wackenfors, A.; Jarvius, M.; Ingemansson, R.; Edvinsson, L.; Malmsjo, M. J.
Cardiovasc. Pharmacol. 2004, 45, 476; (b) Dodick, D.; Lipton, R. B.; Martin, V.;
Papademetriou, V.; Rosamond, W.; VanDenBrink, A. M.; Loutfi, H.; Welch, K. M.;
Goadsby, P. J.; Hahn, S.; Hutchinson, S.; Matchar, D.; Silberstein, S.; Smith, T. R.;
Purdy, R. A.; Saiers, J. Headache 2004, 44, 414.
3. (a) van Rossum, D.; Hanisch, U.-K.; Quirion, R. Neurosci. Biobehav. Rev. 1997, 21,
649; (b) Poyner, D. R. Pharmacol. Ther. 1992, 56, 23.
4. (a) Durham, P. L. Headache 2008, 48, 1269; (b) Edvinsson, L. Expert Opin. Ther.
Targets 2003, 7, 377; (c) Lassen, L. H.; Haderslev, P. A.; Jacobsen, V. B.; Iversen,
H. K.; Sperling, B.; Olesen, J. Cephalalgia 2002, 22, 54.
20. Burgey, C. S.; Potteiger, C. M.; Deng, J. Z.; Mosser, S. D.; Salvatore, C. A.; Yu, S.;
Roller, S.; Kane, S. A.; Vacca, J. P.; Williams, T. M. Bioorg. Med. Chem. Lett. 2009,
19, 6368.
21. Hershey, J. C.; Corcoran, H. A.; Baskin, E. P.; Salvatore, C. A.; Mosser, S.;
Williams, T. M.; Koblan, K. S.; Hargreaves, R.; Kane, S. A. Regul. Pept. 2005, 127,
71.
22. Estimation of the oral bioavailability of alcohol metabolite 31 upon dosing of
ether 30 is calculated as follows: F–OH = (po normalized AUC 31 upon dosing of
30)/ (iv normalized AUC 31 upon dosing of 31).
23. Additions of both amino alcohols and amino ketones to thioamides were rapid,
typically complete in <10 min. Cyclization of the intermediate amidine to 38
required and additional 18 h.
5. Rudolf, K.; Eberlein, W.; Wolfhard, E.; Pieper, H.; Entzeroth, M.; Hallermayer,
G.; Doods, H. J. Med. Chem. 2005, 48, 5921.