Design and synthesis of a novel, achiral class of highly potent and selective, orally active neurokinin-1 receptor antagonists

Article abstract of DOI:10.1016/j.bmcl.2005.11.047

The discovery of a novel, achiral pyridine class of potent and orally active neurokinin-1 (NK₁) receptor antagonists is described. The evaluation of this class is briefly outlined, leading to the identification of netupitant 21 and befetupitant 29, two new proprietary chemical entities with high affinity and excellent CNS penetration.

Full text of DOI:10.1016/j.bmcl.2005.11.047

Bioorganic & Medicinal Chemistry Letters 16 (2006) 1362–1365
Design and synthesis of a novel, achiral class of highly potent
and selective, orally active neurokinin-1 receptor antagonists
Torsten Hoffmann,* Michael Bo¨s, Heinz Stadler, Patrick Schnider,
Walter Hunkeler, Thierry Godel, Guido Galley, Theresa M. Ballard,
Guy A. Higgins, Sonia M. Poli and Andrew J. Sleight
F. Hoffmann-La Roche Ltd, Pharma Research, CH-4070 Basel, Switzerland
Received 11 September 2005; revised 14 November 2005; accepted 14 November 2005
Available online 5 December 2005
Abstract—The discovery of a novel, achiral pyridine class of potent and orally active neurokinin-1 (NK1) receptor antagonists is
described. The evaluation of this class is briefly outlined, leading to the identification of netupitant 21 and befetupitant 29, two
new proprietary chemical entities with high affinity and excellent CNS penetration.
Ó 2005 Elsevier Ltd. All rights reserved.
Moreover, recent clinical trials have demonstrated an
important therapeutic application for NK1 receptor
antagonists in the control of cancer chemotherapy-in-
duced nausea and vomiting and in the treatment of
mood disorders such as anxiety and depression.4 MK-
869 (aprepitant) was the first NK1 receptor antagonist
to receive marketing approval. In March 2003, it was
approved by the FDA for the treatment of chemothera-
py-induced nausea and vomiting.5
Neurokinin (NK) receptors belong to the family of G-
protein coupled receptors and can be divided into three
subtypes: NK1, NK2 and NK3. The endogenous ligand
for NK1 receptors is the 11 amino acid neuropeptide
substance P.1 Following the discovery of the first non-
peptide NK1 receptor antagonist CP-96,345 in 1991,2 a
remarkable number of small molecule NK1 receptor
antagonists were identified by many pharmaceutical
companies in the last decade.3
Following random screening of our corporate library, a
unique compound cluster of achiral aryl-substituted
isoxazole derivatives was discovered which showed
moderate affinity at the human NK1 receptor (e.g., 1,
Ki = 15 nM). We hypothesized that 1 might be able to
adopt conformations which are in agreement with the
pharmacophore model described by researchers from
Merck.6 Moreover this structural class, which displays
simple achiral features, had previously not been de-
scribed in the patent literature either as an NK1 lead
structure or as a pharmaceutically active chemical class
per se. Consequently, we decided to embark on the opti-
mization of this compound class during a focused lead
generation and optimization program.
CF3
O
CF3
O
N
O
NH
N
H
N
F
O
N
N
H
CP-96,345
MK-869 (aprepitant)
Cl
Cl
Cl
H
N
N
N
O
N
Cl
1
During the initial steps of our optimization efforts, it
was discovered that the 5-membered heterocyclic isoxaz-
ole backbone of 1 could be replaced by a simple isocyclic
benzene core (2a–e Table 1). Subsequently the N-(4-pyr-
idyl)-hydrazone substituent, which was associated
with potential toxicity liabilities, was replaced by the
Keywords: NK1 receptor antagonist; G-protein coupled receptor; Drug
design and synthesis.
*
Corresponding author. Tel.: +41 61 6870402; fax: +41 61 6888367;
e-mail: torsten.hoffmann@roche.com
0960-894X/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2005.11.047

T. Hoffmann et al. / Bioorg. Med. Chem. Lett. 16 (2006) 1362–1365
Table 1. hNK1 binding affinity of biphenyl derivatives 2 and 3 Table 2. Optimization of the 3-atom linker L
1363
R1
R2
R1
O
Cl
H
N
L
CF3
CF3
N
N
N
Cl
CF3
CF3
2a-e
3a-d
Compound
L
hNK1 Kia (nM)
O
O
O
O
Compound
R1
H
R2
hNK1 Kia (nM)
3a
51
N
2a
2b
2c
2d
2e
3a
3b
3c
3d
H
933
316
174
49
CH3
Cl
H
4
5
6
7
8
9
>10,000
H
N
H
Br
Cl
H
Cl
—
—
—
—
93
51
H
295
>10,000
562
N
CH3
Cl
17
22
Br
45
N
H
a Displacement of [3H]-labelled substance P from cloned human
receptor expressed in CHO cells.10 Values are means of two experi-
ments with maximal standard deviations below 0.25 logarithmic
units.
N
N
O
O
68
well-documented N-[3,5-bis(trifluoromethyl)benzyl]-N-
methyl-carboxamide group (3a–d Table 1).7 This modi-
fication also led to a significant increase in binding affin-
ity. Notably the potency at the NK1 receptor is
markedly enhanced by increasing the steric demand of
the ortho-substitution (R1 and R2 in derivatives 2 and
3) and thus the torsion angle of the biaryl system.8
H
N
1,380
O
N
10
11
724
372
N
H
The unsubstituted biphenyl derivative 3a was selected
as a starting point to systematically explore a set of
differently functionalized 3-atom linkers L (Table 2).
N-Methyl substitution, which leads to orthogonal
twist of the carboxamide plane with respect to the
aromatic system,9 was found to be crucial for recep-
tor binding (cf. 3a and 4). Geminal dimethyl substitu-
tion (5 and 6) did not lead to increased NK1 receptor
affinity. The N-methyl carboxamide linker can be re-
placed by the inverse amide linker with 10-fold loss
of potency (cf. 3a and 7). However, in case of the in-
verse amide linker, geminal dimethyl substitution
leads to a 10-fold increase in NK1 receptor affinity
and comparable potency (cf. 3a and 8). Again, N-
methyl substitution was found to be important for
receptor binding (cf. 8 and 9). Amine (10 and 11)
and ether (12) linker substitutions were not preferred
over N-methyl-substituted carboxamides or inverse
carboxamides.
12
224
O
a Displacement of [3H]-labelled substance P from cloned human
receptor expressed in CHO cells.10 Values are means of two experi-
ments with maximal standard deviations below 0.25 logarithmic
units.
R''
R''
R''
R''
R''
N
N
R2N
R2N
N
N
A
B
C
D
E
R''=
L
CF3
R''
R''
R''
R''
N
CF3
N
N
N
N
R2N
N
NR2
NR2
NR2
F
G
H
I
Biphenyl derivatives 3a and 8, displaying the N-methyl-
substituted carboxamide and inverse amide linker series,
were selected for further optimization, while conserving
the 3,5-bis(trifluoromethyl)phenyl group, which is char-
acteristic of many potent NK1 receptor antagonists.3 In
parallel, 9 compound classes were explored with ortho-
tolyl substitution and different amine substituents
NR2: Isocyclic biphenyl derivatives A, pyridine classes
B–G and pyrimidine classes H and I (Fig. 1).
Figure 1. Compound classes A–I were explored with ortho-tolyl
substitution and different amine substituents NR2.
In all 9 classes A–I, potent NK1 receptor antagonists
were discovered.10 The structural variations allowed
fine-tuning of: NK1 receptor affinity and subtype selec-
tivity; physicochemical properties, such as pKa value,

1364
T. Hoffmann et al. / Bioorg. Med. Chem. Lett. 16 (2006) 1362–1365
Table 3. Potent NK1 receptor antagonists from classes A–I (Fig. 1)
NR2 = N-methylpiperazinyl
substituent and an ortho-tolyl group. It is worthwhile
to note that without exception all derivatives display
low-nM to sub-nM affinity at the NK1 receptor. Fur-
thermore, NK1 receptor antagonists 1–31 were found
to be at least 100-fold sub-type selective over NK2 and
NK3 receptors as assessed by displacement of the respec-
tive radiolabelled endogenous ligand from cloned NK2
and NK3 receptors in CHO cells, respectively. The func-
tional antagonistic activity of these molecules was dem-
onstrated by using a standard fluorescence imaging plate
reader (FLIPR)-based assay measuring Ca2+ flux
in vitro.
Compound
Class
L
hNK1 Kia (nM)
O
O
13
A
1.4
N
N
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
B
B
C
C
D
D
E
E
F
8.5
3.4
4.6
3.3
3.8
4.0
0.58
0.95
1.7
6.2
2.0
3.5
1.1
1.7
0.44
N
O
O
N
Preferred compounds were tested for their potential to
pharmacologically block central NK1 receptors in ger-
bils. Animals were orally pretreated (2 h) with the test
compound followed by central injection (icv) of the
NK1-selective receptor agonist GR73632.11 Excellent
oral antagonist activity was observed for the pyridine
derivatives from class E as exemplified for netupitant
(21), befetupitant (29) and structurally closely related
compounds such as 30 and 31 (Table 4).12
N
O
O
N
N
O
Pyridine derivatives from class E were synthesized as
outlined in Scheme 1. 2-Chloro-5-nitropyridine 32 was
treated with the respective amine to yield amine-substi-
tuted nitropyridines 33. The nitro group was catalytical-
ly reduced with hydrogen and the resulting aniline
immediately trapped with pivaloyl chloride to yield
derivatives 34. The pivaloyl amide group served as
directing group for ortho-lithiation with an excess of
O
N
N
O
N
O
O
N
Table 4. Pyridines: amine substitution in the 6-position
G
G
H
H
I
N
O
N
CF3
R1
O
N
O
N
N
R2
CF3
Compound
NR1R2
hNK1
Gerbil foot tappingb
Kia (nM) (ID50 mg/kg po)
N
O
O
N
N
21 (netupitant)
0.95
1.0
0.5
0.2
0.5
0.8
N
N
29 (befetupitant)
O
N
I
O
N
30
31
0.40
1.3
a Displacement of [3H]-labelled substance P from cloned human
receptor expressed in CHO cells.10 Values are means of two experi-
ments with maximal standard deviations below 0.25 logarithmic
units.
HO
O
N
S
O
a Displacement of [3H]-labelled substance P from cloned human
receptor expressed in CHO cells.10 Values are means of two experi-
ments with maximal standard deviations below 0.25 logarithmic
units.
b Inhibition of agonist (GR73632)-induced foot tapping behaviour in
gerbils after oral pretreatment of animals with test compounds 2 h
prior to intracerebroventricular (icv) administration of the agonist.11
solubility and permeability; as well as DMPK parame-
ters, such as half-life, clearance, volume of distribution,
phase I metabolism and blood brain barrier permeabili-
ty. Table 3 exemplifies a selection of 16 derivatives
(13–28) from structural classes A–I displaying both car-
boxamide linkers L, with N-methylpiperazine as NR2

T. Hoffmann et al. / Bioorg. Med. Chem. Lett. 16 (2006) 1362–1365
1365
H
Acknowledgments
NO2
N
NO2
i
ii, iii
O
R2N
N
R2N
N
The authors thank Ms. Catherine Muller, Ms. Stefanie
Sa¨nger and Mr. Alain Rudler for biological testing.
Cl
N
32
33
34
I
References and notes
H
H
N
N
iv
v
O
O
1. Almeida, T. A.; Rojo, J.; Nieto, P. M.; Pinto, F. M.;
Hernandez, M.; Martin, J. D.; Candenas, M. L. Curr.
Med. Chem 2004, 11, 2045, and references cited therein.
2. Snider, R. M.; Constantine, J. W.; Lowe, J. A., 3rd;
Longo, K. P.; Lebel, W. S.; Woody, H. A.; Drozda, S. E.;
Desai, M. C.; Vinick, F. J.; Spencer, R. W. Science 1991,
251, 435.
R2N
N
R2N
N
35
36
vi, vii
viii
NH
N
CF3
3. Albert, J. S. Expert Opin. Ther. Patents 2004, 14, 1421,
and references cited therein.
4. Duffy, R. A. Expert Opin. Emerg. Drugs 2004, 9, 9, and
references cited therein.
O
R2N
N
R2N
N
38
CF3
37
Scheme 1. Reagents and conditions: (i) R2NH, THF, reflux; (ii) H2
(1 atm), Pd/C, MeOH, 45 °C; (iii) PivCl, Et3N, THF, rt; (iv) nBuLi,
TMEDA, 2,2,6,6-tetramethylpiperidine, THF, À78 to À30 °C; I2, À78
to 0 °C; (v) o-tolyl-B(OH)2, (PPh3)4Pd, Na2CO3, toluene, 80 °C; (vi)
3 N HCl, reflux; (vii) HC(OCH3)3, TFA (cat.), 130 °C; LiAlH4, THF,
0 °C to rt; (viii) [3,5-(CF3)2C6H3]C(CH3)2COCl, iPr2NEt, CH2Cl2,
0 °C to reflux.
5. Dando, T. M.; Perry, C. M. Drugs 2004, 64, 777.
6. Swain, C. J.; Sewart, E. M.; Cascieri, M. A.; Fong, T. M.;
Herbert, R.; MacIntyre, D. E.; Merchant, K. J.; Owen, S.
N.; Owens, A. P.; Sabin, V.; Teall, M.; VanNiel, M. B.;
Williams, B. J.; Sadowski, S.; Strader, C.; Ball, R. G.;
Baker, R. J. Med. Chem. 1995, 38, 4793.
7. Natsugari, H.; Ikeura, Y.; Kiyota, Y.; Ishichi, Y.;
Ishimaru, T.; Saga, O.; Shirafuji, H.; Tanaka, T.;
Kamo, I.; Doi, T.; Otsuka, M. J. Med. Chem. 1995,
38, 3106.
8. Eliel, E. L.; Wilen, S. H.; Mander, N. L. Stereochemistry
of Organic Compounds; Wiley: New York, 1994, p 1143ff.
9. Ahmed, A.; Bragg, R. A.; Clayden, J.; Lai, L. W.;
McCarthy, C.; Pink, J. H.; Westlund, N.; Yasin, S. A.
Tetrahedron 1998, 54, 13277.
10. (a) Bo¨s, M.; Branca, Q.; Galley, G.; Godel, T.; Hoffmann,
T.; Hunkeler, W.; Schnider, P.; Stadler, H. EP 1035115,
2000; (b) Bo¨s, M.; Branca, Q.; Galley, G.; Godel, T.;
Hoffmann, T.; Hunkeler, W.; Schnider, P.; Stadler, H. WO
2000050401; (c) Bo¨s, M.; Branca, Q.; Galley, G.; Godel,
T.; Hoffmann, T.; Hunkeler, W.; Schnider, P.; Stadler, H.
WO 2000053572; (d) Bo¨s, M.; Branca, Q.; Galley, G.;
Godel, T.; Hoffmann, T.; Hunkeler, W.; Schnider, P.;
Stadler, H. WO 2000073278; (e) Bo¨s, M.; Branca, Q.;
Galley, G.; Godel, T.; Hoffmann, T.; Hunkeler, W.;
Schnider, P.; Stadler, H. WO 2000073279.
nBuLi in the presence of TMEDA at À78 °C. The inter-
mediate lithium organic pyridine derivative was trapped
with an excess of iodine at this temperature to give iod-
opyridine derivatives 35.13 Suzuki coupling with ortho-
tolyl boronic acid under standard conditions gave
4-arylpyridine derivatives 36. The pivaloyl group was
cleaved under acidic conditions and the resulting 3-
aminopyridines were mono-methylated using the
reductive ortho-ester standard procedure to give 37.
Mono-methylated aminopyridines were acylated with
2-[3,5-bis(trifluoromethyl)phenyl]-2-methyl-propionyl
chloride to yield NK1 receptor antagonists 38 of class E
with the inverse amide linker. Detailed experimental
procedures including the first description of 2-(3,5-bis-
trifluoromethyl-phenyl)-2-methyl-propionyl chloride
have been published in series of patent
applications.10,12
a
11. Bristow, L. J.; Young, L. Eur. J. Pharmacol. 1994,
253, 245.
12. (a) Ballard, T. M.; Higgins, G. A.; Hoffmann, T.; Poli, S.
M.; Sleight, A. J. EP 1103545, 2001.; (b) Hoffmann, T.;
Schnider, P.; Stadler, H. WO 2002008232.; (c) Ballard, T.
M.; Hoffmann, T.; Poli, S. M.; Schnider, P.; Sleight, A. J.
WO 2003011860.
13. The procedure was optimized by Dr. Franc¸ois Montavon,
F. Hoffmann-La Roche Ltd, Kilolaboratories Basel,
Switzerland.
In conclusion, rapid multidimensional optimization of
the aryl-substituted isoxazole screening hit cluster (e.g.,
1) gave rise to 9 novel achiral structural classes (A–I)
of highly potent and selective NK1 receptor antago-
nists.10 Pyridine derivatives from class E have been
shown to potently inhibit NK1 agonist-induced foot tap-
ping behaviour in gerbils after oral administration.12