Current Drug Targets, 2006, 7, 975-992 975
` 1389-4501/06 $50.00+.00 © 2006 Bentham Science Publishers Ltd.
`Tachykinin Receptors Antagonists: From Research to Clinic
`Laura Quartara* and Maria Altamura
`Chemistry Department, Menarini Ricerche SpA, Florence, Italy
`Abstract: In this chapter it is described how, starting from different approaches and through extensive medicinal chemistry studies, sev-
`eral discovery compounds were optimized and reached the development stage.
`The first tachykinin receptor antagonist to reach the market in 2003 for chemotherapy-induced emesis has been aprepitant. Other clinical
`candidates (for central nervous system disorders: osanetant, talnetant and saredutant; for irritable bowel syndrome: nepadutant and
`saredutant) are in advanced clinical phase.
`The clinical studies reported in the literature and the destiny of the clinical candidates, where available, will be reviewed.
`Key Words: Tachykinin, antagonist, clinical development, medicinal chemistry, emesis, irritable bowel syndrome, schizophrenia.
`1. INTRODUCTION
`There are more than 500 patents claiming tachykinins (TKs)
`antagonists and their different uses as drugs. Most of them have
`been applied starting from 1990 to date, in correspondence with the
`discovery of the first non peptide antagonists. The patent applica-
`tions, which will be not reviewed here but are a clear sign of indus-
`trial interest, were filed mainly in the years 1994-1996. Almost 80
`% of the existing patents are related to NK-1 antagonists. Patent
`applications describing novel structures or claiming different uses
`(veterinary uses included) and drugs combinations are still appear-
`ing.
`Historically, substance P and NK-1 tachykinin receptors were
`the first to be described. For this or for the number and importance
`of pathologies in which NK-1 receptor could be involved, the first
`tachykinin antagonist to reach the market, more or less 30 years
`after the isolation of substance P, was an NK-1 receptor selective
`antagonist.
`In this chapter it will be described how, starting from different
`approaches and through extensive medicinal chemistry studies,
`several discovery compounds were optimized and reached the de-
`velopment stage. A description of the clinical studies reported in the
`literature and of the destiny of the clinical candidates, where avail-
`able, will be reported. Noteworthy, many companies implemented
`studies on this field through a period of 15 years, and the panorama
`of pharmaceutical companies changed dramatically during this
`time. Therefore, the discontinuation of the development of a drug
`candidate is not necessarily related to a bad performance in the
`preclinical or clinical phase.
`2. NK-1 RECEPTOR SELECTIVE ANTAGONISTS
`The first NK-1 receptor antagonists were developed in the 80s
`and derived from SP sequence. Although they have been quite use-
`ful as pharmacological tools, the drawbacks typical of peptide com-
`pounds as pharmaceuticals (complex structure, poor oral bioavail-
`ability, metabolic instability) let companies to decide to proceed in
`the search for non peptide antagonists. Therefore, no advanced
`development of peptide compounds was performed and the subject
`will be not reviewed here (see [1] and references therein).
`The era of non-peptide antagonists for the tachykinin receptors
`began in 1991, when three distinct research groups divulgated their
`results [1, 2]. As already mentioned since these original reports,
`increasing number of publications and patents describing new
`structures, has appeared. The field of non peptide NK-1 receptor
`*Address correspondence to this author at the Menarini Ricerche SpA, Via
`Sette Santi 3, Florence I-50131, Italy; E-mail:lquartara@menarini-ricerche.it
`antagonists developed as an area of intense competition to reach its
`top in late 90s. As the search for non peptide NK-1 antagonists
`reached its goal, i.e., compounds suitable for clinical development,
`the competition moved to the proof of concept for the therapeutic
`uses claimed. Despite early experiments made by Pfizer, Merck was
`the first to reach the market.
`Since the panorama of NK-1 receptor antagonists is quite com-
`plex and company related in particular in the final phase of devel-
`opment, it seemed convenient for a better comprehension to de-
`scribe the field analyzing each company route, even if some of
`them explored and are continuing to divulgate and claim diverse
`structural series of compounds. Therefore, each company will be
`described in a separated paragraph, reporting briefly the partially
`overlapping structures of discovery antagonists and more in detail
`the compounds which reached the development phase.
`Among the first series divulgated, Eastman Kodak and Sterling
`Winthrop described two classes of steroids derived from the
`screening of natural products chemical collections [3-5]. The series
`were abandoned for their structural complexity and toxicity of
`compounds when administered in vivo.
`Rhone-Poulenc
`The structural motif of Rhone-Poulenc NK-1 receptor antago-
`nists was the perhydroisoindolone ring, present in a series of com-
`pounds selected by screening of a chemical collection and divul-
`gated in 1991 [6]. The prototype of the series, RP-67580 (Fig. ( 1)),
`shows three aromatic residues, the quite rigid perhydroisoindolone
`bicycle and an amidine basic moiety, with strictly defined configu-
`ration at stereogenic centers for good pharmacological activity. RP-
`67580 shows nanomolar affinity for rat NK-1 receptor and good
`antagonist activity in vitro and in vivo . Several SAR studies were
`performed on this class of compounds (see [1] for review). The
`introduction of a further aromatic ring yielded RPR-100893 (Fig.
`(1)) [7], selective for the human NK-1 receptor (IC 50=13 nM on
`human IM9 cells). This higher affinity for human receptor was in
`some extent counterbalanced by the increased structural complex-
`ity. RP-100893 (dapitant) possesses good activity in vivo on plasma
`extravasation models in rats and in models of pain after oral ad-
`ministration. It was developed up to phase II for the treatment of
`migraine (1994) [8], but then discontinued, as happened to other
`NK-1 antagonists, which have been found not active in humans for
`this indication (vide infra).
`Pfizer
`In 1991, Pfizer group disclosed the structure of CP-96345 (Fig.
`(2)) [9]. The compound derived from the screening of a chemical
`HELSINN EXHIBIT 2061
`Azurity Pharmaceuticals, Inc. v. Helsinn Healthcare S.A.
`IPR2025-00948
`Page 1 of 18
`
`
`
`
`
`
`
`976 Current Drug Targets, 2006, Vol. 7, No. 8 Quartara and Altamura
`collection and a first optimization of the lead [10]. The relatively
`simple structure of CP-96345 (a rigid quinuclidine scaffold con-
`taining a basic nitrogen atom, a benzhydryl moiety and an o-
`methoxy-benzylamine group) was the starting point for several
`SAR studies performed from different companies, aimed to amelio-
`rate the pharmacological and pharmacokinetic properties and to
`obtain patentability (see [1] for review). Among the analogs syn-
`thetized, most aimed to mask the strongly basic quinuclidine nitro-
`gen, which was considered responsible for the undesired activity at
`the Ca2+ channels [11, 12].
`Fig. (1).
`A satisfactory improvement in physico-chemical properties was
`achieved with CJ-11,974, a close analog of CP-96,345 bearing an
`isopropyl group on the methoxybenzyl ring. CJ-11,974 (ezlopitant,
`Fig. ( 2)) shows subnanomolar affinity for NK-1 human receptor
`and is able to block cisplatin-induced emesis in the ferret at a dose
`of 3 mg/kg. Its action in vivo seems due to the parent compound
`and one active metabolite [13]. Ezlopitant was developed up to
`phase II in US and Europe for chemotherapy-induced emesis, but
`although the compound resulted well tolerated and effective in
`controlling emesis, it was less effective in nausea and development
`was discontinued. A pilot study was conducted in IBS patients in
`2000 to check the efficacy in the relief of symptoms [13]. Despite
`the encouraging results, the development for this indication was not
`carried further.
`In order to simplify CP-96345 structure, it was settled a phar-
`macophoric model including the proximity and the relative orienta-
`tion among the benzylamino portion, the central saturated ring and
`one of the phenyls in the benzhydryl moiety. Following conforma-
`tional considerations, the quinuclidine ring and the benzhydryl moi-
`ety were replaced by a piperidine and a benzyl group, respectively,
`leading to CP-99994 (Fig. (2)) [14]. The X-ray crystal structure and
`the molecular modeling studies performed on CP-99994 confirmed
`the searched parallel orientation of the two aromatic rings. CP-
`99994 retains high affinity for the human NK-1 receptor (IC 50=0.17
`nM on human IM9 cells). Many other SAR studies were performed
`to investigate further the structural requirements for NK-1 receptor
`affinity, and different kinds of constraints were introduced to look
`for a univocal active conformation [15-20 and 1 for review]. De-
`spite the good results obtained in terms of affinity retention, no
`advantages were achieved for in vivo activity, and the constrained
`structures resulted in general more complex to synthesize. There-
`fore, different studies were devoted to the search of more favorable
`physico-chemical or pharmacological properties which could fa-
`cilitate the development phase. Parameters like undesired Na + or
`Ca2+ channels blockade, blood brain barrier (BBB) penetration and
`oral bioavailability were considered crucial to reach good results in
`clinical phase [21]. But in the meantime companies like Merck,
`Glaxo, Novartis etc profited from the acquired knowledge on the
`field to build their own series of antagonists (vide infra).
`CP-99994 was characterized in a number of pharmacological
`models [22 for review]. The anti emetic activity was checked in
`dogs and ferrets [23]. The inhibition of induced plasma extravasa-
`tion, mucus secretion and bronchoconstriction proved in animal
`models [24] appeared not reproducible in asthmatic patients [25].
`CP-99994 alleviated dental pain in humans, but with a short dura-
`tion of action when administered intravenously at 0.75 mg/kg [26].
`Due to the poor oral bioavailability (< 10 %) [27], the phase II
`clinical trials of CP-99994 in the US were discontinued.
`As mentioned before, several studies were carried on at Pfizer
`to ameliorate the physico-chemical properties of its early antago-
`nists. The introduction of a trifluoromethoxy group in the o-
`methoxybenzyl ring of CP-99994 gave a compound (CP-122721,
`Fig. (2)) [28] with comparable receptor affinity but enhanced up to
`50-fold in vivo potency [29]. CP-122721 blocks plasma protein
`extravasation elicited by aerosolized capsaicin in guinea-pig lungs
`with an ID 50= 0.01 mg/Kg p.o., and antagonizes [Sar 9, Met(O 2)11
`]SP induced locomotor activity in guinea-pigs with an ID 50=0.2
`mg/Kg [28].
`Fig. (2).
`It underwent phase II clinical trials for the treatment of depres-
`sion, emesis and inflammatory diseases including asthma and irrita-
`ble bowel syndrome (IBS). CP-122721 (p.o.) was evaluated for the
`treatment of postoperative nausea and vomiting [30], .in dose-
`ranging trials for acute and delayed emesis following treatment with
`cisplatin [31], and gave interesting trial results in patients with ad-
`N
`H
`H
`O
`NH
`H3C O
`N
`H
`H
`O
`H3C O
`H3C O
`H3C
`RP-67580
`RPR-100893
`(
`dapitant)
`N
`NH
`H3C O
`NH
`NH
`H3C O
`N
`H
`NH
`H3C
`O
`OCF3
`N
`NH
`H3C O
`CH3
`CH3
`CP-96345
`CP-99994 CP-122721
`CJ-11,974
`(
`ezlopitant)
`Page 2 of 18
`
`
`
`
`
`
`
`Tachykinin Receptors Antagonists Current Drug Targets, 2006, Vol. 7, No. 8 977
`vanced solid tumors [32]. Despite the encouraging premises, no
`further development was reported after 2000 [33].
`Glaxo
`The structure of CP-99994 was the starting point for NK-1 re-
`ceptor antagonists search at Glaxo. The screening to assess the
`pharmacological activity of the designed analogs was performed
`either in binding tests or in a model of radiation-induced emesis in
`ferrets, chosen as a likely clinical target. The preferred point of
`variation was position 4 of the benzyl ring of CP-99994, where
`heterocycles like tetrazole variously substituted were placed, like in
`GR-203040 [27] and GR-205171 (Fig. ( 3)) [34]. Both the com-
`pounds showed subnanomolar affinity for NK-1 receptor, metabolic
`stability and activity in the screening emesis test (100 % inhibition
`at 0.3 mg/Kg and 0.1 mg/kg p.o., respectively) [35]. The tetrazole
`ring was supposed to enhance oral bioavailability and to improve
`the pharmacokinetic properties. GR-205171 (vofopitant) has been
`evaluated in clinical phase II for postoperative nausea and vomiting
`[36], migraine and motion sickness, where it resulted ineffective
`[37]. Despite the good results in emesis, the development was dis-
`continued in 1999.
`Fig. (3).
`Merck
`The first class of antagonists published by Merck derived from
`the screening of a chemical collection, from which N-ethyl-L-
`tryptophan benzyl ester was found to bind at micromolar concen-
`tration to the human NK-1 receptor expressed in CHO cells [38].
`This simple lead was optimized to yield compound L-732,138
`(IC50=1.6 nM, Fig. ( 4)), which bears the acetylated amino group
`and the 3,5-bistrifluoromethylated phenyl moiety, a structural motif
`common to many of the non peptide NK-1 antagonists described in
`the literature [39]. Conformational analysis suggested a specific
`reciprocal orientation of the aromatic rings as a key structural motif
`[40], and constrained cyclic structures able to fix this feature were
`designed [41]. The ester bond, susceptible of rapid metabolic deg-
`radation, was then replaced with a ketone and the resulting structure
`was optimized to improve solubility and bioavailability, leading to
`L-737,488 (Fig. (4)) which bears a quinuclidine basic moiety. L-
`737,488 showed high in vivo activity (ID 50=1.8 mg/Kg p.o. in in-
`hibiting plasma protein extravasation in guinea-pig), weak affinity
`for Ca2+ channels and good solubility in water [42]. No further
`development was reported for the series.
`In 1993, Merck started to divulgate series of NK-1 antagonists
`structurally related to compounds previously known in the litera-
`ture. Both quinuclidine (CP-96345) and piperidine (CP-99994)
`templates were chosen as starting point for SAR studies.
`A series of CP 96345 analogs bearing the substitution of the
`amino with an ether function [43], lead to the optimization of the
`benzyl ether group introducing a 3,5-bis(trifluoromethyl) analog
`(compound L-709,210, Fig. (5)) [44]. Applying the simplification
`strategy at a further level (‘toward the minimum pharmacophore’),
`Merck scientists discovered that the quinuclidine or piperidine ring
`are not essential, since acyclic benzhydryl or phenyl aminoethyl
`ethers also preserve a good affinity for the NK-1 receptor [45].
`Moreover, the basic nitrogen could be replaced by an oxygen and
`the insertion of an acetamido moiety allowed to obtain subnano-
`molar affinity and reduced Ca 2+ channels affinity, like in com-
`pounds 1, 2 and 3 described in Fig. (5) [43, 46, 47].
`Fig. (4).
`Fig. (5).
`NH
`NH
`H3C O
`N
`N N
`N
`N
`H
`NH
`H3C
`O
`N
`N N
`N
`CF3
`GR-203040 GR-205171
`(
`vofopitant)
`O
`O
`HN
`CH3
`OHN
`CF3
`CF3
`O
`HN OHN
`CF3
`CF3
`N
`L-732,138
`L-737,488
`N
`O
`CF3
`CF3
`O
`CF3
`CF3
`N
`H
`H2N
`O
`H3CN
`CF3
`CF3
`O
`O
`CF3
`CF3
`NC H3
`H2N
`O
`L-709,210 1
` 2 3
`Page 3 of 18
`
`
`
`
`
`
`
`978 Current Drug Targets, 2006, Vol. 7, No. 8 Quartara and Altamura
`The SAR studies which allowed the selection of compounds for
`development were based on the structure of CP-99994. The com-
`pound L-733,060 (Fig. (6)) [48], shows a 3,5-bistrifluoromethyl
`benzylether piperidine in the place of the 2-methoxy benzylamine
`moiety of the parent compound. The piperidine nitrogen was then
`functionalized with various groups, in order to decrease its basic
`character [49]. The 3-oxo-1,2,4-triazol-5-yl moiety resulted the best
`in terms of basicity modulation and improvement in oral bioavail-
`ability, like in compounds L-741,671 and L-742,694 (Fig. ( 6)) [50,
`51] which showed an ID 50 of 0.037 and 0.009 mg/Kg p.o., respec-
`tively, in inhibiting SP-induced cutaneous inflammation in guinea-
`pig. The morpholine nucleus introduced in L-742,694 was main-
`tained in the further refinement. To avoid possible metabolic deac-
`tivation (debenzylation, oxidation of the phenyl ring at C-3), several
`modifications were introduced, like methylation on the C alfa of the
`benzyl ring and fluorination on the phenyl ring [52]. This afforded
`MK-869 (Fig. (6)) and analogs, which like L-742,694 showed high
`affinity for the NK-1 receptor and high oral activity. MK-869 (or L-
`754,030 or, as it will be referred throughout the text, aprepitant)
`was selected for further studies in pain, migraine, emesis and psy-
`chiatric disorders because of its high potency and central activity.
`While the clinical activities were carried on with aprepitant, it
`was found necessary to improve its pharmacokinetic characteristics.
`In fact, aprepitant shows low solubility in water (0.2 microg/ml)
`and it was almost impossible to find an intravenous formulation
`acceptable in early clinical phases for administration to humans.
`The phosphorylation of the oxotriazolyl ring in aprepitant produced
`a phosphorylated water soluble pro-drug (L-758,298, Fig ( 6), [53]).
`A series of experiments were performed to demonstrate that L-
`758,298 rapidly and univocally converted to aprepitant both in vitro
`and in vivo in the conditions of the pathophysiological models.
`Clinical studies of phase II were performed with L-758,298 in eme-
`sis [54] and migraine [55]. The compound showed favorable toler-
`ability profile and efficacy in the treatment of chemotherapy-
`induced emesis, but not in migraine. Later on, an analog of aprepi-
`tant suitable for both intravenous and oral clinical administration
`has been described (compound 4, Fig. (6)) [56]. In the new antago-
`nist the 3-oxo-1,2,4-triazol-5-yl moiety present in aprepitant was
`substituted with a 5-aminomethyl-1,2,3-triazol-4-yl moiety, isolated
`as hydrochloride. The pharmacological performances were compa-
`rable to those of the parent compound in pathophysiological mod-
`els, but water solubility, oral bioavailability and CNS penetration
`were considerably enhanced. Despite this excellent pharmacologi-
`cal profile, no clinical development has been reported.
`But what happened to aprepitant? As described before, hopes
`were strong to find a new class of analgesic and antidepressant. The
`analgesic hope was dropped in an early stage of development, as
`despite the positive preclinical results in animal models of pain, the
`compound was found inactive in dental surgery derived pain, pe-
`ripheral neuropathy, osteoarthritis and migraine [57]. This result
`was a matter of debate in the scientific community [58].
`Another equally important target was at the horizon: depres-
`sion. Aprepitant in fact had been proven effective in a number of
`pathophysiological models of depression and anxiety [59]. Phase II
`studies on major depressive disorders (MDD) strongly confirmed
`the expectations [60-62], although some difficulties in results inter-
`pretation came out in the early phase of development [63]. Merck
`phase III trial started in 2001. In November 2003, however, it was
`announced that no significative results were obtained with aprepi-
`tant in comparison with placebo in a total of five phase III trials.
`The development for this indication was discontinued [64].
`Fig. (6).
`N
`H
`O
`CF3
`CF3
`N
`O
`CF3
`CF3
`N
`HN
`NH
`O
`N
`O O
`CF3
`CF3
`N
`HN
`NH
`O
`N
`O O
`CF3
`CF3
`N
`HN
`NH
`O
`F
`H3C
`N
`O O
`CF3
`CF3
`N
`N
`NH
`O
`F
`H3C
`H2O3P
`N
`O O
`CF3
`CF3
`N
`N
`HN
`F
`H3C
`N(CH3)2.H Cl
`L-733,060 L-741,671 L-742,694
`L-754,030 or MK-869
`(
`aprepitant) L-758,298 4
`Page 4 of 18
`
`
`
`
`
`
`
`Tachykinin Receptors Antagonists Current Drug Targets, 2006, Vol. 7, No. 8 979
`At the same time of these investigations, Merck was evaluating
`aprepitant in emesis. As mentioned before, vofopitant and ezlopi-
`tant have been evaluated in clinical phase for postoperative nausea,
`vomiting and motion sickness. The phase II studies on aprepitant
`started in 1998. The compound was found active in the control of
`chemotherapy - induced emesis, in combination with granisetron
`and dexamethasone [54]. Phase III was implemented in 1999, and
`in 2003 the results of two phase III trials with a combination of
`aprepitant, a 5-HT3 antagonist and a corticosteroid were completed
`[65]. The compound was launched in 2003 in the US, and in 2004
`in EU with the name of Emend. In November 2004, Ono Pharma-
`ceuticals licensed the development in Japan.
`Sanofi
`Through the random screening of a chemical collection and
`subsequent optimization of the lead, Sanofi researchers discovered
`the highly potent NK-1 receptor antagonist SR-140333 (Fig. ( 7))
`[66] and the potent NK-2 receptor antagonist SR-48968 (Fig. ( 7))
`[67]. The structure of SR-140333 shows a piperidine scaffold sub-
`stituted on the nitrogen with an acyl aromatic group and gem-
`disubstituted in position 3. Three aromatic groups seemed essential
`for high affinity interaction with the NK-1 receptor (Ki=0.019 nM
`on IM9 cells) and selectivity. SR-140333 shows high activity in
`vivo in antagonizing SP-induced hypotension in dogs and guinea-
`pigs and in blocking plasma protein extravasation in rats, also dem-
`onstrating activity at central level [68]. In 2002, Sanofi published
`the pharmacological characterization of a follow up of SR-140333.
`In the new analog, SSR-240600, the central piperidine ring was
`substituted with a morpholine ring and the quinuclidine moiety with
`a piperidine. The 3,5 bistrifluorophenyl group also appears in this
`compound. SSR-240600 (Fig. (7)) shows nanomolar affinity at
`Fig. (7).
`human NK-1 receptor and high selectivity [69]. It is active in vivo
`in citric acid - induced cough in guinea pigs, in reducing micturition
`frequency without effects on micturition pressure, urethral pressure
`or urethral relaxation during micturition and also acts as an antide-
`pressant [70]. Pharmacokinetic studies in guinea pigs showed its
`efficient brain penetration. The compound is in development (phase
`I) for the potential treatment of overactive bladder.
`Ciba Geigy
`Ciba Geigy (Switzerland) developed a class of NK-1 selective
`antagonists through the modification of a lead identified by screen-
`ing of a chemical collection. The 'two aromatic groups five-to-nine
`atoms apart', considered as the minimum requirements for affinity,
`was optimized to obtain CGP-47899 (Fig. ( 8)) [71], which shows
`high affinity for NK-1 receptor. Further optimization with the in-
`sertion of a quinoline group in the place of the phenyl group lead to
`the follower CGP-49823 (Fig. (8)), possessing an IC 50=12 nM at
`the human NK-1 receptor and good oral bioavailability [72]. CGP-
`49823 was found active in the inhibition of intracerebroventricular
`SP-methyl ester-induced thumping behavior in gerbils [73] and it
`was developed through phase I for anxiety disorders, but the devel-
`opment was discontinued in 1998.
`Fig. (8).
`Parke-Davis
`Pfizer (former Parke-Davis) discovered a series of tryptophan
`based compounds through the screening of a dipeptide library [74].
`The dipeptide lead was optimized by methylation to introduce con-
`formational constraints and by selection of the appropriate aromatic
`group to obtain high affinity. PD-154075 (or CI-1021 as it was later
`referred to, Fig. (9)) [75] shows an IC 50=1 nM for binding to the
`human NK-1 receptor and displays a good activity in in vivo tests,
`such as in a guinea-pig plasma protein extravasation model
`(ID50=0.02 mg/Kg i.v.). Further SAR activities were based on the
`rational of an intramolecular hydrogen bond increasing the apparent
`lipophilicity of the compound, in order to enhance CNS penetration.
`In the best compound of the series the alfa methyl group was sub-
`stituted with a dimethylaminomethyl group (compound 5, Fig. (9))
`[76], to obtain enhanced brain penetration after oral administration
`in rats and efficacy at lower doses in in vivo tests. However, no
`further development was described.
`Eli Lilly
`By optimization of N-acetylated tryptophan amides and esters,
`another class of antagonists was discovered at Eli Lilly [77]. LY-
`303870 (lanepitant, Fig. ( 10)), can be considered an N-acetylated
`reduced amide of L-tryptophan. The two piperidine rings were cho-
`sen to enhance affinity for the NK-1 receptor and to increase ba-
`sicity and water solubility. Interestingly, the tryptophan configura-
`N
`O
`O
`N
`Cl
`Cl
`CF3
`CF3
`H2N
`O
`N
`O
`+C l-
`N
`Cl
`Cl
`O
`Cl
`N
`Cl
`N
`CH3
`O
`NHH3C
`O
`(S)
`SSR-240600
`SR-140333
`SR-48968
`(
`saredutant)
`CGP-47899 CGP-49823
`N
`O
`NH
`CH3
`CH3
`N
`O
`NH
`CH3
`CH3
`N
`Page 5 of 18
`
`
`
`
`
`
`
`980 Current Drug Targets, 2006, Vol. 7, No. 8 Quartara and Altamura
`tion is opposite to that present in L-737,488. LY-303870 shows an
`IC50=0.15 nM for the NK-1 receptor on IM9 cells, and possesses
`high potency on an in vivo neurogenic inflammation assay in
`guinea-pigs (ED50=15 ng/Kg i.v. and 91 ng/Kg p.o.), with a dura-
`tion of action longer than 8 hours.
`Fig. (9).
`Lanepitant entered clinical trials by 2000. In a phase IIa study,
`it was used in the treatment of osteoarthritis pain, but no significant
`effect was checked in comparison with placebo or naproxen [78].
`Since Lanepitant showed poor oral bioavailability and caused irri-
`tation upon iv injection, probably due to the dibasic 2-[4-(piperidin-
`1-yl)piperidin-1-yl]-acetyl group [79], Eli Lilly performed further
`SAR work on its structure. The dibasic moiety was substituted with
`various aryloxyacetic and acetylaryloxy moieties, to yield a class of
`compounds devoid of irritant effects, showing ameliorated bioavail-
`ability and microsomal stability (compounds 6, 7 and 8, Fig. (10)).
`Takeda
`In 1995, Takeda described a series of N-benzylcarboxyamides
`developed through a general hypothesis on peptidergic G-protein
`coupled receptors binding site [80]. The early antagonists (structure
`9, Fig. (11)), bearing a naphtyridine ring and the 3,5 difluoromethyl
`phenyl motif, were modified to avoid rotational atropisomerism in
`the exocyclic amide bond. In fact, the separation of the atropi-
`somers was difficult and the use of racemate in clinical develop-
`ment was judged too risky [81]. The exocyclic amide bond was
`therefore included in a 8-member azacyclic ring, and the insertion
`of a methyl group in position 9 (TAK-637, Fig. ( 11)) effectively
`solved the isomerism problem [80].
`Fig. (11).
`TAK-637 has an IC 50 value of 0.45 nM at the human NK-1
`receptor, and good selectivity toward human NK-2 and NK-3 re-
`ceptors. In guinea pigs, ID50 values for capsaicin-induced tracheal
`extravasation were 4.3 microg/kg for TAK-637 administered iv and
`33 microg/kg for po administration [80]. In another in vivo study in
`guinea pigs, TAK-637, at the dose of 0.03 to 0.3 mg/kg iv, inhibited
`bladder contractions [82].
`TAK-637 was found effective after oral administration in a
`number of different models of gastrointestinal disfunctions on
`lower urinary tract function in guinea pigs and cats [83]. The sys-
`temic administration of TAK-637 decreased the number but not the
`amplitude of distension-induced rhythmic bladder contractions in
`guinea pig [81].
`Fig. (10).
`NH
`O
`NHHN
`H3C
`CH3
`O
`O
`O
`N
`H
`O
`NHHN
`CH3
`O
`O
`O
`(H3C
`) 2N
`PD-154075 or CI-1021
`5
`N
`NHHN
`N
`O
`N
`OCH3
`OH3C
`LY-303870
`(lanepitant)
`N
`NHHN O
`OCH3
`OH3C
`6 ( R = OCH3)
`7 (R = Cl)
`O
`N
`N
`N
`NHHN O
`OCH3
`OH3C
` 8
`O
`S N
`N
`N
`N
`O
`R
`CH3
`N
`O
`CF3
`CF3
`CH3
`N
`N
`O
`CH3
`N
`CH3
`O
`CF3
`CF3
`9
`TAK-637
`Page 6 of 18
`
`
`
`
`
`
`
`Tachykinin Receptors Antagonists Current Drug Targets, 2006, Vol. 7, No. 8 981
`TAK-637 was licensed to Abbott Pharmaceuticals for codevel-
`opment. The compound entered Phase II trials in Europe, Japan and
`US for urinary incontinence, depression and IBS in 1999, but the
`clinical development was discontinued in 2003 (Company web
`site).
`3. NK-2 RECEPTOR SELECTIVE ANTAGONISTS
`Peptide or non peptide tachykinin NK-2 receptor antagonists
`have been described since 1992, and review articles on selected
`compounds [84-87], and their potential uses in therapy [88-90] have
`appeared during the last 10 years.
`In the search for novel antagonists, we can recognize at least
`three different general approaches that have been followed by dif-
`ferent research groups. The disclosure of the Sanofi non-peptide
`antagonist saredutant laid the basis for the first approach, with sev-
`eral compounds proposed by other companies and research groups,
`through modifications and optimization of the saredutant structure.
`The proposed definition of the minimal requirements for receptor
`affinity led Glaxo researchers to select and optimize small molecule
`NK-2 antagonist in an in-house collection of compounds. Finally,
`modeling and elaboration on the structure of the first cyclic peptide
`antagonists led to the identification of further peptide and non-
`peptide antagonists with high affinity at the human tachykinin NK-
`2 receptor.
`Sanofi
`The Sanofi non-peptide compound SR-48968 (saredutant, Fig.
`(7)), was described more than 10 years ago as the first highly po-
`tent, selective, tachykinin NK-2 receptor antagonist [91, 92]. In
`vitro, saredutant showed subnanomolar affinity at the human
`tachykinin NK-2 receptor, with high selectivity vs. the tachykinin
`NK-1 and NK-3 receptors. The compound was active in vivo in
`several animal models of NK-2 agonist induced effects [93], tar-
`geted at the respiratory [94], gastrointestinal [95], urinary [96] sys-
`tems and also in a number of CNS preclinical studies [97-99].
`On the basis of these results, saredutant was progressed to clini-
`cal studies, initially in the respiratory field. In a first study, the
`compound demonstrated significant activity in inhibiting bron-
`choconstriction induced by inhalation of neurokinin A in mild
`asthmatic patients [100]. In this study, saredutant was administered
`by the oral route, at the dose of 100 mg, and the challenge with
`neurokinin A was performed at 1.5 and 24 h after dosing.
`Although this study constituted the first evidence of inhibition
`of sensory neuropeptide induced bronchoconstriction in humans by
`a selective tachykinin receptor antagonist, in a following study
`saredutant did not show significant bronchodilatory or bronchopro-
`tective effect in allergic asthmatic patients [101]. The compound,
`orally administered at the dose of 100 mg (as in the previous study)
`for 9 days failed to improve bronchial obstruction and adenosine
`hyperresponsiveness in the patients. At that time, the hypothesis,
`also based on preclinical studies, that contemporary blocking of
`tachykinin NK-1 and NK-2 receptors may be necessary in the
`treatment of asthma, was raised. No further development of
`saredutant in the asthma indication has been reported.
`According to Sanofi-Aventis [102], saredutant is presently in
`phase III clinical trials for depression. Results of a placebo and
`fluoxetine controlled multicenter phase IIb study in 120 patients
`with major depressive disorders were disclosed [103]. Compared
`with patients treated with fluoxetine (20 mg), a greater number of
`patients taking saredutant (100 mg) exhibited a sustained response.
`Saredutant also demonstrated an improved profile as for adverse
`events.
`Further therapeutical indications in which saredutant is being
`evaluated include irritable bowel syndrome (IBS), although no data
`are available concerning these studies.
`The structure of saredutant shares common features with the
`Sanofi selective tachykinin NK-1 and NK-3 receptor antagonists,
`SR-140333 (Fig. (7)) and osanetant (SR-142801, Fig. (20)), respec-
`tively. This general structure provided a suitable model for further
`elaboration, and a number of selective tachykinin NK-2 antagonists,
`together with double NK-1/NK-2 antagonists, have resulted from
`further research.
`Again, Sanofi described the selective NK-2 antagonist SR-
`144190 (Fig. (12)). Together with other changes in the substituents,
`the N-methylated benzamide on the right end of the saredutant
`structure in Fig. ( 7) is "closed", in the structure of SR-144190, in a
`more rigid morpholine ring. Compared to saredutant, SR-144190
`demonstrated an improved profile of selectivity and was found
`more active in several models, particularly in CNS studies, due to
`its improved bioavailability in the central nervous system [104]. A
`clinical development for SR-144190 was initiated, then discontin-
`ued in the asthma and CNS fields.
`Fig. (12).
`Nippon Kayaku
`Another compound that appears to be derived from a con-
`strained form of saredutant structure, although the medicinal chem-
`istry regarding the molecule has not been published, is NK-5807 by
`Nippon Kayaku (Fig. (13)). The compound is reported to be active
`in a bronchoconstriction model in guinea pigs after intravenous and
`oral administration [105] and also showed activity in animal models
`of asthma [106]. In both studies, the effects were similar for NK-
`5807