From basic science to life-saving therapy: the rationale, and drug discovery efforts that led to the direct factor Xa inhibitor eliquis.

Over the past few decades, drug discovery directed at the treatment and prevention of thromboembolic diseases has been challenged by the need to balance robust efficacy with improved safety relative to the standard of care. To this end, the most impactful advance to date has been the discovery and development of oral factor Xa inhibitors. In this essay, a brief account of the program that culminated in the discovery of Eliquis (apixaban) and the commitment to identify a compound with an optimal profile are described.

First-in-human study to assess the safety, pharmacokinetics and pharmacodynamics of BMS-962212, a direct, reversible, small molecule factor XIa inhibitor in non-Japanese and Japanese healthy subjects.

AIMS: The aims of the present study were to assess the safety, pharmacokinetics (PK) and pharmacodynamics (PD) of BMS-962212, a first-in-class factor XIa inhibitor, in Japanese and non-Japanese healthy subjects. METHODS: This was a randomized, placebo-controlled, double-blind, sequential, ascending-dose study of 2-h (part A) and 5-day (part B) intravenous (IV) infusions of BMS-962212. Part A used four doses (1.5, 4, 10 and 25 mg h-1 ) of BMS-962212 or placebo in a 6:2 ratio per dose. Part B used four doses (1, 3, 9 and 20 mg h-1 ) enrolling Japanese (n = 4 active, n = 1 placebo) and non-Japanese (n = 4 active, n = 1 placebo) subjects per dose. The PK, PD, safety and tolerability were assessed throughout the study. RESULTS: BMS-962212 was well tolerated; there were no signs of bleeding, and adverse events were mild. In parts A and B, BMS-962212 demonstrated dose proportionality. The mean half-life in parts A and B ranged from 2.04 to 4.94 h and 6.22 to 8.65 h, respectively. Exposure-dependent changes were observed in the PD parameters, activated partial thromboplastin time (aPTT) and factor XI clotting activity (FXI:C). The maximum mean aPTT and FXI:C change from baseline at 20 mg h-1 in part B was 92% and 90%, respectively. No difference was observed in weight-corrected steady-state concentrations, aPTT or FXI:C between Japanese and non-Japanese subjects (P > 0.05). CONCLUSION: BMS-962212 has tolerability, PK and PD properties suitable for investigational use as an acute antithrombotic agent in Japanese or non-Japanese subjects.

Apixaban Versus Warfarin for Mechanical Heart Valve Thromboprophylaxis in a Swine Aortic Heterotopic Valve Model.

OBJECTIVE: Warfarin is the current standard for oral anticoagulation therapy in patients with mechanical heart valves, yet optimal therapy to maximize anticoagulation and minimize bleeding complications requires routine coagulation monitoring, possible dietary restrictions, and drug interaction monitoring. As alternatives to warfarin, oral direct acting factor Xa inhibitors are currently approved for the prophylaxis and treatment of venous thromboembolism and reduction of stroke and systemic embolization. However, no in vivo preclinical or clinical studies have been performed directly comparing oral factor Xa inhibitors such as apixaban to warfarin, the current standard of therapy. APPROACH AND RESULTS: A well-documented heterotopic aortic valve porcine model was used to test the hypothesis that apixaban has comparable efficacy to warfarin for thromboprophylaxis of mechanical heart valves. Sixteen swine were implanted with a bileaflet mechanical aortic valve that bypassed the ligated descending thoracic aorta. Animals were randomized to 4 groups: control (no anticoagulation; n=4), apixaban oral 1 mg/kg twice a day (n=5), warfarin oral 0.04 to 0.08 mg/kg daily (international normalized ratio 2-3; n=3), and apixaban infusion (n=4). Postmortem valve thrombus was measured 30 days post-surgery for control-oral groups and 14 days post-surgery for the apixaban infusion group. Control thrombus weight (mean) was significantly different (1422.9 mg) compared with apixaban oral (357.5 mg), warfarin (247.1 mg), and apixiban 14-day infusion (61.1 mg; P<0.05). CONCLUSIONS: Apixaban is a promising candidate and may be a useful alternative to warfarin for thromboprophylaxis of mechanical heart valves. Unlike warfarin, no adverse bleeding events were observed in any apixaban groups.

Apixaban versus enoxaparin for thromboprophylaxis in medically ill patients.

BACKGROUND: The efficacy and safety of prolonging prophylaxis for venous thromboembolism in medically ill patients beyond hospital discharge remain uncertain. We hypothesized that extended prophylaxis with apixaban would be safe and more effective than short-term prophylaxis with enoxaparin. METHODS: In this double-blind, double-dummy, placebo-controlled trial, we randomly assigned acutely ill patients who had congestive heart failure or respiratory failure or other medical disorders and at least one additional risk factor for venous thromboembolism and who were hospitalized with an expected stay of at least 3 days to receive apixaban, administered orally at a dose of 2.5 mg twice daily for 30 days, or enoxaparin, administered subcutaneously at a dose of 40 mg once daily for 6 to 14 days. The primary efficacy outcome was the 30-day composite of death related to venous thromboembolism, pulmonary embolism, symptomatic deep-vein thrombosis, or asymptomatic proximal-leg deep-vein thrombosis, as detected with the use of systematic bilateral compression ultrasonography on day 30. The primary safety outcome was bleeding. All efficacy and safety outcomes were independently adjudicated. RESULTS: A total of 6528 subjects underwent randomization, 4495 of whom could be evaluated for the primary efficacy outcome--2211 in the apixaban group and 2284 in the enoxaparin group. Among the patients who could be evaluated, 2.71% in the apixaban group (60 patients) and 3.06% in the enoxaparin group (70 patients) met the criteria for the primary efficacy outcome (relative risk with apixaban, 0.87; 95% confidence interval [CI], 0.62 to 1.23; P=0.44). By day 30, major bleeding had occurred in 0.47% of the patients in the apixaban group (15 of 3184 patients) and in 0.19% of the patients in the enoxaparin group (6 of 3217 patients) (relative risk, 2.58; 95% CI, 1.02 to 7.24; P=0.04). CONCLUSIONS: In medically ill patients, an extended course of thromboprophylaxis with apixaban was not superior to a shorter course with enoxaparin. Apixaban was associated with significantly more major bleeding events than was enoxaparin. (Funded by Bristol-Myers Squibb and Pfizer; ClinicalTrials.gov number, NCT00457002.).

Orally bioavailable factor Xa inhibitors containing alpha-substituted gem-dimethyl P4 moieties.

In an effort to identify a potential back-up to apixaban (Eliquis®), we explored a series of diversified P4 moieties. Several analogs with substituted gem-dimethyl moieties replacing the terminal lactam of apixaban were identified which demonstrated potent FXa binding affinity (FXa Ki), good human plasma anticoagulant activity (PT EC2x), cell permeability, and oral bioavailability.

Discovery and gram-scale synthesis of BMS-593214, a potent, selective FVIIa inhibitor.

A 6-amidinotetrahydroquinoline screening hit was driven to a structurally novel, potent, and selective FVIIa inhibitor through a combination of library synthesis and rational design. An efficient gram-scale synthesis of the active enantiomer BMS-593214 was developed, which required significant optimization of the key Povarov annulation. Importantly, BMS-593214 showed antithrombotic efficacy in a rabbit arterial thrombosis model. A crystal structure of BMS-593214 bound to FVIIa highlights key contacts with Asp 189, Lys 192, and the S2 pocket.

Design and synthesis of bicyclic pyrazinone and pyrimidinone amides as potent TF-FVIIa inhibitors.

Bicyclic pyrazinone and pyrimidinone amides were designed and synthesized as potent TF-FVIIa inhibitors. SAR demonstrated that the S2 and S3 pockets of FVIIa prefer to bind small, lipophilic groups. An X-ray crystal structure of optimized compound 9b bound in the active site of FVIIa showed that the bicyclic scaffold provides 5 hydrogen bonding interactions in addition to projecting groups for interactions within the S1, S2 and S3 pockets. Compound 9b showed excellent FVIIa potency, good selectivity against FIXa, Xa, XIa and chymotrypsin, and good clotting activity.

Apixaban inhibition of factor Xa: Microscopic rate constants and inhibition mechanism in purified protein systems and in human plasma.

Apixaban is a potent, direct, selective, and orally active inhibitor of coagulation factor Xa. Rate constants for apixaban binding to free and prothrombinase-bound factor Xa were measured using multiple techniques. The inhibition mechanism was determined in purified systems and in a plasma prothrombin clotting time assay. Apixaban inhibits factor Xa with a K(i) of 0.25 nM at 37°C, an association rate constant of approximately 20 µM(-1) s(-1), and a dissociation half-life of 1-2 min. Under physiological conditions apixaban exhibits mixed-type inhibition and maintains high factor Xa affinity with a K(i) of 0.62 nM and association rate constant of 12 µM(-1) s(-1) for prothrombinase, and a K(i) of 1.7 nM and association rate constant of 4 µM(-1) s(-1) for the prothrombinase:prothrombin complex. Experiments in prothrombin depleted human plasma showed that the mechanism and kinetics of inhibition are maintained in plasma. The mechanistic detail derived from these experiments can be used to understand and interpret the pharmacodynamic action of apixaban.

Preclinical pharmacokinetics and pharmacodynamics of apixaban, a potent and selective factor Xa inhibitor.

Apixaban is a potent, highly selective, reversible, oral, direct factor Xa (fXa) inhibitor in development for thrombosis prevention and treatment. The preclinical pharmacokinetic (PK) attributes of apixaban feature small volume of distribution (Vd), low systemic clearance (CL), and good oral bioavailability. Apixaban is well absorbed in rat, dog, and chimpanzee, with absolute oral bioavailability of approximately 50% or greater. The steady-state Vd of apixaban is approximately 0.5, 0.2, and 0.17 l/kg in rats, dogs, and chimpanzees, while CL is approximately 0.9, 0.04, and 0.018 l/h/kg, respectively. In vitro metabolic clearance of apixaban is also low. Renal clearance comprises approximately 10-30% of systemic clearance in rat, dog, and chimpanzee. Anti-fXa activity, prothrombin time (PT), and HEPTEST(®) clotting time (HCT) prolongation correlated well with plasma apixaban concentration in rat, dog and chimpanzee. There was no lag time between apixaban plasma concentration and the pharmacodynamic (PD) markers, suggesting a rapid onset of action of apixaban. The PK/PD analyses were performed using an inhibitory E (max) model for anti-fXa assay and a linear model for PT and HCT assays. The IC(50) values for anti-fXa activity were 0.73 ± 0.03 and 1.5 ± 0.15 µM for rat and dog, respectively. The apparent K ( i ) values for PT were approximately 1.7, 6.6, and 4.8 µM for rat, dog and chimpanzee, respectively. The apparent K ( i ) for HCT was approximately 1.3 µM for dog. Apixaban exhibits desirable PK and PD properties for clinical development with good oral bioavailability, small Vd, low CL, and direct, predictable, concentration-dependent PD responses.

Phenyltriazolinones as potent factor Xa inhibitors.

We have discovered that phenyltriazolinone is a novel and potent P1 moiety for coagulation factor Xa. X-ray structures of the inhibitors with a phenyltriazolinone in the P1 position revealed that the side chain of Asp189 has reoriented resulting in a novel S1 binding pocket which is larger in size to accommodate the phenyltriazolinone P1 substrate.

Sulfation of o-demethyl apixaban: enzyme identification and species comparison.

Apixaban, a potent and highly selective factor Xa inhibitor, is currently under development for treatment of arterial and venous thrombotic diseases. The O-demethyl apixaban sulfate is a major circulating metabolite in humans but circulates at lower concentrations relative to parent in animals. The aim of this study was to identify the sulfotransferases (SULTs) responsible for the sulfation reaction. Apixaban undergoes O-demethylation catalyzed by cytochrome P450 enzymes to O-demethyl apixaban, and then is conjugated by SULTs to form O-demethyl apixaban sulfate. Of the five human cDNA-expressed SULTs tested, SULT1A1 and SULT1A2 exhibited significant levels of catalytic activity for formation of O-demethyl apixaban sulfate, and SULT1A3, SULT1E1, and SULT2A1 showed much lower catalytic activities. In human liver S9, quercetin, a highly selective inhibitor of SULT1A1 and SULT1E1, inhibited O-demethyl apixaban sulfate formation by 99%; 2,6-dichloro-4-nitrophenol, another inhibitor of SULT1A1, also inhibited this reaction by >90%; estrone, a competitive inhibitor for SULT1E1, had no effect on this reaction. The comparable K(m) values for formation of O-demethyl apixaban sulfate were 41.4 microM (human liver S9), 36.8 microM (SULT1A1), and 70.8 microM (SULT1A2). Because of the high level of expression of SULT1A1 in liver and its higher level of catalytic activity for formation of O-demethyl apixaban sulfate, SULT1A1 might play a major role in humans for formation of O-demethyl apixaban sulfate. O-Demethyl apixaban was also investigated in liver S9 of mice, rats, rabbits, dogs, monkeys, and humans. The results indicated that liver S9 samples from dogs, monkeys, and humans had higher activities for formation of O-demethyl apixaban sulfate than those of mice, rats, and rabbits.

Comparative metabolism of 14C-labeled apixaban in mice, rats, rabbits, dogs, and humans.

The metabolism and disposition of [(14)C]apixaban, a potent, reversible, and direct inhibitor of coagulation factor Xa, were investigated in mice, rats, rabbits, dogs, and humans after a single oral administration and in incubations with hepatocytes. In plasma, the parent compound was the major circulating component in mice, rats, dogs, and humans. O-Demethyl apixaban sulfate (M1) represented approximately 25% of the parent area under the time curve in human plasma. This sulfate metabolite was present, but in lower amounts relative to the parent, in plasma from mice, rats, and dogs. Rabbits showed a plasma metabolite profile distinct from that of other species with apixaban as a minor component and M2 (O-demethyl apixaban) and M14 (O-demethyl apixaban glucuronide) as prominent components. The fecal route was a major elimination pathway, accounting for >54% of the dose in animals and >46% in humans. The urinary route accounted for <15% of the dose in animals and 25 to 28% in humans. Apixaban was the major component in feces of every species and in urine of all species except rabbit. M1 and M2 were common prominent metabolites in urine and feces of all species as well as in bile of rats and humans. In vivo metabolite profiles showed quantitative differences between species and from in vitro metabolite profiles, but all human metabolites were found in animal species. After intravenous administration of [(14)C]apixaban to bile duct-cannulated rats, the significant portion (approximately 22%) of the dose was recovered as parent drug in the feces, suggesting direct excretion of the drug from gastrointestinal tracts of rats. Overall, apixaban was effectively eliminated via multiple elimination pathways in animals and humans, including oxidative metabolism, and direct renal and intestinal excretion.

Highly efficacious factor Xa inhibitors containing alpha-substituted phenylcycloalkyl P4 moieties.

We previously disclosed a series of highly potent FXa inhibitors bearing alpha-substituted (CH(2)NR(1)R(2)) phenylcyclopropyl P4 moieties in the pyrazolodihydropyridone core system. Herein, we describe our continuous SAR efforts in this series. Effects of the C-3 substitution of the pyrazolodihydropyridone core and the alpha-substitution (R group) of the cyclopropyl ring on FXa binding affinity (FXa K(i)), human plasma anticoagulant activity (PT EC(2x)) and permeability are discussed. A set of compounds obtained from optimization of the R group and the C-3 substituent were orally bioavailable in dogs. Furthermore, representative compounds were highly efficacious in the rabbit arterio-venous shunt thrombosis model (EC(50)s=29-81nM).

Reductive isoxazole ring opening of the anticoagulant razaxaban is the major metabolic clearance pathway in rats and dogs.

Razaxaban is a selective, potent, and orally bioavailable inhibitor of coagulation factor Xa. The molecule contains a 1,2-benzisoxazole structure. After oral administration of [(14)C]razaxaban to intact and bile duct-cannulated rats (300 mg/kg) and dogs (20 mg/kg), metabolism followed by biliary excretion was the major elimination pathway in both species, accounting for 34 to 44% of the dose, whereas urinary excretion accounted for 3 to 13% of the dose. Chromatographic separation of radioactivity in urine, bile, and feces of rats and dogs showed that razaxaban was extensively metabolized in both species. Metabolites were identified on the basis of liquid chromatography/tandem mass spectrometry and comparison with synthetic standards. Among the 12 metabolites identified, formation of an isoxazole-ring opened benzamidine metabolite (M1) represented a major metabolic pathway of razaxaban in rats and dogs. However, razaxaban was the major circulating drug-related component (>70%) in both species, and M1, M4, and M7 were minor circulating components. In addition to the in vivo observations, M1 was formed as the primary metabolite in rat and dog hepatocytes and in the rat liver cytosolic fraction. The formation of M1 in the rat liver fraction required the presence of NADH. Theses results suggest that isoxazole ring reduction, forming a stable benzamidine metabolite (M1), represents the primary metabolic pathway of razaxaban in vivo and in vitro. The reduction reaction was catalyzed by NADH-dependent reductase(s) in the liver and possibly by intestinal microflora on the basis of the recovery of M1 in feces of bile duct-cannulated rats.

Sulfonamidolactam inhibitors of coagulation factor Xa.

As part of an effort to identify novel backups for previously reported pyrazole-based coagulation Factor Xa inhibitors, the pyrazole 5-carboxamide moiety was replaced by 3-(sulfonylamino)-2-piperidone. This led to the identification of a structurally diverse chemotype that was further optimized to incorporate neutral or weakly basic aryl and heteroaryl P1 groups while maintaining good potency versus Factor Xa. Substitution at the sulfonamide nitrogen provided further improvements in potency and as did introduction of alternate P4 moieties.

Achieving structural diversity using the perpendicular conformation of alpha-substituted phenylcyclopropanes to mimic the bioactive conformation of ortho-substituted biphenyl P4 moieties: discovery of novel, highly potent inhibitors of Factor Xa.

Ortho-substituted biphenyl moieties are widely used in drug design. We herein report a successful use of the perpendicular conformation of the alpha-substituted phenylcyclopropyl groups to mimic the aplanar, biologically active conformation of the ortho-substituted biphenyl moieties to achieve structural diversity. This is exemplified by the design and synthesis of a series of highly potent pyrazole bicyclic-based Factor Xa (FXa) inhibitors bearing alpha-substituted phenylcyclopropyl P4 moieties. The designed perpendicular conformation was confirmed by the X-ray structure of FXa-bound compound 2r. The potential structural basis for the high FXa potency in the phenylcyclopropyl P4 analogs and their improved FXa inhibitory activities compared with the biphenyl P4 counterparts are discussed.

Structure-activity relationship and pharmacokinetic profile of 5-ketopyrazole factor Xa inhibitors.

Efforts to further optimize the clinical candidate razaxaban have led to a new series of pyrazole-based factor Xa (fXa) inhibitors. Designed to prevent the potential formation of primary aniline metabolites in vivo, the nitrogen of the carboxamido linker between the pyrazole and proximal phenyl moiety of the razaxaban scaffold was replaced with a methylene group. The resulting ketones demonstrated excellent potency and selectivity for fXa but initially had poor oral bioavailability. Optimization by conversion from a P1 aminobenzisoxazole to a P1 p-methoxyphenyl residue, replacing the 3-trifluoromethylpyrazole with a 3-amidopyrazole, and employing a pyridone P4 group provided a fXa inhibitor with a potency and pharmacokinetic profile equivalent to that of razaxaban and improved selectivity over thrombin.

Structure-activity relationships of anthranilamide-based factor Xa inhibitors containing piperidinone and pyridinone P4 moieties.

Introduction of the phenyl piperidinone and phenyl pyridinone P4 moieties in the anthranilamide scaffold led to potent, selective, and orally bioavailable inhibitors of factor Xa. Anthranilamide 28 displayed comparable efficacy to apixaban in the rabbit arteriovenous-shunt (AV) thrombosis model.

Discovery of 1-(4-methoxyphenyl)-7-oxo-6-(4-(2-oxopiperidin-1-yl)phenyl)-4,5,6,7-tetrahydro-1H-pyrazolo[3,4-c]pyridine-3-carboxamide (apixaban, BMS-562247), a highly potent, selective, efficacious, and orally bioavailable inhibitor of blood coagulation factor Xa.

Efforts to identify a suitable follow-on compound to razaxaban (compound 4) focused on modification of the carboxamido linker to eliminate potential in vivo hydrolysis to a primary aniline. Cyclization of the carboxamido linker to the novel bicyclic tetrahydropyrazolopyridinone scaffold retained the potent fXa binding activity. Exceptional potency of the series prompted an investigation of the neutral P1 moieties that resulted in the identification of the p-methoxyphenyl P1, which retained factor Xa binding affinity and good oral bioavailability. Further optimization of the C-3 pyrazole position and replacement of the terminal P4 ring with a neutral heterocycle culminated in the discovery of 1-(4-methoxyphenyl)-7-oxo-6-(4-(2-oxopiperidin-1-yl)phenyl)-4,5,6,7-tetrahydro-1H-pyrazolo[3,4-c]pyridine-3-carboxamide (apixaban, compound 40). Compound 40 exhibits a high degree of fXa potency, selectivity, and efficacy and has an improved pharmacokinetic profile relative to 4.

Design, structure-activity relationship, and pharmacokinetic profile of pyrazole-based indoline factor Xa inhibitors.

A new series of pyrazole-based factor Xa inhibitors have been identified as part of our ongoing efforts to optimize previously reported clinical candidate razaxaban. Concern over the possible formation of primary aniline metabolites via amide hydrolysis led to the replacement of the primary amide linker between the pyrazole and phenyl moieties with secondary amides. This was accomplished by replacing the aniline with a variety of heterobicycles, of which indolines were the most potent. The indoline series demonstrated subnanomolar factor Xa binding K(i)s, modest to high selectivity versus other serine proteases, and good in vitro clotting activity. A small number of indoline fXa inhibitors were profiled in a dog pharmacokinetic model, one of which demonstrated pharmacokinetic parameters similar to that of clinical candidate razaxaban.